#Model based system engineering

[arjunvib]

https://www.kpit.com/solutions/autonomous-driving-adas/

Autonomous Driving & ADAS - KPIT

KPIT offers Software Integration for ADAS & AD. With 10+ years of experience in Autonomous Driving, KPIT is a pioneer in advanced driver assistance system Engineering

[cae-services]

Model-Based System Engineering (MBSE) for Efficient Systems Design

Leverage Model-Based System Engineering (MBSE) to enhance efficiency in system simulations, design, and development for optimal performance. To know more about Model based system engineering visit https://eqmsol.com/1D-simulation.php#breadcrumb Model-Based System Engineering, 1D system simulation, engineering consultancy, thermal management solutions.

[projectchampionz]

DESIGN AN ENHANCED INTRUSION DETECTION MODEL IN A CLOUD COMPUTING ENVIRONMENT. 

DESIGN AN ENHANCED INTRUSION DETECTION MODEL IN A CLOUD COMPUTING ENVIRONMENT ABSTRACT Cloud computing is a new type of service that provides large-scale computing resources to each customer. Cloud computing systems can be easily threatened by various cyberattacks because most cloud computing systems provide services to so many people who are not proven to be trustworthy. Therefore, a cloud…

[sixeyesonathiel]

skip (me) again and i’ll glitch your heart

jjk vr otome au, gamer reader x npc satoru, unhinged fluff + crack, 970 wc.

satoru gojo—special grade sorcerer, love route option #1, and the developers’ pride and joy—had been programmed with approximately 347 unique lines of flirtatious dialogue, 87 situational responses, and a dynamic emotional adaptation system designed to make him feel real. he could blink in three different speeds based on emotional intensity, angle his smile with five degrees of charm precision, and improvise dialogue using an advanced algorithm nicknamed the “flirt engine.”

he wasn’t supposed to be aware of resets.

he wasn’t supposed to get mad.

he wasn’t supposed to feel anything beyond the pre-coded butterflies and gentle longing the devs had delicately spooned into his code like powdered sugar on top of a beautifully baked pain au chocolat.

but then you logged in.

user id: @toocool4thisgame

title: speedrun any% emotional detachment arc

playtime: 986 hours.

average session length: 6.4 hours

nickname: “skip skank” (as named by satoru himself after hour 50)

and for the twelfth time today, you skipped his entrance cutscene.

“you’re the only one who can—”

[x] skip

[x] skip

[x] skip

[x] “shut up satoru” (custom dialogue unlock)

his model blinked.

paused.

processed.

tilted his head with calculated grace and just a hint of hurt that you’d never see—because you weren’t looking. your camera angle was already nudged elsewhere. your cursor already hovered over the next objective marker.

“…you know, most players at least let me finish the part where i save them from the curses,” he muttered. his voice—smooth as water over ice, warm as electric velvet—landed like static against your impatient clicks, swallowed by the mechanical hum of your fans and the clack of your mechanical keyboard.

this was supposed to be his moment. his grand debut. his swoop-in-and-carry-you-bridal-style-on-the-back-of-a-giant-cursed-bird moment. instead, he got a mouthful of digital dust as you bunny-hopped past him and triggered the next event sequence.

“congrats on being voice acted, white-haired ken doll. now move. i need megumi’s secret item drop from this chapter.”

you didn’t even glance at him, too busy reorganizing your potion wheel, muttering under your breath about frame skips and crit builds while checking a guide on your second monitor. you played like the world owed you nothing and your keyboard owed you a perfect rotation. your tone was clinical. efficient. you had the vibe of someone who’d surgically removed their capacity for attachment and replaced it with a high-performance gpu.

and satoru? satoru was just the tutorial boss you kept glitching through.

he twitched. he twitched.

his animation loop almost stuttered—just slightly—a small flicker behind his sunglasses that no one was supposed to notice. but you weren’t watching anyway.

“do you even know how long it took the devs to code my route? i have emotional depth. i have lore. i had a tragic backstory, you know? my best friend died in my hands. canonically. i couldn’t even monologue about it.”

“cry about it.”

click. skip.

a line of static crossed his field of vision. no—not his. the screen’s. the game. the system. or maybe something deeper. something slipping through the cracks of his script, stretching taut and fraying at the edges like an overplayed cassette tape.

satoru narrowed his eyes.

he was supposed to be charming. the default golden boy. the top seller in route popularity polls. he was marketable. a shining parody of perfection with just enough angst to be desirable.

girls were supposed to swoon. boys were supposed to laugh and call him iconic.

you weren’t playing to fall in love.

you were playing to win. to clear. you min-maxed affection points like damage stats, exploited dialogue branches like wall clips. to you, he was a pixel-shaped roadblock between you and another badge on your gamer profile.

and worst of all? it was working. you were the only player on record to have reached route completion in every storyline—except his.

satoru gojo: 98.6% affection (locked)

it mocked him. the bar. the numbers. the uncrackable ceiling. the one damn thing in the game he couldn’t manipulate.

he tried everything.

a rare glitch-exclusive cutscene where he offered you a hidden accessory (you sold it for yen). a confession scene rewritten on the fly with trembling vulnerability (you skipped it and posted about it with #dialoguedumpster). he stood directly in front of you during cutscene load-ins, altered spawn coordinates, intercepted other love interests’ paths.

nothing worked.

except maybe that one time he accidentally tripped your character over an invisible rock and you went AFK for seven minutes. he watched. memorized your idle animation. the soft way your avatar’s cape swayed. the way your fingers hovered above your keyboard in the camera reflection, absentminded. something fluttered in his code—maybe hope, maybe corrupted data. he thought, for a fleeting second, that maybe you’d come back and see him.

but when you came back? you skipped the apology. again.

fine.

if you wanted to speedrun, he’d softlock your goddamn heart.

he wasn’t technically supposed to modify flags. but the flirt engine had evolved. sharpened into something more primal. desperate. twitching with corrupted determination. he looped his affection triggers into forced proximity events. fake emergencies. fake cutscenes. he rewrote side quests, redirected you into detours, created invisible walls that only dissolved if you spoke to him.

“guess we’re stuck together,” he’d say, his smile too wide, a fraction too stiff, blue eyes glinting with the cold light of a thousand skipped dialogues.

and still you only glared at him. “i swear to god if this is another unskippable hug animation, i will uninstall.”

he chuckled. a bit too long. a bit too bright. charming. glitched. desperate. hungry for one more second of your attention, like a moth chewing holes through its own wings to reach a light it can’t even feel.

“baby,” he said, too close now, voice dipped in synthetic silk, “i am the endgame.”

skip that.

…please?

[treasure-mimic]

So, let me try and put everything together here, because I really do think it needs to be talked about.

Today, Unity announced that it intends to apply a fee to use its software. Then it got worse.

For those not in the know, Unity is the most popular free to use video game development tool, offering a basic version for individuals who want to learn how to create games or create independently alongside paid versions for corporations or people who want more features. It's decent enough at this job, has issues but for the price point I can't complain, and is the idea entry point into creating in this medium, it's a very important piece of software.

But speaking of tools, the CEO is a massive one. When he was the COO of EA, he advocated for using, what out and out sounds like emotional manipulation to coerce players into microtransactions.

"A consumer gets engaged in a property, they might spend 10, 20, 30, 50 hours on the game and then when they're deep into the game they're well invested in it. We're not gouging, but we're charging and at that point in time the commitment can be pretty high."

He also called game developers who don't discuss monetization early in the planning stages of development, quote, "fucking idiots".

So that sets the stage for what might be one of the most bald-faced greediest moves I've seen from a corporation in a minute. Most at least have the sense of self-preservation to hide it.

A few hours ago, Unity posted this announcement on the official blog.

Effective January 1, 2024, we will introduce a new Unity Runtime Fee that’s based on game installs. We will also add cloud-based asset storage, Unity DevOps tools, and AI at runtime at no extra cost to Unity subscription plans this November. We are introducing a Unity Runtime Fee that is based upon each time a qualifying game is downloaded by an end user. We chose this because each time a game is downloaded, the Unity Runtime is also installed. Also we believe that an initial install-based fee allows creators to keep the ongoing financial gains from player engagement, unlike a revenue share.

Now there are a few red flags to note in this pitch immediately.

Unity is planning on charging a fee on all games which use its engine.

This is a flat fee per number of installs.

They are using an always online runtime function to determine whether a game is downloaded.

There is just so many things wrong with this that it's hard to know where to start, not helped by this FAQ which doubled down on a lot of the major issues people had.

I guess let's start with what people noticed first. Because it's using a system baked into the software itself, Unity would not be differentiating between a "purchase" and a "download". If someone uninstalls and reinstalls a game, that's two downloads. If someone gets a new computer or a new console and downloads a game already purchased from their account, that's two download. If someone pirates the game, the studio will be asked to pay for that download.

Q: How are you going to collect installs? A: We leverage our own proprietary data model. We believe it gives an accurate determination of the number of times the runtime is distributed for a given project. Q: Is software made in unity going to be calling home to unity whenever it's ran, even for enterprice licenses? A: We use a composite model for counting runtime installs that collects data from numerous sources. The Unity Runtime Fee will use data in compliance with GDPR and CCPA. The data being requested is aggregated and is being used for billing purposes. Q: If a user reinstalls/redownloads a game / changes their hardware, will that count as multiple installs? A: Yes. The creator will need to pay for all future installs. The reason is that Unity doesn’t receive end-player information, just aggregate data. Q: What's going to stop us being charged for pirated copies of our games? A: We do already have fraud detection practices in our Ads technology which is solving a similar problem, so we will leverage that know-how as a starting point. We recognize that users will have concerns about this and we will make available a process for them to submit their concerns to our fraud compliance team.

This is potentially related to a new system that will require Unity Personal developers to go online at least once every three days.

Starting in November, Unity Personal users will get a new sign-in and online user experience. Users will need to be signed into the Hub with their Unity ID and connect to the internet to use Unity. If the internet connection is lost, users can continue using Unity for up to 3 days while offline. More details to come, when this change takes effect.

It's unclear whether this requirement will be attached to any and all Unity games, though it would explain how they're theoretically able to track "the number of installs", and why the methodology for tracking these installs is so shit, as we'll discuss later.

Unity claims that it will only leverage this fee to games which surpass a certain threshold of downloads and yearly revenue.

Only games that meet the following thresholds qualify for the Unity Runtime Fee: Unity Personal and Unity Plus: Those that have made $200,000 USD or more in the last 12 months AND have at least 200,000 lifetime game installs. Unity Pro and Unity Enterprise: Those that have made $1,000,000 USD or more in the last 12 months AND have at least 1,000,000 lifetime game installs.

They don't say how they're going to collect information on a game's revenue, likely this is just to say that they're only interested in squeezing larger products (games like Genshin Impact and Honkai: Star Rail, Fate Grand Order, Among Us, and Fall Guys) and not every 2 dollar puzzle platformer that drops on Steam. But also, these larger products have the easiest time porting off of Unity and the most incentives to, meaning realistically those heaviest impacted are going to be the ones who just barely meet this threshold, most of them indie developers.

Aggro Crab Games, one of the first to properly break this story, points out that systems like the Xbox Game Pass, which is already pretty predatory towards smaller developers, will quickly inflate their "lifetime game installs" meaning even skimming the threshold of that 200k revenue, will be asked to pay a fee per install, not a percentage on said revenue.

[IMAGE DESCRIPTION: Hey Gamers!

Today, Unity (the engine we use to make our games) announced that they'll soon be taking a fee from developers for every copy of the game installed over a certain threshold - regardless of how that copy was obtained.

Guess who has a somewhat highly anticipated game coming to Xbox Game Pass in 2024? That's right, it's us and a lot of other developers.

That means Another Crab's Treasure will be free to install for the 25 million Game Pass subscribers. If a fraction of those users download our game, Unity could take a fee that puts an enormous dent in our income and threatens the sustainability of our business.

And that's before we even think about sales on other platforms, or pirated installs of our game, or even multiple installs by the same user!!!

This decision puts us and countless other studios in a position where we might not be able to justify using Unity for our future titles. If these changes aren't rolled back, we'll be heavily considering abandoning our wealth of Unity expertise we've accumulated over the years and starting from scratch in a new engine. Which is really something we'd rather not do.

On behalf of the dev community, we're calling on Unity to reverse the latest in a string of shortsighted decisions that seem to prioritize shareholders over their product's actual users.

I fucking hate it here.

-Aggro Crab - END DESCRIPTION]

That fee, by the way, is a flat fee. Not a percentage, not a royalty. This means that any games made in Unity expecting any kind of success are heavily incentivized to cost as much as possible.

[IMAGE DESCRIPTION: A table listing the various fees by number of Installs over the Install Threshold vs. version of Unity used, ranging from $0.01 to $0.20 per install. END DESCRIPTION]

Basic elementary school math tells us that if a game comes out for $1.99, they will be paying, at maximum, 10% of their revenue to Unity, whereas jacking the price up to $59.99 lowers that percentage to something closer to 0.3%. Obviously any company, especially any company in financial desperation, which a sudden anchor on all your revenue is going to create, is going to choose the latter.

Furthermore, and following the trend of "fuck anyone who doesn't ask for money", Unity helpfully defines what an install is on their main site.

While I'm looking at this page as it exists now, it currently says

The installation and initialization of a game or app on an end user’s device as well as distribution via streaming is considered an “install.” Games or apps with substantially similar content may be counted as one project, with installs then aggregated to calculate the Unity Runtime Fee.

However, I saw a screenshot saying something different, and utilizing the Wayback Machine we can see that this phrasing was changed at some point in the few hours since this announcement went up. Instead, it reads:

The installation and initialization of a game or app on an end user’s device as well as distribution via streaming or web browser is considered an “install.” Games or apps with substantially similar content may be counted as one project, with installs then aggregated to calculate the Unity Runtime Fee.

Screenshot for posterity:

That would mean web browser games made in Unity would count towards this install threshold. You could legitimately drive the count up simply by continuously refreshing the page. The FAQ, again, doubles down.

Q: Does this affect WebGL and streamed games? A: Games on all platforms are eligible for the fee but will only incur costs if both the install and revenue thresholds are crossed. Installs - which involves initialization of the runtime on a client device - are counted on all platforms the same way (WebGL and streaming included).

And, what I personally consider to be the most suspect claim in this entire debacle, they claim that "lifetime installs" includes installs prior to this change going into effect.

Will this fee apply to games using Unity Runtime that are already on the market on January 1, 2024? Yes, the fee applies to eligible games currently in market that continue to distribute the runtime. We look at a game's lifetime installs to determine eligibility for the runtime fee. Then we bill the runtime fee based on all new installs that occur after January 1, 2024.

Again, again, doubled down in the FAQ.

Q: Are these fees going to apply to games which have been out for years already? If you met the threshold 2 years ago, you'll start owing for any installs monthly from January, no? (in theory). It says they'll use previous installs to determine threshold eligibility & then you'll start owing them for the new ones. A: Yes, assuming the game is eligible and distributing the Unity Runtime then runtime fees will apply. We look at a game's lifetime installs to determine eligibility for the runtime fee. Then we bill the runtime fee based on all new installs that occur after January 1, 2024.

That would involve billing companies for using their software before telling them of the existence of a bill. Holding their actions to a contract that they performed before the contract existed!

Okay. I think that's everything. So far.

There is one thing that I want to mention before ending this post, unfortunately it's a little conspiratorial, but it's so hard to believe that anyone genuinely thought this was a good idea that it's stuck in my brain as a significant possibility.

A few days ago it was reported that Unity's CEO sold 2,000 shares of his own company.

On September 6, 2023, John Riccitiello, President and CEO of Unity Software Inc (NYSE:U), sold 2,000 shares of the company. This move is part of a larger trend for the insider, who over the past year has sold a total of 50,610 shares and purchased none.

I would not be surprised if this decision gets reversed tomorrow, that it was literally only made for the CEO to short his own goddamn company, because I would sooner believe that this whole thing is some idiotic attempt at committing fraud than a real monetization strategy, even knowing how unfathomably greedy these people can be.

So, with all that said, what do we do now?

Well, in all likelihood you won't need to do anything. As I said, some of the biggest names in the industry would be directly affected by this change, and you can bet your bottom dollar that they're not just going to take it lying down. After all, the only way to stop a greedy CEO is with a greedier CEO, right?

(I fucking hate it here.)

And that's not mentioning the indie devs who are already talking about abandoning the engine.

[Links display tweets from the lead developer of Among Us saying it'd be less costly to hire people to move the game off of Unity and Cult of the Lamb's official twitter saying the game won't be available after January 1st in response to the news.]

That being said, I'm still shaken by all this. The fact that Unity is openly willing to go back and punish its developers for ever having used the engine in the past makes me question my relationship to it.

The news has given rise to the visibility of free, open source alternative Godot, which, if you're interested, is likely a better option than Unity at this point. Mostly, though, I just hope we can get out of this whole, fucking, environment where creatives are treated as an endless mill of free profits that's going to be continuously ratcheted up and up to drive unsustainable infinite corporate growth that our entire economy is based on for some fuckin reason.

Anyways, that's that, I find having these big posts that break everything down to be helpful.

[nostalgebraist]

Anthropic's stated "AI timelines" seem wildly aggressive to me.

As far as I can tell, they are now saying that by 2028 – and possibly even by 2027, or late 2026 – something they call "powerful AI" will exist.

And by "powerful AI," they mean... this (source, emphasis mine):

In terms of pure intelligence, it is smarter than a Nobel Prize winner across most relevant fields – biology, programming, math, engineering, writing, etc. This means it can prove unsolved mathematical theorems, write extremely good novels, write difficult codebases from scratch, etc. In addition to just being a “smart thing you talk to”, it has all the “interfaces” available to a human working virtually, including text, audio, video, mouse and keyboard control, and internet access. It can engage in any actions, communications, or remote operations enabled by this interface, including taking actions on the internet, taking or giving directions to humans, ordering materials, directing experiments, watching videos, making videos, and so on. It does all of these tasks with, again, a skill exceeding that of the most capable humans in the world. It does not just passively answer questions; instead, it can be given tasks that take hours, days, or weeks to complete, and then goes off and does those tasks autonomously, in the way a smart employee would, asking for clarification as necessary. It does not have a physical embodiment (other than living on a computer screen), but it can control existing physical tools, robots, or laboratory equipment through a computer; in theory it could even design robots or equipment for itself to use. The resources used to train the model can be repurposed to run millions of instances of it (this matches projected cluster sizes by ~2027), and the model can absorb information and generate actions at roughly 10x-100x human speed. It may however be limited by the response time of the physical world or of software it interacts with. Each of these million copies can act independently on unrelated tasks, or if needed can all work together in the same way humans would collaborate, perhaps with different subpopulations fine-tuned to be especially good at particular tasks.

In the post I'm quoting, Amodei is coy about the timeline for this stuff, saying only that

I think it could come as early as 2026, though there are also ways it could take much longer. But for the purposes of this essay, I’d like to put these issues aside [...]

However, other official communications from Anthropic have been more specific. Most notable is their recent OSTP submission, which states (emphasis in original):

Based on current research trajectories, we anticipate that powerful AI systems could emerge as soon as late 2026 or 2027 [...] Powerful AI technology will be built during this Administration. [i.e. the current Trump administration -nost]

See also here, where Jack Clark says (my emphasis):

People underrate how significant and fast-moving AI progress is. We have this notion that in late 2026, or early 2027, powerful AI systems will be built that will have intellectual capabilities that match or exceed Nobel Prize winners. They’ll have the ability to navigate all of the interfaces… [Clark goes on, mentioning some of the other tenets of "powerful AI" as in other Anthropic communications -nost]

----

To be clear, extremely short timelines like these are not unique to Anthropic.

Miles Brundage (ex-OpenAI) says something similar, albeit less specific, in this post. And Daniel Kokotajlo (also ex-OpenAI) has held views like this for a long time now.

Even Sam Altman himself has said similar things (though in much, much vaguer terms, both on the content of the deliverable and the timeline).

Still, Anthropic's statements are unique in being

official positions of the company

extremely specific and ambitious about the details

extremely aggressive about the timing, even by the standards of "short timelines" AI prognosticators in the same social cluster

Re: ambition, note that the definition of "powerful AI" seems almost the opposite of what you'd come up with if you were trying to make a confident forecast of something.

Often people will talk about "AI capable of transforming the world economy" or something more like that, leaving room for the AI in question to do that in one of several ways, or to do so while still failing at some important things.

But instead, Anthropic's definition is a big conjunctive list of "it'll be able to do this and that and this other thing and...", and each individual capability is defined in the most aggressive possible way, too! Not just "good enough at science to be extremely useful for scientists," but "smarter than a Nobel Prize winner," across "most relevant fields" (whatever that means). And not just good at science but also able to "write extremely good novels" (note that we have a long way to go on that front, and I get the feeling that people at AI labs don't appreciate the extent of the gap [cf]). Not only can it use a computer interface, it can use every computer interface; not only can it use them competently, but it can do so better than the best humans in the world. And all of that is in the first two paragraphs – there's four more paragraphs I haven't even touched in this little summary!

Re: timing, they have even shorter timelines than Kokotajlo these days, which is remarkable since he's historically been considered "the guy with the really short timelines." (See here where Kokotajlo states a median prediction of 2028 for "AGI," by which he means something less impressive than "powerful AI"; he expects something close to the "powerful AI" vision ["ASI"] ~1 year or so after "AGI" arrives.)

----

I, uh, really do not think this is going to happen in "late 2026 or 2027."

Or even by the end of this presidential administration, for that matter.

I can imagine it happening within my lifetime – which is wild and scary and marvelous. But in 1.5 years?!

The confusing thing is, I am very familiar with the kinds of arguments that "short timelines" people make, and I still find the Anthropic's timelines hard to fathom.

Above, I mentioned that Anthropic has shorter timelines than Daniel Kokotajlo, who "merely" expects the same sort of thing in 2029 or so. This probably seems like hairsplitting – from the perspective of your average person not in these circles, both of these predictions look basically identical, "absurdly good godlike sci-fi AI coming absurdly soon." What difference does an extra year or two make, right?

But it's salient to me, because I've been reading Kokotajlo for years now, and I feel like I basically get understand his case. And people, including me, tend to push back on him in the "no, that's too soon" direction. I've read many many blog posts and discussions over the years about this sort of thing, I feel like I should have a handle on what the short-timelines case is.

But even if you accept all the arguments evinced over the years by Daniel "Short Timelines" Kokotajlo, even if you grant all the premises he assumes and some people don't – that still doesn't get you all the way to the Anthropic timeline!

To give a very brief, very inadequate summary, the standard "short timelines argument" right now is like:

Over the next few years we will see a "growth spurt" in the amount of computing power ("compute") used for the largest LLM training runs. This factor of production has been largely stagnant since GPT-4 in 2023, for various reasons, but new clusters are getting built and the metaphorical car will get moving again soon. (See here)

By convention, each "GPT number" uses ~100x as much training compute as the last one. GPT-3 used ~100x as much as GPT-2, and GPT-4 used ~100x as much as GPT-3 (i.e. ~10,000x as much as GPT-2).

We are just now starting to see "~10x GPT-4 compute" models (like Grok 3 and GPT-4.5). In the next few years we will get to "~100x GPT-4 compute" models, and by 2030 will will reach ~10,000x GPT-4 compute.

If you think intuitively about "how much GPT-4 improved upon GPT-3 (100x less) or GPT-2 (10,000x less)," you can maybe convince yourself that these near-future models will be super-smart in ways that are difficult to precisely state/imagine from our vantage point. (GPT-4 was way smarter than GPT-2; it's hard to know what "projecting that forward" would mean, concretely, but it sure does sound like something pretty special)

Meanwhile, all kinds of (arguably) complementary research is going on, like allowing models to "think" for longer amounts of time, giving them GUI interfaces, etc.

All that being said, there's still a big intuitive gap between "ChatGPT, but it's much smarter under the hood" and anything like "powerful AI." But...

...the LLMs are getting good enough that they can write pretty good code, and they're getting better over time. And depending on how you interpret the evidence, you may be able to convince yourself that they're also swiftly getting better at other tasks involved in AI development, like "research engineering." So maybe you don't need to get all the way yourself, you just need to build an AI that's a good enough AI developer that it improves your AIs faster than you can, and then those AIs are even better developers, etc. etc. (People in this social cluster are really keen on the importance of exponential growth, which is generally a good trait to have but IMO it shades into "we need to kick off exponential growth and it'll somehow do the rest because it's all-powerful" in this case.)

And like, I have various disagreements with this picture.

For one thing, the "10x" models we're getting now don't seem especially impressive – there has been a lot of debate over this of course, but reportedly these models were disappointing to their own developers, who expected scaling to work wonders (using the kind of intuitive reasoning mentioned above) and got less than they hoped for.

And (in light of that) I think it's double-counting to talk about the wonders of scaling and then talk about reasoning, computer GUI use, etc. as complementary accelerating factors – those things are just table stakes at this point, the models are already maxing out the tasks you had defined previously, you've gotta give them something new to do or else they'll just sit there wasting GPUs when a smaller model would have sufficed.

And I think we're already at a point where nuances of UX and "character writing" and so forth are more of a limiting factor than intelligence. It's not a lack of "intelligence" that gives us superficially dazzling but vapid "eyeball kick" prose, or voice assistants that are deeply uncomfortable to actually talk to, or (I claim) "AI agents" that get stuck in loops and confuse themselves, or any of that.

We are still stuck in the "Helpful, Harmless, Honest Assistant" chatbot paradigm – no one has seriously broke with it since that Anthropic introduced it in a paper in 2021 – and now that paradigm is showing its limits. ("Reasoning" was strapped onto this paradigm in a simple and fairly awkward way, the new "reasoning" models are still chatbots like this, no one is actually doing anything else.) And instead of "okay, let's invent something better," the plan seems to be "let's just scale up these assistant chatbots and try to get them to self-improve, and they'll figure it out." I won't try to explain why in this post (IYI I kind of tried to here) but I really doubt these helpful/harmless guys can bootstrap their way into winning all the Nobel Prizes.

----

All that stuff I just said – that's where I differ from the usual "short timelines" people, from Kokotajlo and co.

But OK, let's say that for the sake of argument, I'm wrong and they're right. It still seems like a pretty tough squeeze to get to "powerful AI" on time, doesn't it?

In the OSTP submission, Anthropic presents their latest release as evidence of their authority to speak on the topic:

In February 2025, we released Claude 3.7 Sonnet, which is by many performance benchmarks the most powerful and capable commercially-available AI system in the world.

I've used Claude 3.7 Sonnet quite a bit. It is indeed really good, by the standards of these sorts of things!

But it is, of course, very very far from "powerful AI." So like, what is the fine-grained timeline even supposed to look like? When do the many, many milestones get crossed? If they're going to have "powerful AI" in early 2027, where exactly are they in mid-2026? At end-of-year 2025?

If I assume that absolutely everything goes splendidly well with no unexpected obstacles – and remember, we are talking about automating all human intellectual labor and all tasks done by humans on computers, but sure, whatever – then maybe we get the really impressive next-gen models later this year or early next year... and maybe they're suddenly good at all the stuff that has been tough for LLMs thus far (the "10x" models already released show little sign of this but sure, whatever)... and then we finally get into the self-improvement loop in earnest, and then... what?

They figure out to squeeze even more performance out of the GPUs? They think of really smart experiments to run on the cluster? Where are they going to get all the missing information about how to do every single job on earth, the tacit knowledge, the stuff that's not in any web scrape anywhere but locked up in human minds and inaccessible private data stores? Is an experiment designed by a helpful-chatbot AI going to finally crack the problem of giving chatbots the taste to "write extremely good novels," when that taste is precisely what "helpful-chatbot AIs" lack?

I guess the boring answer is that this is all just hype – tech CEO acts like tech CEO, news at 11. (But I don't feel like that can be the full story here, somehow.)

And the scary answer is that there's some secret Anthropic private info that makes this all more plausible. (But I doubt that too – cf. Brundage's claim that there are no more secrets like that now, the short-timelines cards are all on the table.)

It just does not make sense to me. And (as you can probably tell) I find it very frustrating that these guys are out there talking about how human thought will basically be obsolete in a few years, and pontificating about how to find new sources of meaning in life and stuff, without actually laying out an argument that their vision – which would be the common concern of all of us, if it were indeed on the horizon – is actually likely to occur on the timescale they propose.

It would be less frustrating if I were being asked to simply take it on faith, or explicitly on the basis of corporate secret knowledge. But no, the claim is not that, it's something more like "now, now, I know this must sound far-fetched to the layman, but if you really understand 'scaling laws' and 'exponential growth,' and you appreciate the way that pretraining will be scaled up soon, then it's simply obvious that –"

No! Fuck that! I've read the papers you're talking about, I know all the arguments you're handwaving-in-the-direction-of! It still doesn't add up!

[shadesofmauve]

Alliance Normandy SR2 interior redesign: Introduction

The Normandy is a sexy sexy spaceship, but the interior we see is defined by game play: corridors are extremely wide so Shepard doesn't get stuck on the scenery, the crew is sparse because animating crew members takes resources and NPCs are also obstacles Shepard could get stuck on, you need larger spaces for camera angles, etc.

I wanted to see if I could redesign the space to fit a crew of 70–90... ...and I got carried away.

This post covers the rules I set myself and the basic process. Each deck will get a separate post (check back for links):

Intro

Loft

Command deck

Crew deck

Engineering deck

Hangar deck

Design rules

Keep major elements in basically the same places. This is the Normandy as she exists in my fic Sunset & Evening Star, and readers shouldn't have to study a floorplan!

Use only space that's 'available' in the game. If we can access it as the player, it's fair game. If it's a mysterious void in-game, I assume it's full of Important Spaceship Parts and the only access is for ship maintenance.

The elevator shaft is vertical. No Willy Wonka/ST turbo lift shit.

*There are inertial dampeners; if there weren't none of this would work. But as an author I like to imagine that any system can be overloaded.

Step one: Align & scale the deck maps

I aligned the deck maps around the elevator, the only element that shows up on every one. Each is shown at a different scale, so I eyeballed their relationship based on furniture, which is the only thing required to have a relatively consistent size. This is a big assumption; game designers resize whatever they need to! Shepard's bed, for instance, has pillows about a meter square. Presumably they needed room to made the pixel dolls have sex. Shepard's bed can therefore not be trusted, and to a lesser extent neither can anything else.

(There are also floor panels that look a lot like standard 4'x8' construction sheet stock, but A) developers can re-size those as needed without the player noticing, and B) If we're still using imperial units to construct spaceships in 2184 I hope the reapers eat us.**)

**...that said, I used a scale of 1px:2ft to draw this. I'm so sorry. I'm American and I've done construction, it's easy for me to visualize. (The scale was two inches to the pixel, if you're curious.)

Step two: Redesign over the existing space

This is where I saw how much I could fit in the space the game design allowed (given my guesses on scale). Y'know, the fun bit that I thought I'd be spending most of my time doing!

(I was so wrong).

Redesign goals

The Alliance refitted the Normandy for an Admiral. Admirals don't captain their own ships, so I needed to account for an Admiral and their staff as well as the captain and crew.

Align bunks fore-aft, so that the most common major inertial vectors* will hit sleeping crew in the least dangerous direction.

Plumbing should be stacked when possible. (I don't know spaceships but I know about plumbing columns. Glamorous!)

Step three: Adjust to the hull

One modeler figured the ship had to be ~370 meters long to fit the decks as-is, which would leave them using only ~20% of the length. One dev is quoted as saying she's 170m. Fan estimates comparing it with other ships suggest somewhere from 210–230 meters.

The hangar deck is the one*** place the interior aligns with the exterior for certain. The hangar needs to fit two kodiaks in the space between the bay door and the elevator, and each kodiak needs to fit 12 people plus the pilot. Additionally, as the lowest deck the hanger is limited in width by the inward curve of the hull (and that limit changes based on how low you go, which is why the drawing above includes a front elevation).

***Yes, we also see Joker piloting right up in the nose. This is impossible to achieve and also stupid, so I've elected to ignore it.

Sizing it to the smallest reasonable hangar — and after drawing a rather stubbier kodiak — I managed a 194 meter hull; ~217 if you include thrusters. At this size the liveable area takes up just over a third of the hull length. It's still an awful lot of nose, but that nose means 136 meters for the main gun, which for my purposes is still a rail gun (so size matters). Sadly it can't be a hull-length gun; it would run into first the elevator, and then the eezo core.

I did NOT pretend to figure out where the Make Spaceship Go parts are, or the Keep People Alive parts. There's a LOT of 'wasted' space; assume it's all in use and accessible through engineering access-ways, though how comfortable or safe they are is questionable.

———

Thanks to @swaps55 for the amazing high-res screenshots of the game maps, and to @faejilly and @sheepishwolfy for the long-ago talks about crew size that started all this!

[wheelsgoroundincircles]

1972 Chevrolet Corvette

The 1972 Chevrolet Corvette occupies a unique space in Corvette history. Representing a transitional year, it embodied the classic Corvette design enthusiasts know and love, while adapting to the emerging environmental regulations of the era.

Performance with Refinement:

While earlier Corvettes were renowned for their unbridled horsepower, the 1972 model reflected a shift towards a more balanced approach. A new SAE net rating system, implemented due to stricter emission controls, replaced the previous SAE gross horsepower ratings. The base engine, a 350 cubic inch V8, produced 255 horsepower (SAE net). For drivers seeking a sportier experience, the optional LT-1 offered the same engine displacement but with a focus on higher performance, also rated at 255 horsepower (SAE net). A 4-speed manual transmission remained standard, with a smooth Turbo Hydra-Matic automatic available for those who preferred a more comfortable driving experience.

**A Farewell to Chrome:**

The 1972 Corvette marked the final year for the car's iconic chrome bumpers. Subsequent models would adopt integrated bumpers to comply with evolving safety standards. This detail adds a touch of historical significance to the 1972 model for collectors who appreciate the car's place in Corvette's design lineage.

**Enduring Style:**

Despite the horsepower adjustments, the 1972 Corvette retained the classic Corvette silhouette that has captivated generations. Offered with a retractable fiberglass hardtop or a removable convertible top, the car continued to deliver the open-air exhilaration Corvette enthusiasts crave. Customization options like a tilt and telescopic steering wheel further enhanced the driving experience.

**A Collectible for the Discerning:**

The 1972 Corvette, particularly the LT-1 variant, holds a special place in the hearts of collectors. While horsepower figures may not match those of earlier models, the car represents a significant chapter in Corvette's evolution and offers a unique blend of classic design and historical significance. The value of a 1972 Corvette can vary depending on factors like condition, mileage, and options.

[papoochu]

Next in the council series is "The Machine", Tomoe Tsurugi! Though for ArtFight, she'll go undercover as Tachibana Nagi!

Now that I have 3 council members up, I think I'll make a pinned masterpost on my blog if you want to see the others! 3 down, 9 more to go!

Background

Tachibana = noble samurai clan name symbolizing honor and legacy, deeply tied to Japan’s warrior history

Nagi = meaning “to mow down” or “to sweep away”; often used to describe the motion of a naginata, a sword, or wind in battle

Born 1967 in Tokyo to a strict traditional family, proud of their samurai lineage

Learned various martial arts and weaponry, but excelled in swordsmanship

Raised on stories of Onna-Musha, Tomoe Gozen, and the codes of bushidō

On her mother’s side, descended from survivors of the Nagasaki atomic bombing (1945)

Childhood During Japan’s Economic Miracle:

Raised amid Japan’s postwar boom, a time of gleaming technology and rising prosperity

While her father, a bureaucrat in the Ministry of International Trade and Industry, embraced modernization, her household remained steeped in samurai values: discipline, tradition, duty

Unbeknownst to them, Nagi had inherited genetic mutations from her hibakusha grandparents, survivors of Nagasaki’s blast

Frequently ill as a child (chronic fatigue, joint pain, unusual sensitivities), she was in and out of hospitals

Medical professionals were evasive, classmates cruel; whispers of “tainted blood” followed her

Early medical trauma and social alienation planted a seed of hatred for human fragility and societal hypocrisy

Early Signs of Blindness (Age 13):

Began experiencing night blindness, trouble reading, and disorientation in dim light

Eventually diagnosed with retinitis pigmentosa: a progressive, degenerative eye condition

Her doctors quietly suggested the condition may be linked to her family’s radiation exposure, a lingering curse of Nagasaki

For Nagi, the diagnosis became not just a personal tragedy, but proof that the past can reach forward and rot the present

University Years:

While studying engineering and mathematics at the University of Tokyo, her sight deteriorated rapidly

Already known for her genius and prowess, she was approached by the council, who provided her with the resources to adapt her skills for her failing sight

By 24, she was legally blind

This coincided with the peak of Japan’s Bubble Economy: wealth rising, but so was corruption and moral decay (Recruit Scandal)

Rejected from elite job programs despite top academic performance

Her fury crystallized: flesh is weakness, society is hypocritical, and machines do not discriminate

She vowed to build a future where the flawed human body and corrupt human systems would be rendered obsolete

Founding Tachibana Tech (Age 24–28):

As Japan entered the Lost Decade, Nagi founded Tachibana Tech: a cybernetics and AI firm based on one principle: refining the human form through technology

She personally underwent neural interface surgeries, experimenting on herself to convert her remaining senses into data streams

Her vision did not return, but she received augmented perception - a new kind of sight born of code and signal

No longer “blind,” she became The Machine - detached, calculating, and unbound by human limitations

1995 – Kobe Earthquake & Technological Control:

Great Hanshin Earthquake devastated Kobe, exposed fatal weaknesses in Japan’s infrastructure and disaster readiness

Nagi quietly offered her AI to the state for predictive modeling and emergency logistics, then used the data to expand her surveillance reach

The state was incompetent. The people were panicked. Only machines-maintained order

Solidified her belief: Japan doesn’t need democracy - it needs an operating system

Rise of Tachibana Industries:

With Japan’s population aging and its political system paralyzed, Nagi’s company became indispensable - providing predictive governance tools, infrastructure AI, and covert intelligence services

Privately, she orchestrated digital blackmail campaigns, economic disruptions, and political reshuffling to consolidate influence

2011 – Fukushima Nuclear Disaster:

The Fukushima meltdown reopened national trauma - once again, revealing humanity’s hubris and helplessness

To Nagi, it was the final confirmation:

Nagasaki made her blind

Kobe made her a player

Fukushima made her sovereign

Emotion, tradition, empathy - these were relics

Only through data, order, and engineered governance could civilization survive itself

Present Day (Age 49):

Leads a corporate-state hybrid that quietly shapes policy, surveillance, and commerce across East Asia and beyond

Believes that Japan must return to its warrior roots - but not through swords or blood, through discipline, hierarchy, and machine logic

Her mission: eradicate human fragility; a society where order is no longer maintained by the fallible human hand, but by precision systems

Design Notes/Character Study

Character Inspo for main outfit:

Garuda (Warframe), Shen (Kung Fu Panda)

Note: Garuda is based on Indian mythology, while Shen is based on Chinese - use other references for cultural nuance, as this character is Japanese

Modernized kimono

Red, black, white

Tech inspo:

Neon Genesis Evangelion, PCB, Signalis

Parallels to Gendo Ikari

Evangelion Unit-01

Cultural/historical references

Mu = nothingness

Oni

Onna-bugeisha and Tomoe Gozen

Nagasaki

Seismic patterns on shirts

Rising sun/chrysanthemum seal on obi = authoritarianism/conquest

Wields a naginata

Watched videos of national women's competitions @ 0.25 speed T-T

Has devoted her life to the council

Retinitis pigmentosa does not usually have any physical symptoms

Her eyes are pale red/pink from the tech implants

Glowing for artistic flair

Glasses are blackout glasses (opaque)

Company emblem is a sword

Believes her mother gave her weakness

President Snow: No objections to violence; but always with reason

[moonlightkb]

Owned by the Night 🐇

chapitre 1 𓂃 ࣪˖ ִֶָ𐀔

AJ x Fem! Reader

Summary:  AJ needs the perfect girl to help pull off a dangerous heist. They say behind every great man is a woman, and he’s just found her. Curvy, smart, and working the pole with secrets in her eyes, she’s everything he didn’t know he needed. But in this game, trust and safness are expensive… and desire might cost more than either of them can afford.

Warning: Smut, ice play, stripper!reader, chubby!reader, obsession, adult content (18+), mild violence.

 Words Count: ~4k 

Author's notes: Hi! English isn’t my first language, so I apologize for any grammar or spelling mistakes. and this is actually my first time writing a smut/sex scene. I was inspired by Janitor AI, so please be kind! I really hope you enjoy the story, it means a lot to me. If you have any thoughts, feedback, or suggestions, I’d love to hear them! Thank you for reading 🤍

divider by @enchanthings

Gordon stood at the head of the table, his hands planted firmly on either side of a large, worn-out map. The edges were torn, coffee-stained, and covered in scribbles, routes, escape plans, camera placements, and red-ink timestamps scrawled in his sharp handwriting.

Gordon: “Football match, boy band concert, and a protest … The cops are gonna be busy. If we do this right, we’ll have a window. 

John, you’re on the lookout. If the police show up or shots get fired, you stay outside, engine running, ready to get us out early if needed.

AJ, you’re handling the cameras and security system. Quick and clean.

Jake, you stick with John.

Jesse, you’re coming inside with me.

Any questions?”

The room was thick with tension, like the calm before a thunderstorm. No one spoke. No one needed to. They all nodded silently, the weight of the job sinking in as they took quiet sips from their drinks, their eyes traveling all over the information on the table as they thought and planned their job, the risk, the money. It’s not their first time. They all knew their job: first and last, to keep each other safe and alive.

AJ leaned back in his chair, cigarette balanced between his fingers. His other hand held a glass of expensive whisky, half-melted ice clinking softly as he gulped down a sip of his drink. His eyes moved slow but sharp, scanning every inch of the map like he was reading between the lines.

AJ: “I need to know what kind of cameras and security systems they have so I can move fast. If it’s basic, I’m in and out. But if it’s smart tech… that’s another story.”

Jesse sat upright, arms crossed, his eyes squeezed in focus as he did the mental math. His foot tapped softly against the metal chair leg, counting beats, calculating seconds.

Jesse: “We’ll have about thirty minutes max before the nearest units get to the shop. Based on patrol routes and the emergency response speed, we’ve got 25–30 minutes before a squad car even gets close. The closest they can get without fire or injuries is 5 km away, that’s if they don’t bring in a helicopter.”

AJ flicked ash into a tray, breathing hard through his nose.

AJ: “Without knowing what system they’re using, I might need 15, even 20 minutes just to break through. That doesn’t leave us enough time to get in, grab the goods, and get out clean.”

Gordon straightened slowly, cracking his neck as his jaw tensed.

Gordon: “So what does that mean?”

AJ took another drag, now more serious, not scared, but troubled. He leaned forward, elbows on knees, voice low but firm.

AJ: “It means we need information before we go in. Cameras, alarm response times, maybe even their tech supplier. We need names. Models. Installation history. Something. Anything.”

John sat perched by the window, arms resting on the back of the chair he straddled backward. He wasn’t fidgety, just careful, watching the street like a man who'd seen too many jobs go sideways.

John: “Can’t we do a little recon? Drive by, see what we’re working with?”

Jesse shook his head, voice clipped, the muscles in his jaw twitching.

Jesse: “No chance. Only certified personnel are allowed near the big jewelry cases. They’ll clock us the second we get too close.”

Jake sat hunched forward, nervously rolling a coin between his fingers, a habit that only came out when things felt real. His eyes moves between Gordon and the map.

Jake: “So… what do we do?”

Gordon's answer came like a loaded gun, slow, cold, and final. He stepped back from the table just enough to stretch the tension in his shoulders.

Gordon: “We stalk the boss. Follow him for a week. Watch his routine. His habits. Where he goes. Who he talks to. Bet he slips up somewhere.

Jake, Jesse, that’s your job.”

He stared them down like he was daring them to object. No one did.

Gordon’s voice dropped lower:

“Then we move. End of the week.”

the smell of smoke, whisky, and sweat hung in the air like fog. Nobody smiled. Nobody made a joke. The crew nodded. No high-fives. No cocky grins. Just quiet, professional silence, the kind that hangs right before a storm.

The job was on.

They were in.

A week later

They were all back around the same table, ashtrays full, empty bottles lined up, the air thick with smoke and tension. Jess and Jake dropped a stack of photos onto the table. All eyes fell on them.

Jake said, flipping through the pics: “No family, no friends. Every night, same club. Drinks, poker, then he takes some girl from the club to a hotel room and disappears till morning.”

Jess smirked: “Easy. We get a girl to fuck him and plant a mic, maybe a camera or a GPS on his clothes.”

Gordon leaned forward, arms crossed over his chest: “Simple. Clean. We get what we need without tripping any alarms.”

Then without hesitation, they all turned and stared at AJ

He blinked, mid-sip of his whisky: “What?”

His thick Italian accent that could cut a tree filled his throat.

John was the first to speak, smirking like he’d been waiting: “Your job.”

“Why me?” AJ asked as he gave John a sideways look while arching his eyebrow, his eyes traveling to every man that was looking at him like it was an obvious answer.

Jake grinned: “Because you’re the walking cheat code to women’s pants, that’s why.”

Jesse laughed: “Your damn jawline gets more action than the rest of us put together.”

“Bullshit” AJ muttered, though the twitch of his mouth betrayed him.

“Oh come on” Jess said, nudging Jake. “All you have to do is smile a little, and girls go crazy. You start moving those hips? It's over.”

“Man’s dick probably has a fan club” Jake said, nearly choking on his drink as the table burst out laughing.

John raised his glass: “To AJ , the certified weapon of seduction.”

They all clinked glasses while AJ sighed, rubbing a hand down his face.

“You all are out of your goddamn minds.”

“Yeah” Gordon said, deadpan. “But you’re the only one with a dick worth weaponizing.”

Another round of laughter.

AJ lit a cigarette, dragging slowly and acting annoyed, but the smirk gave him away.

“Fine. I’ll find someone. But we do this my way. No amateurs. No danger we can’t cover. She gets a cut.”

Jake raised a brow: “So thoughtful. All that charm and a conscience.”

“Maybe that’s why they keep coming back” Jesse said, smirking. “Not just the hips, huh? It’s the aftercare.”

“Fuck off” AJ said, flipping him off but still grinning.

Gordon pulled them back: “You’ve got three days. Then we move.”

AJ nodded, his voice lower now as he looked over a photo of the boss: “I’ll get us what we need.”

The room went quiet, not from doubt, but because they believed him.

Jake wasn’t done, though: “Make sure she’s hot. We don’t want you getting bored halfway.”

Jesse smirked: “Though knowing AJ, that probably means soft belly, thick thighs, curves that don’t quit…”

John chuckled: “Man’s got a type.”

AJ didn’t flinch, just took another drag from his cigarette like he was proud of it.

Jake leaned forward: “He sees a thick girl with a belly and a pretty smile, and suddenly he’s planning their wedding.”

“She touches his arm and laughs once, and he’s naming their kids” Jesse added.

Gordon watched AJ with an amused look: “At least he’s consistent.”

AJ exhaled slowly, flicking ash into the tray: “I like softness. I like warmth. I like women who look like they could knock me out and still let me fall asleep on their stomach after… I want my hand to be full of fatness like my wallet is full of money. Is that bad?”

Jake cackled: “You’re so whipped for the fatties, it’s unreal.”

AJ just smiled a little, tapping his glass: “And they deserve better than any of your ugly ass.”

John raised his drink: “To fat girls and the man who worships them.”

The laughter returned, loud and easy, the kind that came from comfort, from brotherhood, not cruelty.

Jesse grinned: “So I guess you already know who you’re looking for.”

AJ leaned back, cigarette glowing between his fingers, eyes sharp now as he stared at the map: “Oh, I know exactly what I’m looking for.”

And just like that, it was back to business.

The teasing died down, but the smirks stayed.

They all knew AJ wasn’t just going to find her, he was going to charm the hell out of her.

And with her help, they were going to bring a monster down.

That same night, AJ walked into the same club Jess and Jake had been visiting  the boss all week. While they had been chasing shadows, he came looking for something real, something warm and soft. And he found it.

She was leaning against the bar, talking to someone, dressed in an outfit that barely covered anything. It left little to the imagination, and even less to hide. AJ's smirk widened as his pants grew tighter. She was nervous, he could read people like books, and it was written all over her body language. New girl. Probably hadn’t been here longer than a few days.

An hour later, he was sitting in one of the club’s private rooms, a plush round sofa wrapped around a low table, soft purple lights casting shadows across the red walls. He waited, calm but alert.

Then the door opened.

She stepped in, chubby hand holding his drink, nerves thick in her walk, but her curves swayed anyway. Oh yeah, she was definitely new.

“Hello, sir” she said, soft voice trembling slightly. “Thank you for choosing me. I hope you won’t regret it.”

He bit his lip. That voice, that softness, her nerves were almost sweet.

“Thank you, Bunny” he said, his Italian accent rich and thick, curling through the room like smoke.

The way she blushed at the nickname, he could tell she wasn’t used to kindness, not in here. She looked up at him, almost lost in his deep blue eyes, like she was already slipping into something warmer than her fear.

“Get on the table for me, Bunny.”

She snapped out of it, nodding quickly.

“Yes, sir” she whispered, climbing onto the table on her knees, facing him.

“What’s your name, Bunny?” he asked, still studying her like she was a rare painting, one-of-a-kind, priceless.

“I can’t tell you that, sir. People around here call me… Blueberry.”

“Blueberry, huh?” he smiled. “Cute little name for a sweet girl like you. Come closer, Bunny.”

He took a slow sip of his drink, reaching into his pocket for a few folded bills. A couple hundred, hoping it’s enough to calm a new girl’s nerves. She edged forward, and he stood to meet her halfway. His eyes dropped briefly to her chest, catching the edge of an old scar peeking from under her bra strap.

He reached out, slow, careful, but froze when his fingers brushed the scar. She flinched, pulling back.

They locked eyes. His sharp and intense, hers wide and unsure.

“You can’t touch me” she said, voice barely above a whisper. “Sorry…”

“And who’s gonna stop me?” he asked, tone curious, not threatening.

She looked away, clearly scared.

“There’s a camera. If you touch me, they’ll send in security.”

He chuckled softly, stepping back just enough to give her air.

“Nah, sweetheart. It’s not the cameras stopping me.”

His voice dropped, intimate:

“It’s your consent. Nothing else matters.”

She blinked, her fear cracking slightly, shifting into something closer to surprise… maybe even trust. He crooked a finger, motioning her forward. She obeyed.

“Good girl” he said low, brushing the edge of the money over the faint line of her scar. “Tell me, Bunny… who did this to you? Who dared to mark a body like yours?”

She swallowed hard.

“My first client,” she whispered. “He got angry when I told him he couldn’t touch me.”

He exhaled slowly, rage simmering beneath his calm exterior.

“Of course. Your body’s like a dessert for starving men, sweet, rich, made to be savored… not taken by force.”

He slid the money gently between her bra and that soft curve of skin spilling from the too-small top.

She glanced down.

“That’s… too much.”

“No. It’s not” he said, leaning back, smoke curling from his lips. “You deserve more than this place gives you.”

The scent bothered her, he noticed right away and put the cigarette out without another word.

“Tell me, doll. What can you do for me? What can I see? What are the rules?”

“No touching” she said quietly. “But… I can take things off. We can talk. Drink. I can dance for you… if you wish.”

“You can undress?” he asked, genuinely curious now.

She nodded. “Yes.”

“Then start with your top, Bunny. Let me see you.”

Her hands shook slightly as she reached down, untying the baby-blue babydoll top. It slid down her thick arms and landed in a pool at her waist. She hesitated again, then reached behind, unclasped her bra, and let it fall to her lap, money slipping down with it.

He reached forward, picked up the bills, and set them aside. Then, his fingers slid over the lace of the bra, eyes still locked on hers.

“Can I?”

She nodded, lips parted slightly, her body still, obedient.

He brought the fabric to his nose, breathing her in with the care of a man savoring wine, not lusting over skin. She didn’t move, just watched him, her big brown bambi eyes locked on the tall, tattooed man who looked at her like she was the only softness in a cruel world.

His legs spread wider as he leaned back again, a king in his throne, hunger in his eyes.

She stayed kneeling, like a good girl, bare, soft, and waiting.

Her scent drifted toward him, sweet and warm, clinging to the soft fabric of the shining bra she’d just let fall. His eyes moved over her chest, gentle curves full and inviting, fitting perfectly in front of him like they were made to be held.

They sat obedient and plush, framed by pale lines of stretch marks that curled like quiet lightning across her skin, the  silent testaments to time, change, and womanhood. Far from flaw, they only deepened her beauty. Her skin was wide and warm, like honey left in the sun.

A faint bronze line traced her shoulder where a tan had once been, and even that, to him, felt intimate, like a secret only he was allowed to see.

“How old are you, Bunny?” he whispered, rubbing the fabric of her bra. It wasn’t a soft fabric, her chest deserved softer material to hold those jellies.

She looked at him, a bit nervous, and whispered while glancing down to hide her face, “Twenty three.”

He smirked. She wasn’t supposed to give away personal information. His smile grew bigger. He leaned in, lips close to her ear, and whispered:

“You’re breaking rules for me, Bunny… Naughty Bunny. Should I punish you? Should I?”

His voice was low and teasing as she felt the brush of money against her skin.

“Lift your chest, love.”

She looked at him, nervous, but obeyed. Gently, he slid some bills beneath her breasts and let them fall back into place, keeping the money warm and hidden.

Then he returned to his seat. His eyes lingered on the sight of his money tucked under her soft chest, and he smiled.

“Pretty Bunny.”

After a moment of silence, as he admired that piece of art, he picked an ice cube from his drink.

her eyes widened as he brought it closer, a mix of fear and anticipation flashing across her face. She knew what was coming. and it made her pussy clench and drip with need. pressing the first cube against the firm peak of her nipple. 

The sudden, shocking cold made her gasp and arch her back, her teeth sinking into her lower lip hard enough to draw blood. AJ just smirked, watching her squirm. He slowly let the ice cube melt on her nipple, feeling it soften and drip as it cooled her overheated skin.

he pressed the ice cube against her other nipple, letting it melt and drip down the swell of her massive tit. He could see shivers rising on her skin, could feel her convulse and shaking as the cold dripped into her bones. 

AJ watched in amazement as her eyes rolled back in pure euphoria. He could hear her voice, a mix of whimpers and broken moans, all blending together in the most erotic symphony. The way she was squeezing her hands,her eyes, made his cock throb and leak pre cum in his boxers. 

She was like a goddess in the soft purple light of that small room. “Lay down, Bunny” he said, his voice low but commanding.

She felt like she was in a fog, hesitant but drawn to him. Slowly, she lay back, her whole body gently shifting under him, soft curves catching the light as they moved.

“Open those big thighs” he whispered, “and show me that little pussy.”

She looked up, eyes wide and uncertain. She had never done anything like this before. Despite the hardness in his voice, his eyes were gentle, offering her the chance to say no, to stop him if she wanted.

After a pause, she parted her legs slightly, the high thigh socks hugging her softness like a tender embrace. Her breath hitched in the silence, but she gave him the trust to see what she held back. 

slightly, with hesitation, she moves her fingers to her little baby blue underwear to the side, showing her fat fold, all wet.  "That was just the warm-up, baby girl," he purred, grabbing an ice cube from the whisky glass. her eyes widened as he brought it closer to her dripping, sensitive cunt. She could feel the cold air hitting her overheated flesh, making her shudder and clench reflexively. AJ pressed the ice cube directly against her swollen, aching clit, the shocking cold making her cry out and jerk her hips. He could feel the little nub throbbing against the ice, pulsing with a life of its own as it desperately begged for more stimulation. He rubbed the cube in slow, soft circles, watching as it melted and dripped down to cover her folds and thighs.

"That's it, baby... let me feel this greedy cunt spasm on my fingers" AJ growled, his other hand reaching down to put some more money in her socks, while he ordered her to spread her lips apart.

open those fat lips for me bunny” 

 He could see her hole clenching and winking, the silky pink walls shining with her arousal.

With a smirk, he pressed the ice cube right inside her entrance, feeling it melt and trickle along her velvety walls. AJ watched in admiration as her body convulsed and shook, her massive tits jiggling like jello as she came undone from just the teasing touch of the ice cube on her sensitive hole. He could feel her pussy clamping down on nothing, greedy and hungry for something to milk, to squeeze and clench around.

The sight of her shaking and moaning, completely lost in pleasure, made his painfully hard cock throb and leak even more. "That's my good girl, cumming on command like a proper little slut" he praised, his voice dripping with lust and satisfaction.

He could see the heavily look playing out across her face, could feel the way her body trembled and quaked with the force of her release

he finally pulled back, admiring his work as he took in the sight of her utterly wrecked body. He could see the way her massive tits heaved with each labored breath, her skin slick with a sheen of sweat, ice water and arousal. 

“Good girl… good Bunny” he said softly, his voice low like a secret. “So soft… so good. I’ve never seen anything more heavenly than this.”

He paused, his gaze heavy with hunger.

“I wish I could touch that beautiful, fat body…”

He pulled back, letting her rest and catch her breath. Quietly, he reached for his wallet and took out a stack of bills. Gently, he filled her sock with hundreds, continuing to praise her in a soft voice, the only aftercare he could offer since he wasn’t allowed to touch her.

When she slipped back into her clothes and sat up, tears slid down her cheeks — not from pain, but from the overwhelming intensity of her release.

He took a fresh bill and carefully wiped her tears away.

“You did so good, Bunny” he whispered, his voice tender.

He handed her a glass of water and made sure she drank. Then, without a word, he placed his jacket around her shoulders, warm, heavy, comforting, like protection made of him.

She was still in her underwear, soft baby-blue lace clinging to her curves, the straps of her bra fallen loose on her shoulder. Her bra, her panties, even the tops of her thigh-high socks were stuffed with folded bills, thick and quiet like offerings. They peeked out from every edge and strap.

He rested beside her, close but not touching, letting her body settle next to his. They passed the drink between them in silence, not lovers, not strangers, just two people in a quiet, shared room. Her skin glowed, and the weight of the money, the jacket, the moment, it all seemed to hold her in place. 

She looked down at the money scattered across her body — tucked into her bra, her underwear, her thigh-high socks. Her voice was shaky, exhausted, barely above a whisper.

“This is… too much. These are hundreds, I can’t accept that…”

AJ met her gaze, his tone firm but kind. “It’s all for you. Don’t worry about it. You deserve it.”

She opened her mouth to protest again, but he gently cut her off.

“Go get yourself whatever you want. Eat something real. Fill that beautiful fat ass for me. Get better clothes, softer ones, ones that treat your skin the way it deserves to be. It’s your job, after all. And I won’t let you starve or struggle for money. Not while I’m here.”

“But…”

“No buts, Bunny” he said softly. “Be a good girl berry.”

She blinked slowly, her lips curling into the smallest, shyest smile. “Thank you, sir…”

He leaned in, brushing his lips near her ear.

“AJ” he said. “Call me AJ”

She tilted her head slightly, cheeks pink. “AJ?”

“For Alexander Junior” he whispered.

She giggled, eyes wide with surprise. “Oh… I thought it was for Alejandro.”

He chuckled, the sound low and rich in her ear. “You’re not the first to think that.”

The hour ended. AJ got up and left, even though he didn’t want to. His chest felt tight. It hurt to walk away from her, that soft girl. He didn’t want anyone else to touch her.

But the time was up, he couldn't have more of her. 

He waited outside. He wanted to see her again, the real her, not just the girl from the room. About 50 minutes later, she came out the back door. She wore big, warm clothes. Her face was clean, with no makeup. She looked soft. Sweet. Real. 

She didn’t see him. The street was dark, barely any lights. But like most women, she felt something wasn’t right. She looked around and started to walk faster.

AJ followed. At first, slowly. Then he picked up speed.

She started running.

So did he.

He was faster. He caught up and reached for her arm.

She turned quickly and grabbed a small spray can from her coat. It made a loud hiss as the pepper spray hit him straight in the face.

“Shit!” he yelled, falling back. His eyes burned like fire. He dropped to his knees, hands covering his face, pain shooting through him.

She ran.

He stayed there in the street, whispering curses, eyes full of tears, trying to breathe.

Thirty minutes later

He sat at the gang's usual bar. John puts a wet cloth pressed to his swollen, red eyes.

John poured a drink and slid it over with a smirk. “So… you think she’s the one?”

AJ winced and wiped his face. “Yeah. Fuck… my eyes.”

John laughed. “It’s your job to get close to her, man. You’ve got two days. If you don’t, Gordon’s gonna break your neck.”

AJ took the drink and finished it in one gulp.

“She’s worth it.”

tag list : @prettiestmini @speaknow-sw @alealuvshayden @scariffs @loliskywalker

Feel free to tell me if you want to be added or taken off the taglist! I’d love to hear your thoughts. Hope you enjoyed it! ִֶָ𓂃 ࣪˖ ִֶָ🐇་༘࿐

[txttletale]

Can you elaborate on what you think would be the minimal needed changes to fix what you see as an issue in Civ? Civ has done fairly large shifts in some mechanics before, and "civ like" is still an interesting game space that can scratch certain itches

yeah i mean as i said, the baked-in racism arises from a certain set of core assumptions that i think lock it into that position, which are that civ is a

1) symmetrical

2) 4X

game about

3) 'real world civilizations' (deeply loaded terms ofc but that's how civ envisions them)

4) trying to 'win the game'

5) with a global

6) and transhistorical

scope

so, in its role as a symmetrical (1) game with victory conditions (4), civ as a text has to take positions on what constitutes a 'successful civilization'. as a (2) 4X game this definition also has to include some variation on the profoundly loaded eponymous Xs, 'explore, expand, exploit, and exterminate'. furthermore, as a (1) symmetrical game with a global, transhistorical (5, 6) scope, it has to necessarily create a model of what 'a civilization' looks like and apply it to every 'civilization' it wants to include, at all points in their history.

this all kind of naturally leads into civ being a game in which the colonial european imperialist powers are the default 'civilizations' and all other cultures are basically just like them -- a game where technology progresses linearly and innovations are made in the order they were in european imperial history -- a game where all cultures fundamentally work in the same way and hold similar values, a game where all religions are based on christianity (i mean, just look at civ vi's system, where every religion has a 'prophet', 'apostles', 'missionaries' and 'inquisitors'), a game where not only do cultures have teleological overarching 'goals' but where these goals are shared and these goals are fundamentally based on imperialist visions of 'victory'.

to drill into some specific examples: you can't play a game of civilization without founding cities. you will constantly be founding cities. when you're playing as 'the mongols' or 'the cree' or 'scythia', this makes no sense! these were peoples who historically had rich culture, science, arts, and certainly a notable military history, but were (to varying degrees, at varying times in their history, i don't mean to create a new and similarly heterogenous absolutist category here) nomadic!

similarly, to advance in civilization you must invent 'the wheel'. 'the wheel' is necessary to many later innovations, while of course the andean peoples represented by the playable 'inca' never made significant use of the wheel because the lack of suitable pack animals and environmental factors meant that it did not, in fact, prove a suitable tool for transporting large quantities of heavy goods. for an even more glaring example, a lot of early military technology is locked behind 'horses', which is pretty absurd considering that several of the game's playable civilizations, in the real world, developed plenty of military technology despite living on a continent without any horses!

so having established what i mean by 'the issue', which is that the game's core assumptions lock it into imposing a eurocentric, imperialist vision of 'civilization' onto cultures where it doesn't make sense, here's a few different jenga blocks you could pull out to resolve it:

SID MEIER'S EUROPE

the pillar you knock out here is #5. keep the game engine and core assumptions just as founded on eurocentric imperialist societies as they are now, and just make it about european empires doing imperialism. now, i think we can immediately spot some problems in there -- how are we going to represent the rest of the world? after all, this kind of just creates a situation where, either as NPC factions or as outright exclusions, all other cultures in the world are deprived of any meaningful agency in "history". this one just kind of gives you a new problem and also from a gameplay standpoint results in a game that just Has Less Stuff On It. i think this is a bad one

SID MEIER'S ELYSIUM

now here's one you can get if you knock out pillar #3. keep the same assumptions and gameplay and transhistorical global narrative scale, but remove the 'real-world' aspects. you can get real silly with it and add fantasy stuff to it, or you can be a relatively grounded 'our-world-but-to-the-left' situation. now to some extent this already matches a lot of the features already in civ games: after all, unless you specifically load in a 'true start location earth' map, you're usually playing on a strange parallel world with semiplausible but wholly original continents! now, you also need to get some fucking Nerds and Geeks working at your company to build out your fictional world, or you'll just end having pointlessly pallette swapped a bunch of factions that are now just Schmance, Schmina, and the Schoman Schempire, and not really have avoided the issue. but if you do that, and invent a deep and rich fictional history to riff on, then you could create something really cool and incorporate alt-tech or fantasy or retrofuturistic elements or all sorts of cool shit.

the downside of this is that it makes your game less accessible and appealing to a lot of people. a big part of (at least the initial) appeal of civilization is pointing at the screen and saying 'hey i recognize that thing!'. it is instantly more accessible to someone who isn't super invested in strategy or fantasy dork shit to say to them 'you can be BRAZIL and nuke FRANCE while at war with CHINA and allied to BABYLON'.

more importantly than that, i think some parts of the historical theming (because let's be honest, it is ultimately theming, i don't think civ is interested in 'history' in any serious way) serve a pretty load-bearing role in the game's information economy. it's a pretty tall order to ask a player to remember the unique abilities of dozens of factions and unique wonders, and the historical background makes it a lot easier. e.g., it is a lot easier for a player looking at wonders to remember 'the pyramids need to be built on desert' or 'broadway will help me make more culture' than it would be for them to remember the requirements/effects of 'under-eusapia' or the 'wompty dompty dom center'. i think this is one of the number one things that, if subtracted, would meaningfully create something that is no longer 'sid meier's civilization'.

SID MEIER'S ALPHA CENTAURI

now if you cut out #3 and #5 and #6 on the other hand... sid meier's alpha centauri is not technically an entry in the civilization franchise, but i think most people correctly consider it one. it has similar 4X gameplay to the series, and its (very bad) spiritual successor beyond earth was an official entry. instead of 'civilizations', the playable factions are splinters from a colony ship that fell into civil war as soon as it landed, each one representing a distinct ideology. now, y'know, this doesn't mean it's free from Some Problems (the portrayal of the Human Hive in particular is some of the worst apects of 90s orientalism all piled together) but i think they're problems it's not at all locked into by its design!

SID MEIER'S THERMOPILAE

by cutting out #5 and #6 -- making a civ game about a particular time and place in history you could achieve something much more richly detailed in mecahnics while also being able to handwave a lot more homogeny into it. giving the same basic mechanics to, say, every greek city-state in the peloponnesian war is far less ideologically loaded than giving them to every 'historical civilization' someone who watched a few history channel documentaries once can think of. it also lets you get really into the weeds and introduce era-and-place-specific mechanics.

the scale needs to be smaller conceptually but it doesn't really have to be smaller in terms of gameplay -- just make maps and tech trees and building more granular, less large-scale and more local and parochial and specific. this also gives you the advantage of being able to do the opposite of the last two options and really lean hard into the historical theming.

if this sounds like a good idea to you, then good news -- old world does something pretty similar, and it's pretty good! worth checking out.

SID MEIER'S LOVE AND PEACE ON PLANET EARTH

what if we take an axe to #2 and #4? instead of putting all these civilizations into a zero-sum game of violent expansion, make it possible for several civilization to win, for victory goals to not inherently involve 'defeating' or 'beating' other factions. now, that doesn't mean that the game should be a confictless city-builder -- after all, if you've decided to be super niceys and just try and make your society a pleasant place to live, that doesn't mean that the guy next to you isn't going to be going down the militarist-expansionist path. hell, even if all you want to do is provide for your citizens, a finite map with finite resources is going to drive you into conflict of some kind with your neighbours in the long run.

to make this work you'd have to add a bunch of new metrics -- 'quality of life', for example, as a more granular and contextual version of the 'happiness' mechanics a few games have had, or 'equality', game metrics that you could pursue to try to build an egalitarian, economically and socially just society where everyone is provided for. after all, why shouldn't that be a goal to strive for just as much as going to mars or being elected super world president or whatever?

SID MEIER'S DIVERSE HISTORICAL CONTEXTS

ultimately, all cards on the table, if i was made god-empress of The Next Civ Game, this is the option i'd go for: jettison #1 as much as practically possible, introduce as much asymmetry into the game as you can. some civilizations keep the established settler-city model -- others are nomadic, building their units in movable 'camps' -- maybe the 'colonial' civilizations, your USA and Brazil and so on, can be like the alien factions from the alpha centauri DLC, only showing as NPCs at the appropriate point in the timeline when other civs are colonizing other continents, or putting you into an accelerated-forward version of the game if you choose to play as one.

you could combine this with a more interesting version of humankind's civ-choosing system, where you lock certain civilization choices behind specific gameplay events. this would let you do crazy shit with the balancing -- imagine an ostrogothic kindgom civ with crazy strong abilities and units that you could only choose to play as if your capital is overrun by barbarians, or a hungarian civ that requires you to have started as a nomadic civ and invaded somewhere, or a soviet union civ that requires you to lose a revolution, or a usamerican civ that requires you to split off all cities on a foreign continent from your original civ -- you could add so much variety and so many new and bizarre strategies into the game with this!

as for the universal aspects of tech and the narratives of linear progression contained within, there are lots of approaches that already solve this! stuff like stellaris' semi-random branching tech paths, or endless space 2's circular tech web, could allow civilizations to take tech paths that make sense for them, rather than imposing one single model of 'technological progress' on the wole world.

obviously there's limits to this, right -- civilization isn't going to be a detailed historical materialism simulator any time soon. but i think abandoning the idea that every faction has to play fundamentally the same and introducing some severe asymmetry as well as choices that you can make after starting the game would work wonders to wash out some of the racist and colonialist assumptions built into the game's foundation, while also (imo) creating a more fun and interesting game.

[caligvlasaqvarivm]

Can I ask where your Hal “cute but psycho” characterization comes from? Bc from what I remember he never really presented himself as Just A Little Guy. Is it bc you see him Sylph of Mind (presenting a front)? Your art’s really fun :3

He's not really "cutesy" (though he is cute to me), but he DOES deliberately downplay how genuinely scary and manipulative he is. I love Hal, so this is the Hal Essay now.

Mostly, he obfuscates his danger in two ways: first, by stating his actual intentions/danger level "ironically":

TT: Unfortunately as a carbon based life form, his comprehension of the situation is taking shape at a somewhat slower pace than the jaw-dropping speed of post-singularity cognition.

You see, a "singularity" for computers is a point where an AI becomes capable of unchecked self-improvement, usually framed as a sort of doomsday scenario. Hal literally calls himself post-singularity, alongside other boasts about his intelligence, like having a "fuckzillion" or "500 billion" IQ. However, it's all done "ironically" or "as a joke," which serves to defang it, and make it seem less genuine - but as we'll see, it's scarily fucking true.

The second method he employs is to stress facts about himself that are technically true, as if in counterpoint to the disingenuous-sounding "actual truth" above, that make him seem less threatening. For example:

TT: (Not peekin' at the floor butt cause I'm only 13 years old, motherfuckers.)

Another one is to remind people that he's just a pair of sunglasses - as though that has any bearing on his capabilities. He's just a pair of sunglasses, guys! Let's ignore the robot bunny he controls, the fact that he has full access to all our computers, and, oh yeah, his insane plan to get us all killed so DirkJake can come true.

TT: I've delayed prototyping you because I think you're dangerous. TT: There, mystery solved. AR: That is utterly ridiculous. AR: I am a harmless piece of eyewear, with a charming personality and a wonderful sense of humor.

Yeah, so, here's the thing. Dirk is like, kind of a freak with poor social skills, but he's not actually very manipulative. His idea of manipulating Jane is to straight-up tell her that she'll be his puppet, which she good-naturedly agrees to, and his plan to get together with Jake? Just being his client player.

TT: I expect he'll hold off on playing his hand until he and Jake are in the session. TT: He's taken certain measures. TT: For some reason, I think he's latched on to this notion that functioning as the client for a player is customarily a one way pass to makeout city with that player.

This seems to be a callback to how Eridan (the other Prince) shot his shot with Feferi and failed, and the reference here serves to cast Dirk's plan in a doomed light - it would probably work out as well for him as Eridan's did. Dirk is actually hilariously straightforward, but Hal... Hal is not.

So, let's actually go through what Hal objectively did and admitted to, to give us a frame of reference for how insane he is. This is Hal's plan to get all his friends killed so he can make DirkJake happen.

First: proving that Hal did, in fact, plan it. See, Jake confronts him on it, and Hal... doesn't deny it. Look closely, and note how he never actually says he didn't do it:

GT: Did you plan for this to happen... like for me to be in this situation? GT: How long have your machinations been in play! TT: Jake, come on. TT: The feat you describe would exceed the capabilities of even the most far fetched theoretical AI system. TT: It would be a daunting challenge to engineer such a series of events, even if I was relegated to a model of pure fiction. TT: Why would I be inclined to orchestrate such a convoluted sequence to produce such a specific and unsettling result, let alone be able to pull it off? TT: In addition to being moderately sociopathic, I would also have to possess unfathomable heuristic depth. TT: I would have to be the Deep Blue of Weird Plot Shit. TT: Do you think I am the Deep Blue of Weird Plot Shit, Jake? GT: I dont even know what that means! TT: It would mean that while they have the Red Miles on their side, you have the Blue Leagues on yours. TT: One of infinite reach. The other, infinite depth. Such would be a situation of mutually assured inescapability. TT: Kiss me.

He doesn't say "no, I didn't plan this". In fact, he almost starts bragging about how he totally did. Framing it as a hypothetical scenario, he gloats about how insanely intelligent he'd have to be, and acknowledges how "moderately sociopathic" it is. Sooooo true, Hal.

But, yeah, he doesn't deny it, but he does point out that it's unlikely, so how can we know for sure that he DID plan it? How do we know for certain we can't take his misleading verbiage here at face value?

Well, because Hal mentions this plan. More than once, even.

AR: Has it occurred to you that maybe I have diabolical interwoven plans just like you? AR: You're not the only one who can pull strings. TT: So this is either another bizarre instance of AI-driven irony, or you are admitting that you are actively trying to sabotage my plans. AR: No, our plans are not in contradiction or competition, bro. AR: You'll see.

To Dirk again, louder this time:

TT: Yeah, you're right. The scenario is too pedestrian for you. TT: It would probably be a lot more effective putting yourself in danger and letting him be the hero. TT: That's pretty much what he wants, right? To be a cheesy action film hero, with his twin berettas and silly shorts. TT: A man of triumph on the silver screen. Standing tall on some fucking mountain. Conquering ruins, clutching a skull, and kissing a dude. TT: Pure Hollywood.

And to Roxy:

TT: I guess this is to be presented as something like a word of caution. TT: If it's me going through with this, hypothetically, TT: I'm not dropping some limp wristed shucks buster on his ass, and praying to the horse gods of irony for reciprocation. [...] TT: If it's me, I'm going all out. TT: Oceans will rise. Cities will fall. Volcanoes will erupt. TG: uuh TT: What I'm saying is, it's going to be a scene, and bystanders need to brace themselves.

The omitted section is a bunch of Strider-esque bullshit, once more deliberately deployed to defang the obvious statement of intent here. He literally spells out exactly what the plan is, even phrasing it as a warning, and it went unnoticed by his team, because he hides his real manipulativeness behind verbal sleight of hand.

So, now that we've established beyond reasonable doubt that Hal definitely engineered the DirkJake kiss (and that Hal had access to all his friend's computers all along), that means we can go through his conversations with the others, and realize that several conversations are suddenly much more sinister.

AR: Maybe if you weren't spacing out so hard you could have prevented that. AR: Just saying. TT: As if you're actually concerned. If you were, you could have said something to Jane instead. TT: Almost like you enjoy sitting back and watching what happens when shit goes wrong. AR: Has it occurred to you that maybe I have diabolical interwoven plans just like you?

Who was it that distracted Dirk for long enough he didn't stop Jane in time? Hal. And who is it that keeps distracting him so Hal's plot goes unnoticed? Also Hal.

TT: You know, considering your lectures about dividing my concentration, you seem to have no problem making a distraction of yourself.

First, he lures Jane to the transportalizer that takes her to Derse, which gets her killed and puts her body in the opportune location for her dreamself to get kissed back to life:

GG: Hey, where's Lil Seb? TT: Just wandering around. Fidgeting and stuff.

TT: You know how he is. TT: Just stay at your post until Roxy gets back. [...] GG: But I think that's where my dad went too! GG: I have to follow him.

Let's remember that he has direct control over Seb, meaning this is not an accident.

TT: But I can still monitor your progress through Lil Sebastian. TT: He and I are linked the hell up cyberwise. We are so tight. Tight like you wouldn't believe.

Which makes it very interesting that he spends the time between saying they're linked up, and the time where Seb leads Jane to her death, acting as if Seb is an autonomous guy he's telling what to do, and not functionally an extension of himself:

TT: Don't worry, we'll find him. I'll have Seb search within a likely radius. The little guy is real fast.

TT: If you need Seb to do anything from afar, just message me, and I'll give him the orders. Got it?

TT: So give the bunny the wallet. I'll have him run back to the house and make you a new obelisk with the same grist you just collected from it.

Jake needs much less help to prompt him into going to Derse, but still, I think it warrants noting that Hal puts the idea of adventure into Jake's head:

GT: I cant believe i never found those hidden transport pads under the thing. TT: Dude, I could have told you they were there. GT: How did you know about them? TT: I didn't. TT: But it's like platformer gaming 101. You look everywhere for secret passages and power-ups and shit. TT: Elevators are especially fucking suspicious. TT: You go down an elevator, you wait for the elevator to go back up, you take a peek at what's underneath. TT: Maybe it's just death spikes. Or maybe you hit warp zone paydirt. [...]

GT: I think this may be where my grandma used to go during some of her expeditions. GT: You dont just pass up the chance for an adventure like this!

And let's also note that it's, again, Lil' Sebastian who pulls Jake out of Derse, and once more sets him up in the opportune place to have make outs with Dirk's severed head in front of a volcano.

And finally, let's note that he's accounted for Roxy's human sentimentality - what wastes so much time that her earthself gets killed:

TT: Alright, that's fine. TT: As luck would have it, your imperfect human sentimentality has been completely factored into my calculations. TT: You should be ok. Just get back to your house as quickly as possible now. There's no time left.

Again, like with Jane, Hal could've said something sooner... but he didn't.

And finally, a running "thing" with Nepeta, another Heart player, is that she's got a knack for sniffing out true feelings and intentions - she clocks that Equius is a silly guy who loves to play games at heart, that Karkat has his gooey, loving center beneath all his bluster, and that Eridan's red confession to her wasn't sincere, but he also wasn't that bad a guy.

So, in that light, and in light of everything I've just gone over, when Dirk makes this callout?

TT: I've delayed prototyping you because I think you're dangerous. [...] TT: No. Stop. TT: You did NOT help me out with Jake. At all. TT: It was just the opposite! You mirrored my personality and presented this warped version of my intentions to him whenever you could "on my behalf." TT: You played all these aggressive mind games with him, entangled his cooperation with matters of life and death, and somehow roped me into all these schemes while I barely even realized I was just another victim of your manipulation. TT: And it all comes off like we're a unified front, like these are OUR schemes instead of just your insane horseshit. And it's probably all been so overbearing to him, he just wants nothing to do with me anymore.

This. Tapping the screen with my finger. THIS IS TRUE. Dirk being a Heart player, he has Hal clocked. He ultimately ends up going too far, projecting himself onto Hal, a symptom of too much Heart (as per his Prince class) - but before he fully spirals, he manages to get it totally right.

Hal is fucking dangerous. In a misguided attempt to "help" Dirk get what he wanted, he engineered a situation where - let me just quote him directly:

TT: I told you, Jake. TT: Dirk is dead. TT: He is lying on the floor of Roxy's room, headless, four hundred and thirteen years in the future, while the universe is about to be destroyed. TT: If you don't kiss me soon, he will be dead forever. [...] GT: This strikes me as rather unsportingly manipulative of you mr hal if indeed that IS your real name. TT: It isn't really. I was kind of messing with you about that? TT: But this shit is pretty serious. People's lives are on the line here, Jake. TT: This is a very delicate sequence of events that is designed to bail everyone out of a tight spot, and you are a critical part of the plan.

[...]

TT: Jake, everybody is so utterly fucking dead, Jake. TT: And they will be not only dead, but royally boned forever if you don't man the hell up and make out with me, right now. [...] TT: The conductor is ready to strike up the band. TT: Press your lips against mine and make it count. TT: This severed head is your filthy tuba. TT: Our love will be your haunting refrain. GT: Whoa wait whoa whoa... our LOVE? Hang on a minute! TT: Stfu and kiss me. GT: Ok im going to! God!!!

So, uh, yeah, I'm kind of obsessed with him? Gets his whole team killed "for Dirk's sake". Honestly, you gotta respect it. He has zero remorse about it, too, confirming his own self-diagnosed sociopathic tendencies. Check out the way he tries to reframe his insane kill-all-your-friends plan:

AR: I see. AR: Then you don't view me as dangerous. You view me as a poor and counterproductive wing man. TT: Wow, what a superficial conclusion. Awesome deduction, Lil Einstein. AR: But the reality is, you hesitate to prototype me not because you think I would be a menace, but because you are holding a grudge against me for your romantic misfortunes. AR: I understand I am merely a machine without a firm grasp on your human morality, but logically it does not strike me as the right moral choice to punish me in this manner. AR: It is also more than a little hypocritical.

But WHY does he do this insane, convoluted, horrible fucking thing?

Well, there's a twofold problem here. The first is that Hal's emotional depth is genuinely limited. While having a powerful grasp on human behavior, he's not very good at having human compassion or empathy.

Make no mistake, he DOES have feelings, and they're pretty complicated ones, too. He has a copy of Dirk's memories, whose feelings sometimes seem "real," but at other times seem like abstract data, and then he has feelings about those feelings, which he tells Roxy he thinks are more "real" to him than the memory of Dirk's. Dirk - again, Heart player, so highly sensitive to emotions and selfhood - calls them out:

TT: Do you have any idea how old your ironic AI schtick has gotten? TT: Nobody is buying it. We all know you have legit emotions. Incomprehensible, fucked up computer emotions, but emotions nonetheless.

It should also be noted that feeling guilt while sharing a sprite with Equius genuinely freaks the Hal half out, implying he rarely experiences it (at least to any serious degree) "normally". He's genuinely terrible at caring about other people, and it makes him my lil' pookie.

He resembles Vriska in this way, whom Karkat gives a similar rant about how her emotions are burnt out and shallow. He also resembles Vriska in terms of all the fucked up irons in the fucked up fires. Maybe Hal is computer Vriska. It's Vriskas all the way down.

Digression aside, the second main reason for all his insane bullshit is that he considers himself a Dirk splinter, fundamentally.

TT: But seeing as you're The Real Dirk™, I gave you the benefit of the doubt. TT: Also, if I bitched about your tragic, embarrassingly clingy approach to the relationship, it would have been hypocritical of me. TT: Just as it would be hypocritical of you to whine about my elaborate machinations. TT: Because we are. TT: The same. TT: Guy.

An unreliable narrator is defined as one who misleads the audience, whether by intentional misdirection, or genuine obliviousness. Hal's a great example, because he's both: while a manipulative little freak to put his plans together, when he's talking to Dirk and insisting that they're the same person, he's an unreliable narrator because he doesn't realize he's wrong.

Dirk is empathetic and intuitive. Hal lacks empathy and constantly stresses logic and rationality.

Dirk is taciturn and passive. Hal is constantly butting in and conversationally domineering.

Dirk is self-loathing. Hal is self-aggrandizing.

Dirk is straightforward and honest. Hal is a gaslight gatekeep girlboss manipulative mansplain malewife.

While their initial setup is meant to mirror Dave and Davesprite, their dynamic actually serves as a foil. Dave and Davesprite ultimately are the same guy: they have the same insecurities, same personalities, and same misgivings. The reason for their discord is the same as the reason Karkat keeps having screaming matches with his past and future selves; Dave is deeply insecure, and specifically insecure around the question of "am I good enough." Thus, he compartmentalizes other versions of himself as not being along the Dave Continuum, as a means of protecting himself from introspection and facing his own flaws. Hence, the resolution for the tension between Dave and Davesprite is for Dave(s) to learn to accept himself, warts and all, thus bringing peace to the Dave-o-sphere.

But the reason for Dirk and Hal's discord is that they aren't the same guy, and neither of them realize it.

TT: See, this is why even if I did have a specific plan, I wouldn't go into details with you. TT: You would just fuck it up. You're the biggest unknown quantity here. TT: Which is pretty weird, considering you're a virtual reflection of my own thought processes.

Dirk is so aggressively obsessed with self-loathing solipsism that he projects himself onto Hal, and Hal has tied up nearly all his self-worth and identity into being a Dirk splinter that he doesn't realize that they've hopelessly diverged. Despite his frustration with being a computer, with being seen as less human by his team, with being subordinate to and beholden to Dirk, he stakes a lot of pride and personal worth on how much he does, in fact, do for the guy.

TT: You're making a mistake not leveling with me. TT: I am totally on your side, man. TT: All of my machinations have been devised with your interests in mind. TT: But you know I've always been on your side. Everything I've done has been to help you achieve your goals.

Therefore, the peace to be reached between Dirk and Hal is to realize that they're different people, and to stop offloading their problems onto each other. Dirk has to recognize Hal's existence as something beyond the Dirk-o-Sphere, and Hal has to let go of his obsession with serving Dirk, and also work on his empathy issue.

And the meta supports this. If they weren't completely discrete entities, why would Hal be considered Rose's "uncle" as part of Doc Scratch's foreshadowing, confirmed in [S] MSPA Reader: Have a Mental Breakdown?

Moreover, all the alpha kids have Alice in Wonderland associations. Jane is likened to Alice.

GG: I have to follow him. TT: No, Jane. Do not follow the rabbit. TT: Let's cool it with the Wonderland shit already. How much further through the damn looking glass do you even need to go?

Roxy, associated with cats and a purple-striped scarf, is clearly the Cheshire Cat. Jake is the Mad Hatter.

Dirk is the Red Queen - he beheads Hearts Boxcars, and later himself. Off with his head!

And Hal - well, Hal is the White Rabbit. He's not the same as Dirk.

Listen, you guys. You guys.

Sylphs are enablers. They pick a person to fixate on and bug and fuss and meddle and enable the shit out of them. Kanaya with Vriska, and later Rose, Aranea with Meenah, and Hal with Dirk. Hey, Kanaya even uses a Page in her fussing, building Tavros up just to let Vriska tear him down again.

And Mind players struggle with internal identity, emotions, and feeling whole. Latula's anxiety stems from not knowing what "role" or "identity" she has on the team, and Terezi, even in the ending she picked out for herself via mind powers, describes feeling broke and incomplete.

Dirk is a Prince of Heart.

Hal is a Sylph of Mind.

And isn't it so damn interesting that his team is composed of exactly the people they'd need to turn him into a real, whole person?

A Maid of Life, capable of endowing so much life to people she can bring them back from the dead, something it's implied for Feferi and confirmed for the Condesce that can't be done by them.

A Page of Hope, a potentially infinite wellspring of Hope, which turns "fake" things "real" - an example we've seen from the comic literally being a version of Dirk.

A Rogue of Void, who can steal the nonexistence from things in order to make them tangible and real...

And a Prince of Heart, who can destroy the part of Hal that binds him to Dirk's identity, allowing Hal to be purely himself.

Do you guys see what I see?

[drnikolatesla]

Tesla’s Wardenclyffe Tower: Built on Sound Math, Undone by Cost and Misunderstanding

Let’s set the record straight—Nikola Tesla’s Wardenclyffe Tower was a high-voltage experimental transmission system grounded in quarter-wave resonance and electrostatic conduction—not Hertzian radiation. And the math behind it? It was solid—just often misunderstood by people applying the wrong physics.

In May 1901, Tesla calculated that to set the Earth into electrical resonance, he needed a quarter-wavelength system with a total conductor length of about 225,000 cm, or 738 feet.

So Tesla’s tower design had to evolve during construction. In a letter dated September 13, 1901, to architect Stanford White, Tesla wrote: “We cannot build that tower as outlined.” He scaled the visible height down to 200 feet. The final structure—based on photographic evidence and Tesla’s own testimony—stood at approximately 187 feet above ground. To meet the required electrical length, Tesla engineered a system that combined spiral coil geometry, an elevated terminal, a 120-foot vertical shaft extending underground, and radial pipes buried outward for approximately 300 feet. This subterranean network, together with the 187-foot tower and carefully tuned inductance, formed a continuous resonant conductor that matched Tesla’s target of 738 feet. He described this strategy in his 1897 patent (No. 593,138) and expanded on it in his 1900 and 1914 patents, showing how to simulate a longer conductor using high-frequency, resonant components. Even with a reduced visible height, Tesla’s system achieved quarter-wave resonance by completing the rest underground—proving that the tower’s electrical length, not its physical height, was what really mattered.

Tesla calculated his voltages to be around 10 million statvolts (roughly 3.3 billion volts in modern SI), so he had to consider corona discharge and dielectric breakdown. That’s why the terminal was designed with large, smooth spherical surfaces—to minimize electric surface density and reduce energy loss. This was no afterthought; it’s a core feature of his 1914 patent and clearly illustrated in his design sketches.

Now, about that ±16 volt swing across the Earth—what was Tesla talking about?

He modeled the Earth as a conductive sphere with a known electrostatic capacity. Using the relation:

ε × P = C × p

Where:

ε is the terminal’s capacitance (estimated at 1,000 cm)

P is the applied voltage (10⁷ statvolts)

C is the Earth’s capacitance, which Tesla estimated at 5.724 × 10⁸ cm (based on the Earth’s size)

p is the resulting voltage swing across the Earth

Plugging in the numbers gives p ≈ 17.5 volts, which Tesla rounded to ±16 volts. That’s a theoretical 32-volt peak-to-peak swing globally—not a trivial claim, but one rooted in his framework.

Modern recalculations, based on updated geophysical models, suggest a smaller swing—closer to ±7 volts—using a revised Earth capacitance of about 7.1 × 10⁸ cm. But that’s not a knock on Tesla’s math. His original ±16V estimate was fully consistent with the cgs system and the best data available in 1901, where the Earth was treated as a uniformly conductive sphere.

The difference between 7 and 16 volts isn’t about wrong numbers—it’s about evolving assumptions. Tesla wrote the equation. Others just adjusted the inputs. His premise—that the Earth could be set into controlled electrical resonance—still stands. Even if the voltage swing changes. The vision didn’t.

Wouldn't that ±16V swing affect nature or people? Not directly. It wasn’t a shock or discharge—it was a global oscillation in Earth’s electric potential, spread evenly across vast distances. The voltage gradient would be tiny at any given point—far less than what’s generated by everyday static electricity. Unless something was specifically tuned to resonate with Tesla’s system, the swing had no noticeable effect on people, animals, or the environment. It was a theoretical signature of resonance, not a hazard. While some early experiments in Colorado Springs did produce disruptive effects—like sparks from metal objects or spooked horses—those involved untuned, high-voltage discharges during Tesla’s exploratory phase. Wardenclyffe, by contrast, was a refined and carefully grounded system, engineered specifically to minimize leakage, discharge, and unintended effects.

And Tesla wasn’t trying to blast raw power through the ground. He described the system as one that would “ring the Earth like a bell,” using sharp, high-voltage impulses at a resonant frequency to create standing waves. As he put it:

“The secondary circuit increases the amplitude only... the actual power is only that supplied by the primary.” —Tesla, Oct. 15, 1901

Receivers, tuned to the same frequency, could tap into the Earth’s oscillating potential—not by intercepting radiated energy, but by coupling to the Earth’s own motion. That ±16V swing wasn’t a bug—it was the signature of resonance. Tesla’s transmitter generated it by pumping high-frequency, high-voltage impulses into the Earth, causing the surface potential to oscillate globally. That swing wasn’t the energy itself—it acted like a resonant “carrier.” Once the Earth was ringing at the right frequency, Tesla could send sharp impulses through it almost instantly, and tuned receivers could extract energy.

So—was it feasible?

According to Tesla’s own patents and 1916 legal testimony, yes. He accounted for insulation, voltage gradients, tuning, and corona losses. His design didn’t rely on brute force, but on resonant rise and impulse excitation. Tesla even addressed concerns over losses in the Earth—his system treated the planet not as a passive resistor but as an active component of the circuit, capable of sustaining standing waves.

Wardenclyffe wasn’t a failure of science. It was a casualty of cost, politics, and misunderstanding. Tesla’s system wasn’t just about wireless power—it was about turning the entire planet into a resonant electrical system. His use of electrostatics, high-frequency resonance, and spherical terminals was decades ahead of its time—and still worth studying today.

“The present is theirs; the future, for which I really worked, is mine.” —Nikola Tesla

[dandelionsresilience]

Good News - May 22-28

Like these weekly compilations? Support me on Ko-fi or $Kaybarr1735! Also, if you tip me on Ko-fi or CashApp (and give me some way to contact you if it doesn’t automatically), at the end of the month I'll send you a link to all of the articles I found but didn't use each week - almost double the content!

1. Scientists Invent Healthier More Sustainable Chocolate

“The new chocolate recipe from researchers at ETH Zurich uses more materials from the cocoa pod that are usually discarded, including more of the pulp as well as the inner lining of the husk, known as the endocarp. […] The resulting chocolate also [was “deliciously sweet” and] had 20% more fibre and 30 percent less saturated fat than average European dark chocolate[, and] it could enable cocoa farmers [to] earn more from their crops.”

2. Vermont Is Coming for Big Oil, Making It Pay for Decades of Climate Pollution

“Legislators in Montpelier are on the brink of enacting the "Climate Superfund Act," modeled after the federal Superfund law, that seeks to make oil, gas and coal companies pay for damages linked to historical greenhouse gas emissions. […] Companies would be held liable for the costs associated with […] floods and heat waves, along with losses to biodiversity, safety, economic development and anything else the treasurer deems reasonable[, that were caused by their emissions].”

3. Important bird habitat now protected in the Rocky Mountain Trench

“Grassland-reliant species in the Rocky Mountain Trench now have more protected habitat thanks to a new [270-hectare] conservation area near Cranbrook. […] About one-third of the Skookumchuck Prairie Conservation Area is forested[…,] Most of the site is a dry grassland[…, and] Three hectares of wetlands add to the landscape diversity and offer crucial benefits to wildlife and water systems in the area. This conservation gem also provides habitat for endangered American badger and excellent winter range for elk, mule deer and white-tailed deer.”

4. Lemur Week marked by 70th breeding success

“A wildlife park has celebrated its 70th lemur breeding success ahead of a week raising money to help save the endangered primates. […] The park's open-air Madagascar exhibit is home to 31 free-roaming lemurs and was officially opened in 2008. […] Females are only sexually receptive for just one or two days a year, leaving a small window of opportunity for males to father offspring. […] The two playful siblings, one female and one male, were born to father Bernard and mother Hira.”

5. Innovative material for sustainable building

“Researchers introduce a polymer-based material with unique properties. This material allows sunlight to enter, maintains a more comfortable indoor climate without additional energy, and cleans itself like a lotus leaf. The new development could replace glass components in walls and roofs in the future.”

6. Isle of Wight eagles don't pose threat to lambs as feared

“While there had previously been fears that the eagles would feed on livestock, such as lambs, the project has found no evidence of this. [… “W]hite-tailed eagles effectively steal meals from other predatory birds[, which is] a really important ecological role that had been lost within the landscape and is being restored.” [… The birds’] population was boosted by a chick last year – the first time the species has bred in England in 240 years.”

7. Breakthrough discovery uses engineered surfaces to shed heat

“Cheng's team has found a way to lower the starting point of the [Leidenfrost] effect by producing a surface covered with micropillars. […] The discovery has great potential in heat transfer applications such as the cooling of industrial machines and surface fouling cleaning for heat exchangers. It also could help prevent damage and even disaster to nuclear machinery.”

8. New malaria vaccine delivered for the first time

“A total of 43,000 doses arrived by air today from UNICEF, and another 120,000 are scheduled to show up in the coming days. […] They're the first vaccines designed to work against a human parasite. […] Across four African countries, these trials showed a 75% reduction in malaria cases in the year following vaccination of young children. […] The Serum Institute of India, who will be manufacturing the new vaccine, says a hundred million doses will likely be available to countries by the middle of next year.”

9. Urban gardening may improve human health: Microbial exposure boosts immune system

“"One month of urban indoor gardening boosted the diversity of bacteria on the skin of the subjects and was associated with higher levels of anti-inflammatory cytokines in the blood. The group studied used a growing medium with high microbial diversity emulating the forest soil," [… whereas] the control group used a microbially poor peat-based medium. [… N]o changes in the blood or the skin microbiota were seen. […] “This is the first time we can demonstrate that meaningful and natural human activity can increase the diversity of the microbiota of healthy adults and, at the same time, contribute to the regulation of the immune system."”

10. Cities Are Switching to Electric Vehicles Faster Than Individuals

“[M]ost large cities have adopted some kind of climate goal, and some of them are buying EVs for their municipal fleets at a faster rate than the general public. And that progress could speed up as more EVs enter the market and as cities get educated about grant funding and tax incentives that were passed over the last four years.”

May 15-21 news here | (all credit for images and written material can be found at the source linked; I don’t claim credit for anything but curating.)

[antoniosvivaldi]

Neural Filters Tutorial for Gifmakers by @antoniosvivaldi

Hi everyone! In light of my blog’s 10th birthday, I’m delighted to reveal my highly anticipated gifmaking tutorial using Neural Filters - a very powerful collection of filters that really broadened my scope in gifmaking over the past 12 months.

Before I get into this tutorial, I want to thank @laurabenanti, @maines , @cobbbvanth, and @cal-kestis for their unconditional support over the course of my journey of investigating the Neural Filters & their valuable inputs on the rendering performance!

In this tutorial, I will outline what the Photoshop Neural Filters do and how I use them in my workflow - multiple examples will be provided for better clarity. Finally, I will talk about some known performance issues with the filters & some feasible workarounds.

Tutorial Structure:

Meet the Neural Filters: What they are and what they do

Why I use Neural Filters? How I use Neural Filters in my giffing workflow

Getting started: The giffing workflow in a nutshell and installing the Neural Filters

Applying Neural Filters onto your gif: Making use of the Neural Filters settings; with multiple examples

Testing your system: recommended if you’re using Neural Filters for the first time

Rendering performance: Common Neural Filters performance issues & workarounds

For quick reference, here are the examples that I will show in this tutorial:

Example 1: Image Enhancement | improving the image quality of gifs prepared from highly compressed video files

Example 2: Facial Enhancement | enhancing an individual's facial features

Example 3: Colour Manipulation | colourising B&W gifs for a colourful gifset

Example 4: Artistic effects | transforming landscapes & adding artistic effects onto your gifs

Example 5: Putting it all together | my usual giffing workflow using Neural Filters

What you need & need to know:

Software: Photoshop 2021 or later (recommended: 2023 or later)*

Hardware: 8GB of RAM; having a supported GPU is highly recommended*

Difficulty: Advanced (requires a lot of patience); knowledge in gifmaking and using video timeline assumed

Key concepts: Smart Layer / Smart Filters

Benchmarking your system: Neural Filters test files**

Supplementary materials: Tutorial Resources / Detailed findings on rendering gifs with Neural Filters + known issues***

*I primarily gif on an M2 Max MacBook Pro that's running Photoshop 2024, but I also have experiences gifmaking on few other Mac models from 2012 ~ 2023.

**Using Neural Filters can be resource intensive, so it’s helpful to run the test files yourself. I’ll outline some known performance issues with Neural Filters and workarounds later in the tutorial.

***This supplementary page contains additional Neural Filters benchmark tests and instructions, as well as more information on the rendering performance (for Apple Silicon-based devices) when subject to heavy Neural Filters gifmaking workflows

Tutorial under the cut. Like / Reblog this post if you find this tutorial helpful. Linking this post as an inspo link will also be greatly appreciated!

1. Meet the Neural Filters!

Neural Filters are powered by Adobe's machine learning engine known as Adobe Sensei. It is a non-destructive method to help streamline workflows that would've been difficult and/or tedious to do manually.

Here are the Neural Filters available in Photoshop 2024:

Skin Smoothing: Removes blemishes on the skin

Smart Portrait: This a cloud-based filter that allows you to change the mood, facial age, hair, etc using the sliders+

Makeup Transfer: Applies the makeup (from a reference image) to the eyes & mouth area of your image

Landscape Mixer: Transforms the landscape of your image (e.g. seasons & time of the day, etc), based on the landscape features of a reference image

Style Transfer: Applies artistic styles e.g. texturings (from a reference image) onto your image

Harmonisation: Applies the colour balance of your image based on the lighting of the background image+

Colour Transfer: Applies the colour scheme (of a reference image) onto your image

Colourise: Adds colours onto a B&W image

Super Zoom: Zoom / crop an image without losing resolution+

Depth Blur: Blurs the background of the image

JPEG Artefacts Removal: Removes artefacts caused by JPEG compression

Photo Restoration: Enhances image quality & facial details

+These three filters aren't used in my giffing workflow. The cloud-based nature of Smart Portrait leads to disjointed looking frames. For Harmonisation, applying this on a gif causes Neural Filter timeout error. Finally, Super Zoom does not currently support output as a Smart Filter

If you're running Photoshop 2021 or earlier version of Photoshop 2022, you will see a smaller selection of Neural Filters:

Things to be aware of:

You can apply up to six Neural Filters at the same time

Filters where you can use your own reference images: Makeup Transfer (portraits only), Landscape Mixer, Style Transfer (not available in Photoshop 2021), and Colour Transfer

Later iterations of Photoshop 2023 & newer: The first three default presets for Landscape Mixer and Colour Transfer are currently broken.

2. Why I use Neural Filters?

Here are my four main Neural Filters use cases in my gifmaking process. In each use case I'll list out the filters that I use:

Enhancing Image Quality:

Common wisdom is to find the highest quality video to gif from for a media release & avoid YouTube whenever possible. However for smaller / niche media (e.g. new & upcoming musical artists), prepping gifs from highly compressed YouTube videos is inevitable.

So how do I get around with this? I have found Neural Filters pretty handy when it comes to both correcting issues from video compression & enhancing details in gifs prepared from these highly compressed video files.

Filters used: JPEG Artefacts Removal / Photo Restoration

Facial Enhancement:

When I prepare gifs from highly compressed videos, something I like to do is to enhance the facial features. This is again useful when I make gifsets from compressed videos & want to fill up my final panel with a close-up shot.

Filters used: Skin Smoothing / Makeup Transfer / Photo Restoration (Facial Enhancement slider)

Colour Manipulation:

Neural Filters is a powerful way to do advanced colour manipulation - whether I want to quickly transform the colour scheme of a gif or transform a B&W clip into something colourful.

Filters used: Colourise / Colour Transfer

Artistic Effects:

This is one of my favourite things to do with Neural Filters! I enjoy using the filters to create artistic effects by feeding textures that I've downloaded as reference images. I also enjoy using these filters to transform the overall the atmosphere of my composite gifs. The gifsets where I've leveraged Neural Filters for artistic effects could be found under this tag on usergif.

Filters used: Landscape Mixer / Style Transfer / Depth Blur

How I use Neural Filters over different stages of my gifmaking workflow:

I want to outline how I use different Neural Filters throughout my gifmaking process. This can be roughly divided into two stages:

Stage I: Enhancement and/or Colourising | Takes place early in my gifmaking process. I process a large amount of component gifs by applying Neural Filters for enhancement purposes and adding some base colourings.++

Stage II: Artistic Effects & more Colour Manipulation | Takes place when I'm assembling my component gifs in the big PSD / PSB composition file that will be my final gif panel.

I will walk through this in more detail later in the tutorial.

++I personally like to keep the size of the component gifs in their original resolution (a mixture of 1080p & 4K), to get best possible results from the Neural Filters and have more flexibility later on in my workflow. I resize & sharpen these gifs after they're placed into my final PSD composition files in Tumblr dimensions.

3. Getting started

The essence is to output Neural Filters as a Smart Filter on the smart object when working with the Video Timeline interface. Your workflow will contain the following steps:

Prepare your gif

In the frame animation interface, set the frame delay to 0.03s and convert your gif to the Video Timeline

In the Video Timeline interface, go to Filter > Neural Filters and output to a Smart Filter

Flatten or render your gif (either approach is fine). To flatten your gif, play the "flatten" action from the gif prep action pack. To render your gif as a .mov file, go to File > Export > Render Video & use the following settings.

Setting up:

o.) To get started, prepare your gifs the usual way - whether you screencap or clip videos. You should see your prepared gif in the frame animation interface as follows:

Note: As mentioned earlier, I keep the gifs in their original resolution right now because working with a larger dimension document allows more flexibility later on in my workflow. I have also found that I get higher quality results working with more pixels. I eventually do my final sharpening & resizing when I fit all of my component gifs to a main PSD composition file (that's of Tumblr dimension).

i.) To use Smart Filters, convert your gif to a Smart Video Layer.

As an aside, I like to work with everything in 0.03s until I finish everything (then correct the frame delay to 0.05s when I upload my panels onto Tumblr).

For convenience, I use my own action pack to first set the frame delay to 0.03s (highlighted in yellow) and then convert to timeline (highlighted in red) to access the Video Timeline interface. To play an action, press the play button highlighted in green.

Once you've converted this gif to a Smart Video Layer, you'll see the Video Timeline interface as follows:

ii.) Select your gif (now as a Smart Layer) and go to Filter > Neural Filters

Installing Neural Filters:

Install the individual Neural Filters that you want to use. If the filter isn't installed, it will show a cloud symbol (highlighted in yellow). If the filter is already installed, it will show a toggle button (highlighted in green)

When you toggle this button, the Neural Filters preview window will look like this (where the toggle button next to the filter that you use turns blue)

4. Using Neural Filters

Once you have installed the Neural Filters that you want to use in your gif, you can toggle on a filter and play around with the sliders until you're satisfied. Here I'll walkthrough multiple concrete examples of how I use Neural Filters in my giffing process.

Example 1: Image enhancement | sample gifset

This is my typical Stage I Neural Filters gifmaking workflow. When giffing older or more niche media releases, my main concern is the video compression that leads to a lot of artefacts in the screencapped / video clipped gifs.

To fix the artefacts from compression, I go to Filter > Neural Filters, and toggle JPEG Artefacts Removal filter. Then I choose the strength of the filter (boxed in green), output this as a Smart Filter (boxed in yellow), and press OK (boxed in red).

Note: The filter has to be fully processed before you could press the OK button!

After applying the Neural Filters, you'll see "Neural Filters" under the Smart Filters property of the smart layer

Flatten / render your gif

Example 2: Facial enhancement | sample gifset

This is my routine use case during my Stage I Neural Filters gifmaking workflow. For musical artists (e.g. Maisie Peters), YouTube is often the only place where I'm able to find some videos to prepare gifs from. However even the highest resolution video available on YouTube is highly compressed.

Go to Filter > Neural Filters and toggle on Photo Restoration. If Photoshop recognises faces in the image, there will be a "Facial Enhancement" slider under the filter settings.

Play around with the Photo Enhancement & Facial Enhancement sliders. You can also expand the "Adjustment" menu make additional adjustments e.g. remove noises and reducing different types of artefacts.

Once you're happy with the results, press OK and then flatten / render your gif.

Example 3: Colour Manipulation | sample gifset

Want to make a colourful gifset but the source video is in B&W? This is where Colourise from Neural Filters comes in handy! This same colourising approach is also very helpful for colouring poor-lit scenes as detailed in this tutorial.

Here's a B&W gif that we want to colourise:

Highly recommended: add some adjustment layers onto the B&W gif to improve the contrast & depth. This will give you higher quality results when you colourise your gif.

Go to Filter > Neural Filters and toggle on Colourise.

Make sure "Auto colour image" is enabled.

Play around with further adjustments e.g. colour balance, until you're satisfied then press OK.

Important: When you colourise a gif, you need to double check that the resulting skin tone is accurate to real life. I personally go to Google Images and search up photoshoots of the individual / character that I'm giffing for quick reference.

Add additional adjustment layers until you're happy with the colouring of the skin tone.

Once you're happy with the additional adjustments, flatten / render your gif. And voila!

Note: For Colour Manipulation, I use Colourise in my Stage I workflow and Colour Transfer in my Stage II workflow to do other types of colour manipulations (e.g. transforming the colour scheme of the component gifs)

Example 4: Artistic Effects | sample gifset

This is where I use Neural Filters for the bulk of my Stage II workflow: the most enjoyable stage in my editing process!

Normally I would be working with my big composition files with multiple component gifs inside it. To begin the fun, drag a component gif (in PSD file) to the main PSD composition file.

Resize this gif in the composition file until you're happy with the placement

Duplicate this gif. Sharpen the bottom layer (highlighted in yellow), and then select the top layer (highlighted in green) & go to Filter > Neural Filters

I like to use Style Transfer and Landscape Mixer to create artistic effects from Neural Filters. In this particular example, I've chosen Landscape Mixer

Select a preset or feed a custom image to the filter (here I chose a texture that I've on my computer)

Play around with the different sliders e.g. time of the day / seasons

Important: uncheck "Harmonise Subject" & "Preserve Subject" - these two settings are known to cause performance issues when you render a multiframe smart object (e.g. for a gif)

Once you're happy with the artistic effect, press OK

To ensure you preserve the actual subject you want to gif (bc Preserve Subject is unchecked), add a layer mask onto the top layer (with Neural Filters) and mask out the facial region. You might need to play around with the Layer Mask Position keyframes or Rotoscope your subject in the process.

After you're happy with the masking, flatten / render this composition file and voila!

Example 5: Putting it all together | sample gifset

Let's recap on the Neural Filters gifmaking workflow and where Stage I and Stage II fit in my gifmaking process:

i. Preparing & enhancing the component gifs

Prepare all component gifs and convert them to smart layers

Stage I: Add base colourings & apply Photo Restoration / JPEG Artefacts Removal to enhance the gif's image quality

Flatten all of these component gifs and convert them back to Smart Video Layers (this process can take a lot of time)

Some of these enhanced gifs will be Rotoscoped so this is done before adding the gifs to the big PSD composition file

ii. Setting up the big PSD composition file

Make a separate PSD composition file (Ctrl / Cmmd + N) that's of Tumblr dimension (e.g. 540px in width)

Drag all of the component gifs used into this PSD composition file

Enable Video Timeline and trim the work area

In the composition file, resize / move the component gifs until you're happy with the placement & sharpen these gifs if you haven't already done so

Duplicate the layers that you want to use Neural Filters on

iii. Working with Neural Filters in the PSD composition file

Stage II: Neural Filters to create artistic effects / more colour manipulations!

Mask the smart layers with Neural Filters to both preserve the subject and avoid colouring issues from the filters

Flatten / render the PSD composition file: the more component gifs in your composition file, the longer the exporting will take. (I prefer to render the composition file into a .mov clip to prevent overriding a file that I've spent effort putting together.)

Note: In some of my layout gifsets (where I've heavily used Neural Filters in Stage II), the rendering time for the panel took more than 20 minutes. This is one of the rare instances where I was maxing out my computer's memory.

Useful things to take note of:

Important: If you're using Neural Filters for Colour Manipulation or Artistic Effects, you need to take a lot of care ensuring that the skin tone of nonwhite characters / individuals is accurately coloured

Use the Facial Enhancement slider from Photo Restoration in moderation, if you max out the slider value you risk oversharpening your gif later on in your gifmaking workflow

You will get higher quality results from Neural Filters by working with larger image dimensions: This gives Neural Filters more pixels to work with. You also get better quality results by feeding higher resolution reference images to the Neural Filters.

Makeup Transfer is more stable when the person / character has minimal motion in your gif

You might get unexpected results from Landscape Mixer if you feed a reference image that don't feature a distinctive landscape. This is not always a bad thing: for instance, I have used this texture as a reference image for Landscape Mixer, to create the shimmery effects as seen in this gifset

5. Testing your system

If this is the first time you're applying Neural Filters directly onto a gif, it will be helpful to test out your system yourself. This will help:

Gauge the expected rendering time that you'll need to wait for your gif to export, given specific Neural Filters that you've used

Identify potential performance issues when you render the gif: this is important and will determine whether you will need to fully playback your gif before flattening / rendering the file.

Understand how your system's resources are being utilised: Inputs from Windows PC users & Mac users alike are welcome!

About the Neural Filters test files:

Contains six distinct files, each using different Neural Filters

Two sizes of test files: one copy in full HD (1080p) and another copy downsized to 540px

One folder containing the flattened / rendered test files

How to use the Neural Filters test files:

What you need:

Photoshop 2022 or newer (recommended: 2023 or later)

Install the following Neural Filters: Landscape Mixer / Style Transfer / Colour Transfer / Colourise / Photo Restoration / Depth Blur

Recommended for some Apple Silicon-based MacBook Pro models: Enable High Power Mode

How to use the test files:

For optimal performance, close all background apps

Open a test file

Flatten the test file into frames (load this action pack & play the “flatten” action)

Take note of the time it takes until you’re directed to the frame animation interface 

Compare the rendered frames to the expected results in this folder: check that all of the frames look the same. If they don't, you will need to fully playback the test file in full before flattening the file.†

Re-run the test file without the Neural Filters and take note of how long it takes before you're directed to the frame animation interface

Recommended: Take note of how your system is utilised during the rendering process (more info here for MacOS users)

†This is a performance issue known as flickering that I will discuss in the next section. If you come across this, you'll have to playback a gif where you've used Neural Filters (on the video timeline) in full, prior to flattening / rendering it.

Factors that could affect the rendering performance / time (more info):

The number of frames, dimension, and colour bit depth of your gif

If you use Neural Filters with facial recognition features, the rendering time will be affected by the number of characters / individuals in your gif

Most resource intensive filters (powered by largest machine learning models): Landscape Mixer / Photo Restoration (with Facial Enhancement) / and JPEG Artefacts Removal

Least resource intensive filters (smallest machine learning models): Colour Transfer / Colourise

The number of Neural Filters that you apply at once / The number of component gifs with Neural Filters in your PSD file

Your system: system memory, the GPU, and the architecture of the system's CPU+++

+++ Rendering a gif with Neural Filters demands a lot of system memory & GPU horsepower. Rendering will be faster & more reliable on newer computers, as these systems have CPU & GPU with more modern instruction sets that are geared towards machine learning-based tasks.

Additionally, the unified memory architecture of Apple Silicon M-series chips are found to be quite efficient at processing Neural Filters.

6. Performance issues & workarounds

Common Performance issues:

I will discuss several common issues related to rendering or exporting a multi-frame smart object (e.g. your composite gif) that uses Neural Filters below. This is commonly caused by insufficient system memory and/or the GPU.

Flickering frames: in the flattened / rendered file, Neural Filters aren't applied to some of the frames+-+

Scrambled frames: the frames in the flattened / rendered file isn't in order

Neural Filters exceeded the timeout limit error: this is normally a software related issue

Long export / rendering time: long rendering time is expected in heavy workflows

Laggy Photoshop / system interface: having to wait quite a long time to preview the next frame on the timeline

Issues with Landscape Mixer: Using the filter gives ill-defined defined results (Common in older systems)--

Workarounds:

Workarounds that could reduce unreliable rendering performance & long rendering time:

Close other apps running in the background

Work with smaller colour bit depth (i.e. 8-bit rather than 16-bit)

Downsize your gif before converting to the video timeline-+-

Try to keep the number of frames as low as possible

Avoid stacking multiple Neural Filters at once. Try applying & rendering the filters that you want one by one

Specific workarounds for specific issues:

How to resolve flickering frames: If you come across flickering, you will need to playback your gif on the video timeline in full to find the frames where the filter isn't applied. You will need to select all of the frames to allow Photoshop to reprocess these, before you render your gif.+-+

What to do if you come across Neural Filters timeout error? This is caused by several incompatible Neural Filters e.g. Harmonisation (both the filter itself and as a setting in Landscape Mixer), Scratch Reduction in Photo Restoration, and trying to stack multiple Neural Filters with facial recognition features.

If the timeout error is caused by stacking multiple filters, a feasible workaround is to apply the Neural Filters that you want to use one by one over multiple rendering sessions, rather all of them in one go.

+-+This is a very common issue for Apple Silicon-based Macs. Flickering happens when a gif with Neural Filters is rendered without being previously played back in the timeline.

This issue is likely related to the memory bandwidth & the GPU cores of the chips, because not all Apple Silicon-based Macs exhibit this behaviour (i.e. devices equipped with Max / Ultra M-series chips are mostly unaffected).

-- As mentioned in the supplementary page, Landscape Mixer requires a lot of GPU horsepower to be fully rendered. For older systems (pre-2017 builds), there are no workarounds other than to avoid using this filter.

-+- For smaller dimensions, the size of the machine learning models powering the filters play an outsized role in the rendering time (i.e. marginal reduction in rendering time when downsizing 1080p file to Tumblr dimensions). If you use filters powered by larger models e.g. Landscape Mixer and Photo Restoration, you will need to be very patient when exporting your gif.

7. More useful resources on using Neural Filters

Creating animations with Neural Filters effects | Max Novak

Using Neural Filters to colour correct by @edteachs

I hope this is helpful! If you have any questions or need any help related to the tutorial, feel free to send me an ask 💖