#computer hardware engineer courses

[thegeorgetelegraph]

Get The Best Computer Hardware and Networking Course in Kolkata 2023

Are you looking for the best computer hardware networking course from the top computer hardware and networking training institute in Kolkata? then visit George Telegraph today and know your course fees details. Visit Now.

[jcmarchi]

For developing designers, there’s magic in 2.737 (Mechatronics)

New Post has been published on https://thedigitalinsider.com/for-developing-designers-theres-magic-in-2-737-mechatronics/

For developing designers, there’s magic in 2.737 (Mechatronics)

The field of mechatronics is multidisciplinary and interdisciplinary, occupying the intersection of mechanical systems, electronics, controls, and computer science. Mechatronics engineers work in a variety of industries — from space exploration to semiconductor manufacturing to product design — and specialize in the integrated design and development of intelligent systems. For students wanting to learn mechatronics, it might come as a surprise that one of the most powerful teaching tools available for the subject matter is simply a pen and a piece of paper.

“Students have to be able to work out things on a piece of paper, and make sketches, and write down key calculations in order to be creative,” says MIT professor of mechanical engineering David Trumper, who has been teaching class 2.737 (Mechatronics) since he joined the Institute faculty in the early 1990s. The subject is electrical and mechanical engineering combined, he says, but more than anything else, it’s design.

“If you just do electronics, but have no idea how to make the mechanical parts work, you can’t find really creative solutions. You have to see ways to solve problems across different domains,” says Trumper. “MIT students tend to have seen lots of math and lots of theory. The hands-on part is really critical to build that skill set; with hands-on experiences they’ll be more able to imagine how other things might work when they’re designing them.”

Play video

A lot like magic Video: Department of Mechanical Engineering

Audrey Cui ’24, now a graduate student in electrical engineering and computer science, confirms that Trumper “really emphasizes being able to do back-of-the-napkin calculations.” This simplicity is by design, and the critical thinking it promotes is essential for budding designers.

“Sitting behind a computer terminal, you’re using some existing tool in the menu system and not thinking creatively,” says Trumper. “To see the trade-offs, and get the clutter out of your thinking, it helps to work with a really simple tool — a piece of paper and, hopefully, multicolored pens to code things — you can design so much more creatively than if you’re stuck behind a screen. The ability to sketch things is so important.”

Trumper studies precision mechatronics, broadly, with a particular interest in mechatronic systems for demanding resolutions. Examples include projects that employ magnetic levitation, linear motors for driving precision manufacturing for semiconductors, and spacecraft attitude control. His work also explores lathes, milling applications, and even bioengineering platforms.

Class 2.737, which is offered every two years, is lab-based. Sketches and concepts come to life in focused experiences designed to expose students to key principles in a hands-on way and are very much informed by what Trumper has found important in his research. The two-week-long lab explorations range from controlling a motor to evaluating electronic scales to vibration isolations systems built on a speaker. One year, students constructed a working atomic force microscope.

“The touch and sense of how things actually work is really important,” Trumper says. “As a designer, you have to be able to imagine. If you think of some new configuration of a motor, you need to imagine how it would work and see it working, so you can do design iterations in your imagined space — to make that real requires that you’ve had experience with the actual thing.”

He says his former late colleague, Woodie Flowers SM ’68, MEng ’71, PhD ’73, used to call it “running the movie.” Trumper explains, “once you have the image in your mind, you can more easily picture what’s going on with the problem — what’s getting hot, where’s the stress, what do I like and not like about this design. If you can do that with a piece of paper and your imagination, now you design new things pretty creatively.”

Flowers had been the Pappalardo Professor Emeritus of Mechanical Engineering at the time of his passing in October 2019. He is remembered for pioneering approaches to education, and was instrumental in shaping MIT’s hands-on approach to engineering design education.

Class 2.737 tends to attract students who like to design and build their own things. “I want people who are heading toward being hardware geeks,” says Trumper, laughing. “And I mean that lovingly.” He says his most important objective for this class is that students learn real tools that they will find useful years from now in their own engineering research or practice.

“Being able to see how multiple pieces fit in together and create one whole working system is just really empowering to me as an aspiring engineer,” says Cui.

For fellow 2.737 student Zach Francis, the course offered foundations for the future along with a meaningful tie to the past. “This class reminded me about what I enjoy about engineering. You look at it when you’re a young kid and you’re like ‘that looks like magic!’ and then as an adult you can now make that. It’s the closest thing I’ve been to a wizard, and I like that a lot.” 

[cosmicpuzzle]

Which Education🎓📚 is right for you?

Mercury rules your interest and consequently which type of course you would select.

Now you have to see how Mercury is placed. For example if Mercury is conjunct Moon it would have same effect as Mercury in Cancer or Moon opposite Mercury.

Mercury -Sun: It is called Budh Aditya yoga. These people can shine in political science, geology, sociology, medicine and they can be good leaders too. They may prepare for competitive exams.

Mercury-Moon: Some changes or confusion in choice of course. Can study more than one subject but both vastly different from each other. Chemical, hotel management, nutrition, chef, psychology, tarot and intuitive studies.

Mercury-Mars: Some obstacles in education, breaks and interruptions (dropping classes), engineering (especially related to machines, drawings, plans, civil, electronics), medicine (especially related to surgery), fire and safety engineering,

Mercury-Venus: Sales, marketing, HR, interior designing, makeup courses, all type of fine arts, vocational courses, acting courses.

Mercury-Saturn: Engineering (like construction , petroleum, mining core subjects), structural engineering, drafting, administrative studies.

Mercury-Jupiter: Finance, CPA, CMA, accounting, teaching, law field, journalism, VJ, pilots, aeronautical.

Mercury- Rahu: Chemical, nuclear subjects, cinematography, software courses, digital marketing, share markets, computer hardware, import export, AI, Machine Learning courses.

Mercury-Ketu: Computer coding, electrical engineering, bio technology, astrology, virology, research oriented fields.

For Readings DM

[ikiprian]

Barbara Gordon's Coding & Computer Cram School is a popular YouTube series. Tucker Foley is a star student.

Barbara Gordon's Cram School posts free online courses for both coding and computer engineering. Think Crash Course in terms of entertainment, but college lecture in terms of depth. Hundreds of thousands of viewers flock to it— students who missed a class, people looking to add new skills to a resume, even simple hobbyists. It’s a project Barbara’s proud of.

Sometimes, when she wants to relax, she’ll even hop in the comments and spend an afternoon troubleshooting a viewer’s project with them.

User “Fryer-Tuck” has especially interesting ones. Barbara finds herself seeking out his comments, checking in on whatever this crazy kid is making next. An app for collecting GPS pings and assembling them on a map in real-time, an algorithm that connects geographic points to predict something’s movement taking a hundred other variables into account, simplified versions of incredibly complex homemade programs so they can run on incredibly limited CPU’s.

(Barbara wants to buy the kid a PC. It seems he’s got natural talent, but he keeps making reference to a PDA. Talk about 90’s! This guy’s hardware probably predates his birth.)

She chats with him more and more, switching to less public PM threads, and eventually, he opens up. His latest project, though, is not something Barbara has personal experience with.

FT: so if you found, hypothetically, a mysterious glowing substance that affects tech in weird and wacky ways that could totally have potential but might be vaguely sentient/otherworldly…. what would you do and how would you experiment with it. safely, of course. and hypothetically

BG: I’d make sure all my tests were in disposable devices and quarantined programs to keep it from infecting my important stuff. Dare I ask… how weird and wacky is it?

FT: uhhh. theoretically, a person composed of this substance once used it to enter a video game. like physical body, into the computer, onto the screen? moving around and talking and fighting enemies within the game?

FT: its been experimented with before, but not on any tech with a brain. just basic shields and blasters and stuff, its an energy source. also was put in a car once

FT: i wanna see how it affects software, yk? bc i already know it can. mess around and see how far i can push it

BG: […]

FT: … barbara?

BG: Sorry, thinking. Would you mind sharing more details? You said “blasters?”

Honestly. Kid genius with access to some truly wacky materials and even wackier weapons, she needs to start a file on him before he full sends to either hero or villain.

[OR: Tucker is a self-taught hacker, but if he were to credit a teacher, he'd name Barbara Gordon's Coding & Computer Cram School! He's even caught the attention of Dr. Gordon herself. She's full of sage advice, and with how she preaches the value of a good VPN, he's sure she's not pro-government. Maybe she'll help him as he studies the many applications of ecto-tech!]

[c1qfxugcgy0]

adventures in QA

(previous post in this series)

My shop in Advanced Midbody - Carbon Wing (AMCW) at Large Aircraft Manufacturer (LAM) is at the very end of the composite fabrication building. Hundreds of people carefully lay up a hundred foot long slab of carbon fiber, cure it, paint it, and then we totally fuck it up with out of spec holes, scrapes, primer damage, etc. The people who write up our many defects are from the Quality Assurance (QA) department.

Every single screw and rivet on a LAM aircraft can be traced back to the mechanic who installed it. Back when even everything was done in pen and pencil, it was joked that the paper used to produce an aircraft outweighed the plane itself. Now that everything is computer-based, of course, the amount of paperwork is free to grow without limit.

(Haunting the factory is endless media coverage of an emergency exit door plug popping out of an Advanced Smallbody - Upengine (ASU) plane during a routine flight a few months ago. Unlike that airframe's notorious problems with MCAS, this was a straightforward paperwork screwup by a line worker: the bolts were supposed to be tightened, and they weren't.

As a result the higher ups have visited hideous tribulations on non-salaried workers. Endless webinars, structured trainings. Here at the Widebody plant we have received a steady flow of refugees from the Narrowbody factory, hair-raising tales of receiving one hundred percent supervision from the moment they clock in to the second they clock out from FAA inspectors who can recommend actual jail time for any lapse in judgement.)

A single hydraulic bracket Installation Plan (IP) is around four brackets. The team leads generally assign two bracket IPs per mechanic, since each bracket set is something like a foot apart, and while working on the plane is bad enough it's much worse to have another mechanic in your lap.

Let me list the order of operations:

One: Find where you're supposed to install these brackets. This is harder than you might think.

Firstly, it's a hundred foot long plank of carbon fiber composite, with longitudinal stringers bonded to it to add stiffness. The stringers are pilot drilled in the trim and drill center, a truly Brobdingnagian CNC mill that trims off the composite flash at the edges and locates and drills part holes for us. But there's a lot of holes, so you must carefully find your set.

A minor difficulty is that the engineering drawings are laid out with the leading edge pointing up, while the wing panels in our cells hang from the trailing edge. Not so bad, you just rotate the paper 180 when orienteering, then rotate it back up to read the printed labels.

A major difficulty is that the drawings are from the perspective from the outside of the panel. But we work on the inside of the wing (obviously, that's where all the parts are installed) so we also flip the drawings and squint through the back of the paper, to make things line up.

Large Aircraft Manufacturer has a market cap of US$110 billion, and we're walking around the wing jig with sheets of paper rotated 180 and flipped turnways trying to find where to put brackets.

Oh well, we're paid by the hour.

Two: Match drill the aluminum brackets to the carbon fiber composite stringer. I can devote an entire post to the subtleties of drilling carbon fiber, but I can already tell that this post is going to be a miserable slog, so I will merrily skip over this step.

Three: Vacuum up all the carbon dust and aluminum swarf created during this process. This step is not optional, as your team lead will remind you, his screaming mouth clouding your safety glasses with spittle at a distance of four inches. LAM is very serious about FOD. Every jet airliner you've ever ridden in is a wet wing design-- each interstitial space is filled with Jet A. There is no fuel bladder or liner-- the fuel washes right over plane structure and wing hardware. Any dirt we leave behind will merrily float into the fuel and be sucked right into the engines, where it can cause millions in damage. No place for metal shavings!

If you are nervous about flying, avoid considering that all the hydraulic lines and engine control cables dip into a lake of a kerosene on their way from the flight deck to the important machines they command. Especially do not consider that we're paid about as much per hour as a McDonalds fry cook to install flight-critical aviation components.

Four: Neatly lay out your brackets on your cart, fight for a position at a Shared Production Workstation (SPW) (of which we have a total of four (4) for a crew of thirty (30) mechanics) and mark your IP for QA inspection as Ready To Apply Seal.

Four: Twiddle your thumbs. Similarly, we have three QA people for thirty mechanics. This is not enough QA people, as I will make enormously clear in the following steps.

Five: Continue waiting. Remember, you must not do anything until a QA person shows up and checks the box. Skipping a QA step is a “process failure” and a disciplinary offense. From the outside, you can observe the numerous QA whistleblowers and say “golly, why would a mechanic ever cut a corner and ignore QA?” Well...

Six: QA shows up. Theoretically, they could choose to pick up the mahrmax you prepared for them and gauge every single hole you've drilled. But since we're three hours into the shift and they're already twenty jobs behind, they just flick their flashlight across the panel and say “looks good" and then sprint away. Can't imagine why our planes keep falling out of the sky.

Seven: Apply the seal to the bracket. P/S 890 is a thick dark gray goop that adheres well to aluminum, carbon fiber, fabric, hair and skin. Once cured, it is completely immune to any chemical attack short of piranha solution, so if you get any on yourself you had better notice quick, otherwise it'll be with you as long as the layer of epidermis it's bonded to. LAM employees who work with fuel tank sealant very quickly get out of the habit of running their hands through their hair.

Eight: Now you wait again. Ha ha, you dumb asshole, you thought you were done with QA? No no, now you put up the job for QA inspection of how well you put the seal on the bracket. Twiddle your thumbs, but now with some urgency. The minute you took the bottle of seal out of the freezer, you started the clock on its "squeeze-out life." For this type of seal, on this job, it's 120 minutes. If QA doesn't get to you before that time expires, you remove your ticket, wipe off the seal, take another bottle out the freezer, and apply a fresh layer.

Nine: Optimistically, QA shows up in time and signs off on the seal. Well, you're 100 minutes into your 120 minute timer. Quickly, you slap the brackets onto the stringer, air hammer the sleeve bolts into position, thread nuts onto the bolts, then torque them down. Shove through the crowd and mark your IP "ready to inspect squeeze out"

Ten: Let out a long breath and relax. All the time sensitive parts are over. The criteria here is "visible and continuous" squeeze out all along the perimeter of the bracket and the fasteners. It is hard to screw this up, just glop on a wild excess of seal before installing it. If you do fail squeezeout, though, the only remedy is to take everything off, throw away the single-use distorted thread locknuts, clean everything up and try again tomorrow.

Eleven: QA approved squeeze out? Break's over, now we're in a hurry again. By now there's probably only an hour or two left in the shift, and your job now is to clean off all that squeeze out. Here's where you curse your past self for glopping on too much seal. You want to get it off ASAP because if you leave it alone or if it's too late in the shift and your manager does feel like approving overtime it'll cure to a rock hard condition overnight and you'll go through hell chipping it off the next day. You'll go through a hundred or so qtips soaked in MPK cleaning up the bracket and every surface of the panel within three feet.

Twelve: Put it up for final inspection. Put away all your tools. (The large communal toolboxes are lined with kaizen foam precisely cut out to hold each individual tool, which makes it obvious if any tool is missing. When you take a tool out, you stick a tool chit with your name and LAMID printed on it in its place. Lose a tool? Stick your head between your legs and kiss your ass goodbye, pal, because the default assumption is that a lost screwdriver is lurking in a hollow "hat" stringer, waiting to float out and damage some critical component years after the airplane is delivered.)

One tool you'll leave on your cart, however, is the pin protrusion gage. There is a minimum amount of thread that must poke outside of the permanent straight shank fastener's (Hi-Lok) nut, to indicate that the nut is fully engaged. That makes sense. But there's also a maximum protrusion. Why?

Well, it's an airplane. Ounces make pounds. An extra quarter inch of stickout across a thousand fasteners across a 30 year service life means tons of additional fuel burnt. So you can't use a fastener that's too long, because it adds weight.

On aluminum parts, it's hard to mess up. But any given composite part is laid up from many layers of carbon fiber tape. The engineers seemed to have assumed that dimensional variation would be normally distributed. But, unfortunately, we buy miles of carbon fiber at a time, and the size only very gradually changes between lots. When entire batches are several microns oversize, and you're laying up parts from fifty plies and an inch thick, you can have considerable variation of thickness on any given structural component. So you had better hope you had test fit all of your fasteners ahead of time, or else you'll be real sorry!

And, if you're really lucky, QA will show up five minutes before end of shift, pronounce everything within tolerance, then fuck off.

And that's how it takes eight hours to install eight brackets.

[melanodis]

don't really have much for a caption but i do have batshit insane rambling and more art under the cut

hardware capable of temporarily. read: TEMPORARILY. substituting for the central core to minimize downtime during maintenance

got wiped like 7 different times because he kept questioning his existence until they finally encrypted his imprint so much they had him asking how to use a doorknob

dual core processor equivalent to a PowerPC 970, capable of clocking up to 2.7 GHz for intensive tasks

always looking for """vintage""" computer parts in the hopes that they could be a compatible upgrade, like more RAM for his beautiful big brain ideas (wrong)

stole a network card out of a scientist's computer as a prank once he remembered humor. forgot to put it back and constantly tortures everyone else with his wifi capability

205X specific :

i don't think anyone is exactly jumping for joy at the words "open heart surgery." Especially not when it's you... on a table... alone with a light and a scalpel.

lights are dimmer for power conservation (dogshit battery)

processing power is also reduced unless necessary for hacking

low power mode has a dual purpose; conservation of battery power as well as keeping temperatures low to prevent overheating (huge headache + lots of broken parts)

clear coat has long eroded due to weathering + gel exposure during chamber maintenance

loss of coolant is nearly a death sentence for him and anyone involved due to the particular chemical mixture being difficult to find + reproduce (insanely hazardous to touch or breathe the fumes of)

yeah his radiator is probably rusted over but he's not touching that. it hasn't completely plugged itself so who cares (not him)

built like an old ass car, completely made of metal and doesn't crumple when impacted, leading to... lots of dents and shaken internals. nokia 3310 headass. he couldn't be assed to go through the cosmetic surgery that is fixing one of the newer models after a little fall (unless they're a patient of course)

on a related note, virgil is the robot equivalent of a beater car with the check engine light on and probably 6 other lights on, the underside is completely rusted out and the tires are bald. but it still runs so whatever, drive it into the ground buddy. he genuinely refuses to perform any work on himself unless its something critical. boo hoo nobody cares about some sensors and --- ah. temperature sensor.

i think

that can wait a little longer.

[canmom]

using LLMs to control a game character's dialogue seems an obvious use for the technology. and indeed people have tried, for example nVidia made a demo where the player interacts with AI-voiced NPCs:

this looks bad, right? like idk about you but I am not raring to play a game with LLM bots instead of human-scripted characters. they don't seem to have anything interesting to say that a normal NPC wouldn't, and the acting is super wooden.

so, the attempts to do this so far that I've seen have some pretty obvious faults:

relying on external API calls to process the data (expensive!)

presumably relying on generic 'you are xyz' prompt engineering to try to get a model to respond 'in character', resulting in bland, flavourless output

limited connection between game state and model state (you would need to translate the relevant game state into a text prompt)

responding to freeform input, models may not be very good at staying 'in character', with the default 'chatbot' persona emerging unexpectedly. or they might just make uncreative choices in general.

AI voice generation, while it's moved very fast in the last couple years, is still very poor at 'acting', producing very flat, emotionless performances, or uncanny mismatches of tone, inflection, etc.

although the model may generate contextually appropriate dialogue, it is difficult to link that back to the behaviour of characters in game

so how could we do better?

the first one could be solved by running LLMs locally on the user's hardware. that has some obvious drawbacks: running on the user's GPU means the LLM is competing with the game's graphics, meaning both must be more limited. ideally you would spread the LLM processing over multiple frames, but you still are limited by available VRAM, which is contested by the game's texture data and so on, and LLMs are very thirsty for VRAM. still, imo this is way more promising than having to talk to the internet and pay for compute time to get your NPC's dialogue lmao

second one might be improved by using a tool like control vectors to more granularly and consistently shape the tone of the output. I heard about this technique today (thanks @cherrvak)

third one is an interesting challenge - but perhaps a control-vector approach could also be relevant here? if you could figure out how a description of some relevant piece of game state affects the processing of the model, you could then apply that as a control vector when generating output. so the bridge between the game state and the LLM would be a set of weights for control vectors that are applied during generation.

this one is probably something where finetuning the model, and using control vectors to maintain a consistent 'pressure' to act a certain way even as the context window gets longer, could help a lot.

probably the vocal performance problem will improve in the next generation of voice generators, I'm certainly not solving it. a purely text-based game would avoid the problem entirely of course.

this one is tricky. perhaps the model could be taught to generate a description of a plan or intention, but linking that back to commands to perform by traditional agentic game 'AI' is not trivial. ideally, if there are various high-level commands that a game character might want to perform (like 'navigate to a specific location' or 'target an enemy') that are usually selected using some other kind of algorithm like weighted utilities, you could train the model to generate tokens that correspond to those actions and then feed them back in to the 'bot' side? I'm sure people have tried this kind of thing in robotics. you could just have the LLM stuff go 'one way', and rely on traditional game AI for everything besides dialogue, but it would be interesting to complete that feedback loop.

I doubt I'll be using this anytime soon (models are just too demanding to run on anything but a high-end PC, which is too niche, and I'll need to spend time playing with these models to determine if these ideas are even feasible), but maybe something to come back to in the future. first step is to figure out how to drive the control-vector thing locally.

[andmaybegayer]

Was looking at my energy use to get context on some Eat More Electrons posting. I used about 1.3MWh of electricity in the previous billable year, which is about 3.5kWh/day or 150 watts continuous average draw, which sounds about right.

That's not all my energy use. My hot water and heat are thermal central. Rough estimate would say I got maybe 1MWh thermal on hot water in various forms, directly or indirectly through my neighbours walls. No active cooling (yet). Emissions on direct thermal are messy because on the one hand direct thermal is a bit more efficient than thermal electric, but also CZ has an okay-ish nuclear mix. Don't really care though.

Then there's daily transport, which is. Hard to estimate. I'm on the metro and the tram and the bus. Mostly pretty efficient.

But yeah like. Electricity wise that makes sense. My house probably has an idle draw around 80W. Most of that is computers, my server and network hardware is about 60W non-stop. Not a lot of passive power consumption in the Thicket. PC+peripherals ramps up to like 150-200W when on but not in heavy use, and like 600W if I power virus it. Main use of power is probably cooking and computer? Which is a funny look at what kind of person I am.

I spend 6-10 hours a day at the office but that's not my problem.

Of course all of this thrown massively off by a long distance flight anywhere. Flying is so energy intensive. Vaguely estimated numbers here: A 777 is putting out like 27MW per engine at cruise, two engines, 360 passengers, 150kW per person, like 2 MWh per person for a 12 hour flight. Not directly electrical comparable but you get the idea.

[aeferkssr]

FORCE OF ATTRACTION

every magnet has a north pole and a south pole. placing two unlike poles together causes them to attract. when you try to place two like poles together (north to north or south to south), they will repel each other.

established relationship (xiao, childe), pining (kazuha), public opinion. obvious xiao favourtism. mentions of assassins/murder. mentions of still births. slight yandere!childe (not really,, kinda?). author started college and thinks they know everything /j

masterlist.

COMPUTER ENGINEERING / ELECTRICAL ENGINEERING

you two were polar opposites.

personality wise, xiao was quiet, aloof, and kept to himself most of the time. you, on the other hand, were the friendliest person on campus, always eager and ready to help if anyone had a problem.

he wore dark shades of black, grey, and the occasional teal to match the accents in his hair, while you were always delightfully decorated with an abundance of colours and patterns. even your majors were black and white.

so why is it that all of a sudden you two are inseparable?

the canteen was filled with idle chatter as usual, but one topic seemed to stick among the students of your college, the topic was discussed in hushed whispers and occasional coos as you broke off half of your cinnamon roll to give to him.

students will gawk at the gentle way you hold his hand to lead him to the location of your biweekly date, will laugh among themselves at the rips of his ears as it glows pink from realising that despite your major, how nice they are to hold, and will reminisce on the small kiss you have gave him prior.

(now that they think on it a little more, his flushing ears may be the result of that)

hardware and software rarely come together. but when they do, you know that good things are bound to happen

ASSASSIN / OBSTETRICIAN

your dynamic was chaotic. one who did all they could to bring life into this world, and the other who never hesitated to take it away. you did everything in your power to ensure safe deliveries while he would massacre in the flick of a wrist.

you sob as you mourn the loss of your patients offspring, while tartaglia laughs at the screams of victims praying for their lives.

two completely different world views.

yet, you cradle his head into your chest as a mother would to her newborn. yet, his blades soil with deep crimson as soon as he finds any person that may have given you a hard time.

yet, despite the stark difference in occupations, you two make it work.

after all, you two are human.

LINGUIST / MATHEMATICIAN

the both of you bonded on your love for your respective fields… while unintentionally hating on the others.

kazuha loved poetry. words were so powerful to the point where they can alter your very emotions, they were colourful, so full of life and meaning. he loved the way how the right words could change the very course of a conversation, and the allure of alliterations that allow ample amusement. numbers just seemed too… stuck up, too boring.

you, on the other hand, has a passion for numbers. numbers were factual, straightforward, to the point. if you counted ten apples, there is without a doubt ten apples. numbers never lied, numbers never had double meanings, numbers never mislead you the way that words and letters did.

kazuha visibly relaxes as he sees your animated expressions explaining your passion, his mind being swarmed with thousands of words and phrases to attempt at describing this feeling, describing you

but alas, nothing seemed good enough for someone as breath-taking as you.

© aeferkssr

[rexomi]

Clair Obscur musing

Okay so I'm not making this anything in depth and I'm kinda just spewing out my thoughts. I'm not done the game either, I just started act 2 but I mention one or two minor spoilers to the plot, more character names. Not the big one at the end of act 1.

Clair Obscure has a ton of themes, a ton of thing that it could easily be metaphors for. The idea of those hitting 33 (so young in the grand scheme of things) and sacrificing themselves for younger generations and just the pervasiveness of tragedy and also the idea that there are those who don't want to bring children into a world to suffer. These are all things that we face today.

However I also think it's about the immortality of art.

Verso and Renoir are immortal and they are also references to two different fields of art. Verso is the back side of a piece of paper, the left hand pages of the book that bear even numbers. It's opposed to the recto side of the books.

Renoir is probably a reference to Pierre-Auguste Renoir (as mentioned to me by moonsomniari) one of the father's of the impressionist style which Clair Obscur takes a lot of inspiration from. (as a side note another game that uses the impressionist style is Zelda's Skyward's Sword who's engine will mimic the brushstrokes for the background) For those who didn't study art history Impressionism is a movement in the 19th century that's characterized by short and visible brush strokes coming from the advancement of paint (not only in thickness but also in pigment) in the century.

These people are dead.

But are they?

Renoir died in 1919.

But also did he?

They're here right now. In this post. Reaching past the screen through in a scan of the brushstrokes. You can go see it today in the Musée d'Orsay in Paris. As long as that painting is remembered and shared and seen and thought about, all those people who visited the Moulin de la Galette on that Sunday afternoon are still alive exactly like they were on that day.

Going back to Verso. Go look at a bookshelf. How many dead authors are sitting there. Art whether through painting or writing or even video games or silly little videos on youtube, they're a form of immortality.

They're slices of the creator's life immortalized exactly as they were the moment they created it. This post is too, for as long as it's shared around and thought about a piece of me will live on.

I don't think it's a coincidence the Paintress takes people at the age of 34, arguably the prime of someone's life. I'm pretty sure that she's taking them to be immortalized, painted as exactly they are at that moment in time.

Is that the only theme? Of course not, it's a packed narrative with so many that touch people in so many different ways. But I do think there's a sort of beauty to make a comment on the immortality of art in an art form that is arguably the easiest to lose.

Video games are often lost, either by lack of access (whether deliberate or not) or by a lack of hardware to play it. Even if you have the game as someone who plays old games, sometimes the only way to play old games is literally to play it on the original hardware. I have an xp computer just for that reason.

The developers are remembered just as they are, immortalized in that moment they created, only as for long as we can remember them. As long as it's on the Steam store, as long as even one those disks survive (an estimate 80-100 years for blurays but that's unsure).

But still they try, but what else can they do? All we'll have to see if Clair Obscur survives or joins the rest of it's comrades in a life of lost media.

[thegeorgetelegraph]

Best Computer Hardware and Networking Course in Kolkata

The leading hardware and networking institute of Kolkata offers the best computer hardware and software course. Join us and get a good hardware and networking job.

[jcmarchi]

Staying in the Loop: How Superconductors are Helping Computers “Remember” - Technology Org

New Post has been published on https://thedigitalinsider.com/staying-in-the-loop-how-superconductors-are-helping-computers-remember-technology-org/

Staying in the Loop: How Superconductors are Helping Computers “Remember” - Technology Org

Computers work in digits—0s and 1s, to be exact. Their calculations, processes, and memories are digital, all of which require extraordinary power resources. As we look to the next evolution of computing and develop neuromorphic or “brain-like” computing, those power requirements are unfeasible.

Distinct circulating current paths in a 4-loop network show the possible switching activity that allow flux to travel between loops. Image credit: Q-MEEN-C / UC San Diego

Some researchers are looking at analog improvements to advance neuromorphic computing. In other words, they are advancing not just software but hardware, too. Research from the University of California San Diego and UC Riverside shows a promising new way to store and transmit information using disordered superconducting loops.

The team’s research, which appears in the Proceedings of the National Academy of Sciences, demonstrates the ability of superconducting loops to demonstrate associative memory, which allows the brain to remember the relationship between two unrelated items in humans.

“I hope what we’re designing, simulating and building will be able to do that kind of associative processing fast,” stated UC San Diego Professor of Physics Robert C. Dynes, one of the paper’s co-authors.

Creating lasting memories

Picture it: you’re at a party and run into someone you haven’t seen in a while. You know their name but can’t quite recall it. Your brain starts to root around for the information: where did I meet this person? How were we introduced? If you’re lucky, your brain finds the pathway to retrieve what was missing. Sometimes, of course, you’re unlucky.

Dynes believes that short-term memory moves into long-term memory with repetition. In the case of a name, the more you see the person and use the name, the more deeply it is written into memory. This is why we still remember a song from when we were ten years old but can’t remember what we had for lunch yesterday.

“Our brains have this remarkable gift of associative memory, which we don’t really understand,” stated Dynes, who is also president emeritus of the University of California and former UC San Diego chancellor. “It can work through the probability of answers because it’s so highly interconnected. This computer brain we built and modeled is also highly interactive. If you input a signal, the whole computer brain knows you did it.”

Staying in the loop

How do disordered superconducting loops work? You need a superconducting material — in this case, the team used yttrium barium copper oxide (YBCO). Known as a high-temperature superconductor, YBCO becomes superconducting around 90 Kelvin (-297 F), which in physics, is not that cold. This made it relatively easy to modify. The YBCO thin films (about 10 microns wide) were manipulated using magnetic fields and currents to create a single flux quantum on the loop. When the current was removed, the flux quantum stayed in the loop. Think of this as a piece of information or memory.

This is one loop, but associative memory and processing require at least two pieces of information. For this, Dynes used disordered loops, meaning the loops are different sizes and follow different patterns — essentially random.

A Josephson juncture, or “weak link,” as it is sometimes known, in each loop acted as a gate through which the flux quanta could pass. This is how information is transferred and the associations are built.

Although traditional computing architecture has continuous high-energy requirements, not just for processing but also for memory storage, these superconducting loops show significant power savings — on the scale of a million times less. This is because the loops only require power when performing logic tasks. Memories are stored in the physical superconducting material and can remain there permanently, as long as the loop remains superconducting.

The number of memory locations available increases exponentially with more loops: one loop has three locations, but three loops have 27. For this research, the team built four loops with 81 locations. Next, Dynes would like to expand the number of loops and the number memory locations.

“We know these loops can store memories. We know the associative memory works. We just don’t know how stable it is with a higher number of loops,” he said.

This work is not only noteworthy to physicists and computer engineers; it may also be important to neuroscientists. Dynes talked to another University of California president emeritus, Richard Atkinson, a world-renowned cognitive scientist who helped create a seminal model of human memory called the Atkinson-Shiffrin model.

Atkinson, who is also former UC San Diego chancellor and professor emeritus in the School of Social Sciences, was excited about the possibilities he saw: “Bob and I have had some great discussions trying to determine if his physics-based neural network could be used to model the Atkinson-Shiffrin theory of memory. His system is quite different from other proposed physics-based neural networks, and is rich enough that it could be used to explain the workings of the brain’s memory system in terms of the underlying physical process. It’s a very exciting prospect.”

Source: UCSD

You can offer your link to a page which is relevant to the topic of this post.

[lovelylotusf1]

Inspired by the lovely @wisteria-wisteria and her Pinterest roulette posts. I present to you an AU:

Lestappen - Coding rivals.exe

Don't know yet if I'll do anything larger with it so I'd love if someone adopts the idea! Give it a new and welcoming home!

Max Verstappen, a brilliant hardware engineer who is mostly focused on robotics, is partnered up to work with programmer and game designer Charles Leclerc. From the moment they meet, he despises his new coworker. Charles is everything he isn't - all smiles and polite questions, a laid-back attitude when it comes to his work and yet somehow constantly praised by others. The predestined programmer, they call him. Max thinks it's all very ridiculous.

But they need to work together to meet the looming deadline. Maybe they'll even learn to appreciate the other's company along the way...

Small scene after this universe's InchidentTM:

"Mr Verstappen. What happened?" His boss's cold voice greets him the moment Max sits down in the overpriced meeting chair. The board of directors stare at him, disdain obvious in their postures, but their faces are carved into polite masks.

Max glares at Leclerc, who is sitting across from him. Even now, he doesn't look like he gives a single damn about the situation he has brought them into, that unnervingly pleasant smile still plastered on his face. Max wants to wipe it off.

Leclerc is slumped in his chair, glasses askew and in his normal working attire consisting of a sweater thrown over a shirt. There is a small coffee stain on front of it, reminding Max of what brought them into this situation in the first place.

"It's just unfair. He pushed a glass of coffee onto the robot, so I pushed it back onto his computer. It's not fair, right?"

Before he can give further explanations, he is shot down, "Thank you, Mr Verstappen. What is your perspective on this matter, Mr Leclerc?"

Leclerc straightens and lets his eyes sweep over the room. Max doesn't miss the fact that he pointedly doesn't look at Max and a small pang of annoyance flashes through him.

Leclerc says, "Nothing. Just an inchident in the lab. We will have the code restored and the robot running by next week."

Max needs to remember all the self-control exercises he taught himself in order to not jump over the table, drag Leclerc towards him by his stupid oversized sweater and yell at him until that guy's ears ring.

Of course, he can say that easily. He just needs to write a few lines of code that the automatic backup hasn't caught. But Max will have to build the hardware from scratch, the delicate electronics fried beyond repair. Oh, how he would like to strangle Leclerc for that.

His boss is apparently pleased by this response. He nods. "Very well then. Get back to work and fill out the provided reports. I don't want to repeat this, the funding is already minimal and you don't want to make your budget even tighter than it already is."

They are dismissed and walk back to the robotics lab in silence, their steps echoing off the blank walls. At least it would be silent if Leclerc's annoyingly smooth voice didn't disrupt it, "I know you don't like me. But at least pretend that you do when other people are around. It will be easier if you let some of my charm work on you, non?"

That fucker has the audicity to wink at him and it is the most awkward thing Max has seen in his entire life.

Max almost snarls. He'd rather be caught dead than enjoy Leclerc's company.

(I don't know why a robotics guy would work with a game dev but just let me fantasize about my AUs without worrying too much about realism)

[coupleofdays]

Why Kevin Flynn no longer visits the ENCOM hardware design meetings

The chief engineer of the hardware division of ENCOM stood in front of the office whiteboard, explaining his plans for the newest version of ENCOM’s line of home computers. At one end of the office table sat his fellow engineers, and at the other end sat CEO Kevin Flynn and his colleague Alan Bradley.

The chief engineer was drawing up schematics on the whiteboard while he talked. “Now, in order to stay competitive, we’re of course going to need more RAM…”

There came a sudden laugh from the other end of the table. The engineer paused, and looked at the CEO apprehensively. “Uh, Mr. Flynn, you don’t agree?”

Flynn waved a hand, holding the other over his mouth. “No, of course, you’re absolutely right. I agree, 100%. You can never have enough Ram.”

The engineer smiled, relieved, and began drawing on the whiteboard again, showing how the new machine would be upgraded over the previous version. “Exactly, and that’s why we’ve redesigned the structure of the machine, since there has to be more places where RAM can be inserted…”

A loud, high-pitched snort came from the other end of the table, causing the engineer to flinch. “Mr. Flynn…?”

Flynn was holding one hand over his face, struggling and failing to suppress a fit of intense giggles. “I agree”, he managed to get out. “I agreeheehee…”

The engineers looked at each other uncomfortably. Alan took off his glasses and pinched the bridge of his nose with his fingers.

[kyousystem]

More GNU Backgammon woes

I've found and fixed two separate bugs in GNU Backgammon over the past two days, both located in the code used to calculate the inputs to the evaluation neural nets (i.e. very critical data).

They were small bugs, but fixing them made a visible change in the analysis output. I strongly suspect that, for certain difficult positions, these fixes might be enough to change what move the engine would score as the highest, even after a rollout.[1]

Do you want to know where those bugs were located in the original version of the code?

In the middle of a rat's nest of a function some 700 lines long; specifically, in obtuse sections of code completely bereft of comments and which almost exclusively used variables with names only 1 or 2 characters long. See for yourself:

Is it any wonder that subtle bugs like off-by-one errors or using the wrong one-character variable for an array index (aka the bugs in question here) were able to slip in here and remain completely undetected for God only knows how long?

----------

[1] Of course, now I'm wondering if the neural nets for GNU Backgammon were trained while being fed these slightly erroneous inputs, or if these bugs crept in later. I very much hope it's the latter; I don't have the know-how or computing hardware at my disposal to train new nets myself.

[mitvpusolapur]

Top B.Tech Courses in Maharashtra – CSE, AI, IT, and ECE Compared

B.Tech courses continue to attract students across India, and Maharashtra remains one of the most preferred states for higher technical education. From metro cities to emerging academic hubs like Solapur, students get access to diverse courses and skilled faculty. Among all available options, four major branches stand out: Computer Science and Engineering (CSE), Artificial Intelligence (AI), Information Technology (IT), and Electronics and Communication Engineering (ECE).

Each of these streams offers a different learning path. B.Tech in Computer Science and Engineering focuses on coding, algorithms, and system design. Students learn Python, Java, data structures, software engineering, and database systems. These skills are relevant for software companies, startups, and IT consulting.

B.Tech in Artificial Intelligence covers deep learning, neural networks, data processing, and computer vision. Students work on real-world problems using AI models. They also learn about ethical AI practices and automation systems. Companies hiring AI talent are in healthcare, retail, fintech, and manufacturing.

B.Tech in IT trains students in systems administration, networking, cloud computing, and application services. Graduates often work in system support, IT infrastructure, and data management. IT blends technical and management skills for enterprise use.

B.Tech ECE is for students who enjoy working with circuits, embedded systems, mobile communication, robotics, and signal processing. This stream is useful for telecom companies, consumer electronics, and control systems in industries.

Key Differences Between These B.Tech Programs:

CSE is programming-intensive. IT includes applications and system-level operations.

AI goes deeper into data modeling and pattern recognition.

ECE focuses more on hardware, communication, and embedded tech.

AI and CSE overlap, but AI involves more research-based learning.

How to Choose the Right B.Tech Specialization:

Ask yourself what excites you: coding, logic, data, devices, or systems.

Look for colleges with labs, project-based learning, and internship support.

Talk to seniors or alumni to understand real-life learning and placements.

Explore industry demand and long-term growth in each field.

MIT Vishwaprayag University, Solapur, offers all four B.Tech programs with updated syllabi, modern infrastructure, and practical training. Students work on live projects, participate in competitions, and build career skills through soft skills training. The university also encourages innovation and startup thinking.

Choosing the right course depends on interest and learning style. CSE and AI suit tech lovers who like coding and research. ECE is great for those who enjoy building real-world devices. IT fits students who want to blend business with technology.

Take time to explore the subjects and talk to faculty before selecting a stream. Your B.Tech journey shapes your future, so make an informed choice.