trahlliw-blog - Tumblr blog

[trahlliw-blog]

Drop-In Drums.

A small group of decided that we should try and book a drop in session in the big studios now that the driving test isn't a problem in order to get some more recording done. We started by planning out what we wanted to record, deciding that it would be a good idea to start work on our 8 track recording. We outlined what we wanted to record on each track starting drums, then bass, acoustic guitar,some kind of keyboard instrument, Vocals and backing vocals on top of the two tracks that were to be made up of loops. Once we had this plan we went into the studios and booked a drop in session in studio 1 for just over an hour and a half.

We started micing up the setting up the drum kit, dampening the snare with some moon gel and the kick with a couple of our hoodies. Once the kit was sounding nice enough we started micing it up. We decided to try 5 microphones: A D112 on the kick drum placed about half way inside pointing at the beater, a Shure Beta SM57A on the snare drum a few centimetres above the head of the drum directed at the point about half way between the edge and the centre and a stereo pair of AKG Perception 170s as overheads roughly a foot above the drummers head and about three times that apart. In addition to this we placed a perception 120 microphone near the hi hat, however after we recorded some initial tests we decided that because of the position of the mic it made the whole kit sound very heavy on the hi hat side but very thin on the other side. Because the overheads picked up enough of the hi hat by them selves and the hi hat mic was throwing the kit out of balance we decided to not use it.

Once everything was set up and we got some levels for the mics and a good fold back level we started recording some drums for our eight track, setting the tempo at 120bpm and recording a variety of different beats that we could piece together into a structure for our eight track once we had some other parts. This went very well and our microphone set up gave us good results and a good over all dry drum sound. 

After we had a good selection of drum takes we decided to try and experiment a little with the drum patterns, setting up a midi track in logic and having some one improvise some keyboard parts while I played drums along. This didn't give us much that we could use in the context of our project but it did work as a good way to transfer musical ideas between the Control Room and the Live Room.

The drums we got were very useful and I have been thinking about trying to make use of some of the beats with in my loop pack to add some variety to the instruments I already have loops of.

[trahlliw-blog]

Working For Cajon Loops.

In the same session as bass and acoustic guitar we also recorded some good cajon parts for our loop packs. This began as an experiment for our loop packs as we already had the bass guitar and acoustic guitar. We had the idea that we could try and get all three instruments playing in the live room and record them all using a stereo pair of AKG Perception 120 condenser microphones. We set up against one wall with the cajon in the middle, the bass amp on the left and the guitar on the right. As we recorded we found that this produced a very live sounding recording, however it didn't really give us anything that could be considered useful for our loop packs. When our tutor came to listen to the recording he suggested that we record the cajon by its self as something for our loop packs. We took this feedback and worked from it to record some cajon loops that could fit in with our previously recorded guitar and bass parts.

We decided to record cajon using a two mic setup with a AKG Perception 120 about a foot from the front of the cajon with the polar pattern set to cardioid and the AKG D112 placed half way into the cajon at the back. The D112 was acting in the same way as a kick drum mic, giving us the means to mix the punchy low frequency sounds differently to the rest of the cajon which was being picked up with the condenser microphone at the front. This technique gave us a really good sound when we listened back to the cajon and as we tried to give the two mics a good balance we found that the condenser mic at the front was also picking up a lot of the low frequencies so we needed to turn the D112 signal down significantly to avoid boosting those frequencies too much. The D112 was useful in adding just a little punch to the sound of the cajon when set at a low level underneath the condenser signal.

At the end of the session I was very happy with how much raw material we had gotten together for the loop packs as we were walking away with potential bass, acoustic guitar and cajon loops. I feel as though we made very good use of the studio time.

[trahlliw-blog]

Working For Acoustic Guitar Loops.

After we got all the bass loops recorded we moved onto recording some acoustic guitar loops, specifically ones that would work on top of some of the bass loops that we got recorded. We got to work doing this by using our ears. I got the performer standing just off centre in the room facing at a slight diagonal. Once he was in what sounded like a good spot he started playing and I walked around the room, covering one of my ears to try and find the place in the room that sounded the best. I eventually found a spot that sounded really good so I set up the Røde NT2A in this position and set the polar pattern to omni in order to get all the sound of the room in. With our room mic in place we set up the AKG Perception 170 as a close mic directed at the guitars 12th fret, a method that we found useful in the acoustic guitar recording task.

We recorded some good sounding guitar parts with this method however the position of the room mic may have been too far away from the sound source as the signal was extremely weak. When we turned up the signal it sounded very unnatural and didn't work well with the close mic. This could have been due to the placement of the mic being out of phase with the other microphone as we hadn't strictly adhered to 3:1 rule of microphone placement. Feed back from our tutor confirmed that the room mic wasn't a success, however our close mic was and held it's own without the aid of a room mic. We decided to use just the signal of the close microphone by its self for our first few acoustic guitar loops.

To experiment a little further we decided later on to see how the guitar would sound with a slightly more aggressive playing style recorded through what we deemed as a slightly less harsh sounding microphone. We set up an AKG Perception 120 about a foot away from the centre of the guitar with the polar pattern set to cardioid and with the bass roll off turned on. This produced a good guitar sound, however I did think that because we turned on the bass roll off the guitar sound wouldn't really work in the context of a sparse mix, especially if there wasn't any other bass instruments. 

We got some good sounding guitar riffs and chord patterns for this session, keeping with the tempo of 120 bpm, the tempo of all our other loops and trying to stick in E minor and D minor to work with the bass.

[trahlliw-blog]

Working For Bass Loops.

In the this studio session we really made the most of our time. Although we were in the small studios we had no specified task today other than to record audio for the studio loops section of our assignment. For this I specifically wanted to demonstrate the different techniques that we had learned in the previous weeks as well as maybe going beyond them and experimenting with things a little more. 

We started off by recording some bass guitar riffs for our loop packs. We went about the recording process in the way we though was best which was to split the signal in order to record both the sound from an amp and a dry DI signal. We started by getting our equipment tuning the bass guitar. We connected the bass to the DI box via 1/4" jack cable and from their routing DI signal into Logic. This is done by sending the signal to one of the desk tie inputs and then using the patch bay to route the signal from the desk tie into one of the pre-amp channels which would then take it to the sound card and into the DAW. The DI box link output was then sent into the live room and into the bass amp using two coaxial cables. We miced up the amp with an AKG D112 microphone, a choice we made because it is a microphone built for kick drum, an instrument that produces a lot of the same frequencies as a bass guitar. This meant that the microphones frequency response would be working in our favour.

We recorded some good riffs for our loop packs which sounded pretty good. As we recorded we experimented with mixing different ratios of DI and Amp signal finding that a ratio of about 3:1 DI to amp signal worked well as well as 2:1 the other way round.

[trahlliw-blog]

The track from the post 'Apple Looping.' made from apple loops and using automation and audio manipulation.

[trahlliw-blog]

Apple Looping.

In this lab session we were given the task of making a demonstration track out of apple loops. The aim of this task was to give us a better understanding of what professional quality loops consist of and how a loop pack such as the one we are making would be used. 

I began the task by finding some cool sounding drums. In my opinion some of the best drums that are included in apple loops are found in the jazz section so that was my starting point. I settled for some really nice Bebop drums with a nice tight hi-hat pattern that I could easily build on. I got the drums playing by themselves for the first 4 bars and then after this began experimenting on top of them. I ended up with alternating between classical piano and synth parts for the next 4 bars before moving onto some vibraphone. The last four bars I tried to get some semi swelling horn sounds so I found some parts that were harmonies of one another to play with a flute sound and some upright bass. The transitions were very harsh so I tried fading in some of the drum parts which helped a little. 

Once I had something that sounded good I starting messing with it, inserting some FX loops and reversing some of them as well as one of the horn parts to give a strange effect of the original and reversed parts playing at the same time.

I still felt like the parts clashed a little so I tried setting up some delay on the vibraphone track in order to try and mask one very obvious transition. I set up a delay on a send and then set up the automation so that I could ride it on a keyboard controller fader. I pushed the delay up at the end of phrases which helped to mask the break.

I also felt that the horns didn't swell up enough so I automated their volume in the same way to get a better sounding transient of them.

I think that my final resulted sounded pretty good so I was happy with the session. If I had more time I probably would have started to mix it a little more maybe applying some compression and EQ.

[trahlliw-blog]

Through The Wall.

This studio session began with me taking the studio competency test. One person from every group was supposed to do it so I volunteered to do it first in my small group. Those of us that volunteered went into studio 1 to take the test, me going first. The test went fine and I performed the tasks asked reasonably quickly and easily.

After this I went back to the group where we began to do the task that had been set for the session. We were recording electric guitar in such a way that the guitarist could sit in the control room alongside the producer, a way that obviously has a lot of benefits. The method detailed using a direct injection box to split the signal, the DI'ed signal being send into the desk mic tie and then being patched through to channel 23 on the desk. The link signal from the DI box would then be sent into a mic'ed amp in the live room using coaxial cables. This is a technique that forms the basis of re-amping. Re-amping is a valuable process as recording the signal from the amp allows for a good monitoring experience, ensuring that the player hears the sound they expect to hear. This likely leads to them giving a better performance than if they could only hear the dry guitar signal. Recording the dry guitar sound alongside the amp signal means that the engineer can tweak the guitar sound later on without needing a new performance. This is done by routing the recorded dry sound out of the DAW and into a new amp/pedal, allowing the engineer to tweak the sound as much as they want while still maintaining the same perfect performance every time.

We got to work quickly on this task, setting up the mic and amp and coaxial cable in the live room before moving into the control room. We tuned the guitar using the rack tuner in the studio and then plugged the guitar into the DI box. We connected the DI out to desk tie 1 and then using the patch bay sent the signal from the desk tie into channel 23 on the desk, ensuring that phantom power was on. We then plugged the coaxial cable into the DI link and sent it through to the control room. 

At first this didn't work very well. We were picking up the signal from the amp but we were getting nothing from the DI. We spent some time trying to trouble shoot, however after nothing coming up wrong we tried switching the DI box as the one we were using didn't look like it was in the best condition. The new DI box worked fine and so we got on with some guitar recording. We recorded some good riffs and chords that can be used in our loop packs. 

While recording the guitar we experimented with getting a good mix going between the amp sound and the DI'ed sound. We got some really good results by putting some light tremolo on the amp and cranking up the gain to get some light overdrive. This sounded really cool when we mixed it in with the dry sound, using the short faders to mix the signals together nicely. For some parts we tried a mix more heavily weighed towards the amp sound, however on other parts a mix with more dry sound sounded better.

[trahlliw-blog]

The track from 'Reverb Matching.' and 'Dynamics Processing.'

Including all the reverb, dynamics and mixing practices detailed in the posts.

[trahlliw-blog]

Dynamics Processing.

In this lab session we found ourselves working on the same rock track as in 'Reverb Matching.' The aim this time though was to introduce some dynamics processing into the track. The first task was to put a noise gate onto one of the snare tracks to eliminate some of the kick drum spill present. I opened up logics noise gate and got to work.

I set the reduction setting to -100dB so that when the gate was activated it would more or less cut the signal completely and then I adjust the threshold appropriately to find the level where it cut the kick spill without tampering with the transients of the snare its self. I found this setting at around -47dB. Once I had put the gate on the next task was to apply some compression to the kick.

I started working on the kick compression by adjusting the ratio to determine a good amount of reduction and then I set the threshold to a point where the kick sound was being suitably compressed and already sounding punchier. I set the ratio to 2.1:1 which is the same as 11:1 and the threshold to -32.5dB. This means that for every 11 dB increase above -32.5dB the signal increases by 1dB. I then set the attack and release settings to control how fast the compression acts and how fast it lets go before adjusting the knee setting to make the compressor come on hard. Once the compression sounded good I moved onto to the third task which was to apply side chain compression to the bass.

I went about the compression of the bass in a similar way to the kick, adjusting the levels to reduce the dynamic range of the sound. The difference being that this time I set the kick drum channel as a side chain, effectively meaning that the compression acts based on what the kick drum is playing, ducking out the bass when the kick drum plays to tighten up the rhythm section.

After finishing all the tasks in the session I continued to mix the track, applying compression to the rest of the drum tracks and the vocal, as well as EQing the guitars, bass and vocal and redoing some of the levels and panning, on guitars and synths.

[trahlliw-blog]

Recording Acoustic Guitar.

In this studio session we were given the task of recording an acoustic guitar and to figure out which ways worked best in the rooms we were in. AFter a quick intro and talk over of the upcoming studio competency test we got to work. We borrowed a mic box and cables along with a guitar and headed back to the studio. 

The first task was to figure out where to put the mics. The technique that we were instructed to use to determine this was an ear test, meaning that someone played the guitar and the others in the grow listened in all the different spots of the rooms to try get the best sounding area. We did this with me playing guitar and the rest of the group listening. In the end we opted to try three mic locations. These were a cardioid dynamic microphone facing the bottom of the body of the guitar, a small diaphragm condenser microphone pointing at the twelfth fret of the guitar and a condenser microphone set the figure of eight at the head height of the guitarist in an attempt to capture what they would hear. After we recorded these positions we came back into the control room to listen back.

The twelfth fret mic sounded great and so did the dynamic mic. The head heigh mic sounded very far away compared to the others and didn't pick up that great a sound. Our main problem was that the mics all sounded very out of phase with each other. To solve this we did the task over but this time we used the 3:1 rule to reposition the mics, ensuring that the microphones were roughly three times as far away from each other as they were from the sound source.

On listening back we found that the mics sounded much better, especially when played together (Although the head height mic still sounded faint and tinny compared to the others). We found that the dynamic mic sounded the most bassy and concluded that that was likely because it was the closest to the sound hole. The twelfth fret microphone definitely gave a sound truest to standing in the room with the guitar and therefore we decided that it was the best position the record the guitar from.

After we did these tests we utilised the extra time in the session to record some acoustic guitar chords and melodic lines for our respective loop packs and saved all our work before zeroing the desk and returning the equipment.

[trahlliw-blog]

The vocal from the post 'Reverb Matching.'

The first time round is the Reference Reverb and the second is the matched reverb from logics PlatinumVerb.

[trahlliw-blog]

Reverb Matching.

in this lab session we were given a track and the task of giving it some very specific reverb. The track was an unmixed rock song with drums, bass, guitars, synths and a lead vocal, our first task was to apply reverb to this vocal track. Along with the vocal we were given a reference reverb track and the brief that we needed to get the reverb that we put on the vocal as close the the reference reverb as possible.

This seemed like a pretty tricky task but I got started by listening to both tracks critically. After this I set up a reverb on a bus from the vocal track. I used logics algorithmic reverb - Platinum Verb as it is a high quality reverb plugin with lots of controls making it possible to be heavily customised.

Next I began changing the controls heavily in between listening back to the reference and then my own reverb. The key controls that I used were the pre delay setting which sets the time before the effect kicks in after the sound, Room Size and Shape, Diffusion which controls reflection clarity, Density to set the intensity and closeness of reflections and Reverb Time to set the length of the reverb tail. Along with this I used the high cut control to EQ the reverb a little. I altered these controls as well as the send level until I got the results below. I found that they produced a reverb very close to the reference.

Once I did this I used the same reverb on sends from the snare tracks to give them a similar sound and keep the reverb consistent across the tracks.

After I finished this task I got to work on mixing the track a little, getting some basic levels and panning sorted before applying so parametric EQ to  some of the drum tracks before the end of the session. This, along with the reverb got the track to sound much better already.

[trahlliw-blog]

Polar Patterns.

In this studio session we were given the task of testing out the way that different microphone polar patterns work by taking a Rode NT2A microphone and having one person from the group walk around the microphone (keeping the same distance each time) whilst counting to thirty. We conducted this test three times, once with the mic's polar pattern switched to Cardioid, once to Figure Of Eight and once to Omnidirectional. Once everything was set up properly and we recorded the counting into logic we all gathered in the control room to listen critically to the three recordings.

The first pattern we listened to was cardioid. When we listened we could hear very clearly that the microphone was picking up from the front the best and then as the counting circled the microphone it was getting rejected more and more until it reached the opposite side where it was very faint. Continuing around we could here the sound being picked up more and more until the sound reached the front of the microphone again. This clearly demonstrated the cardioid shape of the polar pattern, picking up the front and sides but rejecting from behind.

The next pattern we tested was the Figure Of Eight polar pattern. When we listened back we could hear that it had a similar effect to the cardioid, the main difference being that there the rejection zones went by much faster, suggesting that signal was being picked up at the front and the back of the microphone. The rejection was also more severe than that of the cardioid patterned.

With the Omnidirectional polar pattern it wasn't obvious that the sound source was moving at all, meaning that the microphone was picking up from all directions equally with no sound rejection at any point.

Our next task was to use the same polar patterns, but this time move away from the sound source instead of circling it. We did this quickly and listened back to test the effect. The cardioid and figure of eight microphones had a similar result, both producing lots of pops and sounding very bassy when the sound was close to the microphone, however as the sound moved away it became thinner. As these are directional polar patterns the bassyness is caused by the proximity effect. The proximity effect is when the difference in sound pressure levels at the front and back of the microphone are greatly different (like when the source is close to the microphone). The difference causes the bass response of the microphone to increase. 

When we did the same thing with the omni directional pattern we did not notice the proximity effect as it only applies to directional microphones patterns. This is because due to the nature of the omni directional polar pattern, there is no front or back of the microphone.

After we completed the task we miced up an amp in the studio and recorded some guitar chords and riffs that can be used as part of the loop pack. From these raw recordings I need to now go and form loops out of them as well as determining their tempo and key.

[trahlliw-blog]

This is the final audio from the post 'Flex.' complete with retimes and basic mixing. It's a full length song so key points from the post are:

First Hit Retime - 0:19

Second Hit Retime (Hi Hat Problem) - 0:26

Fill Clean Up - 2:24

Creative Retime - 3:10

[trahlliw-blog]

Flex.

Today's task in the labs was focused on flex-time editing in Logic Pro X to fix out of time drum hits and to creatively apply the same techniques to alter drum beats. We were given a track that had a couple of out of time drum hits to ready to be fixed in the flex editor. I started by selecting the 'Rhythmic' setting in the editor and once the track was fully analysed I started to listen through in order to find the drum hits.

The first thing I realised is that for some reason the flex editor had shortened the audio file, effectively time compressing it. I don't know exactly what caused this but it is something that I have encountered as a glitch in some of my past work using flex-time. The way to fix this is to  create a flex marker at the very end of the audio file and then grab it and drag it out until the audio file starts to be dragged out with it. This is what I did until the end edge of the audio file matched with where the others all ended, working to even the audio back out.

  Once everything was how it should be I started to re-listen and find the out of time drum hits. I found the first one quickly and used the flex markers to adjust the position of the hit until it was in time, making sure to be aware of the effect that time stretching could have on the audio. Once it sounded good I carried on listening and found another out of time hit and fixed it in the same way. The problem was that putting the hit in the right place threw the hi-hat playing in-between the hits completely out. I tried to insert some new flex markers and adjust where the hi-hat should be but it wasn't working. When the drum hit had been put out of time ready for this task it had made the hi-hat hit so short that when I was trying to stretch it back out again into time (pictured below) it introduced all kinds of weird beeps and harmonics to it, changing the pitch and the sound of the hi hat drastically. 

In the end I decided that the best way to avoid the obviously out of time hi-hat and the even worse beeping alien hi-hat was to remove the hit completely. I did this by using flexitime to make the hit shorter and shorter until it was no longer audible at all (shown below), and it worked pretty well. In the context of the whole song, one hi-hat note missing is not even noticeable, especially compared to the alternative.

I listened through the rest of the track and couldn't identify anymore out of place drum hits so I decided to move on to the next task which was to use the flex-time editor creatively to add interest. Although this was my intention and reason for listening through again to find a good place to try it out, another listen did identify a particular fill with minor timing errors which I used flexitime to clean up (below).

Once I had fixed the fill I found a good place to use flex-time creatively, putting a snare hit back a little just before a fill to give the note a swung feel (shown below). I listened back and adjusted it until it was just right and I think the result sounded pretty good.

I listened back over the whole thing to double check my retiming and then in the last part of the session tried out some basic mixing to tidy up the track a little bit.

First I mixed the levels to get them a bit more balanced and panned the Guitar and Keyboards out a little left and right. I Eq'ed each track individually and then applied compression to all but the electric guitar because that already sounded pretty well compressed. For the Vocal I also messed around with a speech enhancer plugin and liked the way it gave the vocal presence so I kept it. After this I used a noise gate to cut out the low level sounds of the instruments playing in the background of the recording between phrases. All in all I think it sounded much better.

[trahlliw-blog]

Harsh Microphones.

In this studio session we were given the task of testing how we perceive different levels of sound harshness from different microphones in the small studios. The idea of this is that we can test the different microphones given to us in the mic box by recording the same sound, (in this case the human voice speaking) individually with each microphone and then on listening back to the results, rate the levels of harshness on a scale from 1 to 10. Knowing this could help us when recording other sound sources in the studios based on the theory that soft sounds can be matched with harsh microphones and harsh sounds with soft microphones.

Before we even began with this we ran into a minor problem with setting up. We got the mics set up fine and recording them into the DAW was all ok, however we we're having a hard time getting fold back working. We spent a while trying to figure it out and eventually realised that our problem was that we didn't have the ID22 software open and were trying to set the fold back from inside logic. We opened the software and turned up the Cue fader on the 'DAW Output' channel to get the fold back working. Although it delayed us, I think that sometimes it's better to have problems like this because all the time spent trying to figure it out really helps to jam the solution to the problem into your long term memory.

Once fold back was working we got to work with the tests. The microphones we were using were a Shure SM57A Beta, an AKG Perception 120, AKG Perception 170, AKG D112 and a Rode NT2A. Of these microphones two are dynamic and three are condenser.

We rated the SM57A Beta as having the lowest harshness score of all the microphones in the test with a 4/10. This could be because it is a dynamic microphone meaning that the frequency response isn't as wide as a condenser, therefore the mic doesn't pick up the really high frequencies. On top of this, despite there being a boost in the frequency response towards the top end, the mic start to roll off at around 15kHz.

The next microphone, the AKG Perception 120 scored a reasonably harsh 7/10. As a condenser microphone, it's frequency response stretches up to 20kHz, but like the SM57A, it starts to roll off towards the very top end. What it does do though is boost the frequencies around 10kHz higher and with a much more consistent curve.

The AKG Perception 170 was the next test and it scored the harshest score of all the mics with an 8/10. This is likely because it has a big boost in the top end around 10kHz and as well as this, because of the size of the boost, there isn't really the same roll off we have seen in the other microphones, only just cutting slightly at 20kHz.

The 4th microphone to test was the Rode NT2A, scoring a middle of the road 5/10. Although this mics frequency response has similar boosts to the others, it also, like the 57A the boost starts further back, at around 2kHz. This wider boost could possibly affect the perceived level of harshness as it means that it isn't just the top highs that are boosted. This would also explain why the 170 sounded very harsh, as in its frequency response not only are these lower highs not boosted, but actually cut.

The last microphone that we tested was the AKG D112, scoring a similar mark to the rode with 5/10. This could be because it is a dynamic microphone (and one specifically designed for kick drums). The frequency response has a much smaller boost around 10kHz than the other mics and it has a much more dramatic roll off at around 15kHz. When we started the task I assumed this mic would be the least harsh, however in the test it just sounded a touch harsher than the 57A.

(All Frequency Response Graphs by recordinghacks.com)

[trahlliw-blog]

The final compiled take from the post 'Vocal Compilation'. Made up of the best parts from three individual takes.