In this weeks labs, we have been editing together a piano recording. There are several different takes of the same piece, ranging in tempo. The assignment included taking certain chunks of bars from different takes, depending on the better performance. We tried doing time expansion on some of the parts, but the piano is a complex instrument. The waveform and sounds seem too complex still, for the software to fully and accurately maintain the original sound quality. After completing the assignment, we tried out some drum replacing/enhancing. To do this, create a duplicate track of the one you want to replace. Find a transient that you like, and make sure you cut it where you want it. Bring it down to the new track, and zoom all the way in get the audio to the proper starting point and make sure it matches to the grid. If you don’t stay locked in grid, things can get messy fast and your audio will be all out of line. Taylor and I got some really cool bass samples that we edited into a custom rhythm. We then ran it through the eventide harmonizer and got a few different pitched rhythmic takes. We organized the pattern into a little odd time groove in grid.

Compression settings:

The prerecorded track goes out A1 and the newly recorded track goes in A2. To listen back, they both go out 2 track.
The Millenia’s fastest attack setting is 2ms.
Set settings at 12 o’clock with a 3:1 ratio
Try a fast attack, fast release
Ratio 6:1
Low threshold, attack around 30ms, release 1000ms, 9:1 ratio
The attack lets the initial transient through. The release is slow enough that the compressor has a slow release and the room sound rushes up.
Fastest attack, Slowest release, 10:
Fast attack, slow release, 9:1
Medium fast attack, medium release
Relative snare timbres:
Thud – fast attack
Crack – slower attack
Changing the threshold will keep a consistent snare, with a ratio of 8:1 or higher
Try a fast attack, and a fast release. Then the same fast attack with a faster release.
Fast release, medium attack gets lots of punch out of a snare drum.
Slowest attack, fastest release, threshold at 9:00, and turn the make up gain up. Start with a ratio of 1.4:1, then 3:1, then 6:1.
The 3:1 setting sounds brighter than the 6:1 setting. The compressor also beings acting like an EQ.
Fastest attack, fastest release.
Fastest attack and slowest release, ratio 15:1. Long release will prevent the room sound from rushing up.
Fast attack shaves off the transient.
Low threshold, fully clockwise, everything is being compressed, fast attack and release, 10:1.
Fast attack, .5 sec release, ratio 10:1, low threshold, room sound not very present but overall loudness is up
Same settings with Medium attack -
Threshold fully clockwise, attack 10, release .7, ratio 8:1
A compressor acts like glue, and imparts a sonic characteristic when combining tracks that were recorded from different places.
4:1 ratio, 2 sec release, 2 ms
Release – how long it releases after the initial attack.
Raising the threshold means less audio will be compressed. Because little to no audio is getting compressed, the compressor acts as an amplifier.
Conservative settings make it sound more uniform.
Fast release can release too fast on the low frequencies and it will distort.
Super fast attack, long release, and lower threshold, you’ll get a crushing sound

COMPRESSION

Compressors control maximum levels and maintain higher average loudness. Compressors and limiters are specialized amplifiers used to reduce dynamic range. The dynamic range is the distance between the loudest and softest part of a wave. A flute produces a tone where the difference is 3 dB. The human voice has a 10db dymanic range, while plucked instruments have about a 15dB range. Our ears act as compressors as well. Ears respond to the average loudness of a sound. Compressor are designed to include detecort circuits that responds to an average signal level. A second circuit responds to peak signal levels. A brick wall limiter hits a threshold and stops there, it wont be any louder than the threshold.
Multiband compression: Some compressors and digital plugins can separate the compression into 3 or 4 different bands of lo, lo mid, mids, and highs. This allows compression on only a certain amount or “band” of frequencies.
Compressors: Optical (LA2A, Fairchild 670) – use a photoresistor: signal comes in to a lightbulb. More audio = brighter light. Photocells can recognize the very suttle light changes caused by the incoming audio. FET (1176) – Field Effect Transistor: First transistor that emulated tubes in the way the work. They are fast, clean, and reliable. VCA – Voltage controlled amplifier – most versatile of all the compressors, have a higher level of control to control range. Varigain – doesn’t involve circuits, digital compressors – exaggerated compressors: all settings can be from 0- infinity, and you can get precision from this compressor versus the others.
Ratio – the degree to which the compressor is reducing dynamic range, of the difference between signal increase at the input and output. 2:1 means that for every 2 dB coming into the compressor, it will only sound like 1dB over going out.
Threshold – the level of the incoming signal at which the compressor amplifier changes from a unity gain amplifier into a compressor reducing gain. Everything above the threshold is being compressed . Once threshold is reached, compression happens depending on the amount of signal coming in, and the ratio setting. KNEE – hard and soft, the exact moment the signal reaches the threshold. A hard knee is sudden and abrupt. A soft knee eases on the compression. By manipulating the attack and knee, we change the envelope, mainly the attack and release.
Attack – brightness of character) the time it takes for the compressor to compress after the threshold has been reached. Attack times typically range from 1ms to over 100ms. The attack time effects tone in terms of brightness. Fast attack clamps down on the signal
Release – the time the compressor uses to return to unity gain after the singlahas fallen below the threshold. longer release time creates a darker sound, shorter relase makes it sound brighter. Compressor is released from gain reduction. 20ms – 5 seconds. Depends on the tempo and program material.
Fast release on a bass distorts it.
Slow attack – all the brightness comes through on a snare. Fast attack for snare – dulled sound
Brings loud portion down, and creates make up gain.
Send a snare track from protools out A 1-2 into line 1 inputs. Send that signal out of channel insert sends into the distressor, and then send the signal out of the distressor to channel insert returns. Start with the input, attack, release, and output to 5 on the distressor module.

In the studio this week, Taylor and I practiced sub-grouping and putting the song Jonesy through the MTA 980, on the group bus, to the monitor groups, and then back into the box to record the mix. We used 3 groups, Drums/Bass, Gtr/Horn/E Percussion, and Vocals. That means we had to use 6 tracks- 3 stereo AUX tracks to get out of protools and into the board, and 3 stereo audio tracks to get the mix back into protools. We then got started on Raw Tracks 4 is not a very good track. In fact, if I were in the real world, and had some credibility as a producer, I would not accept this as a paid gig. I would send it back and tell them to get it re-tracked, or hire new musicians. The main thing I noticed was that there were a ton of vocal track, which isn’t the initial problem. In most of these recordings the vocalists’ words don’t line up, and it would take an unnecessary amount of editing/ be a huge waste of time – which is why I mentioned it be re-tracked. We edited the tracks and cleaned the session up.

Groups, Submixing In and out of the box, and Stems

The board has just been fixed and its awesome. Today we learned some new stuff: Sub-mixing and Stemming both in and out of the box. To mix in the box, we still use the 2Tk1 and mix buttons on the master fader. Going through the board, we can use the master fader itself, with the mix button down. Start by grouping a few instruments together, like all of the drum tracks with bass, the guitar with piano, and all of the vocals into 3 separate stereo aux tracks to be labeled as subgroups in ProTools. We get sub groups by sending the original audio tracks to AUX tracks. To record a stem in the box, create 3 stereo audio channels with the inputs set the same as the outputs of the subgroup channels. Mute and record enable. To get the sub mixes out of the box, patch those stereo aux tracks out 1-2, 3-4, 5-6 into line inputs with the 6 channels of your choices on the board (1, 2, 3, 4, 5, 6) and pan the channels L R L R L R. You can send these subgroups on the group bus with via the group bus matrix (the buttons at the top of any channel strip on the board). These are sent to the Red faders at the right of the board, the monitor group channels (Yes, Will… they are monitor and group channels!). Send the drums/bass to 1 and 2, gtr/piano to 3 and 4, the vocals to 5 and 6, and make sure the monitor level pot is dialed all the way up. The each group hard left and right (1L, 2R, 3L, 4R, 5L, 6R). Now that we have these submixes/groups through the board, and we have our submixes in the box to be recorded as stems, we want to send the monitor groups back into ProTools to be recorded out of the box as stems. Create 3 more stereo audio tracks in ProTools and label them the inputs that you patch them to. Patch the monitor groups out of the “group outputs – 1, 2, 3, 4, 5, 6” on the patch bay into ProTools inputs that are not being used as outputs, or you will get a feedback loop. Say, In B-1, B-2, B-3, B-4, B-5, and B-6). In ProTools set the inputs of the 3 stereo audio tracks you created to correspond ( drums B1-2, gtr/piano B3-4, and vox B5-6) and record enable. You should now be able to record your in and out of the box stereo stems from the submixes you created.

“The Inner or Deep Part of an Animal or Plant Structure” Bjork DVD

Bjork collaborates with many different artists. For the album Medulla, she worked with a variety of artists from different part of the world. This interested me a lot, seeing as she did much of her work on her own in the earlier days. I think she is very culturally aware, she thought to incorporate many different styles of music into this project. Bjork used the Icelandic and English languages for the lyrics in this record. A few places she went for the recordings of this album were New York, Iceland, and Brazil. Medulla is a concept album, meaning it revolves around a particular idea or concept – in Bjork’s case this was to be an “all vocals” album. Back when she was a teenager she always knew that she was going to do a vocal album, but she wanted to wait until the right time and the right place. Razhel is a famous artist that has an incredible beat-boxing talent. Bjork heard of/met Razhel in New York came invited him to do work on the album. It was really cool how he made a bass drum sound with the throat, in the pattern of a rhythm that resembled a heartbeat in a song. Mark Bell programmed some electronic beats that eventually became the beats that Razhel imitated with his skills. She worked with Mike Patton because he’s from the rock background, a very experimental, open-minded, and a good guy to work with. Bjork felt she had done all she could do by herself, and started to begin with to work with an engineer and incorporate other peoples’ voices. Voices emulating the land, or animals, or whatever is going on at the time. *The Inuit culture is one that lives in extreme weather conditions in places like Canada, in the northern and southern Arctics, northern Russia and Alaska. These people have to use the environment around them and the resources there (which are minimal) to sustain life. The September 11th tragedy got her thinking and invoked the idea of primitive elements: Back in the day, we didn’t have all of the convenient resources we have available to us today. One of the only things we are born with as humans is a voice, and she wanted to express that we can use what primal resources we might have to create the things we invision. I think Bjork is inspired by this, in that she mentioned she wanted to have an album that contained “blood, bones and meat”, or in other words, all sounds will be derived from human expression. She had made a collection of hundreds of voices and effects, but she only ended up using a few! Despite the technological capacity we have today, Bjork and other artists are making killer sounding records with a DIGI888 interface that is only 16 bit, 44.1KHz. At the time of conceptualizing the album, Bjork wasn’t focused on lyrics at first. Reading through her diary that she had written, she felt that words wouldn’t be as important as the message and emotion that human breaths, noises, and voices are capable of creating. As she gathered more and more vocal tracks, she became more conscious of how the voices were fitting in with each other, and their relationships with one another. Nigel Godrich has produced many albums, including Radiohead’s OK Computer, and has mentioned in an interview that he may dream of producing Bjork, but she’s already so interesting in her own way. Spike Stent produced Bjork’s first album Homogenic, and helped out on the production of Medulla. During the recording process, Bjork wears many different hats. Not only is she the inventor of the concept, and the mastermind behind the production and composition, she acts as a conductor and an engineer. She also acts as a translator when trying to describe to someone the sound that she is looking for.

The Mexican food was great ;)

In class this week we went over possible combinations of inexpensive home studios, and the gear necessary to have one up and running with 16 channels in and out. Starting with the computer, have a powerful enough computer to handle the processes carried out by a DAW like Protools or Logic, with at least 2 GB of RAM memory. For Protools, the DIGI 002 or 003 are good models. They run between $1,000-1,500. A 1/4 inch patchbay will be necessary for routing signals where you want them to go. The API 312 mic pre’s are pretty nice and the API 550’s and 560’s are small, modular EQ’s that you can fit I a rack or on a channel strip. For Logic, using an RME interface will work. Protools will only accept Digidesign hardware.

We have been working on the mix for RawTracks 3 and I still can’t get over how well this song was tracked. All of the tracks are super clear sounding with very little bleed. The coolest thing we’ve done so far in lab was rhythmically rearrange one of the vocal lines at the end of the song and it sounds like the singer is rapping over it. We got our mono mix out of the board recorded into protools. We compressed the kick and snare through the distressors, and the vocals through the millennia. The snare was sent to a cool reverb-delay and we also sent the electronic percussion track there as well. When we did a mixdown, we used the board as an instrument too and we performed live automation! We controlled the reverb at certain sections of the song with the AUX knob. The board was being serviced and repaired for the rest of the week so we weren’t able to do another mix, but we will pick back up on Tuesday. There have been many issues with the board, so it’s great that it will be running solid soon.