Tunes, Ideas, Tasks

Sound Production and Recording for Musicians
Alex Rodgers
0604849
Task 5

Intro
I chose to remix elements from the track I recorded for task 4, ‘Devolution’. I tried the idea of slowing the whole piece down and trying to create a dub mix but found it simply didn’t work at the lower tempo. Instead I went for a drum and bass esque piece with hints of Faithless and The Prodigy from spoken audio and synth sounds.

Method + Arrangement

I started by chopping up loops from the original project folder of ‘Devolution’ including various bass loops, guitars and some other small sections which could be added in. All of these loops were then converted to apple loops.
I laid out the some of the bass loops to create a skeleton of a track with which to work off. The next stage was adding drums, I found a drum loop from an Aphrodite song which features hats, snare and kick sounds. It is a 8 bar loop which I then converted to an apple loop and matched the tempo to the bass tracks from the original song. The basic loop can be heard in the intro to the piece, despite it being effected. I then chopped up the loop to create alternate beats and add in snare rushes, which are a prominent feature of drum and bass.
The kick drum from the loop I used was very feint and did not punch through as I wanted it so I created an Ultrabeat drum machine with some kick samples in. I settled with a deep ringing kick sound and it is a constant feature through the track giving it its drive. For some variation I added a sample of kick/cymbal stab I found as an extra kick drum hit at the beginning of bars. This really beefed up the low end and gave the track some life.
I also added the sample of a snare from the original Devolution track and added it on a separate track to add some variation to the drum line. This track has a reverb on it to add to the contrasting snare sound. Most of the instances of this snare are as snare rushes.
The lead synth was added to supplement the riff on the bass guitar, it adds to the drum and bass feel. I programmed in the midi notes that are exactly the same as the riff played on the guitars and bass. I also added a phaser to this synth to give it some movement and create some space and add to the eeriness of the track.
The second synth is more of a pad sound and is used as chords to fill up the prechorus/chorus sections and create more of a progression, again the chords are from the original track and programmed in.
The guitar tracks I added are double tracked bounces directly from the original track; I did add an EQ to them to thin them out and get them to sit in the track a little better. This goes for both instances of the guitar; both the main riff and chorus lead part.
Lastly I added the voice over which is from a radio show and features Richard Dawkins discussing Evolution and speculating on the vastness of the universe. I thought that this was suitable especially as a remix of the original track, which is based on atheistic and Darwin driven lyrics.
Processing

On the main snare track I added a delay unit, which was automated to come in during the breaks to add a bit more colour to the breaks. I automated the wet/dry signals and the Feedback enable button to create a loop of delay.  I also placed a Fuzz/wah unit on the track, which only comes in during the introduction and the outro snare rush. Again I automated the pedal position and the unit bypass on the channel in order to get it to come in and out where I wanted.
There is a reverse reverb that can be heard in a break in the song. I created it by taking a snare sample and bouncing it with a long reverb. I then reversed this and ended up with a smooth reverse reverb rush, which leads into drops nicely.
I also use a dry reverse snare during the last part of the track which is simply a snare hit bounced and then reversed to add to the drum tracks variation. This technique is common in drum and bass and is very effective at breaking up drumbeats.
During the second ‘verse’ part I chopped up the guitars to only hit on stabs leaving the synth behind to continue the riff. I found this has a very nice effect and acts as a nice build up. I also used this technique during the introduction on the spoken ‘devolution’ combining this technique with the snare rush tactic of looping syllables to create a sampler effect(de-de-de-devolution). I also automated in the panning for this section.
For the voiceover I wanted to bring down the overall tone as it was too bright for the feel of the track so I added a pitch shifter and brought the voice down two semitones to give it a darker feel. I also EQ’ed the voice to bring out only the formant sounds of the voice (boosting mids) and give it some definition.
I added quite a lot of automation on the main Stereo Out track, mainly being a bit crusher and an EQ that acts as a filter, keeping only the phone line frequencies (between 1kHz and 4KHz. Both of these can be heard on the intro and are automated off as soon as the track drops. I found it quite effective, as the contrast from the first part to the drop is very pleasing.

Conclusion

I feel that as a remix it works well as it still retains the feel and mood of the original piece but can also stand alone as an individual piece. I wanted a listener to feel slightly uneasy while listening to it and the odd breaks and drops and varied amounts of frequencies during these drops definitely works to achieve this. I would have liked to include more elements from the original track such as the vocals but when I tried adding them it just did not fit with what I wanted to hear.

Sound Production and Recording For Musicians
Task 4
Alex Rodgers 0604849

Intro
I chose to record one of my own bands songs that was nearly finished in terms of writing. All the musical elements were ready but it was still awaiting vocals to be written. We planned to write the vocals after the instrumentation was recorded. It is a heavy fast paced song with a basic ‘metal’ band line up:
2x Guitars
Drums
Bass
Vocals

I wanted to aim for a big hard hitting sound and used 2 reference tracks to help guide me where I wanted to go.
These were
Arch Enemy – We will rise
Lamb of God – Redneck

Recording

Drums:

I recorded the drums in the Studio Booth in Melish Clark. I used the university kit, which is not in the best shape. It took a few hours to get it sounding in tune via a combination of duct tape and tissue paper. This was mainly on the tom’s that were from two different kits.
For microphones I used the Audio Technica Drum Microphone Kit for the Kick Drum, Snare and Toms. I used a Matched Pair of AKG C414’s for Overheads

The snare and tom mics were all clip ons with gooseneck fittings so I tried moving the mic’s around each drum until I got the desired sound as well as avoiding spill. I ended up with a similar positioning for the snare and tom mics with the microphone about 8cm from the drum skin and angled 45 degrees towards just off centre of the drum skin.
The kick drum microphone, which was both dynamic and a condenser via different cables (I chose dynamic), was placed just off centre of the beater. It was about 4cm away from the drum skin to get as much ‘beater’ sound as possible to get the ‘click’ bass drum sound.
The overheads were placed in an AB pattern about 50cm apart with the kick and snare line perpendicular to the microphones to ensure that they were the central part of the kit in the stereo image.

Bass:

I recorded the bass straight into a Mackie Audio interface and straight into logic. The bass is an active 5 string so I had to put the pad on the input. We adjusted the tone on the bass itself to get a basic sound, which could then be processed later. There was a lot of buzzing on the bass recording but we found that placing the bass facing a corner of the room minimized this. I think it is due to the active bass not being earthed properly. I thought about micing up a bass amp but have had very good results recording bass directly into a desk or interface. I find it gives you better raw material with which to play with afterwards.

Guitars:

The guitars were recorded directly into an audio interface via a PodXT Live effects pedal. The effects pedal has a studio out setting allowing you to take a line output from the Pod and go into a line input instead of a instrument input. I played with the pod settings for a long time before finalizing each of the two guitar tones. They are both very biting and growly but with less gain than you initially think. I planned to double track everything so putting too much gain on the distortion will cause you to loose definition especially when doubling up.
The lead guitar sound has a lot more mid range and gain as well as a delay effect to allow it to sound a bit washing and sit in the mix nicely. Again it is double tracked with harmonies.

Vocals:

I had a lot of different vocals to record and mix with both singing and screaming elements. I started with Joe’s vocal and used a TLM 103 condenser microphone behind a pop shield and with a reflection filter to keep the sound as dry as possible. Joe’s voice is quite raspy and thin but sounds a lot better when double tracked, which is what I did. I used the same principles of microphone positioning for both sets of singing vocals with the microphones tilted slightly downwards to avoid too much bass proximity effect.
For Dan singing I used two microphones, an AKG C414 condenser and a Sure SM 58beta. I liked the combination of the two microphones for Dan’s voice as the condenser keeps the definition and high end while the dynamic mic keeps the meaty growling part. I used both mic tracks for all Dan’s singing parts.
The first round of screaming recordings were for Joe and Oli’s vocals. I used a Shure 57beta that was held in the vocalist’s hand. I played the track as loudly as possible through the foldback system to try and immerse the vocalist in the song. I have found that they tend to hold back if they can hear their own voice too much and don’t get the same sound as when screaming in band practice or a gig. As screaming vocals is based around distortion and weight of the sound I chose to use a dynamic microphone, as well as the fact that screaming right into a condenser microphone can break it.
For Dan’s screaming parts (the verses) I used both a C414 and a Sure SM 58beta. Both of the microphones were mounted and behind two pop shields. Dan’s natural scream is not that loud and is concentrated in the high range frequencies so using a condenser was justified. He also has a very good technique when changing volume of his vocals and moving closer or away from the microphone.

Mixing

Drums:

After getting the recordings into logic and laying them in the arrange window I listened to all the takes and choose the best ones. Because we recorded the song in stages, stopping after each section, it allowed me to ensure all the takes were as good as possible. The next step was to sample delay the tracks to sync up audio. As microphones are placed in different places the time sound takes to reach each microphone is different causing your recording to be slightly out of time and phase. I used the overheads as a reference and sampled delayed all the other tracks to them.
The next step was tidying up all the audio, removing clicks and pops and removing unnecessary parts with nothing playing. As there was not much use of the toms in this track I went through and chopped out all the hits and removed the rest of the audio to avoid cluttering the mix.
I started mixing from the overheads, I firstly added an expander to bring out the punchiness of the sounds. Next in the chain was a limiter to bring up the overall level and increase the density of the sound. Channel Eq was then applied boosting the top end slightly as well as a high pass filter around 100hz to remove rumble. Lastly I added a small amount of compression to even out the sounds and bring out the cymbals.

Kick:
The kick drum was the next track to work with. I spent a lot of time during this project honing in on the kick drum sound I wanted and ended up doing a lot of processing. The final result was as follows: An Expander to bring down the noise floor and the cymbal spill, which was very prominent. A limiter to boost the level and density, this is due to the parts in the song where there is double bass and the volume of the kick is lower than when hitting single strokes of the kick. When this happens the compressor threshold will miss the lower volume double bass parts. Placing a limiter early on helps to compensate for this. The next step was Channel EQ where I pulled down the very low end (below 40hz) and the low mids (between 500 and 150hz) quite drastically to shape the kick drum into the sound I wanted. I also pulled down a bit at 5kHz where there was a prominent cymbal spill. Unfourtunaly this is also the area that the click from the bass drum was so I needed to replace this with the next step in the processing. I added a Clip distortion with very low gain and a filter adding only distortion between 2.5-10kHz in order to give the kick and clicking sound. This dramatically changed the sound and effectively gave me the sound I wanted. I then placed another EQ to pull the low mids and cymbals down again slightly just to round off the sound. Compression was the last part of the chain, I went for quite a lot of compression with a fairly low threshold, this is due to the double bass parts not being as loud as other parts of the recording and the compressor missing them. It took a while to get the correct balance. The compressor has a very fast attack and release to compensate for the fast double basing. I tried putting a gate on the kick drum but could not find a decent setting to cover both the slow and fast parts. I added a limiter at the end to beef up the sound and add density.
I also created a kick drum bus track which had the final kick drum sound sent to it. I then added a HP filter at 300hz to kill all the low end of the bass drum. I also shaped the sound a bit pulling some mids and the high end to just get as much click as possible. I added an expander to reduce the noise floor which was very prominent then a limiter to beef it up. This was then mixed with the origional kick sound.

Snare:

I started with an expander to reduce the noise floor and make it a bit more punchy followed by an EQ to boost the mids and take out the low end rumble. A limiter was put in before going into a compressor, which had fairly high compression setting. I also placed a gate to remove unwanted noise in the gaps. Finally I placed Space Designer reverb plugin with a fairly long reverb but still quite a dry sound.

Toms:

Again I placed an expander to reduce the noise floor and bring them out a bit. This was followed by a limiter then an EQ which was pulling up the fundamentals and harmonics of each different tom. Lastly I placed a compressor on to even out the levels. On the high tom I placed a space designer with a short reverb to give the tom a bit more ring. The other toms did not need this.

2 Main Drum Busses:

I sent all the drum tracks to two separate busses, each with a different purpose. I found this technique online at http://www.allfreeessays.com/student/Mixing_Music.html
The first bus has very fast limiting on it to bring up the overall level of the transients and make it very punchy and ensure every hit is as hard as the last. I call this bus the Drum Transient Enhance. The second bus I call the Drum Elongator, it also has a lot of limiting but also has a space designer with a tight reverb, this is to extend the sound of the hits and make the kit sound bigger. I also added an exciter to enhance the cymbal sound and make the kit sound a bit glossier. The two busses are then mixed together resulting in a very big drum kit sound. I was very pleased with this result and will use the technique again in future.

Bass:

I placed the logic bass amp on it to create the final sound; I went for a gainy distorted sound that most metal bands tend to use on their bass. I EQed the bass to add a bit more low mid and take out the low end which helped reduce muddiness in the mix. Bass playing should almost always be compressed and I chose to compress quit hard to even out all the playing by reducing the dynamic range. Finally I added a limiter to add density to the bass.

Guitars:

All the guitars are double tracked to beef them up, this is common practice in metal bands as it creates a ‘wall of sound’ impression. I did not process the guitars very much due to them already being sculpted in the PodXT during recording, however I did add a channel EQ with a high pass filter at 150hz to remove the low end and give the bass and drums more room to breathe.
I also added a limiter to beef up the sound on some of the guitar tracks.

Vocals:

Singing Joe:
Joe’s voice naturally has a lot of sibilance so naturally I placed a deesser on all of his tracks to quite heavily bring this down but without giving him a lisp. I also placed a channel eq on to remove he low end and shape his voice a bit. A compressor is also added to all of his tracks to even out his singing and beef the sound up a bit.

Singing Dan:
I placed a channel EQ to pull down the low end and boost the mids to retain a bit more clarity in his voice. This went into a compressor to even out the singing followed by a limiter to add density.

All singing tracks were sent to Vocals Bus with Delay from delay designer and exciter to brighten up vocals. As well as limiter to control effect levels from peaking. Delay as opposed to reverb is a genre style, most metal bands favor delay over reverb on singing.

Screaming:

Dan Verses:
As I recorded with two microphones and planned to use the feed from both of them as well as double tracking I needed to use a lot of tracks. Also because some of the lines overlap each other at the ends so each line needed separating. I ended up with 8 tracks for Dan; I processed all of them the same way. Starting with a channel EQ I cut the low end and boosted lower mids to make the sound a bit more forceful. I also added DeEssers to compensate for the multiplied effect of sibilance on the multiple tracks. Finally I added a light compressor as screaming naturally compresses the sound in your throat so not too much was needed.

Other Screaming Tracks (Dan Joe and Oli):
I placed compressors on all of them to even out the levels, there are lots of tracks playing at once and some of them have a wide dynamic range which needed to be reduced in order to bring out all the different parts. This was followed by an EQ to bring down the lower end. I also placed a hard high pass filter at 900Hz on Oli’s screams to keep only the high white noise esque part of his scream which I found worked well in the mix.

All of the scream tracks were sent to a ScreamVerb Bus with large plate reverb and limiter, this was to give the screams a bit more depth and try to get a creamier small grain sound that I was going for.

Stereo Mix:

On the stereo out track I placed an exciter to brighten the whole track up a bit with added harmonics above 12kHz. Also I placed a fairly hard limiter to give the whole mix an added boost of density to create a really hard-hitting sound.

Mastering:

I mastered the sound through a TLA audio Compressor and EQ in the studio. I used my stereo mixdown from my logic project and re-imported it into a new instance of logic that then sent the track out and recorded the sound coming out of the hardware. It took along time to get the settings right on the hardware and I had to take frequent breaks and ask many peoples opinions in order to get to the result I got.
I started with the EQ where I ended up pulling down the very low end (around 30hz) slightly. I also pulled down the muddy frequencies at around 150-250Hz somewhat, this created the stereotypical scooped metal sound. I also pulled down with a very narrow Q setting a point at around 4500hz where there was a lot of cymbal clang cluttering the mix, I had to be careful with this part as I found a lot of the crack of the snare was also in this range. Finally I boosted the top end from around 7000Hz to crisp up the overall sound. You can see some of the settings here. EQ: high end and EQ: Low end

For the compressor I used a high threshold and quite a low ratio; I only really wanted to warm up the sound slightly and not overcook my mix. It took a long time to set the attack and release because of the double bass parts. When the attack and release was too long the hits turned into a sludgy sound, when too fast there was an audible breathing of the compressor. I finally got it sounding right with a slight pumping sound which I feel adds to the big wall of sound style of the track.

I have listened to the track back on many different systems now and feel I have a achieved a good balance of frequencies and has a lot of clarity. Overall I am most pleased with the drum and guitar sounds, which are very close to what I aimed to go for initially. Given the chance I would rerecord the vocals as I feel they were quite rushed, not so much in the recording and sound but in the actually melodies and arrangement. This is the part of the mix I am least happy with.

However overall i have found this to be a very successful project and am very pleased with the final result. It is definetley a step up from my previous productions and taught me many useful tricks which i can use in future.

Stereo Microphone Techniques

The three instruments I chose to record as part of this study were Acoustic Guitar, Piano and Clarinet. Below you can see how each technique was used and applied to the instrument.

Guitar

AB – This technique employs time difference to give you the stereo image, sound waves hit the diaphragms of each microphone at different times depending on where it is in relation to the two microphones. This creates phase and time differences which when played back in stereo a human ear can decipher and localise the sounds in the stereo field.
I used the Octava condenser microphones for this technique and used the omni-directional capsules.
I used a distance of 30cm between the microphones to try and simulate an approximate size of a human head as well as think about how wide it needs to be in order to pick up phase differences of low frequencies with long wavelengths. Human localisation stops working at around 150Hz due to the head being to narrow to fit a whole wave shape there not allowing phase/time difference localisation to work. 150hz has a wavelength of around 40cm this is why I chose to make the distance between the microphones 30m which is in-between the average width of a human head (17cm) and the wavelength size of the lowest frequencies we can localise.
The microphones were placed around 60cm away from the guitar with one of microphones aiming at the hole of the guitar and the other at the neck.

NOS – this technique uses both time difference and intensity level differences to pick up the stereo field. Intensity level difference is where sounds are arriving at the microphone diaphragms at different intensities meaning that there will be an imbalance of level over the stereo field placing objects either to the left or right when you listen to them.
I used the Octava condenser microphones with cardioid capsules for this technique.
The NOS set up spaces the microphones 30cm apart and at an angle of 90 degrees to each other. This angle allows for the intensity level stereo image as each mic is facing a different direction meaning sound will hit each mic at different intensities depending on where the sounds source is. Having the microphones set to cardioid is essential as it allows them to have a narrow field in front of them and extenuate the intensity difference.
The microphones were placed at around 45cm away from the guitar. They need to be placed quite close to keep the stereo field realistic and also to avoid a drop in bass frequencies when this set up is placed at further distances from the source.

MS – This technique involves two different microphones. A cardioid Octava set up in the middle of the setup pointing at the sound source and a microphone set on figure of 8 placed perpendicular to the first mic picking up the stereo image – for this mic I used the SE Electronics condenser. The microphones are placed one on top of each other to get them as close together as possible.
For this technique a matrix needs to be applied to them in order to create the desired stereo field. Logic has a plug-in that does this, which I used for the recording. The plug-in places the figure of 8 microphone across the stereo field and adds the mono cardioid microphone in to fill in the middle. There is a stereo field width control on the plugin which I tweaked to create a balanced stereo image as I found it was slightly to wide and I should have put the microphones in a bit closer to make the field tighter, however changing this parameter solved this problem for me. I placed the microphones at about 1 meter from the source with the cardioid mic pointing at the hole of the guitar.

Overall I prefer the AB recording, as it is much warmer than the other techniques, there is a much better bass response, which causes this. The chords especially sound better with this technique. The MS technique sounds a bit thin but is very smooth and would be good if you were planning to add the guitar into a mix of other audio as it sounds very clear and has a narrower frequency band than the other recordings. NOS seemed to be a balance of the AB and the MS where it had more bass and warmth than the MS but was still very thin sounding compared to the AB.

Piano

AB - As with the first AB pair I used the octava condensers and placed them 30cm apart. As a piano has a much wider sound source than a guitar I placed them further away at around 1m to try and incorporate the whole image. Pianos emit different notes from different parts with the lower notes and one end and the higher notes at the other end. This means that placing a microphone set up is important, as you want to include all of the different notes. I placed the microphones slightly towards the left (higher notes – when looking at back of the piano) of the piano as lower frequencies spread out more than higher ones meaning that although it was pointing more to the higher end, lower frequencies were still being picked up clearly.

Blumlein – This setup incorporates two figure of 8 microphones placed on top of each but with the diaphragms at 90 degrees. This creates the stereo image using solely intensity. The microphone was set up with the 90 degree angle between them pointing directly at the middle of the piano at about 1m away. This mic set up is susceptible to spill coming from other sound sources as it picks up from behind as well. This was recorded in an room with no other instruments, this would however gather more early reflections of the sound giving it reverb quality however the room is fairly dead and the effect is subtle.

NOS (with different mic’s) I used two SE electronics microphones for this set up as opposed to using octavas for NOS like on the guitar. I set the mics on cardioid and spaced them 30cm apart and angled the capsules 90 degrees perpendicular to each other. These mics were placed about 45 cm away from the piano. This set up has low bass response so I wanted to bring it in closer but also ensure that the stereo field remains so I placed it quite close to the piano.

Overall I preferred the NOS recording for warmth and depth, it seemed to be a lot nicer and rounder sounding than the other two. The AB mic sounded quite accurate, the sound quality was almost the same as being in the room and hearing the piano directly but it seemed to be slightly thin and not full enough. The blumlein recording was very tinny and the notes were very jangly, although some people may like this piano sound I personally favour the NOS recording.

Clarinet

AB – This set up is used again as a control for this task. I used two octava condenser microphones spaced 17cm apart (average width of a human head) I placed them pointing towards the bottom third of the clarinet about 30cm away. As the sound source is quite narrow I placed the mics closer together as well as to note there aren’t that many very low frequencies coming from it either.

ORTF – This technique uses two cardioid microphones spaced 17cm apart and angled at the slightly wider 110 degrees apart, it is similar to NOS in the view that it uses both intensity and time differences methods of deriving the stereo field but it has a wider spread than the NOS setup. It is quite a similar set up to the human ears, especially if you but a blind between the two to simulate a human head (I did not do this in the recording). As using this set up far away from the sound souce will create a loss of bass I placed it quite close to the clarinet at about 50cm pointing outwards but level with the centre of the clarinet.

DIN – This set up is like the NOS set up except with microphones spaced closer together at 20cm. They are still 90 degrees to each other creating the intensity difference stereo spread as well as the time difference information. I placed this about 1m away from the clarinet to create what sounds like quite a wide stereo spread.

For the clarinet I prefer the AB recording out of the three. It sounds very warm and realistic and sounds a lot smoother than the other two recordings. It sounds very glassy and glistens especially on the higher notes. The ORTF recording sounded similar to the AB but there seemed to be more breathe hiss noise in the sound. The DIN set up recording sounded a lot thinner than the other two and as with the ORTF there was a lot of wind/breath sounds.

Pictures of some of the mic set ups (click to view fullsize)

Welcome to my first post,

I like metal, douglas adams, africa and traveling.

I play guitar in a cambridge based metal band as well as do some remixes/ productions at home.

Here are 2 examples of my past work

1: A Different Perspective - A potential radio snippet discussing the myths of marijuana, was created using samples of many famous songs with vocals i recorded, is really just for fun…

2My Dubya World - Dub remix of a metal song ‘My World’ by Caught in the Zip. Featuring none other than George W.

enjoy the tunes and let me know what you think…