Monday 13 June 2016

Mix hour #5


The final mix didn't quite go to plan, as when I uploaded the mix to Soundcloud it was immediately taken down for copyright protection purposes. If interested in what I mixed, the track was Bannockburn by Cnoc Un Tursa
Instead of spend ages describing the mix process, without a reference track, i'll skip it and conclude the experiment.

Conclusion

It was a greatly revealing challenge, which highlighted the most time consuming elements of a mix and also showed what parts I was a bit slow at completing.
Mixing drums seemed to consistently be a long part of the mix, taking roughly half an hour in every mix. This wasn't a surprise to me as previously drum mixing has never been a quick process.
One of the most revealing things was genres, and how mixing techniques can vary so vastly between some. The second mix was the prime example of this, showing how an electronically produced track barely cried for any EQing, compression or even time based FX. They could have been added, however it was far from a disaster leaving the tracks bare. My use of panning was interesting, while I had very little time, I still created a fair bit of interest in the track, making up for using compression and EQs to give tracks contrast.
In terms of FX usage, while it didn't show that adding reverb and delay to tracks was a hugely time consuming task, but it did demonstrate that rushing FX is never a good idea, proven by the very mismatched reverb present on various parts of the third mix.

Time wise, it certainly showed that while one can get a very basic mix completed within an hour, if the aim is to have a create a reasonable product that can be taken away after one hour, you should probably focus on mixing the elements of the recording, then add some time based FX, and only apply EQ or compression if you have a lot of time or if a track is in major need of fixing.

Friday 3 June 2016

Mix hour #4

I didn't get round to doing yesterdays mix as I met up with some friends in the evening and was busy for most of the day, so instead I thought I would still complete my planned mix, but do it today (as well as decide on a mix for today).

Mix #4:
Big Mean Sound Machine - Contraband
(https://soundcloud.com/rufus-ambler/mix-hour-4)



This Afrobeat song was a decent mix. It was a live recording as would be expected of such a genre, which helped as I didn't need to fret too much with my reverbs. I haven't done too much mixing with percussion and horn instruments so if there had been any issues with the recordings (There weren't as this particular recording session was conducted by the renowned mic company Telefunken), I wouldn't have necessarily known what to do to fix them.
I enjoyed this mix more than I think I would have something like a big band recording due to the rhythmic aspect of the song, which could be lacking in latter song style.

The mix process

I started off as usual with the drums, giving a quick listen to the overhead tracks to see what instruments they mostly picked up, then muted them and built the sound from the the kick and snare. I added a little compression to the kick, but felt nothing else was really needed. I moved onto the snare, first compressing it a little to bring out some of the body, then used a pultech style EQ to bring out some more of the low end & attack of the snare. Near the end of the mixing session it came to mind that it could do with gating a little, so I applied a gate to the track, so it very gently gated out some of the cymbal bleed. I brought the overheads back in next, adjusting them to give a good overall sound of the kit, panning them wide as due to the genre and live nature of the recording. I then moved onto the toms, panning them a little so they had a spacious sound but were not too wide in the stereo field. Last I added in the percussion elements making them loud enough to be heard in the mix but not dominate the drum sound. One of the conga tracks had both a very resonant sound, and parts where the player seemed to play much harder, so I applied a compressor to the track to balance it, and used an EQ to cut out the resonance at around 200Hz. I dialed in a Distressor style compressor on a parallel compression track, sending the drum bus to it. I left it muted til the end of the mix however, to dial in once the other instruments were balanced.
The bass was simple enough to work on, I compressed the bass amp recording a little and balanced the DI track so it was 1-2 dB lower than the amp, giving the bass a bit of attack.
The guitars required little work. While the first track was sounding good, with a slight touch of compression that could have improved it a bit more, the second guitar track had one section where the volume increased dramatically. I could have used automation for this but for times sake, I put a compressor on the track and altered the settings at the loud part so the gain reduction peaked at around 10dB. This kept the volume balanced within the mix. I applied the reverb to the tracks at this point, using the short reverb for the picky guitar and the long reverb for the shimmery strummed guitar.
The horns were next on the list. I decided to pan the trumpet and trombone fairly wide, leaving the baritone sax in the middle as it could accompany the bass track. I applied a bit of the short reverb to the horns, the sax having less reverb than the others due to the low end response of the instrument. The Keyboard track was simple enough to mix in. I didn't particularly like the sound of the organ so I applied a fair amount of long reverb to it and left it quite low in the mix, sitting underneath the horns.
When it came to the room tracks, I was running very short on time, so I left them all the tracks at unity gain and just adjusted the bus.

To dial in the short reverb, I used to snare track to adjust the time of the plugin, and for the long reverb I used a studio room impulse.

A few minutes from the end of the mixing session, I adjusted all the faders once more to balance out some of the instruments, editing some of the panning on things such the percussion. I had ran out of time before I was able to add the parallel compression track back in, so having learnt my lesson from Wednesdays mix where I rushed the reverb and ended up drowning the vocals a little,
instead of throwing the parallel compression track up at a level I thought might be okay, I decided to cut it out as the mix was sounding alright as it was.

Time management

I found that once again, mixing the drums took me quite a long time, spending nearly 30 minutes in the end getting a nice balance. With that said, I did spend a fair bit of time working with some processing FX on some of the drum tracks, which improved the sound a fair bit, but were not quite necessary. If I was extremely pushed for time, I could have instead just quickly panned and adjusted the levels of the faders.
The reverbs didn't take too long, however by the time I had worked through all the instruments, panned and applied reverbs, I was running quite short on time.
I quickly changed some of the panning a bit, adjusted some of the levels of the buses and time was up.

Conclusion

I feel this mix went well enough, and I wasn't exceptionally pushed for time but I still barely went into any depth with my FX usage on the tracks. The balance isn't great, with the drums being present, but not having much power over the mix, just providing a backbeat to the other instruments. I would have liked them to push through a little more during the busy sections of the song. With that said, the drums in the song are very dynamic which could have contributed to this dilemma. I liked the bass, with a quite loud tubby sound, having decent note definition and plenty of power without taking focus away from the other instruments. In my effort to hide the lines of the Keyboard track that I dislike, I seem to have added too much reverb and kept the track too low, making it sound out of place in the mix. In general the reverb doesn't sound too bad, however there could be minor tweaks made here and there.

I will finish my last mix up tonight, which will be a Death Metal song. I'm looking forward to it, however I know for certain it will prove a hard challenge, despite the fact I listen to similar music often and am fairly engaged with an online community of engineers dedicated to audio work within that realm.

Thursday 2 June 2016

Mix hour #3

Todays mix was a hip-hop song:
Side Effects Project: Sing With Me
(https://soundcloud.com/rufus-ambler/mix-hour-3)





I decided today I would challenge myself again today, and I certainly achieved that with this mix. While the stems themselves weren't challenging, the sheer amount of tasks I had to complete in an hour was kept me going 100 mph throughout the whole mix.
While I haven't had any experience mixing or recording in this genre, I listen to fair bit of early 90s hip-hop so I am fairly familiar with the music style.

The mix process

I knew how pushed for time I would be almost from the start, so throughout the mix I was trying to be conscious as to not spend too much time on things that would only make a minor difference to the mix.
I began by mixing the drums, bringing in the tracks one by one, doing panning on the fly and adjusting the volumes so that the kick and snare tracks were the forefront drum sound. I applied a little EQ and compression to the kick and snare when I was starting the mix, but soon realised I didn't have any time to do this for any other tracks, bar the vocals and maybe the bass.
The bass was simple enough to mix in, adding some slight compression before adjusting the levels, allowing the kick to come through nicely.
I moved onto the keys bus next, harp and synth tracks. The harp wasn't too hard to fit in, but the synth line took a little more to time to figure out how to tastefully add it in.
Now came the vocals, by far the hardest part of the mix to work on. When I worked out what vocal lines were doing what, I found that many were double or quad tracks of the same vocal lines. This made the mixing process a little easier, but didn't reduce the time it took to balance the vocals as much as I hoped it would. I started off by balancing the two main vocal lines, then subtly adding in the double tracks for dynamics. I added in the main vocal chorus lines, panning the 4 tracks, varying the stereo width. They weren't easy to match with the verse vocal lines, as the vocal delivery was very different. The backing vocals were laid out in a question-response format, and sounded closer to the vocal style of the verses, so weren't as hard to mix in. As there were 6 tracks per backing vocal delivery, I took advantage of this panning them 20, 40 and 60% left & right, giving a wide stereo image. When I was adjusting the faders of the vocal tracks, with the main two main vocals aside, I worked by adjusting all the tracks of a certain vocal line at once, which saved a lot of time. Once the vocals were mixed, I gave brief thought to FX, adding a slight bit of compression to the tracks. Because time was such a concern, and there were many similar sounding doubletracks, for the compression of each vocal line, I solo'd an individual track and applied a compressor to that, then copied it over to other relevant tracks. In most mixes, this would be a very lazy approach, but I feel it helped achieve my goals without using up too much time.
I quickly moved onto the samples, adding them in a similar fashion to the tracks on the other buses. Some of the samples were harder than others to mix in, and one I felt had no place in the song so I muted it and left it out.

As by this point, I was running very low on time, I quickly got some rough settings of my reverb and delay tracks. Next I threw up some delay and reverb sends to some of the vocal tracks, the snare tracks and the guitar loop.

Time management

As i've said already, I knew I had barely any time for anything other than things that would change the mix drastically, so I had to focus my time on the important tasks. The drums took a while to mix and pan, spending a good 25 minutes on them alone, without even spending any time thinking about any major usage of processing FX.
The bass and keys tracks took virtually no time but following that was the most complex element of the mix, the vocals.
Due to my time saving methods, I managed to get through them faster than if I had been working with them each as individual tracks. But I was still extremely hurried, with every minute spent being of the upmost importance.
I was left with something like 5 minutes to decide what I felt needed reverb and delay and then apply the effects to them tastefully. With the amount of drum and vocal tracks this was a huge effort and I had to stop before I had even applied reverb to all the tracks I planned, because I was out of time.

Conclusion

I was far from happy with this mix, listening back to it, it's best elements are barely acceptable..the worst a trainwreck.
As can be seen in the picture, the some of the buses were clipping, however I had no time to sort this out, so all that could really be done was to quickly drag the mix bus down to ensure the master would not distort.
I had some minor issues with how the drums sounded at the end. The kick and snare sounded okay and cut through the mix enough but if I had more time, the other elements of the drums could have used some adjustment to give them more presence in the mix. The bass and keys and guitar were okay, but the vocals at some points were dire. The verses were okay, being present, if slightly overpowering, however the choruses were mostly very quiet in comparison. It was not helped by the reverb, which brings me to my next point..
Okay, I did have less than 5 minutes to put reverb on a fair few tracks, but it doesn't stop the fact that the reverb sounds terrible in most cases. The reverb on the guitar loop sounded nice, giving it some ambience, and the reverb on the snare was acceptable, but still too much. The vocal reverbs went from bad to shocking though, with the verse vocals sounded much too airy, but still had a forefront sound. The chorus vocals however..a different story. The main vocals sounded too dry, but this would have not been too much of a problem were it not for the backing vocals, which sounded like they were recorded 15m away in a wide hallway.

While I am not at all pleased with the outcome of this mix, I have to remember that this is part of the challenge and it is a learning experience. I know that I bust a gut to fit in time for all the usual elements of a mix, which I barely had time for, so in a way this was a lesson showing me how a simple composition, with many recordings can be far from a quick mixing session.

I've been sleeping little over the last couple of days (not getting to sleep til 5am mostly) so I think I will choose the next song to mix tomorrow.

Wednesday 1 June 2016

Mix hour #2

Todays mix was an ambient electronic song:
Bravestar - Downtempo
(https://soundcloud.com/rufus-ambler/mix-hour-2)




Holy shit, this was a tough one. I am kind of familiar with the genre of music, though I wouldn't say I listen to it much, and certainly have done no work in a style like this, bar one little side project I worked on about two years ago which incorporated some ambient electronic style synths.

The mixing process

This mix was very different to any other mixes I have worked on before. Due to the composition process that usually goes into electronic music, most, if not all the tracks have had their tones created and EQ & compression applied, which means it saves me a job as an engineer. Sure, some tracks could sometimes benefit from a little additional FX work, but in this particular case I had no time for that and didn't really see the need anywhere (as can be seen in the photos)
Instead of FX work, I only worked with panning, fader levels and time based FX sends.

I started off working on the drums, combining the three kick tracks and making my way down the tracks in the drum bus. Once I had adjusted the levels of the tracks, I set some basic panning up on a couple of the tracks and brought the main loop into the mix. I checked that the different drum elements were decent, before moving onto the bass. I brought the bass in and adjusted the fader then moved onto the keys. I went down the tracks in the keys bus, adjusting the faders and setting panning. I found I didn't find a place in the mix for the synth track, so got rid of it.

Then I went back through all the tracks I had worked on so far, and gave some thought to automating panning, as can be seen in the picture. I didn't automate many tracks, only some elements that weren't consistent throughout the song. I used some pingpong panning for some of the tracks, however when there were multiple 'extra' elements, I tried to keep a balance of interest between the left and right of the stereo field.

Before moving onto the FX bus I sent some of the tracks to the reverbs & delay to add ambience to the song.
I repeated the same process that I took on the other buses when working on the FX. I found partially due to time constraints and partially I didn't think they fitted the compositon, I cut out around half the FX tracks that were in the stems. Finally I added a touch of reverb to some of the tracks.

Time management

Time wise, this was a very challenging mix. I found that once I had finished the panning automation for drums and keys, I had a measly 20 minutes remaining of my allocated time. I chose to do the reverb and delay sends at this point because I felt that would be more important than adding in the FX tracks. I was partially right in this judgement, given how many FX tracks I decided to cut out, however I was pushed for time towards the end, having less than 10 minutes to add in and adjust the volumes of FX tracks.
I just about made the time limit, which is a bit of a miracle.

Conclusion

As i've mentioned already, this was a particularly challenging mix because of it's genre, so I approached it without much of a clue what to do. When I heard that the tracks needed little to no processing FX, I underestimated how long it would take to get the mix into any finished state. In a way, the panning automation was the main 'effect' I used to give light and shade to the song.
The fact I took so many tracks out could be an issue if I were completing a mix for a client, however as this is only a practice the decision is obviously up to me.
I'm fairly pleased with the mix, but I know for certain there are many things that I could have spent more time on, or tried out; the particulars being trying out some processing FX on some tracks, spending more time trying to integrate the FX tracks and I could have spent more time on the reverb and delay sends.

I haven't decided what i'm going to mix for tomorrows challenge, but I know for certain at some point in this week I will work on a metal track, hopefully something along the lines of Death Metal.