July 24th 2014 update - remix progress, hardware and software headaches, and other distractions...

It's been two weeks since the last post, and although I've had a big increase in my music productivity it's not up to optimal. I've been mostly spending time on the Muse Receptor issues as I got the SSD back from Muse Research and it's functional, but now it's getting the software licenses transferred from the original owner (Michael Bluestein, keyboardist extraordinaire) to me. Michael has been an excellent person in this regard - the Receptor was sold by the band through a 3rd party on eBay to me, and after I found his name and email address I just cold called him (email) and asked how we could manage this? He's since gone onto iLok website and transferred the licenses on the system to me, plus is working on getting the s/n and registration info for the version of NI Komplete that was preloaded on the unit to me. He's been excellent in making sure I'm up to date on progress towards getting everything done, and I'm hanging on to the original iLok key that was in the Receptor - can you believe that iLok charges $100 to transfer ownership of the key to a new owner, while a brand new iLok only costs $30?? Anyways, you can guess how I'm dealing with that :)

I'm writing this a bit out of order now since the Receptor item is first and foremost on my mind (in the music space at least). Once that's resolved, and I figure out how to use the Uniwire feature that links PC to Receptor (to offload VST workloads and bandwidth onto the Receptor system), then there should be a marked increase in the output on this side.

Next up - Remix status. I've done some early work on two tunes from The Sound of Thoughts: Lady in a Cage and This Mortal Coil. Both have some rework to do - vocals and guitars to be redone, some bass parts as well, maybe touch on some keyboards as required. With the new VST plug-ins I have from iZotope, Universal Audio, Waves, IK Multimedia and others, being able to really turn the drums into stellar sounding instruments is a big plus, and each of the tracks above have had enormous benefit by it. The iZotope plugs in particular have been fabulous as I can really have a all in one solution for gating, eq, compression, etc... that is tailored to each piece of the drumkit, and in fact for each instrument and vocal part in the projects. They're also very efficient in terms of CPU and memory resources such that I have multiple instances of plug-ins loaded and processing, but my CPU and memory load never exceeds roughly 50% of available. This is especially awesome considering the sheer number of tracks and VSTs loaded for Lady in a Cage - I'm guessing 32 tracks total with 18 tracks both mono and stereo audio, plus 14 MIDI tracks, and then each of the VST card slots representing instruments, effects, etc... which is a considerable load on the system. Back in 2004/05 I had to do a lot of stuff like freezing or pre-rendering individual tracks to get around the bandwidth throttling issues on the PC I had at the time, so to be able to have everything functioning real time and not being a load on the CPU or memory - a godsend.

I have a version of an early remix for "This Mortal Coil" - you can check it out here

SO here's my critical side letting loose:
This song badly needs a new vocal - I tried with the radio voice setting to reclaim the original vocal, but there are so many things that feel wrong about it, mostly in the performance itself. They always say hindsight is 20:20 - the entire delivery and how the phonetic rhythm interplays with the music just doesn't work for me.

There are good things about the track - drums are now really sounding great. I have a very nice Chapman Stick on Sample Tank 2 that is a big improvement from the live track, and I'm processing using my favorite VST and preset for all bass - the T-Racks Compressor with Bass Tube Compressor. I've got Amplitube 3 on the clean guitars through a Vox AC30 on one channel, and a Boutique amp on the other - both have some analog chorus on them to make them sound a bit richer. The lead guitar was originally tracked through a POD Pro rack unit and printed to tape - I've changed up the delays and reverbs for all f/x channel strips and the lead now has a bit more richness to it. Keyboards are all pretty standard fare - when I first tried to work on this track in 2010 I had Paul's Korg Triton rack synth and so printed some of the MIDI tracks to audio from the Triton. This resulted in a lot of thickness of pads that really muddied the 2010 mixes, and so I've been going through and automating and arranging the various tracks to enter and exit at key points in the song. As I said before, this is still an early remix, but it's a significant portion of the way down the road towards a final with the ideas and efforts being tried out continually.

Right now I've got Pentelho Vermelho in Cubase 7.5 and ready to work on. The last remix I did in 2010 was just drenched with reverb and delay - way too much. Also, given that I'd just come out of the Fine Line project and the "more is better" sensibility with respect to keyboards and guitars, it's just got too much stuff in it. My procedure will be to bring the song back to the essential instruments that highlight the melodies and rhythms, then add enough to sweeten and enhance the song without burying the essentials. The TSOT songs tend to be far easier to do this with since PC imposed limitations, but even with new hardware the key to success on the remixes will be to Keep It Simple.

I had started doing some work on The Sound of Thoughts (the track itself) - this was predominantly a MIDI track that used my old JX-8P and Wavestation extensively, along with NI Battery and a bunch of freeware VSTs that I still keep as a part of my arsenal. Crazy Diamonds is a great VST that you can still find online - the amazing strings from Shine On You Crazy Diamond are reproduced spot on, plus there's a Jean Michel Jarre patch from Oxygene that is also a perfect recreation - which is the patch I used on this track. As with Pentelho and many of the other tracks I tried to redo four years ago, I tried recording the Triton audio for several parts, and now I'm going to need to decide to thin things out. The interesting thing I'm thinking about with this track though is for me to actually play the drums myself instead of using the programmed drums. I have the electronic drums and the KAT module which streams MIDI through the USB cable for the kick/snare/toms, and then the microphones to capture the hihat, ride and overhead cymbals. At the least, since the tune is in a 4/4 signature and has a consistent tempo, I should be able to do this fairly easily and get some practice playing the drums as well.

There are 12 total tracks on the CD, so there's plenty of work to do. Given the rate at which I get to stuff, it might be end of year or early next year before the material gets to mastering - I'm hoping that with the Receptor situation being resolved, and then starting to call up my mates to get together to jam, maybe I can get this particular effort done earlier...

Finally - the "other distractions" segment. Day job has me working on doing "presence" for our group and company at Siggraph in Vancouver, I have another event in just under two weeks - Seattle Buzz Workshop has me arranging hardware with Android demos to a developer audience. Other complications have also been taking time away from me - I promise to let people know if any change imposes itself on life.

There's another part of my day job that's been interesting - I have been reaching out to hardware and VST developers on behalf of Intel to discuss partnering up with them for future products. I can't say much more about this except it is a great path for me to go down for my day job as long as I can execute on the switch. Being able to influence the ecosystem in that fashion has always been a hope and a goal for me, so lets see how this will turn out...

Solving Big Issues - revisiting past recordings and setting up new projects for the remixes, and the problems they present.

Sometimes I start to write about a given topic with the intent of doing a simple issue/answer and final summary point posting, only to find that I've essentially written an entire chapter of a book as a result. This is one of those postings - born of the work I've been doing lately in my studio focusing on older tracks I have in archive with the intent to remix and remaster them. As a result though...

Here's some background about the production of certain tracks including the song "This Mortal Coil" from 2005 release The Sound of Thoughts...

Heinz Schuller was the drummer on most of the songs on Sound of Thoughts, and he provided great tracks in terms of sound quality. What we did with them back in 2005 though during mixdown sessions was less than stellar (which to be fair the mixdowns were pretty rushed, wanting to get the CD mixed and mastered the night before the release party, and I still ran over to my workplace and used their bulk CD burner to burn off 100 copies the afternoon before the show). I guess trying to mix down an album until 5 am the morning of a publicized show releasing the CD might not have been such a great idea...

There were three tunes on the CD which were really tough from a playing perspective - "Lady in a Cage" which had a ton of meter and tempo changes over a 19 minute running time; "Pentelho Vermelho" again had really tricky tempo/meter changes all over the place (I wrote in a ton of polyrhythms with places where accent notes played in 4 under a motif riff that was a compound time signature and a section where the guitars played offbeats in stereo against each other); and then there was "This Mortal Coil" with it's 13/16 verse and 7/8 chorus using the arpeggiated guitar as the metronome driving the song. All the contributors on the CD had to play their parts against the heavily sequenced demo songs which I had essentially sequestered myself away and wrote in their entirety during the month my wife and son Thomas had gone to visit family in Brasil (Jan 2005). Not having a band or any collaborators to bounce ideas off of was tough - I have a tendency towards ADHD, meaning I get bored quickly and abandon things easily (or used to anyways). Lady took months of effort to write and put together the 7-8 sections that were mostly individual ideas or audio sketches that were sitting around and then Kragle'ing them together in a fashion that wouldn't sound like they were Kragle'd - the original piece of music was 8:10 long and written in 1987 but I always felt it was incomplete. Pentelho a similar thing - a ridiculous riff that was put into different band contexts to lend dynamic to the tune, with a ridiculous piano bit to open (written with a B3 sound originally trying to be a knock off of Keith Emerson's Karn Evil 9 opening bit). Mortal Coil though - a mood piece intended to have a feeling of dread. Lyrics were (as always) over the top dark - and very awkward to sing. Nothing about the three songs was very forgiving to anyone who would play on them.

Lady and Pentelho both had issues with drum tracks w.r.t. meter - it drifted around a lot, and at the time when we flew the tracks into the projects, instead of trying to adjust the project meters to fit the drums, we performed a submix and render of the drums with eq's and stereo placements, then flew the stereo submix into the project and performed the "morphing" of the time scale of the drums to fit within the projects' rigid meter structure. This meant the drums got gate and compression processed per piece before the submix - any changes in the drum processing would require the entire process of submix, fly in and edit to be repeated if we were to keep the other aspects of the projects the same. For those particular songs (maybe one or two others as well) I have some pretty serious things to consider and choose from the following options:

1. do the same process with submix, import and edit as was done originally?
2. process the original tracks in the project, and edit the tempo track to match the drums?
3. get my friend John H to record new drum tracks for all affected songs using the original tempo map?
4. take the original MIDI drum programming and do some deeper programming to add dynamics and feel?
5. Just grab John, Paul and find a bassist (maybe the original bassist Brian T) to do full on recordings of new bed tracks - needs rehearsals and then studio time?

Option 1 would keep the final song with the same essential arrangements and instrumentation. There is enough goodness with the original keyboards to do this - although I'd like to redo guitars and vocals regardless of what option is done. Editing the drums to fit the tempo sucks because we lose some of the cymbal sustains in the cuts, but (and this will be an issue with #2 below) the drums drift around a lot between being late or early on the downbeat depending on measure... very tough to counter that.

Option 2 - as mentioned the drums drift tempo wise, so moving tempo track to fit means that either the other audio tracks get edited to fit the new tempo track, or we redo all other audio tracks. I can be selective about what will be edited and what is redone, but it's potentially 10x more work than #1.

Option 3 - John has excellent meter, and has a lot of experience playing to click tracked material. He also has his own home studio with some excellent soundproofed space and good microphones. I also have great mikes so between the two of us we can get this done. Two concerns though are #1 he has a proprietary recording system in a Zoom console - 2 actually. The files get recorded as WAVs and would require an import, but for the most part the process is doable. #2 He does have a HUGE kit and it will mean flying in 14 tracks into the project and any deviations in tempo would require revisiting Option 1 or Option 2 again. I don't think that's so much of a problem as with the original drum tracks though - we'll have to see how that goes.

Option 4 - this gives me excellent sounding drums by default as I can choose between any sample kits I have on hand. Being a gear pig and a progressive rock guy - the kit I programmed with was the 9 piece kit that is standard on most GM MIDI keyboard workstations, and some of the step edited parts are relatively impossible for any drummer to play verbatim which would make it obvious that it's a programmed drum track. Again, there can be a compromise solution where sections can be the programmed drums and others the Heinz tracks triggering the samples using Drumagog - that's a reasonable option right now.

Option 5 - rehearsing and recording the song bed tracks live and then picking and choosing keyboard/guitar/vocal parts to fly in or redo. If we were doing this thing as a full time gig, or we'd been playing the material out for months or years as a part time thing, then its a feasible option. None of us individually has the time to dedicate to this as a "band" - wife & kids, work, family... any individual or collective responsibility precludes that. I'd like to put a group together to play out, and that's my midterm plan since it's a ton of fun, but to be able to solve the remix problem with this is not feasible given how I want to use this exercise to kickstart new music work.

Bottom line : there is no simple solution. Right now I'm looking at Option 3 as the most feasible to get this done quickly and with highest quality. I can go to John's home and engineer/co-produce the sessions, we can discuss new arrangements for the material and get that printed to disk. We can also discuss taking the songs out to play live - once John internalizes the songs it will be easier to rehearse a band having him act as musical director. Everything lives and dies by the rhythm section after all...

After this long winded discussion, I now get to "This Mortal Coil".

Why "Mortal Coil" is being called out separately is because this song doesn't have the same issues as the others - the tempo stays the same throughout the song, the meter change actually feels pretty natural and only the most astute person actually identifies the change - usually the musician playing on the piece because of the stumble they experience when the change happens. Heinz played a part that was pretty close to the one programmed with accent variations and fills as appropriate, but the key was he was on top of the click the entire track. No submix export needed - we processed with the available VSTs to the best of our abilities, and with it being relatively easy to mix, it was one of the last tracks to get mixed down and mastered - aural and physical fatigue were the reasons why it didn't come out well in the final mix. 

Vocals were just awful - no time to iterate on editing the lyric to fit the song and being stubborn about it, I just sang what I had and only did 5 or 6 takes from which we edited a comp vocal track out of. Instead of choosing the pieces of each take that had the best pitch, we chose the one which best fit the rhythm - then processed the vocal as if the pitch really didn't matter to the final product. There are songs out there which have done this and it works (for reference: "Hey Man, Nice Shot" by Filter) - but in our case (Jay C and I) it didn't.

Why I am looking at this track right now? It's not the glamour track of the CD - hell, it's in the kinda shitty 2nd half of the CD that followed "Lady in a Cage", lost in the middle of a bunch of filler tracks that were written in haste and recorded with even more haste. "Peace, my Brother" is the only other track which could be called an honest effort in that sad collection of filler tunes. But "Mortal Coil" has more effort thrown into the writing than the production, and it's a very different piece than anything that came before. I can say really that "Mortal Coil" contributed to the skills that made it possible for me to write "Frantic" & "Mesmerize" from Fine Line. There's a lot left on the table with this one - fixing the tone of the drums, doing some arrangement magic by reducing the clutter, redoing vocals and potentially bass parts. Mastering the final mix preserving the limited dynamics but balancing the levels - not using a brickwall limiter for one thing... 

This will be the track that will tell the most about how big an improvement is possible from this exercise overall. Learnings here will transfer to all subsequent songs and CD projects, general issues identified and remedied, and a determination that will help set how long this entire back catalog effort will take. I will get to really learn the new software packages and the new hardware that constitute my studio environment. Playing and singing chops will get dusted off and polished up - make no mistake, this is not simply killing time waiting for inspiration for the next project. The next project will require serious chops to execute on and woodshedding is not an option - there's too much to do at the day job, with the new home, the family, etc... 

Paul and I have discussed on occasion how the mixes for Fine Line tend toward the cluttered side - lots of tracks forming big blobs of information where many keyboard timbres layer with multiple guitar parts to form an indiscernible wall of sound in a wide stereo field. In the last two CD projects I fell into that eyes wide open and oblivious to it because I didn't want to throw away work - it took time to record that keyboard or guitar part, all the takes and what not to get it right... "cant not use it now and waste the effort". This is the wrong way to mix audio material. To this day, I find I prefer the rock recordings from the 60's and 70's because you could hear what everyone is doing and there was space and power in the songs. Rush CDs from the last 3 releases most notably had too many overdubs sweetened with too much reverb and echo - the original "Vapor Trails" CD was vilified by audiophiles for being ridiculously cluttered and LOUD because of poor mixing and mastering decisions. The remix by David Bottrill is phenomenal though - he removed the clutter, mixed it with an eye to what's best for presentation and then mastered to enhance. 

I'm looking to reduce or in some instances remove the clutter from the original CDs. I anticipate it will be a hugely improved set of material that will result and am both excited and anxious for this. More to come in the future including some working versions I'll post to show progress and set the table for the final versions that will come out in the future. Cheers and have a great day!!