When the production of this album started for good in December 2013, Eclectic Blah had already ended more than nine years ago, so this was an interesting thing: working on material from a different era so to speak.
Still, it wasn’t that the material back then got recorded and then stored away in the vault, never to be listened to again. As the idea for an album had always existed, even when Eclectic Blah still existed, there was of course preliminary work already done: from most of the recordings (which were twotrack recordings of sessions and concerts, and multitrack recordings of concerts), some producer’s favourites had already been selected, and some mixing and editing had already been done.
A first step was of course to define a rough goal for the album, to be able to properly select from this large body of work. I decided to limit this to high-quality multitrack recordings, and also to maybe more accessible, song-like structures, as opposed to those epic, long-drawn tracks that sometimes…happen.
With those selection criteria, already a lot of recordings could be eliminated: sessions didn’t play a role (due to being two-track), and all of the concerts preceding The Last Days of Veedolkeller (and some after that) were out. And for the remaining concerts, I had already a set of tracks for which roughmixes existed.
Based on this, I prepared a shortlist with 20 tracks on it, drawn from all the sources that also made it onto the final album. Next, I started a decision process as to which tracks go onto the album.
I personally feel that a good running time for an album is roughly between 40 an 60 minutes, with the optimum being rather at the lower end of that spectrum. So the task was to find tracks that represented Eclectic Blah well in its eclectic spectrum, which worked well in a “more than the sum of its parts” kinda way, and which didn’t exceed a playing time of roughly 60 minutes, but were best in a 50ish area.
It was clear that with 20 more or less good tracks, in their current versions clocking in with a total of almost two and a half hours, choice would be a challenge, as would be to find clever edits to reduce length while leaving the musical message intact.
Clear Must-Haves and Last Picks
There were a few tracks that were clear must-haves, at least for me. These include of course those a little atypical ones for Eclectic Blah’s sound (Dreams of Hesse or, at the other end of the spectrum, Harvey Wallbanger come to mind), as well as those best representations of a typical Blah sound (such as The Porcupine). Including all the considerations from above, I finally rated them all on a scale from 1 to 4, and then made the cut after “2”, leaving me with nine tracks and a playing time of 73 minutes, only to include Driving Home Slowly (another nine minutes) as well, to have at least one track from each source.
This was the time when I started working on the actual mixes for those tracks, and defined the ground rules for editing and mixing as I went along. Choices of stereo positioning were relatively simple; I defined a concept that worked well for all the lineups encountered and didn’t result in musicians jumping over the stereo field from track to track. With regard to mixing, I consciously decided that while tracks from the same concert would obviously be mixed similarily, I would embrace the fact that each concert sounded different from the band sound, and that the album should reflect this.
There were no big surprises in the basic mix: I tend to like both bass drum and snare drum with a two-stage compression, first a compressor with a ratio around 2.3:1, attack in the 12ms range and release around 100ms. Of course, genre-typical boosts, dips and cuts are applied. The overheads get a boost via a high-shelf and nothing else for now. Bass guitar usually gets a raise in the bottom end (and possible sharp dips if the venue had resonance issues), plus a not-so-fast compressor; guitars don’t get a high boost, synths tend to work well the way they are, and percussion and sax is always specific.
There’s a reverb for everyone (typically a REVerence convolution) and another one for the snare (sometimes gated). Everything pretty normal, and using Cubase’s onboard tools where possible.
A little bit of trick is ducking the bass guitar (or rather the bottom end, ca. below 140Hz of it) with the bass drum, a technique which helps to have both a prominent bass drum and a bottom-heavy bass guitar while keeping the rest audible.
Ground Rules for Editing
As for the ground rules for mixing and editing, I consciously decided on a “nothing is banned” approach, but with the goal of keeping the integrity of the live experience in mind.
Now the most radical changes are of course overdubs, and in total, there are three of them (all of them for sound quality issues).
We have the first one right in the opening track, Spheres. From that concert, the drum microphones in general were crap, and the recordings sounded accordingly. This could be solved somewhat in the mix for the overheads and the snare, but not so for the bass drum. After a lot of (futile) attempts to salvage something, using all the arsenal from signal processing, I decided to use the same approach as on A Dance Song (from Verschluckbare Kleinteile): Manually sequence the bass drum to a MIDI track and then use that track to trigger a sample.
Overdubs two and three were similar in nature. On Dreams of Hesse, the otherwise really nice 70s latin rock feel was destroyed by the latin percussion, which sounded like an Eighties’ plastic trigger pad had triggered sounds from an Eighties’ sampler (no surprise, as this was an Octapad II triggering an E-mu Proteus 1/XR+). The same approach as above was used here as well, although this was a little trickier, as there were different instruments in the source (meaning I could not simply “paint” the MIDI part by looking at the waveform), and also because I could make use of the bigger choice of samples in the source used (Battery 2) and e.g. differentiate between left/right hand samples.
The third one was also the simplest one – it was the electronic bass drum in the intro for The Porcupine, which was mainly very noisy. I spent some time looking through the library of the Proteus VX, and that being unsuccessful, took one of the hits in the recordings, denoised it, loaded it into Battery, and off we went.
Edits were first and foremost driven by the goal to get the tracks more coherent and dense, i.e. to make that what sounds great in a live performance of a radically improvised track also great on an album. Most of the time, those edits were linear. Usually this meant taking out another chorus, another repetition of the theme, or cutting that in-between vamping section in half. And by doing just that, the musical appeal was not only enhanced for the album context, but also the total playing time was brought down from more than 80 minutes to 52 – in the case of The Porcupine more than halving the original duration.
There were, however, also trickier edits that left the linear editing world far behind. Perhaps the best example is Luke’s first solo in Horst’s New Condominium: at this point, Luke and I had started to play a solo at the same time, his a rather gentle one, mine a hugely overdriven organ sound that was also too loud. After a few bars, we luckily noticed and moved into a nice trade-four (which, on its own, just wouldn’t work), but what to do with those clashing solos?
Yes, Luke’s solo was much nicer, so simply muting the synth track would be the first idea – which also didn’t work due to the organ bleeding into the overheads. So off to find a passage in the overhead track (best from a part of the original recording that had already been cut out) and paste that in place of the correct overhead signal. Yes, this overhead signal was from a guitar solo, so you also have some guitar bleed – ending up in something which sounds like Luke playing call-and-response with himself!
From the side of the mix, tricky things were applied sparsely. On Spheres, there’s that filtered delay on the left-side snare during the mid-part which makes the interplay between drummer and percussionist all the more interesting. Horst’s New Condominium has some really ugly amplitude-modulation-kinda effect applied to the snare. In the end of the same track, there was digital clipping at the beginning of the coda, as I had been driving that distortion effect on the organ too hard. Fortunately, the dry signal from the synth was recorded separately, so it was easy to choose that instead and apply distortion and leslie to it. In the coda for Harvey Wallbanger, I got a little crazy with the pan positions, as different from the normal “synth left, guitar right” setup, here we have synth, guitar and guitar loop recorded by synth player: the synth here sits more center, with both guitar parts panning around it. Finally, Solid State just called for a drum sound that was a little more edgy. I accomplished that (albeit only during a late mix revision in parallel to the mastering stage) by using a really aggressive compressor on the drum bus, but only bringing that in at -18dB.
Already during the early mixing phase, I had begun to establish a master setup in parallel. At first, this was very helpful to play with track order, track spacing and track levels, and for all of this, I had established a working setup relatively early.
What I wasn’t able to do, however, was to get a proper sound through all of the album and all of the tracks. Caused simply by lack of experience on my side, perhaps the most interesting find for a track not translating well was an earlier version of Spheres which on the studio monitors (which go down to 16Hz) had not enough bass guitar while on a cheap home cinema system, the track was nothing but bass guitar. Also, I sometimes was already ending up in the rather unscientific thoughts like wanting In Deep Water more “warm” or Tiny Bugs more “edgy”, both of which I wasn’t able to accomplish that well.
To make matters worse, I was already hugely happy with the music, so applying second-rate mastering was not that good an idea. So finally, already after the E5 milestone (meaning: mixes are done), I opted to outsource mastering.
A wise choice. Starting from a first reference mastering run, the material immediately sounded more “tight” and more “open”, had more air, a more compact bottom and lost its harshness present in some places. The mastering engineer, Thomas DiMuzio, and myself went through a rather agile period of cooperation, exchanging master revisions, issue tracking lists, mix revisions, and at the same time reviewing the other’s artifacts and discussing the more non-quantifyable aspects such as “should sound more like a building with light-blue walls and tall windows”.
An item that I didn’t discuss so far was the cover art. For the front page/album image, I opted for the Eclectic Blah logo and text, designed way back when by Martin Fischer. The album booklet was loosely based on the design of the Akustik Kies album, bringing in photographs from each of the featured performances. I opted not to employ the services of a professional designer; in hindsight, maybe not the wisest of decisions.
Schedule-wise, I had “Q1/2014” specified for the release date at the project start, and was able to target a “end of February” date by end of January. Moving into the last sequence of milestones (and deciding on Sunday as a good day to release an album), we reached the P6/P7 milestones on February 17th, and that paved the way for a P9 (release) on the following Sunday.
With mixing, mastering and cover art taken care of, and the actual music already taken care of about a decade ago, I was even able to think a little more about marketing in the widest sense. Yes, there’s some print reviews, there’s a proper documentation on the blog (including this text), but the way I feel it, this is still something where I’m lacking – so of course any support (simply by spreading the word) is hugely appreciated!
Eclectic Blah was an interesting project, and also one that people sometimes ask me about. Now, there finally is a proper album, not just a set of roughmixes and low-quality twotracks. That was the goal, and this goal has been accomplished.
While being on the path to do so, I was also able to learn a lot. Mixing music that is different (from a technical standpoint) than most of the stuff I did through all the years was part of that learning experience, as was working very closely with a mastering engineer. Yes, I had done that already on Rückwärtsfließpreßverfahren, but that project was simpler if not from the mastering, then from the producer-mastering-mixing interface.
And on the side, I did quickly define and refine a process description for producing a live album, which can also be used as a basis for other forms of album projects.
All in all: go get the album! ;)