Across the Transom: March 2023
Bad Projection, Ribbon Mic Entryway, Patch Notes
Gahlord Dewald :: 04/02/23 :: Mānoa, O‘ahu, Hawai‘i
On sampling hardware instruments
When I want to digitize a hardware instrument or modular synthesizer patch I set up Logic's autosampler, pick what kind of intervals I think will work, then let it do it's thing while I go make coffee. When I'm done drinking coffee, autosampler is done and I listen to the results. If I like the results, then I'm done. If I think it needs tweaking then I tweak the settings and go make more coffee.
I've done the process a couple times now and I rarely have to do a second pass. In those cases, I make more coffee just because I like coffee.
The fidelity of the samples, the nature of them (especially modulations etc), and personal desires for the end goal make it mostly impossible for anyone other than you to decide what settings are right for you. It's really something you're learn by doing.
The nice thing about autosampling is that the active time commitment is mostly the setup. Then the computer and hardware grind for a little bit while you get to drink coffee or stare out the window or do the dishes or whatever else needs doing.
Here's a good tutorial on using Logic's Autosampler.
Notes on a patch
Following are notes on a patch made in response to the Disquiet Junto prompt #0586.
Three tracks are 1) tones: root and partials from Patch Chord, driven by a simple sequence on René, phased entry via Batumi LFO into NSL-32s, little filtering via Instruo I-ō47 and out via aggressive limiter in Pulp Logic Exit Strategy; 2) reverb of said tones via Recovery Dirty Murals, full wet; 3) a sandpapery shaker via Meng Qi DU-Karp filtered via Bastl Cinnamon bandpass.
Additional modulations:
A pretty slow Sloth 1u tile adjusts the feedback of the reverb, René sequence subtly alters Instruo filter and pitch of the DU-Karp, Patch Chord is modulated by an unrelated LFO (cyclic skew) adjusting the duty cycle of the oscillator driving Patch Chord (winterbloom Castor & Pollux).
Erasing:
I’ve been learning Teletype. So I wrote a Teletype scene to handle things. TT takes in the pulse and controls a gate for each of the tracks. It flips a coin every pulse and sets the gate to open or closed accordingly. Each track is delayed a little bit off the pulse, at different amounts. This helps keep things happening mostly at different, more rhythmic variety.
Additional processing:
Aside from normal mix/eq/etc, I have a filter that I love which is noisy and I even love the noise. So I recorded the noise (and amplified it quite a lot) and hand stitched (it was all I could do to keep myself from dumping it to tape and then doing it all with a razorblades haha) it into all the blank spaces so there’s always some sound happening.
Along the way I experimented with different lengths/sizes of erasure and this was pretty interesting and fruitful. If I was further along in knowing Teletype I probably would have used different lengths of on/off for each track as well.
Conceptually it got me thinking about other methods of erasure (musically but also culturally/societally) and this thinking will almost certainly result in another work over the next few months, which is why titled this a study–it’ll probably lead to something.
Re: Building a high resolution human-computer interface in eurorack
If your current interface has up to 16 spare inputs that accept DC:
- NW2S balanced/transformerless O:16 to DB25 if your interface can take DB25 input. $215
- a bunch of Pulp Logic Att-Off. $30 each
- some pulp logic mix-B: $45 each
- Erica Black Joystick 2: $259
(All prices new, used will probably get a meaningful discount)
Depending on how much you need to do in one pass—i.e. is this a performance system or a studio recording system?—you might do great with just the O:16 module and 1 Black Joystick.
Each Black Joystick is capable of 4 independent channels of 2 variables each, with looping/hold/return to zero. Very performable.
Add a PL 1u row with some Att-Offs to help get the outputs trimmed and boosted/cut to taste (or use unpatched as DC knob interfaces), or Sum, or Mix-B etc.
Re: Audio Interfaces
If I was starting from scratch I think I might just go with a Cranborne, though I prefer keeping converters separate from analog gear because converters have a shorter lifespan before obsolescence. But I honestly think we might be close enough now, certainly 192khz is plenty until more daws can handle dsd.
Much of it comes down to how many channels you need at once. As clock rates go up, channel count goes down on the Cranbornes. I don’t find this to be a problem, but it does sometimes require forethought. At 192khz you get 2ch of I/O—enough for a fun small mastering box if you chain/wire the slots rights.
At lower clocks you can use more channels of I/O for tracking.
It would be great if the fate of firmware was handled responsibly in bankruptcies. Like a court-enforced maintenance order on whoever buys it or it defaults to open source.
So many things become futuretrash due to this very solvable problem.
First attempts and video conversion with a video synth
I have a cheap-o Component-to-HDMI box for output to my Atomos recorder which then duplicates to what must be an early/mid 00s flat panel because it has component, composite, and HDMI in.
I use a mess of cheap-o Composite-to-HDMI into HDMI-to-Component boxes to get from Structure (or the CBV001 I suppose) back into RGB via the VU003B.
I’m going to be trying an RCA-to-1/8th to use the Structure as alpha only to use the “In” jack of the VU003B. Though not documented too clearly, that Syntonie is sort of two things at once, not sure what will happen to sync but I’m going to give it a try.
Ribbon Mic Entryway
The Bumblebee DIY ribbon mic kit sounds very nice and is a super simple build. The AEA RPQ500 are fantastic. Lots of clean gain and the the EQ stage is very well dialed for ribbons. I absolutely love mine. But anything with 60db of gain or so will be ok to get started.
Or get some Cloudlifters--they turn phantom power into clean gain. This has two benefits: clean gain and prevents you from frying your ribbon mics with phantom power. Having a set of Cloudlifters on hand is always nice. I think of them as one of those necessary studio utilities like DI boxes etc, just truly good to have them around.
If you’re going for cheap, in order to save for better gear in the future, then go for a cheap pre-amp and use the Cloudlifter for gain. I might suggest taking a look at the Camden500. It has the silly mojo knob which you can turn off (or leave on because, like who doesn’t want mojo amirite?). Once that’s off it’s an exceptionally clean preamp with 68db of gain. It’s in the same price neighborhood as the Golden Age 1 and I would expect it to be a little cleaner in sound (with the mojo off). They'll still be useful after you upgrade to something like the RPQ.
You can always EQ in post so save that for last in your upgrade path. If you get to an RPQ then you won’t need an EQ for ribbon-specific adjustments anyway; no need buying something you’re going to dump later. Nice EQs are often one of the more expensive things in the signal chain.
That’s how I did it in my own system anyway: bumblebee DIY ribbon + A/D converter’s built-in preamp + cloudlifter, saved for an RPQ, then ever more ribbon mics, didn’t look back. Get the input side of things the way you like, then any EQ you apply will be working from good source. Good EQ on less-than-you’d-like input signal runs the risk of highlighting the things you aren’t into. Silk purses and pigs’ ears etc.
Theater owners were excited because digital projectors could be programmed to run on their own without human projectionists to start them up and switch reels.
The Dithertron is web software that turns images into mangled dithered images.
Reminds me of a trick a friend told me about for working with an improvisation partner.
They would record long jams and then listen back to them each with a midi continuous controller set to record an automation track. As the jam was playing back, they would each manipulate their controller, moving the automation data up when they liked what was happening and down when they didn’t like it.
Then they could quickly visually compare the two automation lanes to see which parts they both liked, which parts only one of them liked and which parts neither of them liked.
Gate-keeping, and the suggestion that something is “for us and not for you,” exists in every musical space that I know. It is not a value that inheres in music per se, to whatever extent it even makes sense to think of music as a bounded object. It is a value that inheres in people, and in the physical, social, and discursive spaces we construct. Music is merely a thing that happens in those spaces.