Studio.h CSR: How to scale your entire system + more!

Discussing some incredible modules that don't quite fit into the other forum categories.

Moderators: Kent, luketeaford, Joe.

Post Reply
User avatar
Kent
Large Member
Posts: 12304
Joined: Thu Mar 06, 2008 10:17 am
Location: Vienna, Austria

Studio.h CSR: How to scale your entire system + more!

Post by Kent » Mon Feb 22, 2021 11:00 am

This is a spinoff from another thread regarding the inferior 210e and whether it is worth buying a used one. The answer is "No" unless incredibly cheap, but even then... :foul:

Alright, tIB, I took a crack at how I would deploy a Studio.h CSR in your system. My post grew hugely from there. Although there are many ways a setup could go, I'm going to approach this largely within the context of your system (see photo below) as the example. I'll also point out the obvious options that would seem to make sense for my use & the most likely user cases.

The most important & powerful thing to get one's head around pertains to one of the many features (along with all of the improvements) that make the CSR superior to the 210e: You get to flash into memory as to what 'unity' (or default) is per node. This default setting per node allows one to scale any source to any destination within the limitations of the products. This setting, after you've done the work, is always there, ready to work with each preset. Scaling nirvana! It is memory that is independent of & additional to preset memory!
:wow:
And the loading of outputs (by individual varying CV inputs [looking at you, 259e... as usual.]) can be compensated for.
The 259e is an example of getting close enough for rock 'n' roll but never exact over 3-5 octaves. The 261e is better but you'll never get closer than 3.5 cents to being true. C'est la vie.

"Default 'unity' per node" means you don't have to save a preset for every change of routing. You just click the corresponding node’s button in the grid, press the encoder, and 'boom', your selected source is in tune and ready to rock with your destination. Or send to multiple destinations. Node scaling is stored in the CSR by a long press of the encoder. This can't be done with the 210e as it is: a jittery mess, and has no flash-memory-per-node onboard.

Where the CSR will unleash your creativity is getting the 223e to scale correctly to all oscillators without changing presets or fiddling with voltage adjustments. Hell, play 'em all at once, in unison or offset from one voltage source!
It will make the 250e's unhelpfully un-quantized "Quantized" feature actually work much nearer to correctly and definitely usable. You could possibly even scale the 266e's Quantized Random Voltages section to actual notes. I've not tried this yet as I'm working on a piece and will give it a go at some point although I've also the idea of using the 251e in S H mode for this. No idea if either will work though.

Tools needed:
• reliable, stable, and exact voltage source for your 1.2V, 2.4V, 3.6V, etc. I'm super lucky in that my 251e actually puts out the correct voltages (into a high impedance destination) without having to fine tune anything. Shocking.
As 1.2V actually equals 1.2V and 4.8V = 4.8 measured Volts, I just set these up as steps on a channel of the 251e. I don't trust it, so monitor with a voltmeter at all times. However, it's been solid. I throw in another step as 6V just to see if I can frustrate myself by asking for the unreasonable.
precise voltmeter. I use a Jones O'Tool and a handheld unit depending upon what is handy for me. Each of these meters is within a few thousandths of agreeing with each other, so all is good.
precise tuner . Again, I'm using a Jones O'Tool for this. Just stick with one tuner for the whole system setup.
patience and brain. I find this sort of thing to be fun. I'm sure some find it boring or frustrating as can be. The brain & patience come into play when scaling the 250e, but it is worth it. Something to keep in mind, and I hope it doesn't confuse the issue, but it occurs in my system: the voltage potential at the black system ground banana jack is different to my case's front panel ground. There is a difference that makes things sharp by enough cents to be audible. There's a longer story there but just be aware of this as you get setup for successful wrangling. Use whichever ground point gives you an accurate reading and stick with it.
• a small screwdriver or 'tweaker'. Non-conductive, if possible. This is for scaling your 258v, 261e, a 262v; if you have one.

You can easily do tonal stuff over MIDI with your 225e (or firmware card in MIDI Mode) & e-series oscillators. Nonetheless, I dedicate the principal oscillators to their own blue banana. See the diagram. You can always have the Mod Osc 'pitch track' the Principal Osc. You can also use MIDI on these and get rather more in-tune FM ratios. Save yourself the hassle.
But, hey, you gotta scale these guys up if you want to use them with the other voltages sources in your system (222e, 250e, etc).

And with that, off we go. One last thing: The CV input attenuators on oscillators can kiss my ass. I'm glad that the 258 has a dedicated input for 1.2V/Oct. I just dial the CV attenuators to 'full' so that I never have to think about these again. Yup, thoughtful Buchla-Tech-Pilot-With-The-Sharp-Mind; depending upon the oscillator, it appears that 1.2V/Oct is now violated; but oh, no... it ain't. Your skills with the 'tweaker' will scale it back under the panel. You could always put the front panel attenuators where you like and save it within the master preset we are creating. Up to you and makes little difference in usability (within reason)... unless you knock one of the attenuators whilst in creative mode. And then you are wondering if you can hit "recall" and whether you saved everything. In my system, I just turn the stupid AnnoyUVerter full open again if I bump into it or a stray banana hooks upon it.

Note that this doesn't work with the 259e nor 261e as the CV attenuators must be knocked back a bit back from full on in order to properly scale without using the CSR (more on that a few posts down). The 259e lacks a 'behind the panel scaling trimmer'; unlike the 261e, 258, 262v, and so forth. The CSR is now your scaling trimmer for the 259e... kinda.
The pair of 261e oscillators in my system are exactly the same. The CV input attenuator must be set to around 3:30/4o'clock in order to scale them correctly. They can't be made to scale as precisely as the Studio.h DPO but they can get to within 3cents of the target note for 4 to 5 octaves, which is pretty damn good.
See a few posts below for a tip on how to reduce the risk of knocking your properly scaled e-series & Preset Management compatible oscillators out of tune during a session.

The basic concept is to scale all of your oscillators, on their own, to be 1.2V per octave using a precision & stable voltage source.
Then put the CSR between them and your system’s voltage sources like: the 222e, 225e external busses, 250e, etc.
The CSR reconciles and scales all sources to the calibrated destinations (oscillators).


• Let the 258 warm up for 15-20 minutes. This applies to all analog core oscillators & filters. Switch on the system and reminisce about playing live shows whilst drinking a coffee.

• You don't need to set up a master preset to calibrate/scale the system but it sure is handy to have a reference point of sanity, comfort and laziness. Grab a blank spot.

Chapter 1. Establishing a Reference Point of Sanity (hopefully).

Scale the 258 first. It will be your reference point. No want for the CSR to be inline at this point.
- Patch in a stable and precise 1.2V to the desired CV input. I think it is the grey one as you have a Verbos unit, right? Awesome; no CV attenuvertor on that jack. I use my 251e because I'm lucky in that it is actually correct.
- Using the big "Freq" knob and its fine tuner pot shaft: Tune the 258 to the lowest note you can get out of it that is stable. Try to compensate by having the fine-tuner somewhere away from the extremes of travel. A1 should be fine. This is 55Hz. I aim for better than 1cent of accuracy here because I'm disordered and like struggle.
Anyhow, 0V should equal A0 in an ideal world. How much lower does one need? You always want to be roughly as low as all of your oscillators can go. But don't worry if one or more goes lower than the others when fed 1.2V. Use whatever starting point works: C1, C#1, D1,E#1... no, wait...
- Switch to 4.8V (or any octave higher than 2.4V). Using your small screwdriver/tweaker on the behind-the-panel trimmer: scale the oscillator to the exact octave frequency. If jumping up to 4.8V, then you should be targeting A4 = 440Hz. Be careful and, if you get frustrated, take a break. Don't smash the trimmer and possible nearby components like an ape. The higher the octave you choose, the more precise your scaling will be but also more frustrating to get 'just right'.
- Go back to 1.2V on the 258's CV input. Oops.. you are now either flat or sharp. Tune 'er back up to A1 with the Big Freq Knob and/or fine-tuner (not the scaling trimmer).
- Repeat the jumping between 1.2V & 4.8V until you are landing on A1 & A4 when switching octaves. Boom! You are now the Master of the Universe 'cause you be like He-Man! :omg:
- Try jumping to 6V in order to see if you land on a precise A5. Don't pull your hair out if it is a bit off. You could also have substituted 6V for 4.8V during the whole process but it could be asking a lot of some oscillators' designs. When you feel like it, you can return to this and scale to 6V. Or do it now, if your osc. can handle a 5 octave span.
- Scale the other 258 oscillator in the same manner. Use the first one as a tuning reference, if desired. Do NOT connect both CV inputs to your voltage source at the same time. This will likely load down the output of your voltage source and throw everything off. It'll drive you mad.
If you need to connect two destinations to one voltage source, use the CSR in between. That is what it is for. However, for this initial setup of the analog oscillators, we are not introducing the CSR.

If other folk reading this have additional analog oscillators in their systems, do this process to all of them first. I did this with my Studio.h DPO and Verbos 262V (thanks, Momo!)

Just remember that 1.2V on the input equates to tuning with the tuning knob(s). 4.8V or 6V = use the behind the panel scaling trimmer & non-conductive tweaker/screwdriver. NOT the Frequency Knob.


Chapter 2. Scaling the 261e & 259e to 1.2V/Oct (lol).

– The 261e & 259e should be scaled to being as close to 1.2V/Oct as is possible given the limitations.
• It is possible to scale the 261e much closer to being in tune with all-analog oscillators, like the 258, than the 259e. Depending upon the patch, this is likely not noticeable in most cases. For slow & evolving drone work with changing pitches, you're stuffed; especially in the low frequency registers.
- Scale the 261e as per the 258. No CSR in the middle. Use a precise voltage source.
• Do the Mod & Principal Oscillators
- The 259e has no scaling trimmer. You will have to do your best with the low resolution and finicky CV input attenuator knob.
• I've gotten nearer success by not having the 259e's CV input attenuator at full open. Somewhere closer to 5 o'clock worked better. I'm going by memory. Try anything from 4 o'clock to 5:30. You'll like have to take a couple of stabs at it. I think that I got 3 octaves, maybe four (sure... ;)) out of it.
• You’ll get somewhat close and you can stop there. The CSR will get you closer, but it’ll never be perfect.
- Be sure to save everything to your master preset.

Chapter 3. CSR Time!
CA656C72-4FA5-4286-8D4C-7D3DA3F9BDEB.jpeg
- Connect all of your desired CV sources to the CSR. In the system diagram below, I've selected:
• 2 channels of the 225e MIDI outs; E & F
• both Primary CV outputs of the 250e
• the final 3 channels could be anything that makes sense for what you use most often. Again, you can always scale things per patch or just wing it for things you don't use often. I use the 223e arpeggiator very often. Thus, mine is connected to the CSR so that I can just dial notes in by number. 480 = 4.8V every time. This is what the CSR brings to the system.
For tIB's system, I suggest as many channels of the 251e, 222e, 225e (for Mod Oscs), the 250e's time range CV :hmm:, and perhaps the 266e QRV section as makes sense. Whatever works for how the system will be used most often.
• One could also insert a 1V/Oct source into an open row and scale it to the more civilized 1.2V/Oct.

- Connect all of your desired CV destinations to the CSR.

In the diagram, I've selected:
• both oscillators of the 258
• 261e Principal Oscillator
• 259e Principal Osc

- Let's start with Bus E on the 225e going to one of the 258 oscillators. It doesn't matter as to which one.

• Connect Bus E (on the 225e) to the 258 via the CSR by pressing the corresponding node. In the sloppy diagram, I've got it connected to input C of the CSR. If the 258 is connected to output 1, then push the button on node C1. It should flash white on/off. Flashing indicates "selected".
• Click the "Attenuate" encoder once. Boom! Connection established more quickly and fruitfully than via Tinder. It should flash green on/off (because it is a connected node [green] and selected [flashing]).

- Send a MIDI A1 note to the 225e. I won't go into how to set up all of that nonsense for near proper conversion. My 225e is very far off and requires a LOT of downward transposition within the channel setup of the 225e. Once transposed lower in the 225e software, my channel E puts out 1.17V for A1 and 4.67V for A4. FFS, Buchla...
- Again, send MIDI A1 to the correct MIDI channel # on Bus E.
- Tune the 258 to exactly A1 (don't worry about the fact that voltage is off)
- Send A4 or A5 (whichever you wish) to the 225e. The 258 will jump to something, hopefully, near the correct pitch. We'll use A4 for our example.
- Use the CSR to scale this voltage.
• Node C1 should be green and flashing.
• Turn the "Attenuate" encoder up or down to get to as close to A4 as is possible. This might be fiddly, but that's how it goes. Get close.
• This should feel familiar now. You will not use the 258 scaling trimmer. You are using the CSR instead.
• Send the A1 note to the 225e.
• Tune 258 to A1
• Send A4
• SCALE to A4 using the CSR only.
• Repeat the process until satisfaction is reached, head explodes, significant other sends search party. You'll get close but the 225e is too sloppy for perfection.
• Cursing B&A for releasing this sloppy mess can occur. :hmm: Scratch head in wonder at a decade and a half of no revisions.

- IMPORTANT! Save this scaling node by pressing and holding the encoder down for a couple of seconds. The node will cease flashing green/white and will stay white for as long as you hold down the encoder. This informs you that it is saved within the CSR. Now, every time that you activate the node, regardless of 200e system preset, your scaling value is the default. You can always change it and save a new temporary offset by not doing the 'press & hold to save' on a per preset basis.
• Remember to ‘press & hold’ to save your default scaling when you are happy with each node.
• You can save your scaling work, on a per-node-basis, at any time by pressing and holding the encoder. Or don't and get back to where you were by pressing "recall" on the Preset Manager.

- Scale each source to the same 258V oscillator

- Do Bus F of the 225e
- Do your 222e if you want to add it to the CSR. We're setting up a master preset within the CSR for what you are likely to use most often.
• I like to separate 'creative time' from 'tech time'. As such, I like 480 on my 223e to equal 4.8V on the output of my CSR. I like 120 = 1.2V. Simple stuff.
As we scaled the 258 to track 1.2V precisely, you can set this up and it should work or be very very close to it. Now, we can just dial in 1/2 steps by numbers and land on them. :sb:
• Scale up your 222e/223e like this if you so desire.
- Scale your 251e if it isn't precise or if you don't wish to use the fine-tuning feature. My 251e is not connected to my CSR because I'm a lucky bastard. I'll just use the fine-tuning for when things are a bit off. I'm fine with that. One could set the system up so that the 251e numbers match the output of the CSR in the same manner as what we've done for the 222e/223e.
- The 250e is a special case as we are going to set it up so that the quantizing feature actually corresponds to real life 12-tone notes.

- Scale each source (row) to each oscillator (column) in your system.

Chapter 4. Special Cases for 250e.

- Make your choice as to which you would like:
• Quantizing to actually work
• "Voltage Range" button to switch octaves
One node can't do both. I haven't given it much thought but I think it is impossible to do both even if using another dedicated CSR source column. I don't see how it is possible to double scale something for both results. Again, I've given the subject no thought. If you are fine with not using the quantizer on the 250e, then it doesn't make sense to use the CSR in the path unless octave switching appeals to you.

See videos below for examples of both. Here is a link to a thread with more on the subject of scaling & quantizing. Lots of good info and examples.





- Scaling the 250e Quantizer
- Connect to CSR as normal for each 250e output. This is all done through the CSR and not direct from 250e to an oscillator. In our example, we'll assume CV1.
- Engage "Edit" mode on the 250e or this will be a gigantic flop
- Enable quantizing for Step 1 or all steps. We'll use step 1. Once scaled for a single step, it works on all of them as it is just digital data. Voltage isn’t flowing through the knobs like on an analog sequencer. Thus, once set up the first time, it works on all steps. Nice! :party:
• I've set all sequencer steps to 'quant' on my system as I've saved it in the master preset for ease of quick use.
• One could also set up a couple of different octaves, one octave per step, but that will be confusing and is best left until the end for verification of success; along with sweet octave jams.
- Select Stage Addressing to "Continuous"
- Use Stage Addressing knob to go to Step 1 (The orange/amber LED on Step 1 should be lit)
- Turn the central encoder so that the blue LED is also on Step 1
- Select CV1 in the Voltage section at the upper right
- Turn on "quant"
- Select 0-2V voltage range in Voltage section
- Turn the large CV knob for Stage 1 (not the small timing one) until the output on the 250e's CV 1 reads something close to 1.2V. Mine lands at 1.19V. Count from 0V as 'zero'.
• Be sure to count the quantization points. Depending upon how far out of whack your 250e is, it could land on octaves at 11 or 14 'quant points' or some arbitrary number of steps apart. I've no idea from here. Count 'em.
• Start from 0V and all the way down on the CV1 knob. Count upwards from zero to 12.
- Verify that you are somewhere near 1.2V
- Tune your oscillator (the 258) to A1
- Select 4-6V voltage range in Voltage section
- Turn the large CV knob for Stage 1 until the output reads something close to 4.8V. Mine lands at 4.76V. Yikes.
• Be sure to count the quantization points. Count 1.2V as 'zero' and go up 1 through 36 until you hit 4.8V. Do the same with 4.8V to 6V on the output of the CSR.
• You should run out of knob travel and have to go up to the next 'voltage range'. Resume counting upwards upon the next step.
- Use the CSR to scale that value so that you 258 is putting out a nice A4. This should read as 4.8V on the CV input of your 258, if nothing drifted.

You should get the point of this. Just count upwards & downwards. Use each octave point to as 'zero' until the next higher or lower octave point. Use CSR to scale the higher octaves and go back to 1.2V to tune to A1 again.
Once you've counted once or twice, your voltmeter can be relied upon to tell you when you've hit the octave points. You won't have to keep counting after a couple of rounds. It makes sense and isn't that confusing once you've actually done it once.

- Scale your CV2 output if desired.

I'm out of time for now and won't get into the octave jumping scaling. By now, if you've done any of the above, it should all be straightforward.


There is one open blue banana in the diagram. This leaves room for experimentation with the 291e, 288V, perhaps the 285e (it was possibly useful on the earlier firmwares that allowed for use as an oscillator), or, more usefully: scale it to 1V/Oct for other inferior formats! Whatever... you can change it later or per preset. :love:
tIB's rig cabline CSR.jpg
tIB's rig cabline CSR.jpg (79.24 KiB) Viewed 840 times
Here's a chart that dougcl put together. It shows how far off his 250e is. I did a similar chart for my 250e, going out to 6V. My 250e was much closer than what this chart shows. YMMV.
Doug's 250e Scaling Chart missingtwin.png

elmerfudd
Wiggling with Experience
Posts: 266
Joined: Sun Apr 14, 2013 6:35 pm

Re: Studio.h CSR: How to scale your entire system + more!

Post by elmerfudd » Mon Feb 22, 2021 12:02 pm

Kent, you are a prince. Thanks!

User avatar
tIB
Stainer!
Posts: 9490
Joined: Mon Jun 08, 2009 5:28 am
Location: UK

Re: Studio.h CSR: How to scale your entire system + more!

Post by tIB » Mon Feb 22, 2021 2:04 pm

You certainly are! That's an incredible load of quality information right there - I'm nowhere near digesting it yet but I will... Many thanks!

User avatar
wyrtti
Muffdiver's Helmet
Posts: 1096
Joined: Sun Aug 02, 2009 2:51 pm
Location: Finland

Re: Studio.h CSR: How to scale your entire system + more!

Post by wyrtti » Mon Feb 22, 2021 11:28 pm

Thank you, Kent! This is a wonderful piece of work!
Eschew obfuscation!

User avatar
jheronymo
Common Wiggler
Posts: 139
Joined: Sat Nov 22, 2014 4:11 pm
Location: Vancouver, BC

Re: Studio.h CSR: How to scale your entire system + more!

Post by jheronymo » Tue Feb 23, 2021 2:00 am

Incredible info, thanks a million for posting this :party: :party:

User avatar
Kent
Large Member
Posts: 12304
Joined: Thu Mar 06, 2008 10:17 am
Location: Vienna, Austria

Re: Studio.h CSR: How to scale your entire system + more!

Post by Kent » Tue Feb 23, 2021 9:15 am

Thanks, all. Once starting upon the original topic, I thought that a more general tutorial of how scaling works, what it does, and how to do it with the CSR; would benefit everyone.

I'll add to this thread as I explain other features.

User avatar
Jericho
Common Wiggler
Posts: 143
Joined: Tue Jan 27, 2015 3:33 pm
Location: Germany

Re: Studio.h CSR: How to scale your entire system + more!

Post by Jericho » Tue Feb 23, 2021 1:48 pm

Great stuff! Thank you Kent for sharing this!
I don't have a CSR and just starting out with Buchla. Hopefully my question is not answered elsewhere. I did check the forums for the last couple of weeks. Does anybody know if you can mix pulses to one pulse train with the CSR?

User avatar
Kent
Large Member
Posts: 12304
Joined: Thu Mar 06, 2008 10:17 am
Location: Vienna, Austria

Re: Studio.h CSR: How to scale your entire system + more!

Post by Kent » Tue Feb 23, 2021 2:20 pm

Gerne!

Yes, you can mix pulses. In most cases, just stacking banana cables already does that as Buchla pulse outputs are diode protected. They basically act as “OR” logic.

The fun thing about mixing them in the CSR is that you can change levels, mute and unmute, randomize, etc.

maxl0rd
Common Wiggler
Posts: 183
Joined: Mon Aug 05, 2013 1:51 pm

Re: Studio.h CSR: How to scale your entire system + more!

Post by maxl0rd » Tue Feb 23, 2021 4:20 pm

Ok, but is Doug still making CSRs? :)


Dreadwvlf
Common Wiggler
Posts: 56
Joined: Mon Aug 31, 2020 11:58 pm

Re: Studio.h CSR: How to scale your entire system + more!

Post by Dreadwvlf » Tue Feb 23, 2021 9:53 pm

He definitely is. I just snagged a 254e from him and he told me all of his modules are available!

maxl0rd
Common Wiggler
Posts: 183
Joined: Mon Aug 05, 2013 1:51 pm

Re: Studio.h CSR: How to scale your entire system + more!

Post by maxl0rd » Tue Feb 23, 2021 10:32 pm

That’s great news. His 254e is really nice!

User avatar
Kent
Large Member
Posts: 12304
Joined: Thu Mar 06, 2008 10:17 am
Location: Vienna, Austria

Question: why bother trying to scale more than 1 osc to 1.2V?

Post by Kent » Wed Feb 24, 2021 2:19 am

Here’s a question cluster that I received that possesses answers of general interest:

Q: Why scale more than 1 oscillator to 1.2V/Oct? Won’t the CSR handle the scaling for the others anyhow? Aren’t you effectively scaling each oscillator once more than absolutely required?
The “CSR Time!” diagram makes this evident.

A: True. The process I described could eliminate the scaling of each oscillator without having the CSR in the path. I don’t recommend this. It also doesn’t make sense to skip the extra work within the context of my Buchla rig.

• I have more voltage sources than I can plug into my single CSR at any one time. On an item like my Kilpatrick K4816 Pattern Generator, I can scale its own output to a proper 1.2V/Oct output and patch it in to any properly scaled oscillator as needed. In most cases, I’ll be ready to go. If using it with the 259e, I can use a channel of the CSR if required and unoccupied. Again, we’ve set up the CSR to default to what we will use most often. It isn’t set in stone and any other scaling job is a simple knob twist away and can be stored to a preset.
The 251e in my system is accurate, so I can just patch it to any correctly scaled oscillator and be on my way.

• Getting the destination oscillators as close as possible to 1.2V/Oct allows for more potential adjustment room on the CSR. Neither a major nor very realistic concern but at least I don’t have to think about it if troubleshooting something.

• Sanity/Conformity. If the majority of coarse tuning knobs are all pointing in the same general direction, it aids in ascertaining what is going on at a glance. Again, the removal of variables.

• This is a test of the Early Warning System. If an oscillator or filter can’t be scaled to 1.2V/Oct on its own (again, looking at you 259e),it can very likely still be “made true” by the CSR. However, it is helpful to know of one’s problem students in the class. If one runs into issues down the road, advance warning can save some hair pulling and stress. It is also nice to know this stuff if you are scaling Eurorack gear to the 1.2V norm.

• When in ‘creative mode’, I don’t wish to think about troubleshooting or any variables or compensations. I just wish to patch and go. Establishing a baseline helps tremendously.

User avatar
Kent
Large Member
Posts: 12304
Joined: Thu Mar 06, 2008 10:17 am
Location: Vienna, Austria

Oscillator Oddities

Post by Kent » Thu Feb 25, 2021 6:59 am

Oscillator Oddities:
(If not mentioned otherwise, all findings based upon my personal gear)

1979 DAO

Calibrate to 1.2V/Oct in accordance with the manual
• Instead of sending 1V and 3V, you send 1.2V and 3.6V.
• Tip: Experiment with where you position the "Coarse" tuning knob in order to get the kind of travel or range that you'd like. You can always get back to calibrating to factory specs rather easily.
• Once scaled/calibrated, this thing is very precise and a joy to use.
• There is a built-in quantizer option in the menu. If you turn this on, it is perfectly in tune although the fine-tuner does have an effect.

B&A 261e

• The CV attenuators on both oscillators need to be @ 3:30 to 4o'clock in order to be properly scaled. (Or slam 'em full open and let the CSR deal with it. However, I like getting things as closely as possible to 1.2V/Oct, without the CSR, in the system as a workflow thing.)
• The Mod oscillators seem to be slightly more precise than the Principal oscillators.
• you can expect to get very close to 5 octaves out of these and then tuning starting to get goofy toward the 5th octave and more as you go up.
• Both of my 261es are just shy of 4cents sharp at each octave. That's pretty good but as good as it gets
• Low impedance inputs load down voltage source

B&A 259e

• Lacks a scaling trimmer.
• Needs the attenuators set to something around 4o'clock to scale correctly. (Or slam 'em full open and let the CSR deal with it. However, I like getting things as closely as possible to 1.2V/Oct -without the CSR- in the system as a workflow thing.)
• Mod & Principal are pretty much identical in terms of accuracy, which is worse than the 261e
• Low impedance inputs really load down voltage source
• Seem designed to work with internal modulation sources very well. MIDI tracks well. Internal FM is accurate and can be tuned. Pitch tracking of Mod oscillator to Principal is accurate, etc.
• The CV attenuator is very, very, very sensitive/touchy/squirrelly. See below for a tip.

Sputnik Modular Dual Defect Processor 289 Oscillator Modes

• It can be done! I've managed to get over 4 octaves of range out of mine (C1 being spot on. C5 being 6cents sharp) when using the CSR. The Multitriangle-Core Oscillator sounds especially good.
• The CSR MUST be used.
• Limitations of the FV-1 chip means that there is always slew (portamento), to some degree, on all CV inputs.
• Pitch CV goes to X input.
• The knob needs to be somewhere around 8o'clock and is VERY fiddly to get just right.

Studio.h DPO

• It is analog so you will need to check your tuning every session after letting it warm up.
• It is very stable and is my trusted tuning reference.

Verbos 262v Harmonic Oscillator

• be sure to let it warm up for 15 mins or more
• can't be tuned as low as other oscillators in the system. It can land just flat of A#0 (when no Pitch CV is present), which is fine enough but inconvenient.
• Rock solid once scaled up with only minor drifting over the days.

B&A 266e Quantized Random Voltages

• A fun experiment. It is rather all over the shop with the 'quantized' voltages. I could get it to pretty much scale for 5 octaves but, as the quantization isn't landing on equidistant voltage values (3.93, 4.95, 6.52, 6.73, 7.87, 8.52, 9.99), 1/2 of the range gets incredibly out of tune. I wonder if running it through a voltage processor, in order to constrain it to a narrower window would work. :hmm:


==Tip for Preset Management Adherent Oscillators==

As mentioned above, I like to have my attenuators fully open when possible. This isn't possible with the 259e and 261e or they just won't scale.
In order to reduce the chance of knocking them out of tune (by half), I simply scale them up, save the preset and then turn the CV attenuators all of the way up or down. Then, I recall the preset. At least they can only travel in one direction now.
As mentioned earlier, one can always leave them fully open, yet totally out of whack regarding 1.2V/Oct., and just let the CSR handle the conversion.

User avatar
anomie
Common Wiggler
Posts: 86
Joined: Tue Mar 05, 2013 2:25 pm
Location: Dublin

Re: Studio.h CSR: How to scale your entire system + more!

Post by anomie » Thu Feb 25, 2021 3:19 pm

Two comments:

First, Kent, this is amazing. Thank you for the thoroughness of your research and for the thoroughness of your documentation of it

Second, it’s also amazing that it’s necessary. Much as I love my Buchla, I still find the sh*tty CV tracking and the crazy variance in the resolution of different modules a pain. I’ve learned to work around it and make music, but I’ve got to think there’s a gap in the market for others to do what Doug Clauder has with the DPO - implement a take on Don’s designs with upgraded components and digital signals used for control, not for sound generation.

Rant over. A weekend of trying to program the 223 to work with both a 261 and the DPO has obviously gotten to me. Maybe I should get a 210e 😁

User avatar
Kent
Large Member
Posts: 12304
Joined: Thu Mar 06, 2008 10:17 am
Location: Vienna, Austria

Re: Studio.h CSR: How to scale your entire system + more!

Post by Kent » Sun Feb 28, 2021 1:08 pm

Thanks, anomie. I enjoy this stuff and enjoy helping others where I can. A TON of the people on this board have helped me over the years.

You can probably scale the 261e to get pretty close to good enough for drifty duophony with a DPO.
Whatever you do though, don't get a 210e and expect to be able to scale your oscillators and have pitch stability. It just WON'T work reliably under any circumstance.

One of my biggest wishes is for someone to revisit the 259(analog)/261e single panel concept with the attention, care, and capabilities of the Studio.h DPO.
I ADORE the sound of the 261e/259 waveshaping section; specifically the Symmetry and Timbre sections. I don’t think that I’ve ever used “High Order”. It is a large part of what brought me into Buchla and can be considered an addiction at this point.

So, Doug, if you are reading this... you licked the 258 and brought it up to e-series with recall and the sweet FM of the 258. Your magic is needed on the 259!
Or even an h-series waveshaping section with recall. Sell that alongside the DPO and sales won't be cannibalized.

===

Added some more info to the Oscillator Oddities section.

User avatar
anomie
Common Wiggler
Posts: 86
Joined: Tue Mar 05, 2013 2:25 pm
Location: Dublin

Re: Studio.h CSR: How to scale your entire system + more!

Post by anomie » Mon Mar 01, 2021 8:13 am

Kent wrote:
Sun Feb 28, 2021 1:08 pm
One of my biggest wishes is for someone to revisit the 259(analog)/261e single panel concept with the attention, care, and capabilities of the Studio.h DPO.
I ADORE the sound of the 261e/259 waveshaping section; specifically the Symmetry and Timbre sections. I don’t think that I’ve ever used “High Order”. It is a large part of what brought me into Buchla and can be considered an addiction at this point.
Hard agree. I mentioned this to Doug a couple of years ago and he said he was more interested in DSP, and was hoping Eric Fox & co might do a 259/259e refresh. I would happily drop $$$ if that came about.

Post Reply

Return to “Buchla, EMS & Serge”