search

eh2k / squares-and-circles

squares-and-circles is an alternate firmware for the Eurorack module O_C, targeting Teensy 4.
https://eh2k.github.io/□●/
168 stars 14 forks source link

CV/V/OCT outputs a lower voltage #57

Closed eh2k closed 10 months ago

eh2k commented 1 year ago

Hi @eh2k ! I have been fiddling around with S&C past versions (circa 2022) for few monthes, but found a strange issue with last version (508de0e) : the CV/V/OCT engine outputs a lower voltage (with an approximative ratio of 0.66 of what it should be) than the one written on the bottom of the screen. On the other hand, when I do the calibration, the output voltage is the same than the one displayed. I investigated on this and I guess something strange is happening in the "push" function of the machine binary, as I traced the cv value and it is coherent until that call. Once again, the hardware shouldn't be involved as the calibration process outputs the displayed voltage. Am i missing something ? Thanks again for the work Best regards j.

Originally posted by @jrachedi in https://github.com/eh2k/squares-and-circles/issues/1#issuecomment-1817572010

eh2k commented 1 year ago

Hi @@jrachedi,

thanks for the hint.

As I remeber, I have also noticed something strange some time ago.

I recently redesigned the CV engines, using the code of the internal modulations. The new scope view is quite nice.

Could you please try to reproduce the problem with the latest version https://github.com/eh2k/squares-and-circles/commit/27df093616253b1892dc6c5f70f53ec705771c24 and let me know if this is still the case.

Here are some additonal information:

jrachedi commented 1 year ago

The problem reproduces with 83cf6aa … sorry :) About the points you mention:

Le 19 nov. 2023 à 11:13, eh2k @.***> a écrit :

Hi @@jrachedi https://github.com/jrachedi,

thanks for the hint.

As I remeber, I have also noticed something strange some time ago.

Modulation did not work properly Finetune was quantized. I recently redesigned the CV engines, using the code of the internal modulations. The new scope view is quite nice.

Could you please try to reproduce the problem with the latest version 83cf6aa https://github.com/eh2k/squares-and-circles/commit/83cf6aad241db43ff9277e9eb1d95e04bd680aad and let me know if this is still the case.

Here are some additonal information:

The quantizer affects the "V_OCT " parameter. Check the transpose in the IO page. If another engine is mapped to the same output, possible the output is calculated to audio scale (probably the case?). — Reply to this email directly, view it on GitHub https://github.com/eh2k/squares-and-circles/issues/57#issuecomment-1817809408, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACE7S4F5SRIRH3BRU465KVDYFHLTBAVCNFSM6AAAAAA7RTY5MKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMJXHAYDSNBQHA. You are receiving this because you were mentioned.

jrachedi commented 1 year ago

It reproduces with 27df093 as well. The scope view is indeed a pretty neat feature !

eh2k commented 1 year ago

Just pushd https://github.com/eh2k/squares-and-circles/commit/7d07b9af2e47948ba3e6322deb6385497a935fb6. This issue should be fixed there. Each engine has an internal volume parameter (IO-page). This is not visible in case of CV engines and should now always be 1.0. Could you please check if it works for you. Thanks for your feedback.

Steps to reproduce:

jrachedi commented 1 year ago

Your last commit fixed my issue ! Thanks a lot Julien

koswir commented 12 months ago

Just pushd 7d07b9a. This issue should be fixed there. Each engine has an internal volume parameter (IO-page). This is not visible in case of CV engines and should now always be 1.0. Could you please check if it works for you. Thanks for your feedback.

Hey! First of all I love S&Q! eh2k thanks for this amazeing firmware! I was on my way to share this bug too, but jrachedi was faster! :)

7d07b9a finally works here and I'm so happy about that but I've noticed another problem. The Slew parameter works right up to about 72/73. If I set it to a higher value - the glide is longer but never reaches the desired pitch. Please check that out! Koswir.

eh2k commented 11 months ago

@koswir -Thanks for your feedback, I'm glad you're having fun with the firmware.

The V/Oct engine may be inconspicuous, but together with the modulation you can do some crazy things with it. It could well be that some usecases arise that I didn't even think of during development. If you like to describe how you use it in more detail, just post them here - maybe I can use the information to include it as documentation in the wiki...

Regarding the Slew parameter. It is actually a simple ONE_POLE filter: ONE_POLE(cv, cv0, powf(1 - slew, 10)); where slew is a value from 0 to 0.5. I'll have to take a close look at this, it may well be that the range from 0 to 0.28 would be better. Then the maximum turn to the right would correspond about to the current 72/73.

koswir commented 11 months ago

Regarding the Slew parameter. It is actually a simple ONE_POLE filter: ONE_POLE(cv, cv0, powf(1 - slew, 10)); where slew is a value from 0 to 0.5. I'll have to take a close look at this, it may well be that the range from 0 to 0.28 would be better. Then the maximum turn to the right would correspond about to the current 72/73.

oh no please do not limit the range! Glide at 72 is really very short. My workaround for longer slew rates is to set 72 than add an envelope to push the value up but only for a desired time. :) Thanks! Koswir