Vember Audio Surge is now open-source

VST, AU, AAX, etc. plug-in Virtual Instruments discussion
KVRist
379 posts since 25 Dec, 2018

Post Mon Feb 01, 2021 10:11 am

Oh and most importantly: This is a surge bug of some form. If I try with an arturia plug I don't get this mistake. I'm just not quite sure *what* the bug is.

KVRist
52 posts since 22 Jan, 2010 from Oregon, U.S.

Post Mon Feb 01, 2021 11:10 am

baconpaul wrote:
Mon Feb 01, 2021 9:51 am
Thanks both. Can one of you either add a link to the bitwig ticket system so I can add a comment? Here's my diagnosis which you could also paste into your ticket...
Thanks for the info! Just pasted it all into my Bitwig ticket (#96874) and will let you know if/when I hear anything back.

User avatar
KVRist
186 posts since 7 Oct, 2020

Post Mon Feb 01, 2021 3:17 pm

In the meantime, a vintage string machine inspired sound:

https://soundcloud.com/user-589036812-2 ... -1-x-surge

It's not supposed to be an exact emulation. I was just trying to see how it would sound like if I emulated the ensemble effect using the oscillators alone. In fact, there is no chorus or ensemble effect on that sound.

Edit: And just for fun, I'll also leave this one here :hihi:

https://soundcloud.com/user-589036812-2 ... -1-x-surge

User avatar
KVRist
186 posts since 7 Oct, 2020

Post Tue Feb 02, 2021 5:54 am


KVRist
379 posts since 25 Dec, 2018

Post Wed Feb 03, 2021 5:18 am

Just an update on bitwig: the bitwig team gave me a profoundly useful hint on what surge is doing wrong and it is indeed a surge bug. I’ll investigate today and report back here but looks good for a 182 fix.

KVRist
379 posts since 25 Dec, 2018

Post Wed Feb 03, 2021 7:27 am

OK so I have the bitwig problem fixed. You can grab the latest nightly from https://surge-synthesizer.github.io/nightly and test if you want, and this fix will be included in our 1.8.2 release in mid march.

If you grab the nightly now you are getting a relatively dynamic bit of code. There's loads of new FX for instance, some of which aren't finished. If you use the new stuff please be careful, use a limiter, no in-ear headphones, etc.... the nightly is not fully production tested software (although it does undergo a lot of tests before release).

But it does fix the bitwig problem.

And also, fwiw, the bitwig team is incredibly developer friendly! Thanks to the folks there for pointing me in the right direction :)

KVRian
1046 posts since 3 Oct, 2011

Post Wed Feb 03, 2021 8:36 am

When modulation the filter using CTRL for steps has a readout in Hz which doesn't make sense to me as the steps seems to actually be in semitones. If I change the frequency of the filter the degree of modulation is seemingly different, as in another value in Hz though the degree of modulation seems to be the same. For instance with a filter value set to C2 it seems 3 "CTRL-d" steps of modulation is 231 point something Hz. Now if I increase the cutoff up to C3 then suddenly the readout for the same degree of modulation now reads 458 point 8 something Hz. Increase main cutiff to C4 and the unchanged level of modulation reads as 913 point something Hz. Etc etc.

Could the readout on the RM FX be given in Hz instead of semitones as that would make more sense to me and most people who are accustomed to values in Hz? Pretty please? Amd it would sure make me happy if a constant value for the detuning in Hz in the RM FX could be an option, it would be so sweet to be able to change the frequency yet have a consistant constant beating freqeuncy. ;-) I totally get it if you'd rather spend you're resources on other things though.

KVRist
379 posts since 25 Dec, 2018

Post Wed Feb 03, 2021 8:44 am

Gone soft wrote:
Wed Feb 03, 2021 8:36 am
When modulation the filter using CTRL for steps has a readout in Hz which doesn't make sense to me as the steps seems to actually be in semitones. If I change the frequency of the filter the degree of modulation is seemingly different, as in another value in Hz though the degree of modulation seems to be the same. For instance with a filter value set to C2 it seems 3 "CTRL-d" steps of modulation is 231 point something Hz. Now if I increase the cutoff up to C3 then suddenly the readout for the same degree of modulation now reads 458 point 8 something Hz. Increase main cutiff to C4 and the unchanged level of modulation reads as 913 point something Hz. Etc etc.

Could the readout on the RM FX be given in Hz instead of semitones as that would make more sense to me and most people who are accustomed to values in Hz? Pretty please? Amd it would sure make me happy if a constant value for the detuning in Hz in the RM FX could be an option, it would be so sweet to be able to change the frequency yet have a consistant constant beating freqeuncy. ;-) I totally get of you'd rather spend you're resources on other things though.
Both reasonable requests. We're neck deep in a pretty big rewrite now as we port to juce, but best thing to do for reasonable small requests like this is to make a GitHub account and open them as (two separate) GitHub issues. If they are GitHub issues they get a milestone, and if they get a milestone they get done.

You are right that modulation depth is in semitones and so the display is relative to the central pitch. We actually thought about that some but definitely see you wanting an option. I have not given any such consideration the RM though. As to the unison detune, I see we allow the control to be able to be set to absolute, but we don't actually do anything with that in the code. Again, dropping that into the GitHub issue would be the most helpful for us!

User avatar
KVRAF
21411 posts since 7 Jan, 2009 from Croatia

Post Wed Feb 03, 2021 9:10 am

The primary filter cutoff value unit is Hz, showing the closest note value is a secondary thing. So the modulation amount is also shown in Hz, which is consistent. I would not want to change that, but we could probably add a secondary readout for value in semitones.

BTW you can see the full swing of the modulation if you enable Menu > User Settings > Value displays > Modulation value readout shows bounds.


Gone soft wrote:
Wed Feb 03, 2021 8:36 am
Could the readout on the RM FX be given in Hz instead of semitones as that would make more sense to me and most people who are accustomed to values in Hz? Pretty please?
Since you asked so nicely... :)

Image

User avatar
KVRist
186 posts since 7 Oct, 2020

Post Wed Feb 03, 2021 10:35 pm

Some Surge drums through the new "Combulator" comb filter effect and some other Surge effects:

https://soundcloud.com/user-589036812-2 ... -5-x-surge

KVRist
52 posts since 22 Jan, 2010 from Oregon, U.S.

Post Thu Feb 04, 2021 12:34 am

baconpaul wrote:
Wed Feb 03, 2021 7:27 am
OK so I have the bitwig problem fixed. You can grab the latest nightly from https://surge-synthesizer.github.io/nightly and test if you want, and this fix will be included in our 1.8.2 release in mid march.
Thanks for checking this out, and glad to hear you got some extra clues from the Bitwig team.

Just did a quick test of the 1.8.nightly.ed3308f in Bitwig 3.3.1, Windows 10. Both the base values and modulation rates appear to be properly saved and retained upon reopening the program and GUI (hooray!), with one caveat:

I wasn't able to enter base values by ctrl-left clicking on the Bitwig modulation knob and typing in the value, as is normally the case. When I try to do it this way, the new typed in value doesn't even register upon hitting return.  (I was able to enter values fine that way w/ a few other plugins.)

So at the moment, if a Bitwig user wants to type in an exact value, that needs to be done in the Surge GUI. Or, you can still use shift-left click and drag to do slow adjustments with either the Bitwig knob or the Surge slider (latter easiest with Bitwig LFO temporarily disabled if necessary). That doesn't put a hard limit on the usability since there are a few workarounds, but this remaining piece is probably worth correcting eventually. Many thanks, and hope that helps!

User avatar
KVRAF
21411 posts since 7 Jan, 2009 from Croatia

Post Thu Feb 04, 2021 2:48 am

We'd like to focus on porting the back-end of Surge to JUCE, then all this stuff you're seeing should be working nicely. Quick test would be: can you do the same thing with Chow Tape? If yes, then you can expect the same to work with Surge once Surge XT (next version of Surge that will come after the upcoming 1.8.2 beginning of March) is out there.

That said, DAW only knows the normalized value of the knob (range 0.0...1.0), it doesn't know the actual parameter range and units used. What you see as parameter value is just a text string reported to the DAW.

KVRian
683 posts since 27 Dec, 2003

Post Thu Feb 04, 2021 3:35 am

EvilDragon wrote:
Thu Feb 04, 2021 2:48 am
That said, DAW only knows the normalized value of the knob (range 0.0...1.0), it doesn't know the actual parameter range and units used.
This would be nice to have ...

User avatar
KVRAF
21411 posts since 7 Jan, 2009 from Croatia

Post Thu Feb 04, 2021 3:56 am

Sure, but this is just how VST/AU are, AFAIK. They all just work with normalized values.

KVRian
683 posts since 27 Dec, 2003

Post Thu Feb 04, 2021 4:04 am

No, in REAPER it should work (see Voxengo plugins for example) via https://www.reaper.fm/sdk/vst/vst_ext.php

See https://forum.cockos.com/showthread.php?t=27424

Return to “Instruments”