The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Synths for sale  Search Gearslutz Go Advanced
Volca (sample) firmware hack? Drum Machines & Samplers
Old 5 days ago
  #151
Lives for gear
link is no longer providing the firmware

Anyone got a new link?

Thanks
Old 4 days ago
  #152
Gear Head
 

Quote:
Originally Posted by terrible.dee View Post
link is no longer providing the firmware

Anyone got a new link?

Thanks
Hi, I am putting together a new version with additional features and some fixes to ancient bugs in S&H LFO and DX7 patch transfer. I will post it in this thread when I feel like it is in a good shape.
Old 4 days ago
  #153
Gear Head
 

Quote:
Originally Posted by whatever17 View Post
quick re-cap on the Sample MIDI CCs (11 per sound) vs available MIDI CCs (looks like there might be 120):

https://cdn.korg.com/us/support/down...ation%2Fpdf%3B

http://www.indiana.edu/~emusic/etext...r3_MIDI6.shtml

not sure if some of those 120 have limited data range, or if it just doesn't matter.
if it doesn't, maybe there isn't a problem keeping all parameter CCs (110),
unless it's likely some problem might occur with other equipment if accidentally
sent on wrong channel (... plenty of scope for that anyway).
Hm, all CC:s are 0-127. As the table suggests, there are a bunch of CCs that are "unoffically earmarked" for certain functions. If I were developing a proper product I might want to respect this allocation scheme. If we are discussing a special hacked version then maybe one can do whatever one wants.
Quote:

would Velocity be possible/necessary? would it require 1x CC per voice? can
imagine weird stuff happening with all voices adopting same velocity value
- this is no doubt why velocity on the FM is common to all voices.
This could perhaps be skipped. Velocity sensitivity on individual notes are in the note on message. This highlights the fact that using the Sample as a DAW-fed sound module (maximum MIDI control surface) versus using it as a stand-alone groove box (maximum knob control surface) are two slightly different things?


Quote:
@ pajen : do you have an overview of the FM sysex handling, and whether
single parameter address might be possible ? (would be ace if you could
provide a kind of block diagram of how a firmware is organised, for the
non-coders)

bump for ModWheel/CC#1 reception. ...
The sysex reception in the FM and Sample are somewhat similar in the design. It is a classical state machine - a switch statement with three big switch statements nested within. The main problem of patching in individual params it is that I would need to repurpose some states to make it work, it is a bit messy to do.
Modwheel to pitch mod depth on the FM perhaps?
Old 4 days ago
  #154
Lives for gear
 

found out that an old VST FM plugin called Hexter can also load *.syx files
for DX7
was able to load banks made with Dexed into Hexter.

Last edited by whatever17; 3 days ago at 07:09 AM..
Old 4 days ago
  #155
Here for the gear
 

Quote:
Hi, I am putting together a new version with additional features and some fixes to ancient bugs in S&H LFO and DX7 patch transfer. I will post it in this thread when I feel like it is in a good shape.
I have been telling people about these bugs for years but no one else seemed to care. It's not just the S&H LFO, but any of the LFOs that have abruptly changing values, like square, saw, etc. I know because when I upload the patches to my TX7 through a midi out I soldered to the Volca FM, the patches sound as they should. For a while I was using the Volca to edit patches that I would then upload to my TX7, but the buggy implementation really turned me off. Turned me off of the whole Volca line, frankly, that Korg never bothered to fix the firmware.
Old 3 days ago
  #156
Gear Head
 

Quote:
Originally Posted by FunkStar View Post
I have been telling people about these bugs for years but no one else seemed to care. It's not just the S&H LFO, but any of the LFOs that have abruptly changing values, like square, saw, etc. I know because when I upload the patches to my TX7 through a midi out I soldered to the Volca FM, the patches sound as they should. For a while I was using the Volca to edit patches that I would then upload to my TX7, but the buggy implementation really turned me off. Turned me off of the whole Volca line, frankly, that Korg never bothered to fix the firmware.
Yes well you were 100% correct and at least I do care :-) the S&H bug is pretty severe, consisting of a) a scaling error of 256x, and b) an error in the timing comparison, resulting in roughly 25% of the LFO period it will resample the value 50000 times a second, giving that noise "beat". Baaad programmer, no cookie for you!

Could you please point me toward a good example of a (say) square LFO patch that sounds fundamentally different in dexed than on the FM? I noticed the S&H behaviour very prominently once I was made aware of its existence, by a reddit post (perhaps by you?)
Old 3 days ago
  #157
Lives for gear
 

Quote:
Originally Posted by FunkStar View Post
... when I upload the patches to my TX7 through a midi out I soldered to the Volca FM, the patches sound as they should.
are you saying that the volcaFM transfers patches via sysex with the same
function that sends them via the sync out/in jacks? (cool)

Last edited by whatever17; 2 days ago at 09:06 PM..
Old 3 days ago
  #158
Here for the gear
 

Quote:
Could you please point me toward a good example of a (say) square LFO patch that sounds fundamentally different in dexed than on the FM? I noticed the S&H behaviour very prominently once I was made aware of its existence, by a reddit post (perhaps by you?)
From what I remember, problems occured whenever you routed a discontinuous LFO (square, saw-up, saw-down, s&h) to the amplitude of a modulator. Sample and hold is totally screwed up, as you found out, but the others - in this configuration, create audible clicks and pops that are not heard on the original equipment. Since most people stick to sine and triangle, they don't run into these problems. I have switched most of my FM to Reface DX and OG Microkorg, so it would take me some time to verify this was the exact situation and find patches.

Quote:
are you saying that the volcaFM transfers patches via sysex with the same
function that sens them via the sync out/in jacks? (cool)
I never used the sync in/out jacks to transfer patches, but I would presume those are the same functions. Korg included Midi out solder points on the pcb. At some point people were even selling solderless midi out mods, I believe. Oddly enough, individual patch sends never worked with my TX7, but the bulk patch send would work. Go figure.

BTW, for anyone wanting a practical guide to programming the DX7, get The Complete DX7, by Howard Massey. It covers many of the synth engine nuances that other sources omit, and does so from a musician's perspective.
Old 2 days ago
  #159
Lives for gear
Quote:
Originally Posted by pajen View Post
Hi, I am putting together a new version with additional features and some fixes to ancient bugs in S&H LFO and DX7 patch transfer. I will post it in this thread when I feel like it is in a good shape.
OK, thank you!
Old 2 days ago
  #160
Lives for gear
 

Quote:
Originally Posted by FunkStar View Post
Go figure.
probably the way the volca memorises on/off state of Ops in place of the
checksum on single patches.
Old 2 days ago
  #161
Lives for gear
 

Quote:
Originally Posted by terrible.dee View Post
OK, thank you!
sent you a pm with a link to the 2 wav files. worth checking out in current version. update as per v103 instructions.
Old 1 day ago
  #162
Gear Head
 

Quote:
Originally Posted by whatever17 View Post
probably the way the volca memorises on/off state of Ops in place of the
checksum on single patches.
Yes, that makes sense. If the FM sends OP on/off instead of checksumming it, the receiving device might (rightfully) interpret it as a buggy transfer... I shall investigate.
Old 5 hours ago
  #163
Gear Head
 

Quote:
Originally Posted by FunkStar View Post
From what I remember, problems occured whenever you routed a discontinuous LFO (square, saw-up, saw-down, s&h) to the amplitude of a modulator. Sample and hold is totally screwed up, as you found out, but the others - in this configuration, create audible clicks and pops that are not heard on the original equipment. Since most people stick to sine and triangle, they don't run into these problems. I have switched most of my FM to Reface DX and OG Microkorg, so it would take me some time to verify this was the exact situation and find patches.
Yes I see now what you are referring to. I don't have any original equipment, but I tested a bit with Dexed vs. FM.
Here is the thing: the Volca FM LFO has "infinite attack"; it will happily go from 0 gain to full gain from one sample to the next. Classical "digital popping".
If you look at the Dexed source code, you will find this (simplified):
Code:
void FmOpKernel::compute(int32_t *output, const int32_t *input,
                         int32_t phase0, int32_t freq,
                         int32_t gain1, int32_t gain2, bool add) {
  int32_t dgain = (gain2 - gain1 + (N >> 1)) >> LG_N;
  int32_t gain = gain1;
  int32_t phase = phase0;
      for (int i = 0; i < N; i++) {
        gain += dgain;
        int32_t y = Sin::lookup(phase + input[i]);
        int32_t y1 = ((int64_t)y * (int64_t)gain) >> 24;
        output[i] += y1;
        phase += freq;
      }
where N is 64. This means that it will take Dexed 64 samples to get from zero to full gain; 1.3 ms.
There it is; not exactly a bug, but definitely a different implementation.
As a side note, the FM has a more aggressive max attack in the normal amp EG. That is part of the "character", and might be seen as a positive thing.
The discontinous LFO amp mod is hard to fix for me. I would need some extra variables to keep track of a "current gain" for each operator, and there is no extra space, and tricky to shoehorn in. If we leave the amp EG as-is, then it might be easier to add some modification on the LFO output itself, and that would only need one extra variable per playing note (since every playing note has its own LFO). I do agree it would be nice to fix.


Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump