Hi Alex,
Does the sample selection make a difference? Also, what is a “short time”, like 2 to 3 seconds or so? Does the sample begin to play before Rack crashes?
Thanks a ton for reporting this!
What codex/format/signature was on the sample that caused the crash?
I’m guessing that a pointer is being referenced which is temporarily unset, or an array is out of bounds when a new sample of a different length is loaded. I’ll have to try out some situations myself to see if I can recreate the issue.
I can’t get it to crash for the life of me now. I don’t think sample selection made any difference but I’m not sure which sample it was that caused the crash. It crashed after maybe 10 seconds or so, it could have been less but somewhere in that region. I didn’t get any sound out when this happened because I hadn’t connected the outputs yet. I believe the sample was wav but not sure about sample/bit rate, shouldn’t have been anything outside of 16/24 bit 44.1k/48k.
Sorry I can’t be more precise but if it crashes again I will try to take note of what I was doing.
This is a total shot in the dark, but I see two possible trouble spots:
Perhaps one of these variables are 0, which would cause a divide by 0 error. I’ll go ahead and add some code to prevent this from happening.
2 Likes
That’s probably it, I can get it to crash with connecting a clock, not loading a sample, changing ratchets. I don’t remember clicking on the ratchet page previous to the original crash but I may have. This explains the delay as well, I probably changed ratchet at the end of the bar so it played nearly once and crashed.
The BPM divide-by-zero might not be the issue. I overlooked a line of code:
And I already check if elapsed_time is > 0. Hmm…
Can you walk me through how to reproduce the bug? Showing me a screen capture of the patch so that I can recreate it would be helpful! 
I just tried the same thing again and I can’t get it to crash now, so strange. If I can recreate it I will show a screenshot but I had ZZC clock going into clock and reset, clicked on the ratchet page and put the value up on the last step, started the clock. It crashed when it got to the last step. I’ll keep trying tomorrow to see if I can narrow it down a bit.
Ok, thanks for the info. I’ll try that myself as well. 
Quite a few problems I have encountered recently in various modules appear to be related to uninitialized data. I’ve found that I can reliably cause such a crash only if I reload the patch. If I restart Rack and the autosave is loaded, I can seldom cause the crash, until I reload the patch.
You might try that.
1 Like
That’s bitten me lately as well. I’ll check all my arrays and pointers. My shared “Sample” class should be fairly bulletproof by now. 
1 Like
I tried to torture test the module but wasn’t able to get it to crash. I forgot, are you on PC or Mac? If you’re on PC, maybe I can send you a version with a few things disabled to see if it still crashes?
Thanks!
I get your doubts about putting the step buttons inside the lower screen. I still think it’s a good option to have the on off buttons that close to where you’re going to be with your mouse anyway. I can totally see myself turning steps on and of while playing with the parameter locks (of course that can be said for having the step buttons below the screen as well)
1 Like
It seems like it’s one of those bugs that doesn’t have any discernable pattern to reproduce. I’m using linux and can build if there is anything you want me to comment out etc.
Autobreak Studioooooooo!! This is cool. I cant wait for updates
1 Like
Hi @XtremeSounds ! Did you have anything in mind? I recently received a feature request for setting percentage per trigger (so that triggers don’t fire all of the time).
Playing with groovebox, pretty cool drum sampler! Some things which would be awesome: Module Randomization includes generating steps for all sounds. Option to loading files use alphabetical order of filenames vs random; for example, I have drum kits from other apps organized using filenames like 01-…, 02-… etc. An expander that includes trigger inputs, so internal sequencer doesn’t have to be used.
Hello @auxmux, Thanks!
My apologies, but this looks like a bug. The files are supposed to load in alphabetical order, and on Windows they do. But I forgot to add special sorting code to ensure the proper loading order for other operating systems. I’ll get this done ASAP.
I’ll try to add this as well. Seems logical. I often overlook the randomization and initialization features in rack. Thanks for the reminder.
I’ll see what I can do about external trigger inputs. I’m fairly busy right now, so I don’t know if this will make it into the queue anytime soon though. 
@auxmux Quick update:
- I fixed the loading of files to ensure alphabetical order
- I updated the randomization to include both steps and all parameters – which is pretty crazy! I may need to add some other ways of randomizing that produces less chaos.
These fixes will roll out with 2.23.1 (2.23.0 is still being rolled out)
Cheers, Bret
1 Like
Awesome, thanks so much! Hoping for trigger inputs somewhere down the line.
I love randomization options. There could be different random options in right click menu. Randomize steps, randomize parameters, etc.