@slicetwo I know this problem, I ran into it yesterday. Loaded a sample from the card to static machine, trimmed it, set the base tempo, and then no time-stretching.
Finally I saved (overwrote) the sample attributes and voilá, time-stretching started working properly 
Maybe a bug in Octatrack.
whoa! fascinating. Cheers for the potentially required workaround
So is it because you trimmed too short , as I pointed out, or because of unsaved attributes ?
I’m pretty sure it’s because of sample length, not attributes saving.
I saved a sample trimmed to 6143 sample chunks.
Timestretch still didn’t work after.
Trimming shouldn’t have mattered. I set trim back to the whole sample but that didn’t help.
Attributes were available as it was a loaded sample, so an owerwite was not expected to make time-stretching work. But that helped only.
Oh snap. I will totally try that. Thanks!!!
Ok. Let us know if it works for you.
I can’t reproduce your issue. Could you elaborate in order to check ?
Does anybody agree there is a minimum sample length (6143 sample chuncks/139ms) for timestretching ? Feel lonely on this…
Didn’t work for me. Super bummer. It works for other samples in that same folder, though, so it must just be something with that sound. Maybe I’ll try re-encoding the sound.
What is your sample length, please ? :
The sample length is .11 seconds for the one that’s not working, and .25 for the one that is working.
0.11 sec is 110 ms, so it seems below 139ms.
I bet this is why it doesn’t work.
Hmm. Ok. I’ll try smaller files and see what happens.
Hi,
Yes, I can confirm that TSTR stops working for sample lengths shorter than 6143 samples. But this seems to be the case for Static machines only. For Flex I’ve got different results so far, ranging from around 8804 to 12575. Was tweaking various parameters and re-recording samples so can’t determine what’s making the difference.
Another finding was that once TSTR ceases working, if Loop mode is set to PIPO, sample playback would only loop the very last portion of the playback range, thus producing a high-pitched squeal.
Hi thanks.
It’s weird I found 6143 for Flex and Statics.
Further tests would be welcome.
Pretty sure it’s the TRIM Loop point. 
(Loop point works with LEN=OFF)
For slices, I had the same 6143 result.
(Loop point works with LEN=SLICE)
Ok, I can accept that minimum sample length for time-stretching is what you have measured. Thanks for that info.
However my sample didn’t get time-stretched even with full length (2-3 seconds) until I saved its attributes.
i wonder if the same scenario is repeatable with another sample of similar length of about two seconds?
Yeah, it may be difficult to reproduce. But it definitely happened Will let you know if I run into it again.