This is almost certainly a visual scaling bug, good eye.
EDIT: actually, I think this is the same as it was on the DT1 after all. The values seem a bit arbitrary but I think they’re supposed to represent the proportion of feedback (with the high range almost doubling upon each repeat). Skipped values are normal when scaling up 0-127, of course.
This is not considered a bug, it bugs users for sure and has been discussed before … having some workaround which helps users would be a much needed FR … it’s a vital request imho, I have requested it
I’ve had a response to a bug that I logged that suggests something weird IS happening with multiple files with the same hash. Bug not resolved yet, I will post when I have more info.
In my case the symptoms was that I could see multiple samples allocated to the same slot. This happened after loading a sample pack that had multiple failures (like 5-10%).
Doesn’t seem like a bug, they have to skip values to get the double range into oscillation
Any theoretical ‘gaps’ are essentially meaningless in the real world
They stopped using needlessly high resolution on many other parameters on DT2 fwiw
I’d have to look back at the DT2, but I’m simply suggesting it’s a display bug with the scaling. Maybe the range is the same as it always was, I do agree that when they scale things this way of course they have to skip values visually. It’s just interpolation.
There’s 128 values with 7bit ranges or 16k for 14 bits, but that’s overkill for such a parameter… so if the range is almost 200 there has to be gaps in those 127 integers
I haven’t used pitch bend yet and all the values are at 0 where they should be
Thank you for the suggestion though
Indeed, I’m a programmer; I know how interpolation works. I just don’t recall the top of the range being that number in the DT1 - maybe it’s the same.
Thanks for confirming you’ve seen this too. I don’t know how many people are actually trying to load older projects and audition them for errors. I certainly understand they won’t necessarily sound 1:1 with new algorithms for the sample rate reduction etc., but the file system thing is definitely not expected behavior.
In my case, not an older project, just a soundpack
Ahhh right. So maybe when a file isn’t successfully transferred, the hash assigned to it is reused. When you attempt to reimport, it’s unaware of the collision. Just my best guess.
Curious. Is the DT receiving midi clock?
I had a problem where my timestretched samples were totally warped and out of tune. I noticed the tempo was fluctuating because the box was receiving clock. So I turned that off. Samples sounded fine after that.
I’ve experienced some bugs, transfer fails and now the screen has gone bad as some have posted above. I’ll have to return it and I don’t know if I’m confident in a replacement.
FUNC + NO in perform mode not return to the performed kit but to the current pattern kit
Yes, from an E-RM midiclock.
No suggestion here. It is a bug @BigJ mentioned !
Once you used pitchbend and leave it in the middle, value is 64 instead of 0. And the range is 0 / 128 instead of - 128 / 128.
Probably easy to fix.
Should be easy to code
Was playing around with the PWM wavetable and trying to actually make a PWM bass using the lfo on SLICE but I’m getting an unexpected behavior… it’s like the lower the pitch I play, the more lopsided and saw-like and short the motion of the sample playhead gets. Bug or help me understand?
Sorry to hear about your screen. I am hoping my replacement doesn’t have the same issue as probably from the same batch at the dealer (shout out to Anderon’s for the great service).
Just heard back from support that this will be fixed in the next firmware (no date yet).