Take a breath man! I agree with you and your methodology. What I’m asking is, overall, what is the average or range of jitter for each track? I can’t tell from just zoomed in pics.
2 Likes
If I had to make a completely uneducated guess without any knowledge whatsoever, I’d guess that audio is processed in small blocks (maybe 32 samples) and they trigger samples at the start of a block.
Would be max 0.7 ms jitter for that example block size (@ 48kHz)
1 Like
hum this seems very weird to me, let me put my computer and plug my dt2. i think if this was the case i would have noticed . i think your computer set up is al over the place not DT2
Again, I designed this test to be done on device. No PC required.
I only included the screenshots from the DAW because it’s easier than taking multiple photos on my phone of the Slice screen on the DT2.
I have now added photos directly from the DT2 to the first post. You can clearly see the issue on device no DAW required.
I’d be curious to see the maximum displacement from expected timing.
If it’s on the order of a millisecond I think it’s unlikely you would be able to hear the jitter, even if playing alongside a perfectly timed device.
1 Like
@LackofRAM in the thread here Describes the issues with this, you need to record everything at once and if you want stuff ‘off time’ but with all tracks synced so they are all ‘off’ by the same amount on the same beat, and then need to warp it in a DAW to get it bang on time
If you record tracks separately via overbridge they will each have their own timing and you then need to warp each stem individually:
and here
which shows the exact issue, People thinking they can solve the problem by buying expensive sync gear.
sorry i made the test with a tight hihat sample and i can tell you even i did not upload the sample on the computer that it can’t be so fucked up as your exemple up there just by hear
Ok, now load it into the slice machine and create 16 linear slices (no transient detection) and zoom in on the start and see if the transient hits at the same time for each.
You also realize I’m trying to convey the issue that is reported in other threads I’ve linked above in an easy to understand ascii diagram right?
1 Like
i created a 4 slice but it’s dead on time , gonna retry with 16
remember use the zoom.
yes there is jitter on max zoom, definitely, but was aware of that because i created a sample pack with lot of grid samples.
Right that’s what this entire thread is about.
1 Like
@ripsaw Have you written an email to Elektron support with your findings?
I wanted input from others first to know if this was a known issue, something only my unit was doing, or an unknown issue that is wide spread.
Instead I’ve had responses talking about a known recording start timing inconsistency and others where they pick apart an ascii diagram I posted purely to differentiate the recording start timing inconsistency vs the actual issue I was reporting.
4 Likes
ok so i had 1ms late on first slice and 1ms to early on slice 2 and 3 and again 2ms late on fourth slice, took me 1 minute to correct on dt2. there is jitter but you can correct as soon as you know , signal it to elektron , personnaly i decided to stop slicing with digitakt 2 or live with it
1 Like
The issue is less slicing on device and per the other thread I quoted earlier that when recording into a DAW even via overbridge this jitter remains.
This means if you record different tracks at different times even ones from the same project they will not sync up with each other as they will all have different amounts of jitter happening at different times.
If all I was creating was ambient tracks this would not be an issue.
1 Like
Im probably going to regret mentioning this, but doesnt the slicing do a zero snap based on the left channel in slice machine? For perfectly even slices, dont you need grid machine?
Kind of interesting that DT1 is unaffected.
When you use a SLICE machine on the DT2 you get the option of having transient detection ON or OFF
If you choose OFF it just creates equispaced division the same sort you get from GRID mode (where GRID is limited to multiples of 2). Having this OFF is exactly what I said to do in the first post
The only reason we are using slice mode for this is because it has a zoom function to see the issue on device.
and as I said before if you load a loop created in a DAW to the SLICE machine above without touching the slices you will see the transients line up with the grid perfectly.
i don’t think i can reproduce that with overbridge , i already checked because i’m a kin of obsessed by this . with one shots playing and recorded with overbridge i was really clean on live grid. i think this comes from resampling the device more
- How are you clocking the device?
- What sync settings are you using in over bridge?
- Are you on windows or mac?