Thanks yes you sorted me out! Will do it the long way for now haha cheers
How old an interface is doesn’t really matter. Saffire and RME are both higher end interfaces. RME are known for their incredibly low latency interfaces.
Comparing to lower end interfaces is fairer in my humble opinion. Let’s not forget that for less than the price of an average RME interface we get an audio interface which has an entire groove box built in 
1 Like
Must admit that Im surprised at the low round trip latency from something like an NI interface. I guess Mac has better numbers than windows machines on that front.
I mean, I know it’s a decent interface, that’s why I bought it. 
But it was only £200 new. Nowhere near RME money. I suppose they don’t really make an equivalent now, the Clarett 4Pre is a lot more expensive and the Scarlett 18i8 is USB only.
1 Like
Yeah. I am pretty curious as to why these differences are that great. My A&H db4 has low latency like RME interfaces or at least near that and it’s usb. My guess is it’s hardware related since it would explain it cost wise. But that’s just a guess…
The driver looks useful, even if latency is high. can always track out while monitoring off analog outs. Utilitarian, stuck at 48khz, thou is it 24bit?
Now the plugin is where things get interesting, and potentially groundbreaking. we shall see. i’m 50/50 it will work.
I’ve just checked my Roland MX-1 over USB: 6.2ms roundtrip, 2.4ms out with 128 sample buffer @ 96kHz. With 64 sample buffer it’s 4.9ms and 1.8ms. If I’m recording on ≥10 input channels at once (max is 18 in / 18 out) I do get the odd crackle even at 128. But I would expect Overbridge on an Elektron box to do at least as well - I don’t see why it shouldn’t.
13" Macbook Pro, late 2016, 3.3Ghz i7 by the way. Sierra 10.12.6.
Yeap but at 96khz you get lower latency than at 48khz. I’ll test this this week to see what results I get with OB. Still, with 7ms input latency I don’t see what the problem is exactly. Lower is better but there is a point where it doesn’t really matter for production purposes. It doesn’t affect my production / workflow at least.
3 Likes
Yeah, I need to check what it looks like at 48. Latency isn’t everything but I guess it depends on your personal setup and workflow. I tend to do everything either ITB or OTB, I rarely combine, so it’s not a big deal for me. Just observing that it seems a bit below par, but of course, this is only based on one person’s experience with a beta build.
1 Like
Because you monitor off the direct outs? I’m just envisioning a scenario where you are monitoring round trip and the latency is too high to make adjustments on the fly, at least in a responsive way.
It’s safe to assume the latency for the plugin version is going to be [gulp] even higher?
If anything I’d assume the latency will be lower, considering we’re still in early stages of the beta.
1 Like
I monitor direct, not round trip. I’m an instrumentalist as well and recorded/produced several bands in the past. Never done otherwise than direct monitoring as that is just the way to go when recording stuff live (imho).
No idea how the plugin latency will be, but I don’t see why it would be much different or at all. 
For the Analog Heat it will be different in that you can send audio from DAW through Heat back to DAW, so there the roundtrip latency does apply. 18 ms wouldn’t be that bad in that case either. At the very least an immense improvement to OB1.
It’s one thing to have a driver piping audio into your computer, and it be the master clock…it’s another thing to be feeding, via USB, audio into a DAW that is clocked by a third-party interface. You can’t really have two drivers running simultaneously, with any reliability w/out clocking. This is the ‘holy grail’ of OB, and I’m super interested to see how it works out.
Eventually I’d like to track out an entire session with: 8 channels of DT, and 16 channels of audio from my Apollo soundcard, all in one pass.
1 Like
Sounds awesome really hope it works out.
Quick question:
When you sequence with DT some VST plugins using Midi over USB, there are timing problems(jitter).
Using OB 2, can DT send midi to VST plugins without such midi timing problems??
Hi Agua,
If you mean that the midi sequence on the DT suffers from latency, then I guess so. But by setting the buffer range to 64 samples the latency is so short that it’s hardly noticeable. I have to delay compensate tracks a couple of mili seconds sometimes but that’s it. Nothing out of the ordenary in my experience.
I don’t want to say ‘latency’, but jitter: When I record midi out from DT (USB or MIDI Out) in my PC, the notes have a small randomization to its start and lenght. And unwanted humanization. I think it happens with all midi devices.
Take a look to this picture. This has been recorded from DT, which has 8 steps with a length around .313. I was expecting a result like the red notes, but instead what I record is pretty different. Does this happens when recording MIDI with OB? Or at least, does it improve?
Looks like quite heavy midi jitter to me. You need a stabile jitter free midi clock.
And this is when you go buy an external midi sync box to your DAW, like the E-RM Multiclock.
Or, you press Q on your DAW keybord to quantize the recorded notes and wait for Overbridge to see what it brings.
I think the problem doesn’t relate with midi clock, but with the way computer work with buffers. A test with OB2 would be nice!
1 Like
Yeah, it might be audio buffers (latency) but latency is usually a constant delay when recording. You’re getting fluctuations in midi note on and midi note off messages to you DAW.
A way to test if midi clock is giving you problem is to:
- Set your DAW and Digitakt tempo to 120 BPM.
- Disable the midi clock out from your DAW and also disable the Digitakt midi sync.
- Press record in your DAW and hit the play button on your Digitakt.
- Stop recording and move the entire set of midi notes so the first note is on grid.
So now, how does the other midi notes look? First one is dead on the grid, but do the the others line up on grid or are they early/late and with different or the same note lengths?