asked Nov 30 '18 at 13:07 by Round (1)

Howdy! This is something that's been bugging me for a while, especially as I'm quite the perfectionist! Here's my dilemma...

So, I've noticed, over the years that I've been using Bitwig now, when I bounce both midi and audio tracks to audio, the new bounces can sometimes be out of time from the original source that I bounced down. That's probably a confusing statement, so I included a screen shot for your reference. I only noticed because I like to bounce a lot of my processed midi parts to audio for my workflow, and I've not seen anyone else mention it on the forum or online anywhere, so I thought I'd draw attention to it, to see if anyone else has noticed this.

The screen shot I have taken, is a bounce of the same Kickdrum taken from SampleMagic's Stacker, which I use to create my kick drums. I have increased the clip gain by +20db to highlight the start of the waveforms and I have also zoomed as far in as I can to demonstrate to you guys. The different bounces are colour coded, if they're the same colour, it means they've been bounced with exactly the same amount of plug-ins following Stacker in the process chain. If they're different colours, it means they have different plug-ins added to the chain of the midi track.

Now my question is... is Bitwig bouncing samples slightly out of time, or is it just a graphical misrepresentation of the bounced waveforms? I'm assuming it's caused by latency of adding more plug-ins, but sometimes I create multiple bounces of the same part and each bounce varies in their timings. These Kick drum bounces I have created to demonstrate for you have been relatively well behaved and quite close together in their timings. Maybe I should create another example using another synthesiser... Or maybe because it's a demonstration project and the only plug-ins used are the ones for the example, meaning it's not a CPU heavy project, unlike a project with 30+ tracks loaded with plug-ins? I don't know...

Anyway, let's begin - the red bounces are the first unprocessed bounces I created, with only Stacker being used, the yellow bounce was my final bounce of the orange midi clip containing the most amount of plug-ins to process Stacker.

The Orange midi clip at the top is perfectly snapped the to grid at Bar 2.1... It is triggering the synthesised kick being made by Stacker. You can see the only midi note I have entered is triggering on Bar 2.1.

The three red clips on tracks Audio 9, 8 & 7 are the first three separate but identical bounces of the orange midi clip, with no further plug-ins or processing applied. You can see that the resulting samples all start marginally before Bar 2.1 - which shouldn't be the case surely, seeing as there is no triggered midi happening UNTIL bar 2.1, NOT before...

The two blue audio tracks (6 &5) now have Softube's Tape added to the signal chain after Stacker on the midi channel. This has added a slight Latency compensation to the midi channel, as it no doubt requires more processing power now that another plug-in has been added to the chain. Again, if you look at my screen shot, you can now see that the 2 blue bounces of the same orange midi clip now start even sooner!

On the purple Audio 4 track - I added Fabfilter's Pro L2 to the midi tracks chain to create even higher latency compensation. So now there is Stacker, Tape and Pro L2 being processed on the orange midi track. Again, you can see the resulting bounce down of the midi track is at a different timing, but still starts before the Bar 2.1.

On the green Audio 3 track, I then also added a Fabfilter Pro C2 with Lookahead turned on to try and create an even higher latency compensation. Again, the sample now starts in a different place, but still before Bar 2.1...

On the final bounce I created, the yellow Audio 2 track - I added NI's Bite plug and also their VC160 compressor, again, just to try and increase the latency compensation. You can see that this has had the biggest effect on the bounce down, the sampled kick now starts even sooner, before any of the other audio bounces of the orange midi track.

Now, obviously, it's not a major concern, and I was zoomed in on those waveforms as far as I could go so the timing misalignment would most likely not be audible; but it does bother me, knowing that every bounce down I do, could potentially be out of alignment, even if it is barely noticeable. I notice greater variances when using different plug-ins, such as other synthesisers and I bounce those synth parts down to audio. Some bounces start before their triggered midi part, some bounces start AFTER the triggered midi part (again, only very marginally). There have been occasions where I have bounced down synth parts and I can hear that they don't sound quite in time like the midi part did, so, even though they can be subtle misalignments, they can sometimes be noticeable.

Does anyone have any definitive answer as to why this happens? Is it down to latency? It's a gripe having to go and check to see if your samples are starting exactly in time and then potentially having to re-adjust them. To date, this has been the only issue that's continually bugged me since I've owned Bitwig since it's release.

NOTE: I couldn't upload the image, so here's the link to view it instead:

Have you looked at more than just the first beat? I mean, is each track consistently offset? I dont know why it would ever bother you though. How many bounces would you need to do before a consistent offset built up enough to be audible? Judging by how close all of you examples are, it looks like you'd never need to worry about it. I guess combining two of those tracks could lead to some phasing artifacts - but even if that was noticeable it would probably add character anyway!


answered Nov 30 '18 at 14:41 by djx (315)

Yeah, every track I bounce to audio is always offset by some margin, no matter how small. For the purpose of the demonstration, I thought a strong transient waveform would be best for demonstration purposes. Like I said, I cope by re-aligning what I can, as I normally have to when recording my hardware synths, but I'm curious as to why this happens. In most cases the alignments are so marginal that you can't notice it with your ears. I've noticed fluctuations in timing/alignment in kick loops I've bounced too. I'll post another demonstration of a kick loop when I get chance later.

  — (Nov 30 '18 at 15:19) Round

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

Markdown Basics

  • *italic* or __italic__
  • **bold** or __bold__
  • link:[text]( "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported



Asked: Nov 30 '18 at 13:07

Seen: 717 times

Last updated: Nov 30 '18 at 15:19