An audio "crackle" not present in any other DAW or audio system on my system

[Milkman]

For a while now, Ive been working between Cubase (currently 15) and Bitwig (5 and beta 6).

I use a Focusrite 18i20 gen 3 interface. Windows 10 and windows 11. This is a recently rebuilt workstation (2 months), and I am a senior net/sysadmin with experience in live event streaming and audio production. There are no ground loops in my channels.

Right now, I hear a "crackle" or "snap" in Bitwig's audio, since around version 5.x. This crackle isnt present in Cubase, the lite version of Ableton I tested last year, or any standalone application or instrument. This "crackle" does not record into audio, indicating it is being picked up... outside Bitwig?

But in Cubase 15 and standalone, there is no crackle. I use nearly identical templates, all the same plugins, all the same basic mixing philosophy on both. Cubase....... is set to 64 bit DSP and Bitwig is 32bit float. This is the only difference I can see.

Does anyone else... ever hear a 'SNAP' (very short, almost drum transient sound, almost sounds like a crackling fire but shorter) in their BWS audio, any versions? This has followed me across rebuilds, across 2 interfaces (replaced a 6i8 with a 18i20), and again is NOT in the recorded audio. Its driving me nuts and making me use C15 more than BWS right now just to escape that crackle.

[Milkman]

I should add- CPU usage is identical between both. Interface buffer is the same for both. Inputs/outputs routing is the same for both. Everything I know of is the same for both. No hum, no buzz, no ground loop sound, just occasional 'crack'.

No 'cracks' in windows audio. No cracks anywhere... but Bitwig. Its incomprehensible to me.
It doesnt record into audio, so I usually ignore it but I have people in the room with me, listening to what I do live... so the crack is audible.

[Milkman]

Ill have to create a ticket for this. It makes no sense to me, and its becoming so annoying that its killing my live sessions and forcing me to use cubase more than I want to.

Cant wait to make a ticket for "crackling fireplace sounds in bitwig audio but not cubase or any other audio on this system". The reply to this should be fun. Vendors love random crap like this.

[Broken]

That's a weird one. You'll probably need to provide support a link to a video of it happening.

Does it happen at the same points within a project or is it random? Does it happen at all buffer settings?

[Milkman]

That's a weird one. You'll probably need to provide support a link to a video of it happening.

Does it happen at the same points within a project or is it random? Does it happen at all buffer settings?

Its random. Sounds like what you would expect a mild, main channel ground loop to sound like (very light, occasional crackle/snap, like a small fire burning). Any buffer settings. Any output channels. Headphones, monitors.

Zero presence of that sound in cubase. Frequent presence in bitwig.

Yeah, videos are all the rage these days. No customer service team takes customers for their word anymore, even if their customer is qualified to train their own CSRs and build their own infrastructure, lol. We all get to jump through the same hoops (record a VIDEO of the issue, upload it to some surveillance happy 3rd party host).

I rarely actually record video for vendors because a) I usually solve it myself, b) the assertion is offensive from the onset and usually involves 3rd party hosts, which I dont use.

No, this will be a fun one, for sure. In my decades of working in tech, Ive sometimes run into really weird issues, like this one.

[Broken]

Think of it from the perspective of the developer trying to troubleshoot this. Would you want to read a subjective description or watch/listen to an actual demonstration of the problem? It's less an issue of them not taking you at your word and much more about the efficient use of time for everyone involved.

[Milkman]

Think of it from the perspective of the developer trying to troubleshoot this. Would you want to read a subjective description or watch/listen to an actual demonstration of the problem? It's less an issue of them not taking you at your word and much more about the efficient use of time for everyone involved.

Ive been in tech for close to 40 years. I have been the person reading subjective, customer composed problem descriptions, including descriptions shared with me across language barriers and in the middle of the night during a 24/7 on-call shift, more times than I could possibly count.

That experience has *never* changed me into the type of person who would first accuse customers of lying unless they jumped through hoops for me, shifting the onus of customer service to them instead of me. Im not confused about what this is. This is understaffed, underpaid corporate support teams deciding, as a matter of principle, regardless who they are talking to, that everyone is a liar and that everyone has to jump through hoops.

Again -- we all know about email attachment size limitations, right?
And we know the solution is to use 3rd party, non private file transfer systems like dropbox to upload video because almost every single company has abandoned the FTP upload methodology we used to use because, well duhhh: that, too, is the customer's fault somehow. Since I am ethical and smarter than that, I dont use 3rd party surveillance systems to transfer files, so I am immediately seen as "refusing the offer for support". lol I 100% get this, and Ive watched as this same philosophy has spread to EVERY industry sector where it makes sense.

[Broken]

As far as I know, no one here or at Bitwig has accused you of lying. It's just unfortunate that with these weird cases, where it's likely hardware related and not reproducible by the developers, it will require some hoop jumping on your end.

Anyway, let's have a crack logically going through this to find a possible culprit.

1. Problem appears from 5.x (it would help to know what version, specifically)
2. The description of the noise reads like some type of electrical noise making its way into your audio hardware, as internally recorded audio is clean.
3. You have rebuilt your system and the problem persists (were any new parts used? what are the specs of the system?)
4. Problem occurs at all buffer settings and at random points in a project (so unlikely CPU related)

Without knowing anything hardware specific, let's start with the GPU. A loaded up GPU inducing noise where it's not wanted isn't unheard of. If the noise has been present since >=5.2, when the graphics engine was rewritten to use the GPU, that would also line up.

It could be as simple as Bitwig being harder on the GPU than Cubase or Ableton. Are you in the position to swap out some GFX hardware?

[ManCat]

First thing I would think about, is try the repair button, if you haven't done it already, then a complete removal and reinstall. Don't forget to backup, but since you already have 80+ years of experience and know Cobol or Fortran, you probably already knew . I had to fix the internal bitwig DB a couple of times, because when I wanted to add a VST instrument or something simular, it was terribly slow. Had to rebuild the DB. Usually after an upgrade as far as I can remember.

Windows might also somehow be the culprit, have you checked the high performance setting? Maybe, due to an upgrade, they changed it back to balanced? These would be the first things that comes to mind.

[Milkman]

As far as I know, no one here or at Bitwig has accused you of lying. It's just unfortunate that with these weird cases, where it's likely hardware related and not reproducible by the developers, it will require some hoop jumping on your end.

Anyway, let's have a crack logically going through this to find a possible culprit.

1. Problem appears from 5.x (it would help to know what version, specifically)
2. The description of the noise reads like some type of electrical noise making its way into your audio hardware, as internally recorded audio is clean.
3. You have rebuilt your system and the problem persists (were any new parts used? what are the specs of the system?)
4. Problem occurs at all buffer settings and at random points in a project (so unlikely CPU related)

Without knowing anything hardware specific, let's start with the GPU. A loaded up GPU inducing noise where it's not wanted isn't unheard of. If the noise has been present since >=5.2, when the graphics engine was rewritten to use the GPU, that would also line up.

It could be as simple as Bitwig being harder on the GPU than Cubase or Ableton. Are you in the position to swap out some GFX hardware?

I dont recall claiming someone from bitwig accused me of lying. I was talking about a larger problem, and replying to your comment about uploading videos to 'prove' it to companies. Ive been asked to record video of... logged, server-side issues (the vendor had all the data...) more times than I can count -- this is intentional customer service sabotage more than it is verification.

Ive been doing systems stuff for a long time, and this sound does sound like an electrical 'pop' from a ground loop or some form of EM bleeding from somewhere, but the way it vanishes when Im doing *anything* but running Bitwig is a confounding reality for me. Its not a 'swish', its not a 'hum' or digital glitch pattern sound from rendering(GPU), its short, mostly single crack. Occasionally I get 2-3 in a row.

I am not in a position to swap out GPUs, nor is that a reasonable solution even if I could. If I was hearing a 'crack' from heavy GPU usage (rendering the UI shouldnt be that heavy), I would expect to hear that same crack in many GPU-heavy applications, in GPU stress tests, in video games, etc. The sound is ONLY present in BWS 5, 6. I will need vendor support for this, but Im waiting for the 6.0 release craziness to die down before adding another ticket to their pile.

Thanks for trying to troubleshoot with me here.

[Dionysos]

Especially for hard-to-reproduce bugs that might depend a lot on context or the user's system (and this seems to be the case here), video recordings are a godsend and can often give a programmer a vital hint as to what the cause of an issue might be. This has NOTHING to do with not taking the customer for their word.

[Milkman]

Especially for hard-to-reproduce bugs that might depend a lot on context or the user's system (and this seems to be the case here), video recordings are a godsend and can often give a programmer a vital hint as to what the cause of an issue might be. This has NOTHING to do with not taking the customer for their word.

Again, talking about a larger problem. A larger problem that has affected almost every industry sector in some way, which serves to sabotage and put up barriers to good support via the internet, such as the necessity to use 3rd party companies just to upload a file.

My description of this should be adequate for any serious support engineer, to start off the support. A video recording of my monitors producing a recording of the sound would do *just as much* to aid troubleshooting as my specific description of that problem, and would only serve to prove Im not making it up, lol.

Ill have an email chain with bitwig.

[questionaire]

I had similar problems between Ableton and Bitwig where Bitwig under performed against Ableton. I run Windows 10 home edition and had no "high performance plan" available in power settings, that has been solved by executing the command: powercfg /s SCHEME_MIN
I could adjust all to best performance. I also have a script where "processor_performance_boost_mode" can be added to the power plan and this one should be adjusted from aggressive to off.
And recently i have without Lenovo support upgraded to latest gpu drivers and a lot of issues are gone now.

And under "device manager" switch off "microsoft battery ACPI compliant method"

Maybe it is of no help but at least some ideas which helped to improve performance for Bitwig on my system.

[Broken]

I dont recall claiming someone from bitwig accused me of lying. I was talking about a larger problem, and replying to your comment about uploading videos to 'prove' it to companies.

You claimed that a request for video is "offensive" and that, "No customer service team takes customers for their word anymore". That's what I responded to, and then you doubled down with more. It needed to be made clear that Bitwig hasn't accused you of lying. And if support does eventually request a video, that won't be them assuming you're lying and trying to get you to "jump through hoops". Those assumptions are yours. Your opinions on general corporate support are irrelevant to what helps with an efficient Bitwig bug hunt.

The info you've provided is pretty useless. Without being able to replicate it, hey won't be able to do much about.

Anyway, here's the way I'd go forward:

1. Install Bitwig on another computer. You've worked in IT for 40 years, so that shouldn't be anything that comes close to what could be described as having to jump through hoops.

2. If the problem can be replicated on a second system, create a minimal Bitwig project (Bitwig devices only) that can also produce it.

3. Provide a concise bug report along with the minimal Bitwig project.

If they can't reproduce it and you don't want, or refuse for other reasons, to provide any more assistance, that's the end of it.

Hopefully it's an easy fix, but I doubt it. No one else seems to be having the issue, and it's unlikely that an audio bug in a professional DAW would have no reports, particularly one that has been around since 5.x.

Good luck.

[gurulogic]

That's a weird one. You'll probably need to provide support a link to a video of it happening.

Does it happen at the same points within a project or is it random? Does it happen at all buffer settings?

Its random. Sounds like what you would expect a mild, main channel ground loop to sound like (very light, occasional crackle/snap, like a small fire burning). Any buffer settings. Any output channels. Headphones, monitors.

What you are describing sounds to me like some bad bits/clock/sample rate mis-match in a digital audio stream. Any adat optical or spdif in play? Or maybe could be an incompatibility between Bitwig and the Focusrite driver.
I would suggest trying the interface with an alternate ASIO driver, or seeing if you can recreate the issue using any other interface.
Also, you could consider routing the analog outputs back into inputs on the interface and see if you can capture an example of the audio that you are hearing.