CPU issues

[MaxC]

I typically get audio drop outs at 50% core use on Windows 10. So that is "maxed out" for me. I don't think you are going to get to 100% on a core before something breaks... real-time audio is too sensitive to CPU latency.

DPC (Delayed Procedure Call) latency is the key, no matter how fast your CPU is. Run LatencyMon http://www.resplendence.com/latencymon to test your system for DPC latency.

Ahhhh, so there is more to the story than raw CPU horsepower then. I ran DPC Latency Checker for about a minute at it peaked at 20 microseconds, which seems pretty darn good when the graph deems anything under 500 to be suitable for real-time audio.

Oh well, even if the CPU is running at half capacity, I can at least live with it if I know that's all anyone can really expect for this purpose. I just wanted to make sure the issue wasn't specific to my rig. Thank you for taking the time to respond!

[zzz00m]

I typically get audio drop outs at 50% core use on Windows 10. So that is "maxed out" for me. I don't think you are going to get to 100% on a core before something breaks... real-time audio is too sensitive to CPU latency.

DPC (Delayed Procedure Call) latency is the key, no matter how fast your CPU is. Run LatencyMon http://www.resplendence.com/latencymon to test your system for DPC latency.

Ahhhh, so there is more to the story than raw CPU horsepower then. I ran DPC Latency Checker for about a minute at it peaked at 20 microseconds, which seems pretty darn good when the graph deems anything under 500 to be suitable for real-time audio.

Oh well, even if the CPU is running at half capacity, I can at least live with it if I know that's all anyone can really expect for this purpose. I just wanted to make sure the issue wasn't specific to my rig. Thank you for taking the time to respond!

Since you are still using Win 7, you may also want to take a look at 'DPC Latency Checker'. https://www.thesycon.de/eng/latency_check.shtml
It is not supported on newer Windows yet, but the site also has some great tips for analyzing dropouts, and why they occur.

And this is the best video I have ever seen on the subject!

CPU Performance vs. Real-Time Performance in Digital Audio Workstations (DAW)

[MaxC]

Thanks for sharing that video, zzz00m. It was very informative, even if the cause of reduced performance remains something of a wildcard after you've ruled out DPC latency issues. At the very least, it gave me a better perspective of what is going on behind the scenes.

[zzz00m]

Take a look at this related post from the Tracktion dev:

Tracktion and CPU Optimization, part 2

viewtopic.php?f=22&t=501926&p=7039859&h ... n#p7039859

• The number of cores available to W9 are logical cores, this basically takes in to account hyperthreading to double the number of physical cores. Take a look at Windows Task Manager and you'll see the same thing. It's just how modern CPUs work.

• We parallelise processing by track (where possible) so essentially multiple tracks can be being processed by different cores at the same time. There are some exceptions to this such as submix tracks which need to processed in series.

• Audio file reading is done by memory mapping the files and using the OS to page in the sections about to be read.

• If you want to free up resources, the best thing to do is simply freeze some tracks.

• In our experience, it's almost always plugins that use the most CPU so maybe prod them to optimise a bit if you've got some particularly hungry ones.

[chico.co.uk]

Best easy fixes on windows I've found have been to turn off any WiFi or Bluetooth adapters while using a DAW (the interrupts cause issues) and to use SSD drives. Both made noticeable improvements.

[MaxC]

Take a look at this related post from the Tracktion dev:

Tracktion and CPU Optimization, part 2

viewtopic.php?f=22&t=501926&p=7039859&h ... n#p7039859

• We parallelise processing by track (where possible) so essentially multiple tracks can be being processed by different cores at the same time. There are some exceptions to this such as submix tracks which need to processed in series.

Does this mean frequent use of submixes will lower the threshold for CPU bottlenecking? I send almost all of my tracks to numerous group buses, which are then sent to a master bus with a boatload of plugins on it. Perhaps this is a factor for me...

• Audio file reading is done by memory mapping the files and using the OS to page in the sections about to be read.

Interesting. I currently have a tiny page file (512 MB) allocated to the same physical drive that my audio tracks are rendered to. This has me wondering if I should increase the page file size or reallocate the page file to an SSD. This might also explain why Latency Mon shows Tracktion outputting thousands of pagefaults when playing audio (though curiously the pagefault count caps at 8300 no matter how long a session I run). I may just have to try fiddling with the page file settings and see if I notice an improvement.

[MaxC]

Best easy fixes on windows I've found have been to turn off any WiFi or Bluetooth adapters while using a DAW (the interrupts cause issues) and to use SSD drives. Both made noticeable improvements.

I have my OS on an SSD and my sample libraries on a separate SSD. I'm still using an HDD to store my Tracktion sessions and rendered audio because I've seen conflicting reports on whether to use SSD to record long audio files. Has anyone made the move from HDD to SSD in this regard specifically and can attest to its benefits?

Thanks for the tips!