12 cores not being used equally (cpu errors)

[musicman691]

Originally Posted by stevedresser83

Not talking about production or running VI's. Just purely mixing, sessions with lots of plugins. All audio with plugins. I wonder if anyone has ever done a DAW comparison in how each daw utilizes cpu usage, cores, ram, etc

They all suck at balancing cpu usage. You definitely should look into VEPro; I think there's a demo but you'll need a physical Steinberg or Vienna key as it won't work with the s/w version.

[Darryl Ramm]

If this is a 6 core hyperthreaded system you also don’t have 12 actual cores here. Obsessing about what CPU meters show with hyperthreading can be a trap.... you should not necessarily expect flat virtual core use.

[stevedresser83]

im not obsessing about what cpu meter shows. I'm noticing that 3 out of 12 cores whether hyperthreading or not are being used, 15% of ram being used, and i'm getting cpu errors. I have a machine clearly not being utilized but protools is acting as if its completely maxed out. So i'm wondering what can be done about that.

Looking into VEpro now. Not sure if its really what I want for my workflow. I'd imagine to access all those other machines I will have to be working within VEpro interface. Could be nice though production

[Darryl Ramm]

Originally Posted by stevedresser83

im not obsessing about what cpu meter shows. I'm noticing that 3 out of 12 cores whether hyperthreading or not are being used, 15% of ram being used, and i'm getting cpu errors. I have a machine clearly not being utilized but protools is acting as if its completely maxed out. So i'm wondering what can be done about that.

Work at a lower sample rate, lots of folks here have articulated why 192kHz is likely not needed. Fully optimize the system, every last thing done. Use freeze/commit. Test different plugins to find any bad actors. You can try disabling hyperthreading (results are very plugin dependent). Avoid long bus chains/deep stacking of plugins. Look at the specific AAE error you are getting and what other folks have done. VE Pro is an option (would only use for VIs).

You are getting CPU latency errors, the meters do not directly show processing latency. I would be doing all the above and not paying attention to what the meters show. Changes that reduce CPU errors may or may not change what you see in the meters.

[LDS]

Originally Posted by Emi

...not even Pro Tools is needed, if you go this route.

I don’t get the implications. You don’t need Pro Tools to run 192kHz sessions.

[7Seas]

Originally Posted by musicman691

They all suck at balancing cpu usage. You definitely should look into VEPro; I think there's a demo but you'll need a physical Steinberg or Vienna key as it won't work with the s/w version.

Isn’t VEP mainly for VIs? I never dwelled into it because I understood that VIs were VEP’s thing. Will it work on plugins as well? Macos here...


Sent from my iPad using Tapatalk

[the.engineer]

https://m.youtube.com/watch?v=jGP7-bxPJXU

This is a good video to watch!


James Welch
Recording Engineer | Producer
Masterlink Productions

Mac Pro 5,1 (Mid 2010) | 3.46GHz Six-Core Intel Xeon | 32GB Ram | AMD Radeon R9 280X 3072 MB | 256GB Samsung SM951 SSD Blade (PCIe) System Drive | 1TB WDC Sample Drive | 4TB RAID 0 Seagate Audio Drive | UAD-2 Quad PCIe
Sierra 10.12.6 | Pro Tools Ultimate 2018.10 | HDX | Avid Omni | Avid 16x16 HD I/O x2 | Digidesign 192 | Digidesign C24

[Darryl Ramm]

Originally Posted by Emi

You also don't need pro tools to record, mix, run at 48 or whatever sample rate you want, you don't need pro tools to edit...you don't need pro tools, period...and I repeat, if you go that route.

If it's in its specs and I want it to run at 192Khz...it should run...and it does run...I don't understand having a feature that I want and the software gives it to me, but everyone else tells me you don't need it....what the....

I buy a Ferrari and damn it I cannot drive it everywhere at 300 km/h. It is obviously the fault of the car.

If you want to convince other folks here you need 192 kHz because it sounds better then provide some link to blind tests showing that. Surely with this debate running for decades that must have been demonstrated by now.... Oh there is no convincing proof? Well expect people to point that out when you claim there is a difference.

Problems operating at 192 kHz are largely a self inflicted problem.... want to try to get work done, I would be dropping the sample rate to 96kHz and testing stuff.

[arche3]

Google Lavry papers on sample rate.

Sent from my SM-G975U using Tapatalk

[Eric Lambert]

Originally Posted by arche3

Google Lavry papers on sample rate.

This is the one; what we've all been referring to:

http://www.lavryengineering.com/pdfs...lity_audio.pdf