steinberg power scheme and cpu core parking ?

Can someone tell if when using the steinberg power scheme it disables cpu core parking ?

I found out about cpu core parking on another forum.

I checked my windows 8.1/I7 2760QM and indeed i had 2 cores parked… aaargh…

I found this tool on the net that made it possible to unpark the cores.

It is a clean tool, and actually it looks like it does indeed add extra resources in cubase and the cpu balances in taskmanager seem better between the different cores, but i’m not very techy so judging based on a wobbling asio-bar and a task manager playing some sounds isn’t very accurate.

so my question is: does using the steinberg power scheme actually lift cpu core parking ?

kind regards,
R.

No…Steinberg power scheme does not disable core parking.

This requires an edit to the registry…I guess if you found a utility to do this it is simply writing the registry changes for you.

seems important performancewise, not?
Half of the cpu was snoring in my laptop if my assumption is correct then…
Maybe an add to the tweaklist they put on the site?

seems important performancewise, not?

Have seen a DAW manufacturer state that they tested with no actual difference to performance.

If your cores are parked maybe the computer isn’t loaded that heavily?

If your 2 operational cores were overloading then there is certainly a problem.

Don’t be too quick to wake-up the sleepy cores. I’ve just discovered that disabling 2 of the 4 cores on my desktop DAW actually gives a slight improvement in performance. Perhaps there is some hidden overhead in distributing tasks around the cores (?)

it’s not only the cakewalk forum that is suggesting to use it, as i see that for “studio one” they even provide a system tweak that includes disabling the core (and more) parking in a custom power profile.

http://support.presonus.com/entries/203607-Studio-One-Optimizing-Windows-Vista-7-Power-Settings-for-Studio-One

but it is stated as not usefull for win8…
confusing…

in reply: my system is working fine and indeed when i looked for the parked cores, the system was nearly idle.

it’s not only the cakewalk forum that is suggesting to use it

A forum doesn’t suggest anything in that sense…what you mean is a person who posted on the forum suggested it. This is no different from saying my mate down the pub suggested it. Doesn’t qualify it as good advice although I’m also not saying it’s bad advice

FYI it was on the cakewalk forum that cakewalk themselves said they had tested and found no difference.

but it is stated as not usefull for win8…
confusing…

Doesn’t seem that confusing…you’re on Win8 + it’s not useful on Win8 = stop worrying about it and make some music.

Microsoft’s description of the ‘Processor performance core parking min cores’ in their ‘Processor Power Management in Windows 7 and Windows Server 2008 R2’ document from October 2012 is:

The minimum percentage of logical processors (in terms of all logical processors that are enabled on the system) that can be placed in the unparked state at any given time. For example, on a system with 16 logical processors, configuring the value of this setting to 25% ensures that at least 4 logical processors are always in the unparked state. The Core Parking algorithm is disabled if the value of this setting is not less than the value of the Processor Performance Core Parking Maximum Cores setting.

The Steinberg power scheme sets the “Processor performance core parking min cores” value to 100% (that’s maximum)
So it’s safe to say the Steinberg power scheme disables core parking. :wink:

The concrete info on these kinds of topics is spotty at best. I wish there was a place that people could get definitive facts on things like CPU parking, ASIO, etc.

Alas, we’re left to fend for ourselves. Re the OPs ?: disabling coreparking and hyperthreading helped me greatly on my system.

Word to the wise: if you’re using VEP or something: do NOT run an instance on your host computer if you can help it. Both apps fighting for the CPU bottlesnecks very quickly. OTOH, if I run it off my slave, I can play 64+ tracks simultaneously over 32 stereo channels w/o break-up. This number is greatly impacted by the CPU speed on your slave system. I would advise you disable core-parking and HT on it, too.

  1. I have always run VEP on the same machine as the host. Zero bottleneck problems and it allows for lower latency.
  2. Hyperthreading being enabled is no problem. You can disable it, but you will lose 30% or so of your computer’s power.
  3. I run around 80 stereo returns, and many plugs with no problems

DG

Can I see pics of this, if it isn’t too much trouble? Preferably a gif (http://blog.bahraniapps.com/?page_id=21) of the 80 stereo tracks running w/ the ASIO meter in view. (Non-gif) Steinberg device setup w/ your buffer, VEP prefs. please.

Perhaps your dual xeon machine loans itself well to VEPs multithreading, more so than my overclocked 3930k, which, from what I’ve gathered, should have equal processing power.

If that’s the case, that is very very good knowledge to have for the community.

I’m actually on a different machine now, but some time next week I’ll get my assistant to sort something out for you. However, doesn’t your machine only have 6 cores, 12 threads, which is half of the number that I have/had?

DG

Thanks DG, I think that will be a big help

That’s correct - PassMark - [Dual CPU] Intel Xeon X5675 @ 3.07GHz - Price performance comparison

Almost identical benches but 2x cores, which I’m sure VEP takes advantage of beautifully compared to the ASIO engine. Although, somehow the two are interconnected when you’re conneted via the VST. Again, not too much concrete information on this topic

Interesting discussion on this, here: https://community.vsl.co.at/forums/p/32919/207730.aspx Looks like their findings mirror my hypothesis.