16 posts
Page 1 of 1
I don't know why Studio One 3.3.4 doesn't use the CPU cores like other DAW. We live in a time of multicore and i'm working on a 6 core Mac Pro 2013. Studio One uses first 1 core til its maxed out and the the others. I realized that on many "audio tracks only" projects. Plugins are spread around all tracks...but one core is heavily working and the other doing nothing.
Why not like Logic or Ableton or Cubase per track CPU core balance? Its really wired, because it makes multicore useless!
My multicore support is on, if somebody ask.
User avatar
by CTStump on Sun Apr 02, 2017 8:20 am
Yes there many threads asking basically the the same thing, on this page and the next page of topics. Instead of starting a new topic why not view through those and reply rather then starting a new thread that deals with the same or similar issue. By now it seems to be a little overwhelming when perusing only to find that the issue is already being discussed.

As to your issue I believe it may get some attention in a latter update or upgrade other than that you may find more info in the other threads as this is an ongoing debate now and in the future until it is addressed in a satifactory manner for those suffering issues.

Is it stopping you from using Studio One? Hopefully you can atleast get use out of it until they address it.

Edit: here are some links to those threads for your information,
https://forums.presonus.com/viewtopic.php?f=213&t=23956

https://forums.presonus.com/viewtopic.php?f=213&t=24534

https://forums.presonus.com/viewtopic.php?f=213&t=24224

https://forums.presonus.com/viewtopic.php?f=213&t=22791

Studio One 2.6.5
Various Old Toys
***WARNING***Anything I post is subject to scrutiny as necessary or not... being human I'm prone to error, forgive and forget as necessary...OR NOT***WARNING*** ;)
User avatar
by Funkybot on Sun Apr 02, 2017 10:02 am
violetpow wroteI don't know why Studio One 3.3.4 doesn't use the CPU cores like other DAW. We live in a time of multicore and i'm working on a 6 core Mac Pro 2013. Studio One uses first 1 core til its maxed out and the the others. I realized that on many "audio tracks only" projects. Plugins are spread around all tracks...but one core is heavily working and the other doing nothing.
Why not like Logic or Ableton or Cubase per track CPU core balance? Its really wired, because it makes multicore useless!
My multicore support is on, if somebody ask.


What you're describing is NOT how S1 works. Here's an explanation on what should be hapening, and hopefully this will clear up the expected results for you. If you're seeing something otherwise, then please post detailed step by step instructions on what you're seeing, how to reproduce, and what you're expecting to happen.

So here's what SHOULD be happening:

1. S1 absolutely distributes plugins across multiple cores, and no, it's not designed to only use 1 core.

2. The built-in Studio One Performance Meter shows the CPU load of only the most stressed core. It does not show the other cores. This means the meter is showing more than the average load across all cores. This is probably why you're thinking it only uses 1 core.

3. The way Studio One distributes core usage is as such: anything on a single track ends up on the same core. Example: you load up an instance of RePro-1 as track 1, and your CPU usage is 12%. You add a compressor, EQ, and Delay to track 1. CPU usage is now 16%. It doesn't distribute instruments/effects on a single track across multiple cores. But let's say now on track 2 you add another instance of RePro-1. You add the same compressor and EQ but skip the delay. Result: CPU usage in the Performance Meter is still only 16% even though you practically doubled your instruments/effect. How is that possible? It's because the Performance Meter is only showing the performance of the most stressed core (Core 1) and Track 2 is operating under a less stressed core (Core 2), which you're not seeing at all in the main meter.

4. Multi-instruments, and I believe their effect channels therefore, will all end up on one core. This means if you want to get a more even balance, instead of loading up 1 instance of Kontakt with 16 instruments/outputs assigned to it and effects on each channel, you'll probably be better off using multiple instances of Kontakt so S1 can park them on different cores. [Note: double-check this one, but I'm confident in the expected behavior in 1-3)

Now, you can test this out yourself. Load up a soft synth that's heavy on CPU like RePro-1 or Diva. Look at the CPU load for one instance. If the CPU load is 20% that means you can only run 5 instances right? Now Duplicate that track 5 times. Is your CPU load at 100%? The answer should be no. The S1 Performance Meter will still show 20% (assuming you're using a CPU that your OS sees as having 8 cores). Why? Because those 5 instruments will be distributed between the various cores. No single core will be utilizing more than 20%, which is what Studio One's Performance Meter will show. Now, if your OS is seeing 8 cores, and you load up 9 instances, then guess what? One core will be operating at 40% and the other cores at 20%. What will the Performance Meter show? 40%. The maximum.

Now, I've tested this on Windows and that's definitely how it works. If you're seeing something otherwise, then feel free to provide more details.

Last thing I'll say on the subject: I ABSOLUTELY agree that Studio One could be made more CPU efficient and operate with lower round trip latencies. It's not as efficient as say Reaper, or even Cubase. But that said, the difference isn't as huge as people make it out to be, and I think a big reason as to why is the confusing behavior of the Performance Meter. There's some Feature Requests that deal with these items, so please feel free to upvote them on the FR database:

http://answers.presonus.com/12855/enhan ... 855#q12855

http://answers.presonus.com/12667/impro ... efficiency

http://answers.presonus.com/10836/multi ... nstruments

http://answers.presonus.com/12683/true- ... nstruments
User avatar
by sangamc on Sun Apr 02, 2017 11:21 am
This is what actually happens on my computer. In the screen shot Studio one has just crashed. Nothing can be clicked on and it sounds like a CD skipping.

To reproduce: I started studio one. Loaded a song and pressed play. After about 20 seconds, CPU spikes, Audio becomes choppy as CPU redlines, and Studio one goes down in flames.

This happens to me every day. If I am lucky I can stop the song before a complete crash save exit and reboot.

To get around this Issue, I restart my computer before working in Studio one. After I am done, If I want to come back and work on something later. I restart my computer again before starting studio one. This works about 80% of the time.

Fora while I thought it was too many sound plugins (nexus, kontakt etc) and began bouncing everything as I worked on a song. but even with just a bunch of audio tracks with FX this still happens.

I wish studio one had actual crash logs so we can try and track down what specifically causes this bad behavior.

studio one CPU.PNG
cpu mess

OS: Windows 10 Pro / CPU: AMD A8-3870 / RAM: 16GB /
Presonus Studio One v 3.3.4.41993 Professional Win x64
User avatar
by robertfreeze on Mon Apr 03, 2017 8:05 am
My actual utilization from S1 not running - Start S1, load/play project, close/exit. S1 is clearly utilizing all cores/threads (i7-6700). The initial spike is startup of Task Mgr.

This actually looks pretty darn good to me.

Image

Studio One 3 x64, SONAR Platinum x64, Reaper 5 x64,
Notion 6 x64, Win10Prox64, i7-6700 @3.8Ghz/32G
FA06/FaderPort
User avatar
by brettgoodkin on Fri Apr 07, 2017 11:01 pm
Funkybot wrote
violetpow wroteI don't know why Studio One 3.3.4 doesn't use the CPU cores like other DAW. We live in a time of multicore and i'm working on a 6 core Mac Pro 2013. Studio One uses first 1 core til its maxed out and the the others. I realized that on many "audio tracks only" projects. Plugins are spread around all tracks...but one core is heavily working and the other doing nothing.
Why not like Logic or Ableton or Cubase per track CPU core balance? Its really wired, because it makes multicore useless!
My multicore support is on, if somebody ask.


What you're describing is NOT how S1 works. Here's an explanation on what should be hapening, and hopefully this will clear up the expected results for you. If you're seeing something otherwise, then please post detailed step by step instructions on what you're seeing, how to reproduce, and what you're expecting to happen.

So here's what SHOULD be happening:

1. S1 absolutely distributes plugins across multiple cores, and no, it's not designed to only use 1 core.

2. The built-in Studio One Performance Meter shows the CPU load of only the most stressed core. It does not show the other cores. This means the meter is showing more than the average load across all cores. This is probably why you're thinking it only uses 1 core.

3. The way Studio One distributes core usage is as such: anything on a single track ends up on the same core. Example: you load up an instance of RePro-1 as track 1, and your CPU usage is 12%. You add a compressor, EQ, and Delay to track 1. CPU usage is now 16%. It doesn't distribute instruments/effects on a single track across multiple cores. But let's say now on track 2 you add another instance of RePro-1. You add the same compressor and EQ but skip the delay. Result: CPU usage in the Performance Meter is still only 16% even though you practically doubled your instruments/effect. How is that possible? It's because the Performance Meter is only showing the performance of the most stressed core (Core 1) and Track 2 is operating under a less stressed core (Core 2), which you're not seeing at all in the main meter.

4. Multi-instruments, and I believe their effect channels therefore, will all end up on one core. This means if you want to get a more even balance, instead of loading up 1 instance of Kontakt with 16 instruments/outputs assigned to it and effects on each channel, you'll probably be better off using multiple instances of Kontakt so S1 can park them on different cores. [Note: double-check this one, but I'm confident in the expected behavior in 1-3)

Now, you can test this out yourself. Load up a soft synth that's heavy on CPU like RePro-1 or Diva. Look at the CPU load for one instance. If the CPU load is 20% that means you can only run 5 instances right? Now Duplicate that track 5 times. Is your CPU load at 100%? The answer should be no. The S1 Performance Meter will still show 20% (assuming you're using a CPU that your OS sees as having 8 cores). Why? Because those 5 instruments will be distributed between the various cores. No single core will be utilizing more than 20%, which is what Studio One's Performance Meter will show. Now, if your OS is seeing 8 cores, and you load up 9 instances, then guess what? One core will be operating at 40% and the other cores at 20%. What will the Performance Meter show? 40%. The maximum.

Now, I've tested this on Windows and that's definitely how it works. If you're seeing something otherwise, then feel free to provide more details.

Last thing I'll say on the subject: I ABSOLUTELY agree that Studio One could be made more CPU efficient and operate with lower round trip latencies. It's not as efficient as say Reaper, or even Cubase. But that said, the difference isn't as huge as people make it out to be, and I think a big reason as to why is the confusing behavior of the Performance Meter. There's some Feature Requests that deal with these items, so please feel free to upvote them on the FR database:

http://answers.presonus.com/12855/enhan ... 855#q12855

http://answers.presonus.com/12667/impro ... efficiency

http://answers.presonus.com/10836/multi ... nstruments

http://answers.presonus.com/12683/true- ... nstruments


I concur... Windows has built in CPU monitoring (so does my motherboard)... These two systems essentially show exactly the same core useage, but the studio one CPU monitor is on it's own program lol

That it's not the most efficient daw coupled with this internal monitoring is the #FAKENEWS of s1 haters...

I've been doing this longer than you... Win10... S1 pro latest version... 64 bits... Rme latest drivers... All the ram and i7 monster... I'm only adding this info to speed up answers to my queries as it's torturous to have a thread stop in its tracks because the 1st response is always "give me all your info before I then tell you to pound sand". Xoxo
User avatar
by sirmonkey on Sat Apr 08, 2017 12:07 am
brettgoodkin wrote
Funkybot wrote
violetpow wrote.......

That it's not the most efficient daw coupled with this internal monitoring is the #FAKENEWS of s1 haters...


Good post, mostly. But I think that the impression that Studio One doesn't spread CPU load over several cores is understandable.
I'm a big fan of S1, and I myself thought that S1 tended to not spread tasks across multiple cores.
However, this thread helped me understand things a bit better.
I don't think "FAKENEWS" is going on here about this issue, usually (but sometimes it does!).

NOW I know that: S1 only displays the usage of the most taxed core. I totally didn't know that before. I even knew about windows performance meter, and compared it to S1's performance meter. My conclusion was: "What the heck? How do I know which one is true?"

The windows performance meter gives a better picture. This should have been obvious to me, in retrospect.

Anyway, good post. Fake news does exist, but here, there's occasional misunderstanding (90%), and only 10% is FAKENEWS. (* I think!)

Atari 5200, 64K RAM S1PRO Radio Shack Cassette Recorder w/internal Mic,
User avatar
by CTStump on Sat Apr 08, 2017 8:32 am
sirmonkey wroteAnyway, good post. Fake news does exist, but here, there's occasional misunderstanding (90%), and only 10% is FAKENEWS. (* I think!)


Even with misundersting and deliberate attemps at spreading FAKENEWS does this meter watching really stop people from using Studio One under Normal circumstances i.e. a real session not bench testing I think not...mosty.

That question needs to be asked and emphasized.

Studio One 2.6.5
Various Old Toys
***WARNING***Anything I post is subject to scrutiny as necessary or not... being human I'm prone to error, forgive and forget as necessary...OR NOT***WARNING*** ;)
User avatar
by Funkybot on Sat Apr 08, 2017 9:49 am
CTStump wrote
sirmonkey wroteAnyway, good post. Fake news does exist, but here, there's occasional misunderstanding (90%), and only 10% is FAKENEWS. (* I think!)


Even with misundersting and deliberate attemps at spreading FAKENEWS does this meter watching really stop people from using Studio One under Normal circumstances i.e. a real session not bench testing I think not...mosty.

That question needs to be asked and emphasized.


I think it does actually. At least some people. I've seen enough threads and posts between here and KVR of people pointing to the CPU results in Studio One's Performance Meter compared to say Reaper's with complaints of "why is Studio One so bad with CPU" to know that the current state is confusing versus intuitive, and results in people not jumping on the Studio One train because of the perception of high CPU usage just based off that. So in my experience, it's a problem of perception, but a problem nonetheless.

Now, all that said, Studio One definitely should still improve the CPU performance and is NOT as efficient as say Reaper. There's lots of actual evidence to support this using the same project/plugins on both hosts and comparing track/plugin counts at various latencies. It's just nowhere near as bad as the Performance Meter makes it out to be.
User avatar
by Lawrence on Sat Apr 08, 2017 11:34 am
I intentionally stay out of these threads but I will say this below which I think is relevant (while not weighing in on the actual subject matter) ...

As a practical matter, It kinda doesn't matter in the near term if it's true of not. Saying it over and over and over and over again and typing caps and exclamation points won't speed up or change anything. At some point people have to at least temporarily let it go and maybe wait for the next major update and if nothing changes that satisfies them, bring it up again, bump it again.

This idea that multiple threads and shouting and repeating the same thing about the same thing into perpetuity is going to change something in the short term (or make someone official come in here and talk to you about it, which never happens in the "user" forum) is kind of an illusion (or delusion). If they haven't heard you by now, after multiple repetitive threads about the same thing .... ? ... or just wait and see if they maybe have.

My $0.02.

P.S. I should add, as relates to KVR, Gearslutz and similar. I do take the discussions about this here as being more legit, knowing that - with no exceptions - everyone posting here in the main user forums about it actually owns the application. Can't say the same for some other places. :)

Windows 7 64 SP1 - Intel Core i7 860 2.8GHz 8GB - ATI Radeon HD 5770 1 GB (8.850.0.0) - iPad 2 IOS 8.3
User avatar
by Skip Jones on Sat Apr 08, 2017 12:09 pm
OVER AND OVER AND OVER AGAIN.

Well said Lawrence.

http://www.presonus.com/support
Windows 10 X 64 : AMD Phenom II X 4 945
ATI Radeon 5450 / 512 RAM
8GB RAM / 1T SATA

Mac Mini (Late 2014)

Faderport and Faderport 8
Yamaha S-08
Fishman TriplePlay
Logidy Controller

Assorted PreSonus Gear

To add your software and hardware specs to your signature to make it easier for us to help you, click HERE.
Answers at http://answers.presonus.com/
User avatar
by Skaperverket on Sun Apr 09, 2017 3:53 am
Skip Jones wroteOVER AND OVER AND OVER AGAIN.

Well said Lawrence.


I too do agree with Lawrence. Usually do. However, there are some things I'd like to add to that that might explain why this is an ongoing topic:

1. Presonus has never acknowledged the implications and potential confusion regarding this single core performance meter, so no one knows whether it's actually being taken seriously and addressed.

2. Ari has specifically asked users to prove that Studio One is less CPU efficient than other DAWs. But once tests indicate that this indeed is the case: no further comments, nothing.

3. In the manual or other official papers there is no thorough walk-through on how to balance core usage and there is not sufficient explanation of how one single channel routed to a bus, the master channel and through sends (even when the channel's sending utility is muted) can not be parallelized by Studio One, so everything is thrown onto the same core.

4. And most importantly: More and more is taking place ITB and both 3rd party VST/VSTi and Studio One's own synths and audio processing tools become more and more advanced and CPU demanding, while computing power in recent years has stagnated a bit and for the most part only seen significant gains from adding cores and not CPU core frequency clock rate. We can probably expect more of this trend, with higher resolution screens, faster memory and storage, more powerful GPUs and a shift to more cores also for consumer products. But not much faster cores. And, as developers finally begin to pay attention to the musicality of nonlinearities and the need for oversampling: more demanding plugins. There will probably not be much to gain beyond 5 GHz OCed CPUs and some improvement in IPC (instructions per cycle) and possibly new extensions for quite some time. We'll see Games will probably find a way to overcome the difficulties of their real-time demands and finally be able to utilise more cores; audio production software will have to follow that trend, and some DAWs are already showing signs of capability of such parallelization (Logic f.i. seem to give each bus its own thread instead of adding it to the channel; Diva is able to process its internal voices in parallel, something that Mai Tai and Multi Instruments would benefit greatly from). That is at least the trends I'm seeing.

5. Lastly, I'd like to add that Studio One is the best DAW for me overall and I love it, but even though most people are happy with the CPU efficiency for their use, that doesn't mean that other people need as much efficiency as possible and that the challenges are indeed real when working professionally ITB with low latency real-time monitoring of recording of audio and software instruments, mixing with "un-freezable" aux returns and master bus processing, oversampling and more.

YMMV.

My 0.02, your mileage may vary.

OS X 10.9.5, S1 3.3.3
User avatar
by Lawrence on Sun Apr 09, 2017 3:46 pm
Everything above might be true. What's also true is doing the same thing over and over and expecting a different result.

It's in Answers as - quite literally - if you add up the votes for the two CPU FR's which are really the same thing, the most highly voted thing there. If all people want is feedback on a user forum where that never happens, to feel better, good luck with that. No idea where that expectation comes from exactly since they've never done that in the user forum.

The other sensible thing to do is just wait and see if "the most highly voted FR" in the official feature request system is acted upon.

Windows 7 64 SP1 - Intel Core i7 860 2.8GHz 8GB - ATI Radeon HD 5770 1 GB (8.850.0.0) - iPad 2 IOS 8.3
User avatar
by Skip Jones on Sun Apr 09, 2017 3:48 pm
Dang, spot on.

http://www.presonus.com/support
Windows 10 X 64 : AMD Phenom II X 4 945
ATI Radeon 5450 / 512 RAM
8GB RAM / 1T SATA

Mac Mini (Late 2014)

Faderport and Faderport 8
Yamaha S-08
Fishman TriplePlay
Logidy Controller

Assorted PreSonus Gear

To add your software and hardware specs to your signature to make it easier for us to help you, click HERE.
Answers at http://answers.presonus.com/
User avatar
by CTStump on Sun Apr 09, 2017 4:33 pm
I think it's time to consolidate all these threads and close this topic. It really has been over posted in my opinion.

That probably won't stop someone from starting another thread unless it's pinned or the Mod's are more vigilant if it's not but as has been stated there is the Answers section already.

It gets tiresome to be bombarded with this topic being regurgitated over and over and over.... Again. :?

Edit: I also noticed that the user who posted this topic has not replied. Is he/she busy? Is the topic answered to satisfaction? Were the links helpful? Are the replies rude and unnecessary?

I'm beginning to believe this post is another attempt at keeping this subject on the first page and NOT a reqiest for real support.

OP please reply so I can be proven wrong and if so my apologies for this accusation. If not well then I guess it stands(sadly).
Last edited by CTStump on Sun Apr 09, 2017 6:31 pm, edited 1 time in total.

Studio One 2.6.5
Various Old Toys
***WARNING***Anything I post is subject to scrutiny as necessary or not... being human I'm prone to error, forgive and forget as necessary...OR NOT***WARNING*** ;)
User avatar
by Jemusic on Sun Apr 09, 2017 4:54 pm
CTStump wroteIt gets tiresome to be bombarded with this topic being regurgitated over and over and over.... Again. :?


Especially while some of us are busy and don't have any issues either producing tons of music some of it being super large sessions not even coming close to taxing the full resources of our CPU's. Just choosing a different workflow and taking due diligence in terms of rendering a few things here and there and being smart about it. It has never stopped me.

As Hans Zimmer so wisely pointed out the best DAW for you is the one you know He did not say the best DAW for you is the one that is the most efficient CPU wise. He said the one you know....And if that happens to be Logic or Reaper then you are lucky enough to have a great DAW you know that happens to be more CPU efficient.

It is interesting that there has always been workarounds and ways of solving problems. Back in the early days it might have been limited tracks etc a but somehow they got around it and achieved the end result. Today we still have problems to solve for sure and they are different. But there will be always a way to get to the end and still somehow convey your intention. That has not changed.

Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport - Studio One V3.5.2
iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface and Scarlett 18i20

Poor minds talk about people, average minds talk about events, great minds talk about ideas - Eleanor Roosevelt

16 posts
Page 1 of 1

Who is online

Users browsing this forum: No registered users and 11 guests