Cpu usage in Reaper vs Logic X

[producer]

Please find a video showing cpu usage in Reaper vs Logic X
64 bit buffer size
50 tracks with two plugins ; Fabfilter C & Q

CPU Reaper vs Logic X https://www.youtube.com/watch?v=Vytsf5onHkM&sns=tw

[pixeltarian]

"This video is private."

[marsmellow]

Please make it public. Thanks.

[producer]

Please try now

[philait]

Sadly while Reaper is very responsive and scales well, this high CPU utilisation means I have to use logic exclusively on the road. My battery life runs to less than an hour using reaper while I'm getting 4+ composing in Logic.
My Reaper use is exclusively when I'm plugged into power at home for collaborative projects now. Even then the fans are running constantly to cool the cpu's which is annoying

[producer]

Originally Posted by philait

Sadly while Reaper is very responsive and scales well, this high CPU utilisation means I have to use logic exclusively on the road. My battery life runs to less than an hour using reaper while I'm getting 4+ composing in Logic.
My Reaper use is exclusively when I'm plugged into power at home for collaborative projects now. Even then the fans are running constantly to cool the cpu's which is annoying

There must be a problem with the audio implementation on OSX. I am trying to get the developers attention, but not sure I will suceed.

[Splaaat]

I can confirm the CPU Load issue mentioned by 'producer'.

I ran a test a few weeks back on Logic's 10.0.4 release vs Reaper on an 8-core Mac Pro and a MacBookPro.

If I remember correctly, it was an 90 track mix session and on just initial load with no plugins,

Logic X was cruising at a mere 4%,

Reaper was at 12-13% on playback

I had previously switched from Logic 9 to Reaper 4 a few years back, because Logic struggled at higher loads. I haven't mixed in Logic since then, but Reaper works brilliantly at higher loads, better than any other DAW. I'm not sure if Logic X fixed these issues, but it surely seems to run a lot more fluid now. I'm still worried about running a full session mix on Logic, so I stick to Reaper.

I have previously heard that Reaper works a lot better on PC compared to Macs. Devs care to jump in? Thanks.

[Seventh]

Originally Posted by producer

There must be a problem with the audio implementation on OSX.

Actually, IIRC, inefficient GUI drawing has been one of the main performance culprits on Macs.

[reapercurious]

i bet reaper beats logic if you did the same test on windows.

[Janne83]

There is a little problem with this test youve made.

Logic uses a hybrid buffer system- the 64 io buffer is used only for record armed tracks, other tracks use the processing buffer (stationary at 1024 or 2048 buffer), so comparing both softwares at 64 buffer isnt exactly "fair".

Its true Logic is still a tad more efficient, but craps out much sooner at higher loads.

J

[chriscomfort]

Originally Posted by producer

Please find a video showing cpu usage in Reaper vs Logic X
64 bit buffer size
50 tracks with two plugins ; Fabfilter C & Q

CPU Reaper vs Logic X https://www.youtube.com/watch?v=Vytsf5onHkM&sns=tw

Is there a performance difference between using VST plugs vs Audio Units in Reaper?

[Splaaat]

Originally Posted by Janne83

Its true Logic is still a tad more efficient, but craps out much sooner at higher loads.

J

Is this still true with Logic X?

As I mentioned earlier, Logic 9 did struggle a lot at higher loads, where Reaper would sail smoothly.

[producer]

Originally Posted by Seventh

Actually, IIRC, inefficient GUI drawing has been one of the main performance culprits on Macs.

Why; if you click in another window (application) and Reaper become inactive (the audio part go to sleep), the cpu goes down to allmost zero. It must be the audio implementation. ( i.e. No hybrid buffer system like Logic and Pro Tools).

[producer]

Originally Posted by Janne83

There is a little problem with this test youve made.

Logic uses a hybrid buffer system- the 64 io buffer is used only for record armed tracks, other tracks use the processing buffer (stationary at 1024 or 2048 buffer), so comparing both softwares at 64 buffer isnt exactly "fair".

Its true Logic is still a tad more efficient, but craps out much sooner at higher loads.

J

In this test, Logic use 1/3 of the cpu load of Reaper. I am pretty sure you will be able to run bigger project in Logic due to this fact.
However, let me test in a few days

[producer]

https://www.youtube.com/watch?v=oHVlRQYE6bY

[Splaaat]

I should also be able to comment on Logic X's load handling at higher loads in the coming weeks.

1st track on the album was mixed on Reaper, but I'm going to give Logic X a try for track 2.

[esosotericmetal]

Hmmm... You guys might have convinced me to buy Logic X. The performance improvements in those tests is not subtle. The question is if this is consistent with all macs. I used to use Logic 8 and while I liked the program quite a bit, it was not very stable. However with newer computers and the latest Logic version, i'm guessing it is dramatically better than when I last used it.

Producer, can you list your system specs? And if anyone else with Logic X is able run some tests and report back that would be much appreciated.

[producer]

Originally Posted by esosotericmetal

Hmmm... You guys might have convinced me to buy Logic X. The performance improvements in those tests is not subtle. The question is if this is consistent with all macs. I used to use Logic 8 and while I liked the program quite a bit, it was not very stable. However with newer computers and the latest Logic version, i'm guessing it is dramatically better than when I last used it.

Producer, can you list your system specs? And if anyone else with Logic X is able run some tests and report back that would be much appreciated.

Model Name: MacBook Pro
Model Identifier: MacBookPro10,1
Processor Name: Intel Core i7
Processor Speed: 2,7 GHz
Number of Processors: 1
Total Number of Cores: 4
L2 Cache (per Core): 256 KB
L3 Cache: 6 MB
Memory: 16 GB
Boot ROM Version: MBP101.00EE.B03
SMC Version (system): 2.3f36

[producer]

Originally Posted by producer

https://www.youtube.com/watch?v=oHVlRQYE6bY

Some further comment

the test was executed running 64 bit buffer. Reaper will benefit from a big buffer size i.e. 1024 and will be on par with Logic - maybe slightly better, so Reaper rescue some of its honour.
But no one will like to mix with a cpu load > 60% as the fan will kick in and make a lot of noise. So in that respect, Logic will be more pleasant as it will work form 0 to 65% cpu load.
That said, still some work to do with Logic' s audio engine, so it can run with close to 100% cpu load on the computer, today it bails out approx. 65% cpu load.
Also note that Reaper audio engine always run at full load, it does not stop if your not playing, all tracks are also using cpu. A brut force approach, which is not elegant.

[Doc Shay]

Have you tried changing from 4 to a lower number on the buffering page in Reaper, that brought my cpu way down before when I was having cpu usage issues...but then again I'm on windows but its worth a try on mac.

[Splaaat]

Originally Posted by producer

Some further comment

the test was executed running 64 bit buffer. Reaper will benefit from a big buffer size i.e. 1024 and will be on par with Logic - maybe slightly better, so Reaper rescue some of its honour.
But no one will like to mix with a cpu load > 60% as the fan will kick in and make a lot of noise. So in that respect, Logic will be more pleasant as it will work form 0 to 65% cpu load.
That said, still some work to do with Logic' s audio engine, so it can run with close to 100% cpu load on the computer, today it bails out approx. 65% cpu load.
Also note that Reaper audio engine always run at full load, it does not stop if your not playing, all tracks are also using cpu. A brut force approach, which is not elegant.

Yes, I usually run around the 1024 region while mixing in Reaper.

While it's nice to have a smooth sailing 0 - 60, this was exactly the reason I left Logic 9, then Cubase 6.5 ... Reaper was the only DAW which could survive high loads without much fuss. Others, usually after I surpass the 50% mark, I have to split the session into 2 or 3 new sessions and mix and match levels between sessions. They also get very unstable and I have keep asking clients time to fix and revive the session. :/ In Reaper, it all stays in one session. But, that's my experience, could be different for others. I'm on a 2009 and 2010 Mac Pro.

[producer]

Originally Posted by reapercurious

i bet reaper beats logic if you did the same test on windows.

run the test on my "older" PC, i7 though

The load was approx: 18% i.e. more or less same as Logic

Again, this shows that Reaper on OSX is not efficient. Thats my whole point......

[Primitivity]

Originally Posted by Janne83

There is a little problem with this test youve made.

Logic uses a hybrid buffer system- the 64 io buffer is used only for record armed tracks, other tracks use the processing buffer (stationary at 1024 or 2048 buffer), so comparing both softwares at 64 buffer isnt exactly "fair".

Its true Logic is still a tad more efficient, but craps out much sooner at higher loads.

J

WOW, I always wondered why the buffer settings made no sense to me in Logic. So which one of those buffers am I adjusting in the Logic Settings?

[Janne83]

Originally Posted by Primitivity

WOW, I always wondered why the buffer settings made no sense to me in Logic. So which one of those buffers am I adjusting in the Logic Settings?

You can adjust both:

The io buffer setting corresponds to record armed tracks and has a numerical setting.

The Processing buffer setting corresponds to all other unarmed tracks and has large, medium and small settings- Its not exactly determined but it somehow related to 1024 or 2048 numerical buffer setting.

J

[Jeronimo]

You lost me.

Put both at 1024 which is a realistic mixing setting, and let me know which one craps out first...
The way Logic deals with buffers is a little weird.

My experience with Logic X (9 and 8) is that it craps out waaaaay sooner than Reaper. But I could be wrong, since I never really tested it on the same computer.

Also, which OS are you using?

[producer]

Originally Posted by Jeronimo

You lost me.

Put both at 1024 which is a realistic mixing setting, and let me know which one craps out first...
The way Logic deals with buffers is a little weird.

My experience with Logic X (9 and 8) is that it craps out waaaaay sooner than Reaper. But I could be wrong, since I never really tested it on the same computer.

Also, which OS are you using?

Pls see post #19 above

[Triode]

I'm on OSX 10.6.8 and I find that tracks register CPU usage on tracks with no audio at that point in the arrangement. Sometimes it's tracks with no items that cripple the CPU at points with no items in those tracks. Is this common behaviour?! In logic tracks with no audio have no CPU AFAIK.

[Triode]

In case it's relevant, here is another thread relating to CPU in reaper. It didn't solve my problem but it might help others:

http://forum.cockos.com/showthread.php?t=75025&page=2

[blackenedheart]

I don't agree with any of you. All math aside, Reaper is ALWAYS faster than Logic. Period.

[hermitcrab]

I know reaper likes one thread for one core but logic can use two threads per core. I don't know much about this sort of thing but will that not have an effect on the efficiency ?

[Sibben]

Originally Posted by blackenedheart

I don't agree with any of you. All math aside, Reaper is ALWAYS faster than Logic. Period.

Hear, hear. It also sounds "warmer".