Avid Pro Audio Community

Avid Pro Audio Community

How to Join & Post  •  Community Terms of Use  •  Help Us Help You

Knowledge Base Search  •  Community Search  •  Learn & Support


Avid Home Page

Go Back   Avid Pro Audio Community > Pro Tools Software > macOS

Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 09-26-2018, 09:31 AM
skizzo skizzo is offline
Member
 
Join Date: Aug 2010
Location: United States
Posts: 84
Default 4-core, 6-core, 12-core Mac Pro systems

I am on a 2009 Mac Pro with 5,1 firmware and have gone from the following CPUs

1) 2.66ghz 4-core (stock CPU)
2) 3.46ghz 6-core
3) Dual CPU tray with 3.33ghz 6-core

I saw a MAJOR improvement going from the stock CPU to the best CPU this machine can install. Literally cut CPU usage in half at least.

So one would be lead to believe that putting in a dual CPU tray (even if the speed difference is .1ghz lower) would result in another very noticeable improvement.

However I am pretty disappointed I am seeing the same CPU usage percentages being reported in Pro Tools system use window. Even offline bounces didn't seem to speed up.

Can anyone on here who has gone through similar upgrade steps comment on the real life difference they saw with Pro Tools in regards to using two CPUs? I am now considering selling my 2nd Mac Pro since the only reason I bought it was for the dual CPU tray and give me better performance in Pro Tools. My thoughts are if I am seeing the same real life performance in Pro Tools whether its a 6 core or 12 core system it is not worth the investment. I didn't want to pay $500 ($400 for Mac Pro, $100 for two X5680s) to simply up my GeekBench score, which went from like 14,900 to 22,000 with a 6 core to 12 core as far as multi-core score goes.

Please let me know what your observations and experiences have been when running the same sessions with different CPU's
__________________
Mac Pro '09 - 5,1 - Dual Hex Core X5680 3.33 ghz CPU - 48GB RAM
970EVO 500GB (10.13.6) - 1TB WD Black HDD (Samples) - 6TB Toshiba HDD (Backups) - 500GB Glyph GT050Q (Legacy sessions & backups) - 3TB Glyph Studio (Main audio/sessions)
HD5770 GPU

OSX 10.13.6 - Pro Tools 2018.12 - Presonus Studio 192 / Digi 003 Rack+
Focusrite Clarett OctoPre - AKG HPE6 - DBX160A - WA76
Reply With Quote
  #2  
Old 09-26-2018, 10:38 AM
Balanced Jack Balanced Jack is offline
Member
 
Join Date: Jun 2004
Posts: 64
Default Re: 4-core, 6-core, 12-core Mac Pro systems

I’m in the market for a new Mac... would also be very interested to hear people’s thoughts on Mac Pro vs iMac Pro vs MBP to run ProTools natively with a Presonus Quantum (Thunderbolt 2). Wonderful interface btw. Thanks!


Sent from my iPhone using Tapatalk
Reply With Quote
  #3  
Old 09-26-2018, 10:08 PM
BobbyDazzler's Avatar
BobbyDazzler BobbyDazzler is offline
Member
 
Join Date: Apr 2000
Location: Australia
Posts: 1,074
Default Re: 4-core, 6-core, 12-core Mac Pro systems

I went from a dual quad 2.4 Westmere to the dual hex 3.46.
It is a little bit more stable, but the only definite speed improvement I've noticed is when using iZotope RX spectral de-noiser. The 2.4 used to buffer with the D + Extreme algorithm, but the hex core can play it in realtime without buffering.

I suspect that I'm now hitting the GPU ceiling on my rig because of the old 5770 card.
__________________
HP Z2 Xeon 6 Core, Blackmagic Decklink HD Extreme 3D Win10, PT ver.. Latest
Reply With Quote
  #4  
Old 09-26-2018, 10:25 PM
off the wall's Avatar
off the wall off the wall is offline
Member
 
Join Date: May 2004
Location: NYC-NJ
Posts: 662
Default Re: 4-core, 6-core, 12-core Mac Pro systems

I went from a stock 8 core setup in my 2010 Mac Pro to a 3.46Ghz 12 core. The performance of PT 12/2018 increase was astounding.
__________________
offthewallproductions
Mac Pro 5,1 3.46 Ghz 12 core, Open Core 0.9.8, Radeon RX580 GPU, GC-TITAN RIDGE TB3 PCIe card, M.2 NVMe SSD boot drive on PCIe card, 48GB RAM, macOS 12.7.4
Pro Tools Ultimate 2024.3/HD Driver 2023.3 HDX, UA Octo PCIe card in external TB3 chassis, MTRX Studio I/O, C|24 control surface.
Genelec 1031AP/7070A 5.1 monitoring.
Reply With Quote
  #5  
Old 09-27-2018, 01:31 AM
LDS LDS is offline
Member
 
Join Date: Jul 2014
Location: Sydney, Australia
Posts: 1,495
Default Re: 4-core, 6-core, 12-core Mac Pro systems

I noticed a considerable improvement too when going from 2x 2.4ghz 4 core processors to a pair of 6 core X5680s. There is a limit to what can actually be improved in these machines though. We are still talking about machines that are firmly rooted in 2010. 1333MHz ram - modern computers are now up to 3200Mhz for example. SATA2 drives, compared to SATA3 or exponentially faster m.2 formats being used in new computers. There is a limit to what these old beasts can do.

I would really be surprised if a GPU was restricting the speed of Pro Tools in some way. I am not sure it is that heavy on graphics. Swapping out your CPUs for higher core number & higher base clock rate versions will definitely increase the size of the PT sessions you can run. When upgrading, also check and see if your ram is 1333MHz - most 4 and 8 core models shipped standard with 1066MHz ram, as it was all the CPU could take advantage of. 6 or 12 core CPU setups can run 1333MHz. If you haven't looked into faster hard drives, I would definitely give that a peek into as well. Something like the Samsung SM951 AHCI drives will achieve outstanding speed compared to anything running on the SATA2 busses. SATA3 PCIe solutions will help... but it doesn't increase the snappiness of the system in any hugely noticeable way. Transferring any files between SATA3 drives will be quicker though.

I have zero regrets upgrading my old Mac Pro to 12 core, but with that said... it will never feel as snappy as even some of the most specification-light modern machines due to other limitations it has.
__________________
Pro Tools Ultimate 2024.3. OSX 13.6.5. Win 10. HD Native. Lynx AES16e. Lynx Aurora 16. i9-13900KF. ASRock Z690 Steel Legend. 64GB Ram. AMD Vega 64. BM Decklink. Dolby Atmos Renderer 5.2. Trinnov D-Mon. D-Command.
Reply With Quote
  #6  
Old 09-27-2018, 05:17 AM
skizzo skizzo is offline
Member
 
Join Date: Aug 2010
Location: United States
Posts: 84
Default Re: 4-core, 6-core, 12-core Mac Pro systems

thanks for the replies.

can anyone talk about more details in regards to CPU usage % ?

For example, when I ran a particular session with a 2.66ghz 4-core CPU the CPU % may have been as high as 90%

When switching to the 3.46ghz 6-core CPU the CPU % dropped to 40%
So this is where I said my CPU use was cut in half basically. A very noticeable improvement

However, I cannot friggin believe that putting in TWO 3.33ghz 6-core CPUs and the same session is running at 40% CPU. There is no noticeable real life change in my CPU usage % and offline bounces did not speed up. So this just doesn't make sense to me but could very well just be Pro Tools doesn't necessarily utilize the two CPUs much better than one....at least when we are talking about the X5690 and X5680.

You all went from 8-core setups to 12-core setups, with increased clock speed. I wonder if there isn't as big of a difference, or potentially no difference, going from a 6-core setup to a 12-core setup while keeping the same clock speed? (I don't think the .1ghz difference is a big deal at all between X5690 and X5680)
__________________
Mac Pro '09 - 5,1 - Dual Hex Core X5680 3.33 ghz CPU - 48GB RAM
970EVO 500GB (10.13.6) - 1TB WD Black HDD (Samples) - 6TB Toshiba HDD (Backups) - 500GB Glyph GT050Q (Legacy sessions & backups) - 3TB Glyph Studio (Main audio/sessions)
HD5770 GPU

OSX 10.13.6 - Pro Tools 2018.12 - Presonus Studio 192 / Digi 003 Rack+
Focusrite Clarett OctoPre - AKG HPE6 - DBX160A - WA76
Reply With Quote
  #7  
Old 09-27-2018, 09:08 AM
BScout BScout is offline
Member
 
Join Date: Mar 2007
Location: Los Angeles, CA
Posts: 4,144
Default Re: 4-core, 6-core, 12-core Mac Pro systems

A lot of audio processing is serial and can’t be run parallel. So there are limited gains in more cores vs higher clock speeds.

Think of it this way (simplified analogy) if there are 4 notes in a bar, you can’t play note 2 until note 1 is played.

If you get past your bottleneck in processing parallel things for your Pro Tools session, the gains by having more cores will drop off.
__________________
Pro Tools Ult 2024.3, HDX 2, MTRX/SPQ, RME BBF Pro + MADIface ProS1 x 2, Fire HD10 + Max11, Dock, iPad Air5 Mac Mini 14,12, 12 core, macOS 13.6.5RAM 32GB, SSD 4TB, GPU 19 coreQNAP TVS-872XT 148TB TB3
Reply With Quote
  #8  
Old 09-27-2018, 03:34 PM
LDS LDS is offline
Member
 
Join Date: Jul 2014
Location: Sydney, Australia
Posts: 1,495
Default Re: 4-core, 6-core, 12-core Mac Pro systems

Quote:
Originally Posted by skizzo View Post
thanks for the replies.

can anyone talk about more details in regards to CPU usage % ?

For example, when I ran a particular session with a 2.66ghz 4-core CPU the CPU % may have been as high as 90%

When switching to the 3.46ghz 6-core CPU the CPU % dropped to 40%
So this is where I said my CPU use was cut in half basically. A very noticeable improvement

However, I cannot friggin believe that putting in TWO 3.33ghz 6-core CPUs and the same session is running at 40% CPU. There is no noticeable real life change in my CPU usage % and offline bounces did not speed up. So this just doesn't make sense to me but could very well just be Pro Tools doesn't necessarily utilize the two CPUs much better than one....at least when we are talking about the X5690 and X5680.

You all went from 8-core setups to 12-core setups, with increased clock speed. I wonder if there isn't as big of a difference, or potentially no difference, going from a 6-core setup to a 12-core setup while keeping the same clock speed? (I don't think the .1ghz difference is a big deal at all between X5690 and X5680)

It seems that you are in the best position to judge CPU usage? You have the boards in front of you, and you know what you want to achieve. Opening a session and hoping that the CPU usage is going to change in any great fashion is a little simplistic as a test. Why not put your foot on the gas a little... stretch her legs... see how it pans out? What happens if you select all your tracks in the session and duplicate them? How does the session run with a single CPU vs dual CPU? What happens if you double your sample rate?

From my experience, offline bounce doesn't really vary all that much between Macs in general. I use a Mac Pro, 2012 MacBook Pro, 2017 MacBook Pro and a 2015 Mac mini, and the differences in speed are far, far smaller than any CPU speed would allude to. It may well be that it is a single core, serial function. If that is the case, the X5680 is slightly slower than X5690. It seems to even apply to Reaper. Reaper sessions on PC or often bounce 10x faster or more, despite often running on slower machines.

The differences are there, but as I said... there is a limit to what these machines can actually achieve. And it does largely depend on what you are trying to do or achieve. Considering the cost, and the age of these computers, it really is impressive how relevant they still are for day to day studio work.
__________________
Pro Tools Ultimate 2024.3. OSX 13.6.5. Win 10. HD Native. Lynx AES16e. Lynx Aurora 16. i9-13900KF. ASRock Z690 Steel Legend. 64GB Ram. AMD Vega 64. BM Decklink. Dolby Atmos Renderer 5.2. Trinnov D-Mon. D-Command.
Reply With Quote
  #9  
Old 10-01-2018, 05:52 AM
skizzo skizzo is offline
Member
 
Join Date: Aug 2010
Location: United States
Posts: 84
Default Re: 4-core, 6-core, 12-core Mac Pro systems

Quote:
Originally Posted by BScout View Post
A lot of audio processing is serial and can’t be run parallel. So there are limited gains in more cores vs higher clock speeds.

Think of it this way (simplified analogy) if there are 4 notes in a bar, you can’t play note 2 until note 1 is played.

If you get past your bottleneck in processing parallel things for your Pro Tools session, the gains by having more cores will drop off.
this explains what I believe to be experiencing with this particular upgrade path I took. I cannot imagine I am the only person to notice this, adding a 2nd CPU to your system to discover it has virtually zero benefit


Quote:
Originally Posted by LDS View Post
It seems that you are in the best position to judge CPU usage?
You have misunderstood what I was after. Asking others to judge my CPU use makes zero sense. I have explicitly asked what other's experiences have been when upgrading to a 2nd CPU system, or just in general, their observations when making any type of CPU upgrade. If others can collaborate with what I am seeing then it reassures me that is just how these Mac Pros and Pro Tools perform with each other. If I was the only person to see such "issues" then it could lead me to fix something with my particular system.

It appears from what I am seeing is to be expected based on this feedback from others. Especially given that serial/parallel statement above from BScout
__________________
Mac Pro '09 - 5,1 - Dual Hex Core X5680 3.33 ghz CPU - 48GB RAM
970EVO 500GB (10.13.6) - 1TB WD Black HDD (Samples) - 6TB Toshiba HDD (Backups) - 500GB Glyph GT050Q (Legacy sessions & backups) - 3TB Glyph Studio (Main audio/sessions)
HD5770 GPU

OSX 10.13.6 - Pro Tools 2018.12 - Presonus Studio 192 / Digi 003 Rack+
Focusrite Clarett OctoPre - AKG HPE6 - DBX160A - WA76
Reply With Quote
  #10  
Old 10-01-2018, 02:27 PM
LDS LDS is offline
Member
 
Join Date: Jul 2014
Location: Sydney, Australia
Posts: 1,495
Default Re: 4-core, 6-core, 12-core Mac Pro systems

Quote:
Originally Posted by skizzo View Post
this explains what I believe to be experiencing with this particular upgrade path I took. I cannot imagine I am the only person to notice this, adding a 2nd CPU to your system to discover it has virtually zero benefit

You aren't the only person to notice that serial processes don't take full advantage of all the available cores. That is a reality of any serial process, and any multicore CPU. If you are going to base your tests solely of improvements to serial processes, you are going to be disappointed with any multicore system that you use.


Quote:
You have misunderstood what I was after. Asking others to judge my CPU use makes zero sense. I have explicitly asked what other's experiences have been when upgrading to a 2nd CPU system, or just in general, their observations when making any type of CPU upgrade. If others can collaborate with what I am seeing then it reassures me that is just how these Mac Pros and Pro Tools perform with each other. If I was the only person to see such "issues" then it could lead me to fix something with my particular system.

It appears from what I am seeing is to be expected based on this feedback from others. Especially given that serial/parallel statement above from BScout

Sure, but you do seem to also want your cake and to eat it too. The benefit of the upgrade has never been a simplistic, system-wide doubling of speed for everything. It seems that is what you were expecting? Even with the newest iMac Pro, you aren't going to be increasing the performance of serial processes with higher-core CPUs. It is one of the reasons single-core CPU benchmarks exist side by side multi-core CPU benchmarks.

Pro Tools ISNT a single core based application. I suggested some tests for that precise reason - you aren't going to experience performance increases with serial processes when adding more cores. You are going to experience performance increases with parallel processes - Running larger session sizes, bussing audio around your system (between CPU and DSP, and between DAW and external outboard), maintaining greater system stability at low sample buffers, and using higher sample rates - Pro Tools will take advantage of the extra cores and extra power.

If you haven't investigated your system usage via activity monitor, I would take a peek at it. You may well be running some background applications that are chewing up a lot of juice - and that could include your audio interface. Testing it in comparison with the built-in audio can often highlight some pretty resource sucking audio drivers. As soon as I upgraded from 8 to 12 cores, Pro Tools distributed the load across all 12 cores and the CPU usage dropped by about 25%. It all depends on how you use Pro Tools, what peripherals you use, and what areas you look at to see improvement.
__________________
Pro Tools Ultimate 2024.3. OSX 13.6.5. Win 10. HD Native. Lynx AES16e. Lynx Aurora 16. i9-13900KF. ASRock Z690 Steel Legend. 64GB Ram. AMD Vega 64. BM Decklink. Dolby Atmos Renderer 5.2. Trinnov D-Mon. D-Command.
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Macbook Air 2014 (1.4ghz Dual-core Intel core i5 Vs. Mid 2010 MBP i7 Quad-Core and PT TheGoodhusbands Virtual Instruments 2 09-11-2015 10:49 AM
HDX-2 on (slightly revised) 12-core 2 x 2.4GHz 6-core Intel? / RTAS in PT10HD MixerGuy Pro Tools HDX & HD Native Systems (Mac) 10 06-18-2012 07:10 AM
Intel Core i7 2.0GHz Quad-Core Processor or Intel Core i7 2.7GHz Dual-Core Processor? okodub Pro Tools 9 3 09-23-2011 07:33 AM
Single core vs. dual core vs. quad core John Peeples Tips & Tricks 9 11-07-2007 03:28 PM
D24 Core vs. Mix Core Systems Jeff Claus Digidesign Hardware & Software 1 04-01-2002 08:32 AM


All times are GMT -7. The time now is 06:34 AM.


Powered by: vBulletin, Copyright ©2000 - 2008, Jelsoft Enterprises Limited. Forum Hosted By: URLJet.com