Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Audio Decoding KILLS MacPro (2009) green factor

I sent an email to ANAND (AT) ANANDTECH.COM

He covers PC mostly.. Has great contacts. Has covered major MacPro releases and unofficial upgrades. It comes down to hitting a major hardware site to get exposure. Im surprised ARN, our great Grad Poobah, from MacRumors / TouchArcade hasn't given us Page1 or Page 2 coverage..



VirtualRain said:
Are you volunteering to contact the major media? I'm all for throwing my support behind this. Feel free to PM me if you want my email or something further.
Click to expand...
 
handheldgames said:
I sent an email to ANAND (AT) ANANDTECH.COM

He covers PC mostly.. Has great contacts. Has covered major MacPro releases and unofficial upgrades. It comes down to hitting a major hardware site to get exposure. Im surprised ARN, our great Grad Poobah, from MacRumors / TouchArcade hasn't given us Page1 or Page 2 coverage..
Click to expand...

If you need any further data I can help as my computer does it all too.

If it was a laptop I'd take it to an apple store but I really can't be bothered to lug it somewhere to be told "Its not a problem, when it quite clearly is!
 
These are results of temperatures by testing my System on friendly request by nanofrog

System Specifications:

Mac Pro 4.1
8-Core 2.26 GHz Xeon Nehalem
12GB APPLE ECC Memory
Graphics card: nVidia EVGA 285 GTX Mac Edition & GT120 connected each to one 24" 60HZ Screen.
OS X 10.6.2 installed on 3x1TB Caviar Black's in RAID 0
*external firewire soundcard used in testings : Apogee Ensemble

I don't have anything to measure the room-temperature but I guess it's around 20-25° C.

My my temperatures are measured with the iStat tool. I measured the CPU A and CPU B temps. = CPUA / CPUB
USB devices were disconnected during testings.

All measures in degree Celsius (°C)

Here are my Test results:

Idle with some programs running for 80 min : 57 / 47
Idle with no programs running for 60 min : 37 / 30

iTunes Music playback for 15min with *Apogee Ensemble : 50 /42
iTunes Music playback for 15min with internal speakers : 51 / 42

1080p movie (VLC Player), iTunes music playback, Quicktime MP3 playback, Quicktime movie playback (608x336) all at the same time for 15min with internal speakers : 53 / 42

1080p movie (VLC Player), iTunes music playback, Quicktime MP3 playback, Quicktime movie playback (608x336) all at the same time for 15min with *Apogee Ensemble : 55 / 43

CPUTest results : (like VirtualRain did)

4 Core CPUTest: Stress test using CPUTest with Huge test type, 10 repetitions, and 4 instances (to load 4 physical cores).

Temps at start : 54 / 43
Temps at end : 60 / 48

CPU A was during the test always around 59-61°C !

8 Core CPUTest: Stress test using CPUTest with Huge test type, 10 repetitions, and 8 instances (to load 8 physical cores).

Temps at start : 56 / 46
Temps at end : 64 / 51

My personal conclusion:

I am not impressed by these final measure datas. When I am working on the desktop with pages, safari, etc. without using media I have temperatures around 55 / 45 . These temperatures look fine to me.

Attached screenshot 1 : HW Monitor while listening music in iTunes

Attached Screenshot 2 : CPUTest using 8 instances
 

Attachments

  • Screen shot 2009-12-15 at 15.28.14.png
    Screen shot 2009-12-15 at 15.28.14.png
    61.7 KB · Views: 172
  • Screen shot 2009-12-15 at 16.07.52.png
    Screen shot 2009-12-15 at 16.07.52.png
    1 MB · Views: 158
seisend said:
Idle with some programs running for 80 min : 57 / 47
Idle with no programs running for 60 min : 37 / 30

iTunes Music playback for 15min with *Apogee Ensemble : 50 /42
iTunes Music playback for 15min with internal speakers : 51 / 42
Click to expand...

So, if you start your Mac Pro without the Apogee connected... I assume you would have temps around 30-deg?
If you then plug-in the Apogee - not playing audio - do your temps go up to the 50's?
 
Seisend, they do seem a little high, but there's no CPU% utilization to get a sense of the load (save the 8 core testing, which is high, as Intel rates full load 8x 100% = 67C in their thermal profile), and the ambient is high too (33C).
 
VirtualRain said:
So, if you start your Mac Pro without the Apogee connected... I assume you would have temps around 30-deg?
If you then plug-in the Apogee - not playing audio - do your temps go up to the 50's?
Click to expand...

----------------------------------------------------------------------
Mac Pro:
----------------------------------------------------------------------
The temps are always around 35 / 30 in Idle mode.

CPU A / CPU B :

Idle : around 35 / 30
Idle with Apogee Ensemble on : around 45 / 37
Idle without Apogee Ensemble again : 38 / 30

I say, with my external soundcard. My Mac gets around 10°C hotter !
----------------------------------------------------------------------
MacBook Pro 3.1 OS X 10.6.2 (2007) 4GB RAM 2.4 GHz Core2Duo
----------------------------------------------------------------------

CPU:

Idle: 40
Idle with Apogee Ensemble on : 49
Idle without Apogee Ensemble again : 40

----------------------------------------------------------------------

EDIT:

@ nanofrog:

Well, as long these temps don't damage my system, I don't care much about them. The processor is built to get temperatures very high temperatures. For me , the temperatures are in normal condition. It could always could be less and I would be happy with less heat.
 
seisend said:
----------------------------------------------------------------------
Mac Pro:
----------------------------------------------------------------------
The temps are always around 35 / 30 in Idle mode.

CPU A / CPU B :

Idle : around 35 / 30
Idle with Apogee Ensemble on : around 45 / 37
Idle without Apogee Ensemble again : 38 / 30

I say, with my external soundcard. My Mac gets around 10°C hotter !
----------------------------------------------------------------------
MacBook Pro (2007) 4GB RAM 2.4 GHz Core2Duo
----------------------------------------------------------------------

CPU:

Idle: 40
Idle with Apogee Ensemble on : 49
Idle without Apogee Ensemble again : 40

----------------------------------------------------------------------
Click to expand...
So the audio interface is pushing the temps up, just plugged in. Good to know, as it confirms what VR's experiencing.

There's definitely something up with OS X and audio, and seems to be located in the Quicktime X code. :(
 
nanofrog said:
So the audio interface is pushing the temps up, just plugged in. Good to know, as it confirms what VR's experiencing.

There's definitely something up with OS X and audio, and seems to be located in the Quicktime X code. :(
Click to expand...

The application "LineIn" does it too when passing audio through.

Very very big bug in the OS X audio subsystem, any use of the audio output makes it increase in temps!
 
Concorde Rules said:
The application "LineIn" does it too when passing audio through.

Very very big bug in the OS X audio subsystem, any use of the audio output makes it increase in temps!
Click to expand...
I'd have thought that was part of the Quicktime X code.

At any rate, it makes sense if the issue occurs when a USB (presumably FW too, but we need confirmation) audio device is plugged in.
 
nanofrog said:
So the audio interface is pushing the temps up, just plugged in. Good to know, as it confirms what VR's experiencing.

There's definitely something up with OS X and audio, and seems to be located in the Quicktime X code. :(
Click to expand...

Yes that is consistent with what I'm seeing... I'm not sure what software components may be involved with audio in OSX (CoreAudio drivers, Quicktime, etc.) but it's definitely bizarre in the extreme!
 
handheldgames said:
Its been verified in 10.5.8 with Quicktime 7. Looks like its something that's been around. It makes me wonder if its Quicktime or Core Audio. As Quicktime X was supposed to be rewritten from the ground up.
Click to expand...
Was there any changes to Core Audio late in Leopard or with the release of SL? Or just the Quicktime X code (which I've understood was only changed with SL)?
 
Not only is this whole thing totally unacceptable, what is really peeving me is that thanks to Apple's incompetence, I will be paying for this every month (electricity bill). I also have a workstation that is 25% less capable than it should be if I choose to have any audio playing back while working.

I am honestly considering cutting my losses and contacting Apple Customer Care to arrange a return / full refund.

Three screen shots below.

1)Idle
2)1min of Audio
3)5min of Audio

Machine: Quad 2.93, ATI4870
 

Attachments

  • Idle.png
    Idle.png
    111.6 KB · Views: 166
  • After 1min of Audio.png
    After 1min of Audio.png
    105.7 KB · Views: 154
  • After 5min of Audio.png
    After 5min of Audio.png
    101.1 KB · Views: 154
I agree. Apple promised that "It simply works". That clearly is not the case. It works badly. I'm pissed off that my boost fans always run more than 1000 rpm and that my PCIe fan runs more than 800 rpm when I have the machine in idle with media playing in the back ground. My old MP1,1 X5365 octad was completely silent with a 3870 card and my new MP4,1 with a lower power GT120 is always audible. I used the same Arc-1210 card and the same drives in it. Now I'm paying a higher electricity bill for a machine which is thee years ahead in technology and runs 45 nm CPUs instead of 65 nm of the old machine. It is ridiculous that Apple cannot fix this in a decent time frame.
 
It is ridiculous that Apple cannot fix this in a decent time frame.
Click to expand...

Apple can't fix it in any timeframe. The official answer received from Apple engineering is that this is all normal and within "design limits". They are essentially sticking their head in the sand on this one. I guess I'll have to build my own workstation....
 
Geekbench / iTunes

Anybody try running Geekbench alongside iTunes after the 20 minutes and compare results to Geekbench with no apps running after idle. Curious to see if you get ~25% hit in performance.
 
ByteOfApple said:
Anybody try running Geekbench alongside iTunes after the 20 minutes and compare results to Geekbench with no apps running after idle. Curious to see if you get ~25% hit in performance.
Click to expand...

Confirmed.. 20%-30% hit anytime audio is playing.
 
nanofrog said:
Then the audio device aspect of the issue is only with USB or FW interfaces, and PCIe isn't an issue.

Good to know. Thanks. :)
Click to expand...

I don't see how that works, as onboard speaker does it, optical out does it, analog out does it and any FW/USB attached output does it.

Whats so special about the PCi-E sound card? :/
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back