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

Concorde Rules said:
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? :/
Click to expand...
The difference is PCI vs PCIe in the chipset (then the data is routed via QPI to the CPU). The internal audio is attached to the PCI links in the chipset, as is USB and FW (Audio ->USB via the chip used is also possible).

As we get into this, its beginning to seem like the Core Audio is the culprit, in how it interacts with the new chipset for the LGA1366 CPU's.
 
Passed on a link of this thread to the publisher of MacRumors / TouchArcade. :)

Add Walt Mossberg to the list...

Just sent it to the editor of MacLife...

Also... Macworld..... ;)

And engadget....

AND TUAW...

And Macnn...
 
handheldgames said:
Passed on a link of this thread to the publisher of MacRumors / TouchArcade. :)

Add Walt Mossberg to the list...

Just sent it to the editor of MacLife...

Also... Macworld..... ;)

And engadget....
Click to expand...
Looks like the snowball does run down hill after all. :eek: Let's hope it gets enough momentum to knock a big enough "hole" in the Apple campus attention span to get noticed. ;) :p
 
handheldgames said:
Passed on a link of this thread to the publisher of MacRumors / TouchArcade. :)

Add Walt Mossberg to the list...

Just sent it to the editor of MacLife...

Also... Macworld..... ;)

And engadget....

AND TUAW...

And Macnn...
Click to expand...

Sadly, a lot of those Mac publications are in Apple's back pocket and are afraid to highlight anything that could be negative to the mothership. I hope you can prove me wrong.
 
Hopefully we can get this picked up. There has been an unbelievable support from the MacPro community here @ MacRumors.

What everyone documented highlights a serious flaw that garners the attention of the OS X development team as it invalidates major marketing points of the platform.

Guess it's time to rename my MacPro to Heat Miser!

hmiser.jpg


http://www.youtube.com/watch?v=yon2YuXssvo (1/2 way in... cant find the direct link..)
 
I have a feeling this is just architectural growing pains. Anytime you make dramatic changes to things, there are always going to be bugs. If Apple is going to stick with these new Intel processors for the next few iterations, they will no doubt address this issue in the near term.

Hang in there folks. Apple will fix this. I can feel it. :cool:
 
handheldgames said:
Passed on a link of this thread to the publisher of MacRumors / TouchArcade. :)

Add Walt Mossberg to the list...

Just sent it to the editor of MacLife...

Also... Macworld..... ;)

And engadget....

AND TUAW...

And Macnn...
Click to expand...

Good work man! I sure hope the effort you've invested in this problem pays off! :)
 
Is the inference that we could get a fix/bypass the issue by using a PCIe sound card (to get around using the internal sound chipset)?

If so, has anyone recommendations (cheap is preferred please)?

We've waited 3/4 year for Apple to respond, and there seems no evidence that they even acknowledge the issue yet. Perhaps we just have to hack these systems for ourselves.
 
Just been to an apple store, genius had no idea about the problem but I demonstrated it infront of him on the instore quad mac pro and he has confirmed it. He is submitting it to engineering and keeping me updated.

See what happens!
 
cl-user said:
Is the inference that we could get a fix/bypass the issue by using a PCIe sound card (to get around using the internal sound chipset)?

If so, has anyone recommendations (cheap is preferred please)?

We've waited 3/4 year for Apple to respond, and there seems no evidence that they even acknowledge the issue yet. Perhaps we just have to hack these systems for ourselves.
Click to expand...
RME has a few PCIe cards that work in Macs, but they're nowhere near "cheap". Unfortunately, the least expensive devices I've noticed are either USB or FW interfaced units. :(
 
Hi!

I haven't read the whole thread yet, but thought I'd post that, using iStatMenus for measuring, playing music in iTunes has no impact on temperature on my machine. I tested only for about 5 minutes, though. Temperature was constantly at 51° for cpu1 and 41° for cpu2.
This is on a Mac Pro 2009 2,26GHz via USB audio interface.
 
Taipan said:
Hi!

I haven't read the whole thread yet, but thought I'd post that, using iStatMenus for measuring, playing music in iTunes has no impact on temperature on my machine. I tested only for about 5 minutes, though. Temperature was constantly at 51° for cpu1 and 41° for cpu2.
This is on a Mac Pro 2009 2,26GHz via USB audio interface.
Click to expand...

I believe the problem happens after approx 15-20 minutes of audio playback.
 
Taipan said:
Hi!

I haven't read the whole thread yet, but thought I'd post that, using iStatMenus for measuring, playing music in iTunes has no impact on temperature on my machine. I tested only for about 5 minutes, though. Temperature was constantly at 51° for cpu1 and 41° for cpu2.
This is on a Mac Pro 2009 2,26GHz via USB audio interface.
Click to expand...
Let it idle for a bit, take a screen shot of iSTAT, re-run the test (say 20 min), and take another screen shot of iSTAT. Then compare. You might be surprised, as 51C is too warm, if the CPU % utilization is low.

Also, a USB or FW audio device will push the temps up. If you have one, give it a few minutes to coold down, pull it, take the screen shot, plug it back in, wait a couple of minutes, and do another screen shot. For those that thought they didn't have the issue, it's being discovered they had audio devices plugged in, and the idle temps were around the 51 - 56C mark.
 
Taipan said:
Hi!

I haven't read the whole thread yet, but thought I'd post that, using iStatMenus for measuring, playing music in iTunes has no impact on temperature on my machine. I tested only for about 5 minutes, though. Temperature was constantly at 51° for cpu1 and 41° for cpu2.
This is on a Mac Pro 2009 2,26GHz via USB audio interface.
Click to expand...

I use to think the same way, but found that 51-deg is not normal... something is having an impact on your CPU temps and it's likely related to your audio system. What are you using for sound playback?
 
VirtualRain said:
I use to think the same way, but found that 51-deg is not normal... something is having an impact on your CPU temps and it's likely related to your audio system. What are you using for sound playback?
Click to expand...

I think his USB interface is causing the machine to be stuck in this strange state instead of truly idling. The worst part is, his machine is always hogging extra power and 25% slower because of this...
 
smacman said:
I think his USB interface is causing the machine to be stuck in this strange state instead of truly idling. The worst part is, his machine is always hogging extra power and 25% slower because of this...
Click to expand...

My MP is about a month old. I see temps of +14 C within 5 minutes of playing iTunes (40-41C). During this period, iTunes consumes between .7 and .9 % load. As soon as iTunes is stopped (not playing a song), the temp drops to 26 C within two minutes.

I have observed that the temps continue to climb during playback. I will check to see what my observed temp is over the weekend, when I have time to play music for an extended time period.

-bill
 
I can confirm this strange behaviour on my machine as well.
25°C idle, 53°C after 10 minutes of iTunes music playback.
Cools down to 25-30° after 2 minutes when I close iTunes.
 
I guess some of you have to consider this:

People who are running a Mac with an external Soundcard (or most of them) device will get the heat increase anyway (at least under 10.6.2) ! As you see, I tested it on my MacBook Pro and posted results earlier in this thread. So IF Apple would fix the problem with the Mac Pro itself, it won't change anything for people with external devices except PCI cards like nanofrog means. It's an OS X problem too.
 
seisend said:
I guess some of you have to consider this:

People who are running a Mac with an external Soundcard (or most of them) device will get the heat increase anyway (at least under 10.6.2) ! As you see, I tested it on my MacBook Pro and posted results earlier in this thread. So IF Apple would fix the problem with the Mac Pro itself, it won't change anything for people with external devices except PCI cards like nanofrog means. It's a OS X problem too.
Click to expand...

Maybe I am misunderstanding you, but I think the problem with your Macbook Pro may be something totally different. I did a lot of testing on my 17" Macbook Pro (unibody model), and I could not reproduce this at all. Playing audio had no effect on CPU Temp, CPU Power Consumption, or CPU Benchmarks. I confirmed the same with an i7 iMac as well. So the problem that this thread is detailing seems to be exclusive to the Mac Pro (09 model in OSX).
 
smacman said:
Maybe I am misunderstanding you, but I think the problem with your Macbook Pro may be something totally different. I did a lot of testing on my 17" Macbook Pro (unibody model), and I could not reproduce this at all. Playing audio had no effect on CPU Temp, CPU Power Consumption, or CPU Benchmarks. I confirmed the same with an i7 iMac as well. So the problem that this thread is detailing seems to be exclusive to the Mac Pro (09 model in OSX).
Click to expand...

I am speaking of running a Mac with an external soundcard device.Not internal souncard. Look at my testings.
 
smacman said:
I did see your previous post, but again, I don't think these two issues are related.
Click to expand...

I am not sure , too . But there is a good reason why there could be a relation.

If I connect my Souncard, the temps increase by around 10°C from the "normal temps". And when I then playback audio, there is no heat increase again at ALL.

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

When I shut the external soundcard off and still playback audio, the temps stay exactly the same. So I think it could be the same culprit ! CoreAudio ? Therefore I don't think it's just a Mac Pro problem, it's a problem with the OS itself too.
 
seisend said:
People who are running a Mac with an external Soundcard (or most of them) device will get the heat increase anyway (at least under 10.6.2) ! As you see, I tested it on my MacBook Pro and posted results earlier in this thread. So IF Apple would fix the problem with the Mac Pro itself, it won't change anything for people with external devices except PCI cards like nanofrog means. It's an OS X problem too.
Click to expand...
In the '09 MP's, the chipset has both PCI and PCIe buses.

PCI = USB, FW, Audio chip, and NIC's. Any audio device on the PCI bus has the heat issue in the '09's.

PCIe does NOT have the heat issue. But I'm only aware of RME producing PCIe based audio devices that will work in a MP (took some time for them to perfect them too).

This issue is definitely within OS X though (10.5.8 on), and Core Audio seems to be the most likely culprit, not Quicktime X.
 
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