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

Success MacPro 2009 Audio Fix.

I have already thanked all those responsible for helping with this problem I have had since March 2009 when I took delivery of this 2.26 2XQuadCore core Nahalen in the thread opened today on the main page.

I am impressed for the first time and I cannot feel the heat from his computer, it all appears very stable, as previously iTunes caused it to consume more power than it is currently doing processing 234 events and 85000 images in Aperture3, and itunes.

Thank you Apple, thank you all for making this happen MacForums and the publicity provided by ars and cnet....

It has now been running for in excess of eight hours running iTunes and Aperture3 and the Bresink measurements are 45 and 47 respectively - previously they would have been in the low 60's.... Wow I am happy....
 
Same here - running iTunes now for 2 hours, browsing the web and doing some work in Pages. CPU shows 29°C, it used to be in the low 60's.

FINALLY I can listen to music on my Mac Pro again while working :)

Thank you Apple for fixing it and also SHAME ON YOU APPLE that we basically had to force you to do so...
 
i just tested cinebench and actually got the best cinebench score i've ever gotten with my mchine while running itunes. [edit: i later topped it, but it was at the time]

the tests i ran were a baseline with itunes open: 22963 (0)

then after selecting a song, playing it, and pausing it before running cinebench: 23347 (22963)

then another multiple CPU render test after playing a couple minutes of Giant Steps at 320kbps:24977 (23347)

that's the fastest i've ever run it... ever! i paused the song, waited about a minute and ran cinebench with itunes open but with no music playing. the results are: subsequent tests were 23206, 23451, 23127

i opened safari to post this, and with the same album playing i ran two more tests: 23422 and 23350.

finally, with itunes closed, i ran another test: 23082


update: after having itunes closed for a little while, though... i topped my score with it open... 25121. haven't gotten it that high again.
 

Attachments

  • Screen shot 2010-02-12 at 1.28.43 AM.jpg
    Screen shot 2010-02-12 at 1.28.43 AM.jpg
    37.9 KB · Views: 119
Thanks to all Forum posters here, and thanks to Apple for finally listening to us!

Feels good to listen to audio and staying under 10watts power consumption on CPU :)
 
HT4034 seems a bug fix for my audio frying symptoms (but not exhaustively checked yet).

Stress testing the system, my CPU A temperatures still get into the mid/high 80C range in 10 minutes without fan speed increase. CPU B shows high 60C range.

I wonder if on dual systems, Apple is monitoring only CPU B (for fan control feedback).

As to temperature safety ceilings for these Nehalems, I hear strident opinions in various directions. Intel seems to be clear it is 70C, the overclocking brigade say 100C. My concern is hardware damage/longevity.

Perhaps a subject for another thread.

In all this, I hope I have no axe to grind in any of this (just some steam and exasperation with Apple's secrecy and AppleCare incompetences), and have gleaned some improved sense of where and where not to listen for useful information. Also disappointment at the ego laundry that airs so readily, but much respect and gratitude to some on these forums.

Now how to lay to rest Facebook and web site material ... it's been a learning experience.

You've got to love people power.
 
cl-user said:
As to temperature safety ceilings for these Nehalems, I hear strident opinions in various directions. Intel seems to be clear it is 70C, the overclocking brigade say 100C. My concern is hardware damage/longevity.
Click to expand...

It really depends which temperature we are referring to.

70C (67.9 for Quad core machines) is the max Tcase temp which we can't exactly measure (approximately CPU Temp Diode temp).

100C is the approximate max Tcore temp (CPU core temp).

Generally speaking, as long as your CPU Temperature Diode is staying below 68C, you are "within limits".
 
smacman said:
It really depends which temperature we are referring to.

70C (67.9 for Quad core machines) is the max Tcase temp which we can't exactly measure (approximately CPU Temp Diode temp).

100C is the approximate max Tcore temp (CPU core temp).

Generally speaking, as long as your CPU Temperature Diode is staying below 68C, you are "within limits".
Click to expand...

CPU A temperature as measured by Temperature Monitor on CPU A Temperature diode, which I take to approximating Tcase.

I am on dual 2.66 MP

Smacman, do you get into the 80Cs under stress test with the patch and no other mods?
 
cl-user said:
Smacman, do you get into the 80Cs under stress test with the patch and no other mods?
Click to expand...

I haven't tried a stress test yet (post patch). Will test when I am home later. That does sound about right though. I am going to continue using Fan Control to keep a slightly more aggressive fan profile.
 
bootstrapjesse said:
Yes. No one is talking about firewire. Can someone please evaluate this or confirm my findings? I am now seeing a power usage of 4-5 watts per CPU (above idle conditions) when transferring files to a drive via firewire with a 1-2 deg C temp increase. Example 3-4 W @ idle and 7-9 W with firewire in use. I just want to be sure that this behavior is normal.

I am really happy that the excessive power drain and increased temps due to audio are gone. I just want to make sure that the whole problem is fixed before I do the victory dance.

--Thanks
Click to expand...

Yup, thats fine. The system uses a few CPU cycles to calculate transfer data etc.

As for thanking Ars Technica, it took them long enough, and besides its users on this forum that found the bug, did the testing and spent shedloads of time dealing with it.

I think we should have a Mac Pro 2009 Band of Brothers User title :p
 
I'm really happy for everyone that got their problems sorted out.

I really hope this patch fixes everyones problems.

I've been rooting for you guys since the beginning of this thread.
 
This wasn't really an issue for me (we haven't any speaker connected to the MP ) but I'm glad you finally have the problem solved ;)
 
So what temperature should a Mac Pro (quad) be idling at? Mine idles in the upper 30s or lower 40s, is that normal? Mine would spike up to around 74 when playing music before which is higher than most (I think a lot of people spiked up to around 65). The fix works great, but I'm just wondering if my temps are still a bit higher than normal.
 
I just installed 10.6.2 and software update now found a new firmware update (1.4), is it recommended to install this as well?
 
grey17 said:
FYI:
The info on the patch says an extension called AudioAUUC version 1.0 is installed. The date listed in system profiler for AudioAUUC version 1.0 is 11/3/09 2:24 PM. It looks like Apple was working on this for a lot longer than this thread would suggest. Unless Apple's programmers are backdating their extensions...

...and yes it seems to fix the problem.

P.S. First post, long time lurker...
Click to expand...
Apple's HT4034 article says `AudioAUUC version 1.0 should be listed as an extension'.
As another person has noted, the installer package has only AppleHDA.kext and IOPlatformPluginFamily.kext, not AudioAUUC.kext, you can see this yourself by using Pacifist.
I didn't save a System Profiler MacPro.spx from before the update, but don't remember seeing AudioAUUC.kext, so it appears that a change to one or both of the two kexts causes it to load now.
 
Concorde Rules said:
Yup, thats fine. The system uses a few CPU cycles to calculate transfer data etc.

As for thanking Ars Technica, it took them long enough, and besides its users on this forum that found the bug, did the testing and spent shedloads of time dealing with it.

I think we should have a Mac Pro 2009 Band of Brothers User title :p
Click to expand...

I will stand along side you, brother...
 
YEEEEEEEEEEEEEEEEEEEEEESSSSSSSSSSSSSS!!!!
Finally I can play music while photoshopping without having a heater by my legs

thank you apple and everyone who has helped in this thread to make it known and try and fix
 
thank you all for your support and we made it .... WE MADE IT , NOT APPLE :)

my current temps while listening to iTunes through external interface.

Ambient is always around 27-32 °C.
 

Attachments

  • Screen shot 2010-02-12 at 17.57.08.png
    Screen shot 2010-02-12 at 17.57.08.png
    59.6 KB · Views: 150
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