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

nanofrog said:
I don't think it is multi-threaded, so TB is likely to kick in. But it's not the source of the heat.

That seems to be the Quicktime (multimedia code in OS X) that iTunes,... relys on. Even EyeTV it seems is affected by this (rather drastically), from a post by gugucom. He was pulling 60C, and the CPU was low % utilization at the time. Same story as everyone else it seems. :(

Sucks, but it seems SL isn't that wonderful for multimedia use on Nehalems at this point. It still needs work for the Nehalem architecture, and no idea how long it's going to take Apple to sort it out.
Click to expand...

so QT isnt multithreaded then. how wonderful! i thought the whole purpose of SL and GCD was to make applications more resourceful and able to be distributed amongst threads/cores?

EyeTV does indeed utilise QT for playback, a lot of programs do including HandBrake and whatnot. it would be interesting to see HandBrake results whilst encoding.

is there any indicator at all of extra CPU usage when the computer reaches higher temperatures? is there "System" CPU usage instead of "User" CPU usage?
 
DoFoT9 said:
so QT isnt multithreaded then. how wonderful! i thought the whole purpose of SL and GCD was to make applications more resourceful and able to be distributed amongst threads/cores?

EyeTV does indeed utilise QT for playback, a lot of programs do including HandBrake and whatnot. it would be interesting to see HandBrake results whilst encoding.

is there any indicator at all of extra CPU usage when the computer reaches higher temperatures? is there "System" CPU usage instead of "User" CPU usage?
Click to expand...

I really don't think it's qt. It was completely rewritten for 10.6. Do the optional qt pro install from 10.5.x. You'll see the performance difference, its amazing.. It's something that's not new to 10.6 as it was present in 10.5.x.

So the question is. What code foundation wasn't upgraded in 10.6 from 10.5.8 that deals with audio and/or video?
 
handheldgames said:
I really don't think it's qt. It was completely rewritten for 10.6. Do the optional qt pro install from 10.5.x. You'll see the performance difference, its amazing.. It's something that's not new to 10.6 as it was present in 10.5.x.

So the question is. What code foundation wasn't upgraded in 10.6 from 10.5.8 that deals with audio and/or video?
Click to expand...

disappointing to hear, but glad to hear that the old one still works properly!

have you tried running it with Perian installed? does that make any differences?

whilst just searching the Googles i came across a forum that said QTx does not support AC3 audio very well either (i.e. doesnt play it). can anybody confirm?
 
Further to my earlier post, it seems that iTunes DOES have an issue with decoding MP3s.

It kicks the Mac Pro out of full power saving which increases temps by 25C and power usage by 50W, JUST TO PLAY MP3s!

I have submitted a bug report.

Shall see what they say!
 
gugucom said:
My Nehalem Octad has the same problem when I let it idle in EyeTV. I do nothing but let the app run and it will kick the Boost fans up to 1900 rpm and CPU A temp goes to 53°C. I'm not using iTunes at all and I'm not surfing or doing anything else on the machine. I guess it is a bug in the Quicktime software. I'm pretty sure EyeTV uses Quicktime. I'm not very knowledgeable on software so I can't really tell. But I have shut down all other apps and the bug is definitely coming from EyeTV.
Click to expand...

I had regretted getting the hauppage instead of $100 more for the eyetv as the eyetv does it's own transcoding instead of abusing my CPU. Aside from converting video I don't see the same problem you are having with hauppage when idle.
If I had it my way I would have put eyetv software on a spare laptop as it interferes with disk usage as well, but I need it on all the time to remotely view/record from my iPhone. Very nice gui software, poorly written backend code.
 
It seems to top out in the mid 60º Celsius range, no matter what I have running.

It'll be interesting to see if the i7 iMacs have this problem. An all-in-one will have a much harder time dealing with the excess heat.
 
Cynicalone said:
It seems to top out in the mid 60º Celsius range, no matter what I have running.

It'll be interesting to see if the i7 iMacs have this problem. An all-in-one will have a much harder time dealing with the excess heat.
Click to expand...
Yes, they will, and be interesting to see what happens. Maybe the complaints from the i5/i7 iMac owners will be enough to make Apple address the issue when combined with the reports they've already gotten from '09 MP owners.
 
nanofrog said:
Yes, they will, and be interesting to see what happens. Maybe the complaints from the i5/i7 iMac owners will be enough to make Apple address the issue when combined with the reports they've already gotten from '09 MP owners.
Click to expand...

what if there is no problem and apple still ignores the MP issues? :confused: :eek:
 
DoFoT9 said:
what if there is no problem and apple still ignores the MP issues? :confused: :eek:
Click to expand...

If that happens, I guess it would point to a logic board / chipset specific issue. I think all 09 MP owners would have good cause for a class action law suit against Apple. Heat kills computer components, and almost any internal component failure could be linked to this unnecessary heat output.
 
DoFoT9 said:
what if there is no problem and apple still ignores the MP issues? :confused: :eek:
Click to expand...
It's possible, but I don't think that's the case. Intel's done the board work, and all the PC boards follow the reference design to some extent if not completely. Those boards aren't having problems with other OS's, so it leads me to think the issue is software.

So I really am expecting the iMac owners with Nehalem chips (i5 and i7 models), will end up reporting the same thing.

smacman said:
If that happens, I guess it would point to a logic board / chipset specific issue. I think all 09 MP owners would have good cause for a class action law suit against Apple. Heat kills computer components, and almost any internal component failure could be linked to this unnecessary heat output.
Click to expand...
Quite possible this would happen. America...Land of the Law suit. :p

Keep in mind though, Intel specs the parts out at 67C under stock settings, so technically, they're still passing, despite the lack of core use with the temps seen on the '09 MP's.

Hard to say what the outcome would be, but I don't trust the court system to rule in favor of consumers these days. :rolleyes: :(
 
nanofrog said:
It's possible, but I don't think that's the case. Intel's done the board work, and all the PC boards follow the reference design to some extent if not completely. Those boards aren't having problems with other OS's, so it leads me to think the issue is software.

So I really am expecting the iMac owners with Nehalem chips (i5 and i7 models), will end up reporting the same thing.
Click to expand...

hmm so if the design work by the board makers is followed exactly does this mean that iTunes on a Windows computer would experience the same increased temperatures?

i am watching the iMac shipping thread like a hawk - if there is a problem, ill find it ;)
 
DoFoT9 said:
hmm so if the design work by the board makers is followed exactly does this mean that iTunes on a Windows computer would experience the same increased temperatures?
Click to expand...
In general Yes, but it's harder to make a direct comparison, as the coolers and airflow are different in any system.

However, if a comparison is made between the same operation between OS's on the same system, you'd see a difference, and it's quantifiable.
 
So I just hung up after a 2hr chat with Apple about this issue. The first person I spoke to was absolutely clueless but eventually I was able to speak to a 2nd tier support person. To make a long story short, I got the usual "Apple is not aware of any issue with the Mac Pro" line. Due to the fact that the Activity Monitor shows iTunes using a normal amount of CPU usage, as far as they are concerned, there is no issue.

I hope everyone else with a 09 MP calls in and does their best to escalate this...
 
smacman said:
So I just hung up after a 2hr chat with Apple about this issue. The first person I spoke to was absolutely clueless but eventually I was able to speak to a 2nd tier support person. To make a long story short, I got the usual "Apple is not aware of any issue with the Mac Pro" line. Due to the fact that the Activity Monitor shows iTunes using a normal amount of CPU usage, as far as they are concerned, there is no issue.

I hope everyone else with a 09 MP calls in and does their best to escalate this...
Click to expand...
The response isn't surprising, but I'm sure they're they're at least aware of it.

More complaints could help to escalate it in consideration (i.e. increase the urgency to gain developmental resources), and would be assisted by the i5/i7 iMac owners.
 
nanofrog said:
In general Yes, but it's harder to make a direct comparison, as the coolers and airflow are different in any system.

However, if a comparison is made between the same operation between OS's on the same system, you'd see a difference, and it's quantifiable.
Click to expand...

well yea, the cooling systems will of course give different results but if it is across the entire set of boards and not just an Apple issue then a temperature increase would still be measured there somewhere (even though the CPU is still at ~0%).

has anybody tried digital audio? or just analogue? or just digital? lol. :D
 
DoFoT9 said:
well yea, the cooling systems will of course give different results but if it is across the entire set of boards and not just an Apple issue then a temperature increase would still be measured there somewhere (even though the CPU is still at ~0%).
Click to expand...
Yes it would.

My point was, it may be less recognizable in PC's, especially as some may have gone to the extent of copper coolers, and lapping the base as well. So even with the same software behavior, the max temp could be 15C lower with a large, lapped cooler (i.e. the Copper version of the Thermalright 120 Xtreme). So it wouldn't be as apparent unless it was compared to the same task between Windows and OS X on the same system. There'd still be a temp increase, but the max temp could be lower than those seen in MP's. It would still indicate that there's a software issue though. ;)
 
nanofrog said:
Yes it would.

My point was, it may be less recognizable in PC's, especially as some may have gone to the extent of copper coolers, and lapping the base as well. So even with the same software behavior, the max temp could be 15C lower with a large, lapped cooler (i.e. the Copper version of the Thermalright 120 Xtreme). So it wouldn't be as apparent unless it was compared to the same task between Windows and OS X on the same system. There'd still be a temp increase, but the max temp could be lower than those seen in MP's. It would still indicate that there's a software issue though. ;)
Click to expand...

i was under the impression that the MP heatsyncs/cooling systems were pretty high quality and efficient! i guess not :(
 
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