1 34 35 36 37 38 Previous Next 600 Replies Latest reply: Mar 27, 2010 12:59 AM by svkrzn Go to original post
  • 525. Re: Temperature Rises with audio
    The hatter Level 9 Level 9 (58,880 points)
  • 526. I guess my suspicions were right on the Audio Update...
    xlr8yourmac.com Level 1 Level 1 (40 points)
    Just a FYI for those that missed my comments on this update yesterday (didn't see it listed in 10.6.2 ASP extensions list before the updater applied (i.e. not loaded/used before update) - although the same extension/version (and same last modified date) was present in 10.6.2 already.)

    Anyway this morning Marcel Bresink (the author of hardware/temperature monitor) saw my comments on that and wrote about what was really changed by this update. I've added that in Friday's news at my site (http://www.xlr8yourmac.com)

    But regardless, as I said yesterday I'm just glad I no longer see 110W+ increase from just playing audio in OS after this update in 10.6.2. (Literally no rise after update) And as I said yesterday, I hope they post an update for 10.5.8 also, as my tests last Nov were using 10.5.8, but I guess their priority was to fix SL first.

    (And for those that think this was a hardware issue, as some others here have said - I already tested/posted results of tests in Bootcamp/Windows last November - even with cores running in turbo mode I saw only a few Watt increase in Bootcamp/Windows from audio playback. I need to update that article to note the fix for 10.6.2 at least...)

    I won't repeat my comments (since last Nov) on the SL boot log warnings on "turbo enabled but no turbo P-state found" - but back in Dec I heard the 10.6.3 beta didn't have those (not sure if just supressed or actually fixed...)
  • 527. Re: I guess my suspicions were right on the Audio Update...
    RaySkater Level 3 Level 3 (735 points)
    xlr8yourmac;

    this bit from Marcel caught my eye:

    "The temperature behavior of the CPUs (which has never been a problem) does not change, you only need different programs (not audio) to trigger a rise of temperature values now.

    Best regards, Marcel Bresink"

    ------


    I'm left with this nagging question: WHICH programs will now trigger the rise in temps?

    I hope it's not anything I use...

    hate to say it, but it sounds like they did the software equivalent of hiding the creamed spinach under the sofa cushions!


    ray
  • 528. Re: I guess my suspicions were right on the Audio Update...
    Samsara Level 4 Level 4 (1,320 points)
    The temperature behavior of the CPUs (which has never been a problem) does not change, you only need different programs (not audio) to trigger a rise of temperature values now.
    Best regards, Marcel Bresink

    A nice outcome for the good Doctor. At first he said that I hitting core temps of 90c sometimes was of no concern because these machines were made to "+run hotter+"
    Now that Apple has come out for a fix for something that was never a problem, he still has it right, my temperatures, at least now, are no problem.
    From a norm of mid-80s to now 53. Quite a change for something never to have been a problem. Side effect maybe?
    Maybe his plain writing confuses me...
  • 529. Re: I guess my suspicions were right on the Audio Update...
    xlr8yourmac.com Level 1 Level 1 (40 points)
    Reply to Ray's comment
    "....hate to say it, but it sounds like they did the software equivalent of hiding the creamed spinach under the sofa cushions!"
    -------------

    I disagree - read the comments (the AUDIO issue is FIXED - 100W+ ramp up is GONE)
    but of course there are going to be cases where the core/CPU temps will ramp up (jesus - does that even need to be mentioned??? hello?)

    As I posted before - I've seen as high as 92C cores (or maybe more accurately, 3 C under the throttle down limit) with repeated HB conversions. (which had nothing to do with core audio)
    So he's just saying the obvious - there are cases where temps will ramp up but this AUDIO only (insane wattage ramp up) has been fixed.

    I'm sorry that I have no time to do the usual back and forth forum post debate on this (which is why I rarely post in these forums - even when I see many times things I would like to reply to - incorrect assumptions/logic and worse).

    And since I'm sure many here missed it earlier - yes, I wish apple included some user adjustable fan speed controls (without needing smcfancontrol or similar mods) - I already mentioned that I saw very little (100rpm or so) increase from my idle rpms of the booster fans (inside the CPU heatsink fans shown in my article on 2009 mac pros) even when temps ramped up. They did ramp up about 200rpm when some of my cores hit 3c below where TCC should kick in...)

    (And like some other users here (but I won't brag on it) - I've built homebrew PCs since the 80s and wish Apple had tweaks that are available in the bios of PC motherboards for decades.)
  • 530. Re: Temperature Rises with audio
    dryjoy Level 1 Level 1 (0 points)
    Thanks for the suggestions, Ray.
    Actually, I am already using SMC Fan Control and temperature monitor to monitor and control the situation, so I'm not concerned that my machine is in any immediate danger or anything.
    The thing is though that increased fan speeds are actually pretty BAD news for anyone doing music - we all want the quietest computers available, you see! We recordists are VERY sensitive to noise, so having my fans at higher speed is not ideal and there are times when I have to turn them down to default. Mac Pros are under normal circumstances very highly regarded in musical circles for their low noise factor.
    But I'm cool, though (pun intended), as I said I have only recently contacted Apple and I have found them to be very understanding and helpful, and I remain hopeful that they will find a solution for Leopard users.
  • 531. Re: Temperature Rises with audio
    dryjoy Level 1 Level 1 (0 points)
    I have to say, that with all due respect, the stuff that Marcel Bresink is saying about this update does NOT ring true for me. I know he is very knowledgeable and well respected, but I think he has not understood the full gravity of this issue or what exactly has been happening. For example, he says '+The second extension contains a collection of drivers for audio chips complying with the Intel High Definition Audio standard. After the update, playing audio over the integrated audio chips will no longer create the unexpected change in processor performance.+'
    He seems to be implying that the problem was limited to the onboard audio, but this is clearly not the case, as any audio interface connected via USB or Firewire is affected. Also, he says that the temperature rises were never a problem, but actually they were a problem to some users who experienced hardware failure, and it was quite clear that some users machines were operating right at or above safe limits.
  • 532. Re: Temperature Rises with audio
    Mike Connelly Level 4 Level 4 (1,785 points)
    2bigfeet wrote:
    Define early '09 MacPro...? I downloaded this updater and got a prompt that says "this software does not support this hardware" or vise versa I forget.


    The current generation of MP was released in early March, if that helps. If you bought before that you have the previous model and didn't have this issue.

    RaySkater wrote:
    I'm left with this nagging question: WHICH programs will now trigger the rise in temps?


    I assume he means programs that heavily tax the CPU. It's normal for heavy cpu use to raise temps, the problem in this case was that it was happening with audio playback that barely required any cpu at all.
  • 533. Re: I guess my suspicions were right on the Audio Update...
    RaySkater Level 3 Level 3 (735 points)
    xlr8yourmac.com wrote:
    Reply to Ray's comment
    I disagree - read the comments (the AUDIO issue is FIXED - 100W+ ramp up is GONE)
    but of course there are going to be cases where the core/CPU temps will ramp up (jesus - does that even need to be mentioned??? hello?)


    Yes. Hello and Welcome!

    I guess the way I read his comments was that there would be another culprit, not to merely raising the temps, but doing so out of proportion to the expected CPU load.

    It also sounded like he said that the loading order of a couple of components was changed, rather than a particular component being fixed or re-written.

    It seems to have a functionally good outcome, yet it also sounds like a short cut.


    I'm glad the weekend is coming up so that all of the experts can relax over a beer or three.

    cheers


    ray
  • 534. Re: I guess my suspicions were right on the Audio Update...
    xlr8yourmac.com Level 1 Level 1 (40 points)
    Sorry - rushed here, but as some people (privately) know, I have not always agreed with his comments/opinions (I won't get into that).

    But on the lighter side - I'm finally able to get out for food and to see other life forms (after over a week due to east coast snowstorms). **** it feels good!!!
    (but almost burned up the $!!%$# saturn POS Vue VTI trans trying to get back up the driveway - but that's another story (with NO happy ending...)
  • 535. Re: I guess my suspicions were right on the Audio Update...they were
    Samsara Level 4 Level 4 (1,320 points)
    of course there are going to be cases where the core/CPU temps will ramp up

    I guess one could launch Hardware or Temperature Monitor and one other app on their HD, and test till all one's apps have been tested. Of course each app would have to be actually doing something.

    Good thing for someone with a not too pressing Saturday afternoon to do.
    (Me? Um...lint folding...)

    (PS- I know,.. Johnny Carson didn't always agree with his guests either. _But MAN, did I trust that guy!_... Lol, hey, it's the news, you have to report it as it comes. And, I still have respect for the guy anyway. Use his products enough to say so. But if I ever know exactly what he is saying when he talks about Macs and PCs, I'll let everyone know.

    but that's another story (with NO happy ending...)

    "When it rains, it snows" Or the other way...

    Message was edited by: Samsara
  • 536. Re: I guess my suspicions were right on the Audio Update...they were
    RaySkater Level 3 Level 3 (735 points)
    Could it be a language issue?

    Marcel seems like a pretty bright guy and his English is way better than my German..but, might he phrase things in a way that reads differently to a native reader of English than he intended?



    r
  • 537. Re: Temperature Rises with audio
    The hatter Level 9 Level 9 (58,880 points)
    I give Marcel credit for responding, maybe not weighing in with a full dissertation.

    The processors were functioning as they should is what I took some of his comments, and that there was no hardware failure or aberrant behavior.

    Seeing that Apple didn't really provide a good write up on what that update was and would do, all we do is try to guess, reverse engineer in our mind based on the behavior and numbers before/after and compare. And jump to conclusions.

    Mac Pros have never done a good job at proper cooling, and always came up short.

    Apple provided audio drivers. How those are used, maybe by other devices? and the role that they played....? Apple provides IOKit and libraries that other devices tap into. Jumping from A to D, with inference and 'best guess?'

    So there may be more work to do, put out the fire, and deal with the next step(s) (if any).

    I once read, and this was 3-4 decades ago and the gap has only widened, that 2% of the engineers in computing understood the full scope and range of technology involved. Most understand their backyard territory that they know. And Intel does not publish everything, Marcel has spent time trying to find all the what/when how/why of sensors, digital readouts, whatever.

    High processor usage, whether 110% on one core/thread alone, or using 800% or 1200%, and seeing an increase in temperature is what is to be expected.

    Can or is it possible to write a script or driver that is faulty? sure. We've see that. ATI has found some bugs in their 55/57/5800 graphic cards and 2D, in areas that they never had a way before to test or look for. Putting GPGPU and OpenCL has and will have unexpected and unknowns.

    +New hardware always involves a learning experience.+
  • 538. Re: I guess my suspicions were right on the Audio Update...they were
    Samsara Level 4 Level 4 (1,320 points)
    I was using his products and writing to him for many years. I didn't consider him a bright guy, I considered him to be a Mac genius, and not the kind you find in an Apple Store either. The guy is very accomplished. His software has always been topnotch, I have Tinkertool also, I think the best of its kind of software out there, and he answers every email he is sent.

    So no, he is at least a very, very smart guy, who probably like most Europeans, speaks English better than most Americans. For me it's his technical discussions, and indeed it may all be me and no one else who thinks the same.
    But being thought of a fool isn't a big worry on my list anymore. So since I'm questioning Marcel it's only right the he speaks for himself.
    Try to find where only he speaks in the following pages, if you wish, and see if you think what he says is clear, and clearly explained. And keep in mind the fix that has come out since.
    http://www.xlr8yourmac.com/feedback/2009mac_pro_cputemps/2009MacPro_cputemps.html#storytop
    Second box of Italics is where he starts (first box was mine, ho, ho)
    And below on todays front page:
    http://www.xlr8yourmac.com/
  • 539. Re: Temperature Rises with audio
    Samsara Level 4 Level 4 (1,320 points)
    no hardware failure or aberrant behavior

    That's what I don't understand. Before my temps were in the mid-80s, and Dr. Bresink said that was fine, hence no hardware failure or aberrant behavior.
    Now, my top temp hits 53c.
    Hardware failure, aberrant behavior... Is it happening now or was it happening then? Or, is there another way of looking at it?
    I wonder if he's just so highly knowledgeable that he finds it hard to discuss things with those who know less than he does. One certainly can't say he's not treating you as an equal in that sense. He explains things in ways that only people with his skills would understand.
    You probably have a much easier job at understanding him, Hatter, than the majority here.

    PS: What happened to TurboBoost and the rest, the reasons for the high temps...? Don't hear it now..
1 34 35 36 37 38 Previous Next