Skip navigation

Frequent Panic NVRM[0/1:0:0]: Read Error 0x00000100

152966 Views 743 Replies Latest reply: Jan 6, 2014 6:39 AM by Peter Lorraine RSS Branched to a new discussion.
  • dmdimon Level 3 Level 3 (840 points)

    Can we get something going for this??

    yes.

    Go iFixit for your model, disassemble it up to heatpipes/heatsynk dismount, wipe thermopaste from GPU, CPU and heatsinks, add small amount of good thermopaste, assemble carefully and tight.

     

    90% that problem will go.

    99.5% that problem will go for users who get this errors only after update to ML.

  • MrSparkle75 Level 1 Level 1 (5 points)
    Currently Being Moderated
    Nov 13, 2012 1:36 PM (in response to dmdimon)

    Lol, obviously dmdimon and I are talking about different faults. The "Frequent Panic NVRM[0/1:0:0]: Read Error 0x00000100" I am talking about is a known hardware fault (see my earlier posts in this thread) and is not related to heat. It is publicly documented here;

    http://support.apple.com/kb/TS4088

     

    The way to tell if your machine is affected (as more than one issue can generate this kernel panic) is;

    1) Your machine is a MacBook Pro 15-inch, Mid 2010

    2) The problem goes away if you disable the nVidia card

    and bonus points if;

    3) It only started happening after upgrading to ML (seems that there was a work around in the video drivers for the older versions that couldn't be implemented in ML).

     

    Ta,

  • dmdimon Level 3 Level 3 (840 points)

    MrSparkle75 wrote:

     

    Lol, obviously dmdimon and I are talking about different faults. The "Frequent Panic NVRM[0/1:0:0]: Read Error 0x00000100" I am talking about is a known hardware fault (see my earlier posts in this thread) and is not related to heat. It is publicly documented here;

    http://support.apple.com/kb/TS4088

     

      Sadly no.

    We're talking about same problem. In Apple support message you pointed to not stated cause of problem.

     

    see yourself:

     

    1) If you'll do a bit deeper search, you'll find that nVidia had problems with thermal dissipation on mentioned GPUs and with packaging quality. there even been clash between Apple & nVidia on that. So this error is thermal-bound.

     

    2) If you'd know nVidia drivers ideology (unified driver architecture) you'd know that equal error report means equal problems independently from hardware since first GeForce chips with UDA drivers, so it is thermal problems on ANY nVidia GPU

     

    3) If you'd watch Apple strategic movement in software development, you'd know that everything goes GPU offloaded - so ML just gives more load to GPU than L, thus - more heat from GPU.

     

    Thus, everyone, who worked under 10.7 happily and got troubles under 10.8 just worked on the edge of GPU thermal stability - and crossed that edge.

     

    Thus, lowering GPU temp will help if GPU not degraded.

     

    To make long story short - there was multiple reports on not exellent quality of GPU-heatsink thermal contact in MBP. So here from my recommendation. Also you can try to spin-up coolers with fan control or something, should help - but worse. Also theoretically you can underclock GPU core/memory - or wait until it'll be done in some drivers version or tweaking utility.

  • MrSparkle75 Level 1 Level 1 (5 points)
    Currently Being Moderated
    Nov 14, 2012 2:41 AM (in response to dmdimon)

    I respectfully choose to disagree. As Adam Savage would say, "I reject your reality and choose to substitue my own." :)

     

    Firstly the concept that previous OS were "on the thermal edge" is unlikely. The tiny amount of additional 3D load that ML puts on the GPU vs the amount that a 3D game does, would have resulted in Loin systems crashing every time a game was played and couldn't be resolved with a software patch. Also let's not get started on what would happen in hot weather.

     

    Secondly a real thermal issue is unlikely (although not impossible) to behave the same every time. Thermal faults result in random bit swapping so one moment the GPU might try to access a memory location that doesn't exist. The next time it might read the wrong value back for a colour in a texture map resulting in video glitch.

     

    Thirdly, pulling your own machine apart would be a silly thing to do (in terms of voiding your warranty and potentially damaging the machine) when you consider that Apple have a free replacement program for this fault which doesn't even require the owner to have AppleCare on the machine.

     

    What I do agree with is that should you have a heat related issue caused by faulty conduction between the GPU and heatsink, then replacing the thermopaste would likely fix the problem. But I would only do this on a machine that is out of warranty and not covered by a replacement program.

     

    I also agree that Apple have had heat issues with nVidia in the past. http://support.apple.com/kb/TS2377

    This fault was actually heat related and would cause random issues. It was easy to trigger just by doing anything that made the GPU work hard enough.

     

    Just my 2c.

  • dmdimon Level 3 Level 3 (840 points)

    lets try to find some consensus )

    sorry for my english level, I'll try as I can. So:

    Firstly the concept that previous OS were "on the thermal edge" is unlikely. The tiny amount of additional 3D load that ML puts on the GPU vs the amount that a 3D game does, would have resulted in Loin systems crashing every time a game was played and couldn't be resolved with a software patch. Also let's not get started on what would happen in hot weather.

    well, how cooling works? thing is simple. you have some heating element, some heat dissipation element and some heat transfer element (frome heater to dissipator). For example - heater - GPU, transfer - thermopaste, dissipator - thermotube. Or heater - thermotube, dissipator - air cooled radiator, transfer - zone of contact between tube and radiator. Hope I give you idea.

    So there are some temperature steps on each transfer element, as heat can be transferred with some speed and there are such characteristic as thermal resistance for that.

    Now going to problem - if we have high thermal resistance between GPU and thermotubes we get high thermal step between GPU and cooling system, say 5-10 C more than it should be. (if you take a look at overclockers forums you'll find that it goes 5-15 C). So Mac adaptive cooling system should run coolers say 5-10% faster at same weather and load level to compensate that enormous thermopaste layer.

    Here we go with system almost on edge - it will work ok at cold and probably overheat at warm weather and full load. Looks like situation you described.

     

    BUT. It will be so if cooling system will use only and exactly on-die thermosensor. In real life it is not so - lots of sensors are involved and all of them exept one (on die) lies to us due to high thermal resistance between GPU and thermotube. And our well-balanced and precalculated (for normal T-resist) adaptive cooling with all cooler control algorithms and coefficients go offsync with reality and undercool GPU at ANY weather. So our GPU will in reality be (not so) slightly overheated all time. I'd say up to 5-7 C

     

    Here we are exactly 3-5 C lower than edge now - at least for Intel CPU difference between safety shutdown and 100% load is like =< 10 C. And we are almost on edge for ANY weather - as our fooled cooling system drives us there.

     

    Now we go for more sophisticated thing.

     

    All modern GPUs have asyncronous independently floating speeds for subblocks. It is done to conserve power/lover heat dissipation. Every block in GPU tries to run at lowest possible speed, just enough to fulfill current task. If that is not enough - speed goes up for that block.

     

    Take a look at modern GPU bottleneck analysis and you'll see that most heavy working part in GPU in almost any situation is memory controller. So if we add a bit of memory-intensive things (ML offloaded parts) to GPU we easily can flop memory controller at given GPU speed thus enforcing it to go higher.

     

    And if we have only memory-intensive tasks with low load on other blocks - we have HOT memory controller and cold everything else. And median temperature will be not so high, and cooling system will try not so hard to cool it down.

     

    Got the idea? We have already fooled cooling system (undercooling), and we fool it again due to spot heating and low median temperature. Here we immediately get memory controller overheated and what ? read error. What else can you expect?

     

    For this error in condition of normal cooling we should overheat mem controller for say 10 C, so if we already have 5-7 C gap on cooling it lowers to 3-5 C, i.e. we have 2x - 3x less thermal reserve.

     

    And now we go to time factor. For GPU run faster we need some nanoseconds, for cooling system run faster we need seconds. see the gap? and we have 2x-3x lovered reserve to survive that gap.

     

    This is living on the edge. And to flop over that edge +5-10% specific GPU-memory operations easily can be enough.

  • dmdimon Level 3 Level 3 (840 points)

    Thirdly, pulling your own machine apart would be a silly thing to do (in terms of voiding your warranty and potentially damaging the machine) when you consider that Apple have a free replacement program for this fault which doesn't even require the owner to have AppleCare on the machine.

    Fully agree with you on that. But question was - what we can do?

  • fndtomas Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 14, 2012 10:24 AM (in response to dmdimon)

    Hi folks,

     

    I have following this discussion for quite some time. I had the same "panic problem" as described and, after months of "coming and go", it was finally solved.

     

    Yesterday I got my computer from the repair center. The solution ?

     

    Answer: Logic board replacement.

     

    This is the only way to fix the machine, no way. Insist on TS4088.

     

    Good luck.

  • alekb Level 1 Level 1 (5 points)
    Currently Being Moderated
    Nov 23, 2012 5:19 AM (in response to fndtomas)

    Hello fndtomas,

    can you say me how much does your repair cost you and where are you, please?

     

    Please! if someone fixed this issue with Apple Service under this plan:

    http://support.apple.com/kb/TS4088

    or anything else please, post a note about it here,

     

    thanks

  • fndtomas Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 23, 2012 6:18 AM (in response to alekb)

    Hi alekb,

     

    I got it the logicboard replaced for free in Frankfurt/DE. It took quite long time talking with all the hierarchy of genius, repair center and apple care.

     

    First of all: This problem doesn’t seems to be localized. Several people are reporting the same issue ... since jul/2011.

     

    Most Important thing: NEVER follow any advise which tells you to do stupid things with your hardware such as open it, put thermic pasta, change disk, memory and etc...

     

    Good luck for you.

     

    <Edited By Host>

  • apple242 Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 23, 2012 6:33 AM (in response to fndtomas)

    Unfortunately Apple Germany is a different matter than Apple USA and most other Apple countries.

     

    Save your time, money and health and go another EU country and resolve the problem ASAP.

  • fndtomas Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 23, 2012 6:45 AM (in response to apple242)

    1. I love this country .

    2. The problem is the business model not the algorithm .

    3. My suggestions is constructive.

    4. Only understand who can...

     

    <Edited By Host>

  • apple242 Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 23, 2012 6:55 AM (in response to fndtomas)

    As I said, if you want to save time and money go to France or Holland and repair it without the unprofessional and inferior service from Apple Germany.

     

    I know what I'm talking about. We spent in excess of 6 months trying to resolve it.

    And that was with Apple USA all over them.

     

    They hadn't even heard of TS4088 when the rest of the world was already doing these repairs for a year.


    Go to France or Holland where the service is as polite and competent as in the USA.

    There is absolutely no need to indulge the unprofessionalism of Apple Germany.

  • apple242 Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 23, 2012 7:03 AM (in response to fndtomas)

    fndtomas - I wasn't criticising you. Nevertheless your last post is emblematic. 

     

    For those interested in actually repairing their laptop instead of debating endlessly with Apple Germany,

    better service is available just across the border.

     

    There's no excuse for wasting 6 months with Apple Germany.

  • fndtomas Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 23, 2012 7:05 AM (in response to apple242)

    You have your own opinions that are not congruents with mine.

     

    Thanks for the last emails. You could very well be a genius.

  • apple242 Level 1 Level 1 (0 points)
    Currently Being Moderated
    Nov 23, 2012 7:11 AM (in response to fndtomas)

    It is not important whether our opinions are congruent.

     

    You yourself have mentioned the absurd amount of time required to resolve the problem with Apple Germany.

     

    It's completely unnecessary.

     

    Across the border, one receives more polite and competent service,

    and what requires 6+ months in Germany is resolved in 1 week.

     

    What is important is that people get their laptops repaired ASAP,

    not learn and comply with the German way of doing things.

1 ... 37 38 39 40 41 ... 50 Previous Next

Actions

More Like This

  • Retrieving data ...

Bookmarked By (28)

Legend

  • This solved my question - 10 points
  • This helped me - 5 points
This site contains user submitted content, comments and opinions and is for informational purposes only. Apple disclaims any and all liability for the acts, omissions and conduct of any third parties in connection with or related to your use of the site. All postings and use of the content on this site are subject to the Apple Support Communities Terms of Use.