Well I'm back in business for now, which is great. Everything is a lot smoother than I had been used to. Also this machine is "a lot" quieter than the last one which was replaced. The fan on that machine was noticeably louder than any of the other 3 nMp's I've used - don't know if that means anything but just observing. Managed to get a few hours work in so far with Final Cut Pro X without any issues - only time will tell - just hope that this issue can be solved generally rather than fighting isolated cases10.10.4 gpurestart crash. Hulu with flash, fullscreen. nope, 10.10.4 did not solve it for me. Granted, the time between crashes are further apart, but still.. The pattern for me is video,flash,fullscreen. My flash is on lastest update. This newmacpro is the most unreliable computer ever and the cliche that apple is style over substance is gonna stick if they dont sort this out soon.
My exchange with Apple over my bugreport on nMP suspend=>panic issue seems to have come to a standstill; for last two months (dunno what really triggered it, it was post 10.10.3, but not by much; no other major software updates) my nMP started panicking more and more frequently when resuming from long suspend. I provided them with the panic logs, but they wanted coredumps, which seem insanely slow to produce + I am not sure I want to leak e.g. encryption keys and whatever may be in the memory at the time anyway. So I guess next time, morale of the story is simply not to buy Apple but build hackintosh.. sigh.
I am pretty sure this is a bug in the ATI driver, triggered by combination of HDMI projector + 2 very different 4k displays; especially the Dell 4k one has actually some bugs and does not always recognize signal _at all_ post-resume and requires power cycling. Given the stacktrace, I would have imagined fixing it would be possible, but apparently not.
Well I'm back in business for now, which is great. Everything is a lot smoother than I had been used to. Also this machine is "a lot" quieter than the last one which was replaced. The fan on that machine was noticeably louder than any of the other 3 nMp's I've used - don't know if that means anything but just observing. Managed to get a few hours work in so far with Final Cut Pro X without any issues - only time will tell - just hope that this issue can be solved generally rather than fighting isolated cases
No problems so far - running smooth. The real test will come when I start rendering 3D animation in the next couple of weeks. I did have a 17" Macbook Pro die 2 days after this one arrived though!Hows the newnewmacpro so far. Is it still behaving badly?
No problems so far - running smooth. The real test will come when I start rendering 3D animation in the next couple of weeks. I did have a 17" Macbook Pro die 2 days after this one arrived though!
No flash - I don't even use this machine for internet at all. Just one monitor connected by Apple Thunderbolt to DVI-D adapter - same configuration as with 2 previous machines. In the middle of a massive job right now so not pushing anything too hard. Curiously I think it was flash that killed my Macbook Pro - I don't install it anymore!Do u have flash installed? monitors?
One more interesting thing to add to the list:
I had one more hang (everything frozen, mouse cursor still alive) recently. Out of pure curiosity i hooked up my old monitor via hdmi. You would not believe it: The Mac did not crash at all. It was still running via hdmi, showing the frozen desktop on the TB monitor. Although there were some glitches in the picture, it was still usable as if nothing happened.
I am more and more conviced that this is purely a driver problem that only affects TB.
Oh yes please do! I wonder whether you get a usable but partly flickering hdmi-picture as well. Could be caused by wrong timings do to "hotplugging" - interfering with settings stored in the PRAM or the crashed TB.Have a good HDMI display arriving tomorrow, and will see if I can get similar results.
I'm very curious about that too. And to see if I just run HDMI and get rid of any TB displays altogether (until El Cap comes out anyway), and see if we can see results there.Oh yes please do! I wonder whether you get a usable but partly flickering hdmi-picture as well. Could be caused by wrong timings do to "hotplugging" - interfering with settings stored in the PRAM or the crashed TB.
...or just run our TB display connected via hdmi *cough* - just came to my mind. They usually do have both connectors.I'm very curious about that too. And to see if I just run HDMI and get rid of any TB displays altogether (until El Cap comes out anyway), and see if we can see results there.
One more interesting thing to add to the list:
I had one more hang (everything frozen, mouse cursor still alive) recently. Out of pure curiosity i hooked up my old monitor via hdmi. You would not believe it: The Mac did not crash at all. It was still running via hdmi, showing the frozen desktop on the TB monitor. Although there were some glitches in the picture, it was still usable as if nothing happened.
I am more and more conviced that this is purely a driver problem that only affects TB.
Powercycling the monitor usually works for me. I did not notice a problem with the os before yosemite. I just hope people with this problem files a bug report or a tech news site picks this story up.
No GPU hangs since having my machine replaced so far...I filed an extensive bug-report and was told that "this is not normal", have the nMP checked at Apple and the report was closed.
...just using that Mac right now running Windows - with no problems.
I have got that request about more logs as well. Regarding the final answer, it now appears just like some kind of automated reply...I got an answer on my bug report requesting more information and more logs the next time it happens. That's almost two weeks ago, and the machine has worked perfectly since. Typical!