I don't know, but I don't think so, as this is probably related to NVIDIA's graphics cards that use CUDA.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple Seeds Build 12E30 of OS X Beta 10.8.4 to Developers
- Thread starter MacRumors
- Start date
- Sort by reaction score
I don't know, but I don't think so, as this is probably related to NVIDIA's graphics cards that use CUDA.
Probably it hasn't been a priority to implement it, for whatever reason. It is implemented in the Windows drivers, however, see for example here:
http://www.geeks3d.com/20120427/intel-ivy-bridge-hd-graphics-4000-gpu-opengl-and-opencl-tests/3/ .
When I compile and run this example http://stackoverflow.com/questions/7892955/how-can-i-test-for-opencl-compability
, I get
Device Intel(R) Core(TM) i7-3615QM CPU @ 2.30GHz supports OpenCL 1.2 .
As I understand it, if the _GPU_ supported OpenCL on MacOS, there would be a second line. The device that is shown to be OpenCL 1.2 compatible here is the _CPU_ device. This could explain the screenshot posted by MrNomNom. I'm guessing here, but I'm quite sure I'm right ;-)
Does the damn thing shut down in under a day now? If they do fix this I'm not sure what I'm going to do; I've gotten used to just taking my shower while OSX reboots into Windows. 
Does the damn thing shut down in under a day now? If they do fix this I'm not sure what I'm going to do; I've gotten used to just taking my shower while OSX reboots into Windows.
I still get com.apple.coreservices.appleevents timeout but is it really that big of an issue for you? I understand it's a problem and getting it fixed would indeed be nice but all you have to do is change the timeout durration to something that you find reasonable and boom, you have a 5 second or less shutdown. If there wasn't such a simple user fix for the slow shutdown it might be a little more serious of a problem.
Last edited:
I saw this on another site regarding slow shutdowns. This seems to help some there maybe worth a try.
Anyway, I found a trick that helped me:
Shut down with Reopen windows when logging back in CHECKED.
Turn the computer back on.
Shut down with Reopen windows when logging back in UNCHECKED.
This returned the Mac to an almost instantaneous shutdown.
Anyway, I found a trick that helped me:
Shut down with Reopen windows when logging back in CHECKED.
Turn the computer back on.
Shut down with Reopen windows when logging back in UNCHECKED.
This returned the Mac to an almost instantaneous shutdown.
I saw this on another site regarding slow shutdowns. This seems to help some there maybe worth a try.
Anyway, I found a trick that helped me:
Shut down with Reopen windows when logging back in CHECKED.
Turn the computer back on.
Shut down with Reopen windows when logging back in UNCHECKED.
This returned the Mac to an almost instantaneous shutdown.
Tried on my 2011 MBP (i7/16GB/SSD) and no real noticeable difference for me...but glad this worked for you. Ever since upgrading to ML (clean install) system takes ~20 seconds to shutdown whereas on Lion it was very quick (~5 seconds).
I was at the genius bar last week to investigate TB display startup issue and they noted shutdown seemed slow but not unexpected with spinning media. Once I informed them it was SSD they agreed it was a problem. FWIW - they admitted they had not yet upgraded to ML given some of the issues they are still seeing.
Nothing to lose sleep over but would eventually like to see it fixed
If there was a way to take out and install the updated nvidia drivers from the ML 10.8.4 builds and force install them on Lion, then that would be a great solution for now! But I have yet to see that working? though I haven't actually searched for the drivers myself yet, as I just run the latest build.
But would it be possible?
That is the most likely explanation - I have a feeling that OpenCL sitting on top of AVX is probably faster than it sitting on top of HD4000. Maybe that is the reason why OpenCL 1.2 GPU support hasn't been added or maybe it is something that requires more testing hence going to be held off till 10.9.
But it doesn't explain the snapshot you posted which clearly showed Intel HD4000 listed as a OpenCL 1.2 capable device in December 2012? I have OpenGL extension viewer 4.0 on my rMBP and I can't have the same snapshot generated by it...
Good point - I wonder why, Apple dedicating the whole GPU to the display given the amount of work that is required for it to do?
Does the damn thing shut down in under a day now? If they do fix this I'm not sure what I'm going to do; I've gotten used to just taking my shower while OSX reboots into Windows.
How will you know - or haven't you tried the alleged fix provided by razmarino yet?
It worked for me so far, but now of course I won't see any difference with 10.8.4 - but I doubt Apple will focus on that after all these month...
I use the fix (and actually posted it a long time ago in another thread).
But you guys are missing the point.
It's the fact this "blatant" in-your-face bug has seemingly escaped Apple's comprehension that bothers me. The coders are either incompetent or saying this bug shouldn't matter. I can't accept either. It's disheartening to see OSX relegated to the basement.
At this point, I fear it's here to stay. No evidence to indicate otherwise.
I actually think it will be a part of 10.9 - it surely won't be rewritten completely (and as the focus seems to lie on user interface related stuff). And as we have learned the bug wasn't removed so far, so I think it will be passed on as it seems to be some more complicated than we probably know so far.
Indeed. Simple shut down or restart should be faster.
That's just userspace. Whole thing takes about a minute on an SSD latest iMac. PATHETIC.
Can you check something for me in 10.8.4? What is the webkit version now? Check: http://www.mybrowserinfo.com/ Thanks a lot
Can you check something for me in 10.8.4? What is the webkit version now? Check: http://www.mybrowserinfo.com/ Thanks a lot
Your Browser User Agent String is
Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_4) AppleWebKit/536.28.10 (KHTML, like Gecko) Version/6.0.3 Safari/536.28.10
Why don't you just use the webkit nightly if you want updates? I find it for the most part way faster in every day browsing than safari (at least in 10.8.2 - 10.8.3, dont really use safari anymore).
Downside is that for a few hours sometimes you can run into a build that is a bit buggy.
But my point is, if you want your safari constantly updated (literally every couple of hours) then just use the webkit nightly.
just updated to 10.8.3. another update to 10.8.4 soon? dont feel like to download another 500MB for another update....
dont feel like to download another 500MB for another update....just updated to 10.8.3. another update to 10.8.4 soon?
If 500mb ruins your day you should probably just unplug yourself from the Internet. I'm sorry but that is just silly. complaining about a tiny file size for an update you probably shouldn't have access to yet anyway.
If you don't like it, sit back and wait for your app store to pop-up with an update for your OS X.
anyway I expect weekly or fortnightly updates for 10.8.4 as well just like 10.8.3.
dont get me wrong, I love to use the latest one with some good reason. It's just the area that I am living now has much slower Internet access. no choice.
just updated to 10.8.3. another update to 10.8.4 soon?
If you use delta updaters, you're only downloading what has changed since the last update. So the size is relative to how much they have changed.
We also don't know how soon it's going to be, and when it is released you can wait on updating or skip an update (although it's possible that .4 is the last one before 10.9...also possible that it's not).