I just want to know if Intel is working on problems around their Intel HD Graphics 2xxx, 3xxx, 4xxx (if any).
I have a Mac mini ( I know. You will say I need to contact Apple. I did. I did report problems to their bug reporting page at first. Because I thought it was their S/W issue or framework issue. However, now I have a rigid idea it's at least device driver or Intel's CPU issue. - HD Graphics xxxx is an integrated module in their Core i* processors. )
Let me first briefly talk about the symptom I see on my Mac mini with HD graphics 3000 on Mac OS X and Windows 7 x64.
Firstly, on Mac OS X, when I handle video, the system crashes and hangs with regular rectangular mosaic white dots arranged on a screen.
<Fig. 1> Crash : iMovie
<Fig.2> Flash video
It happens with
- flash video on a browser ( Adobe's own code? )
- sliding "time" while editing video with iMovie ( uses QuickTime or their AV Foundation )
- playing video with VLC ( uses ffmpeg )
- H.264 video on YouTube
So, whatever S/W programs are used, it shows the same problem.
It suggests that the problem is not in any specific library functions or framework methods/functions.
It says, at least, the problem is in a device driver level or chipset itself. I'm pretty sure that they will call similar or same op-code when it crashes.
The monitor is connected through an HDMI port.
No, let's talk about Windows 7 x64.
I installed the latest driver for the graphics module. ( about a 2 months ago? )
The symptom is a little different but I think it's the same in their nature.
The top portion of screen becomes like "blurred" image of whatever existed on the screen.
About 3/4 portion after the top portion becomes blank. Then Windows crashes and goes to BSOD or reboots.
It looks very similar to the screenshot embedded in this thread.
It's very interesting, because the person who reported that had Intel HD Graphics 2000 chip.
In this thread, some people say that it's due to device driver. But some other person said that it's grounding problem.
Some said that older device driver worked but new one didn't.
I think those problems are related one another and it has been existed since HD Graphics 2000.
You can find posts related to the problem I described so far easily on this site. So, it doesn't look to me that it's only happening on Apple's H/W.
Apple may not have soldered the CPU ( ground leg ) properly for Mac mini ( which is what I have ). However, other PC vendors do the same mistake all together? It sounds odd. If it is grounding issue, somehow all H/W engineers working for different companies do the same make at the same time. Then probably Intel's chip documentation didn't explain about it properly or misguided? Or is it chip design issue?
I don't know. However, I think Intel should be involved in and solve this problem.
Crash ratio? If I play 40 min of video, ( especially on-line video ) it surely crashes at least once.
I think Intel should definitely work on this.