People have the same problems when running Windows on the 2012 Mac Mini, so it's not OS or Driver related... meaning it's either the firmware or hardware at fault.
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.
Random Black Screens
- Thread starter harvester32
- Start date
- Sort by reaction score
People have the same problems when running Windows on the 2012 Mac Mini, so it's not OS or Driver related... meaning it's either the firmware or hardware at fault.
So this is only an issue with HDMI connections, correct?
If I get a new Mini and connect it to my older 24" ACD using the mini displayport connector it shouldn't be an issue as I understand it.
If I get a new Mini and connect it to my older 24" ACD using the mini displayport connector it shouldn't be an issue as I understand it.
Wouldn't those that run windows also need to install the windows Intel drivers for it to work in windows? I don't see why windows suddenly means it could be hardware related.
----------
NO, that is NOT correct. Around 99% of people having issues are using HDMI but there are a few using TB or displayport that have reported issues too. However, your chances of having good luck with TB or displayport seems to be much greater. Many people who switched to TB or displayport said the problem went away. Why some using TB or displayport are still having issues is not known, but I wouldn't rule out human error as some are not as computer literate as others. Without being there and seeing it for yourself you really don't know. All I can tell you is that I plan to get another mini and use TB, and I'm very confident that it will work, but just know there are no absolutes from reading the reports.
Remember that the Ivy Bridge CPUs with the HD 4000 iGPU have been around a decent while and shipped by many system integrators already (it's only the Mac platform to which they are new)... one would hope that Intel already have a decent driver for Windows!
It could indeed just be new drivers need (both for Windows and OS X), but as Intel have already fixed the issue in Windows for other HD 4000 iGPU systems, I can't see why those running Windows on their Mac Mini would need a different driver to those running Windows on any other HD 4000 iGPU system.
Say what? It is to my understanding from reading the Apple and Intel site that neither Intel windows driver solved the issue. If that's the case then I'm not surprised that Mac users running windows are seeing the issue. I used to like Intel, but after reading all the issues with the HD3000 and HD4000 I have lost respect for the company and would prefer Apple not use them for graphics anymore.
Thanks. Maybe I'll just wait until they get this sorted out then. I've had a system before that would randomly get black screens and found it to be infuriating. Since Apple/Intel have acknowledged the problem I would image a fix will be coming in a relatively short time frame.
they have known about it since June. i wouldnt call that relatively short time frame
how does firmware fixes work? software fixes understandable... its just code. but firmware implies a physical alteration. how does that work via downloading a firmware fix?
Firmware is typically stored in flash memory embedded in a device's hardware. You can flash (erase) the memory and write new firmware to it.
thanks.
seems like firmware is the same as software in that its more a set of specific instructions to control a physical piece of hardware then... rather than this HD4000 having a physical design flaw that causes the problem... like a polarized capacitor being wired wrong way round or something
thanks.
seems like firmware is the same as software in that its more a set of specific instructions to control a physical piece of hardware then... rather than this HD4000 having a physical design flaw that causes the problem... like a polarized capacitor being wired wrong way round or something
Yes, exactly. Firmware is the software that the hardware requires to do anything at all (like turn on, etc.).
Since memory and microprocessors are so cheap these days, hardware is designed to be programmable (i.e., rely on firmware) as much as possible, so if there's a problem, it can be fixed by reflashing the firmware instead of replacing the hardware. Of course there is always the possibility of a hardware problem that can't be fixed with changes to the firmware but you can usually work around a lot of problems with firmware changes.
This problem with the screen going black and/or showing garbage sounds like a timing problem to me, i.e., the output of the computer gets out of sync with what the monitor expects/understands. I strongly suspect that anything to do with timing is controlled by firmware and thus can be fixed with a patch to the firmware. Notice that if you have an aftermarket motherboard in a PC, you can go into the BIOS user interface before the computer boots and change all sorts of things to do with CPU and memory timing, so I bet everything to do with graphics timing is programmable too.
I was under the impression they just recently publicly acknowledged the issue and promised a fix. But I wasn't even aware this was an issue until I started shopping for a new mini this week so you guys would have a better feel for it than I do. Who knows, it may be many months until they address it or it is entirely possible they may never fix it. Regardless, I'm glad I found out this is not just an HDMI issue before I went through the hassle of buying a new mini and then having to return because of defective graphics. I'll wait until they do fix it or the next gen mini comes out in 2013 or 2014.
Maybe for Intel PC's but for Macs the first I saw it mentioned on their forums was Oct/2012.
This is still going on with my mini over Thunderbolt.
But when running single channel ram(1 x 8gb), all is mostly fine.
Last edited:
More and more I get the impression that it might be a databus conflict or overload.
Had no problems with an Intel X25M-G2 or a Apple Toshiba SSD (different Mac mini's) but then a Samsung 830 caused reproducible lockups under Windows 7 while copying from firewire backup to Samsung SSD.
Investigated these lockups and switched to AHCI mode and can no longer reproduce the problem. Performance improved considerably under Windows 7.
Perhaps my occasional black screens will be resolved too, time will tell.
Attached picture: top result is AHCI, bottom is IDE. Under Toshiba and Intel SSD's there was hardly a difference but this Samsung has a large difference.
Attachments
If you check your Thunderbolt bus Port Status in Systems Information, does it show a device connected?
I did not read all 22 pages, but i don't believe this is HD4000-related, as i have those random black screens on my rMBP (Nov-2012) as well and those were happening when running on discrete graphics - both when using an external monitor attached via HDMI and when running solely on the internal monitor.
I thought it was a problem with OSX randomly switching graphic modes without need (due to whatever bug still hiding inside OSX), but as the mini is running on HD4000 only there might be another problem.
At least i can state the problem is not limited to the mini...
People say it's a 4000 graphics issue because the Intel HD 3000 and HD 4000 have had these issues across all platforms. 10.8.2 also has a separate HDMI issue which is what you might be experiencing.
Changed to a mini displayport to displayport adapter connected to my U2412m yesterday and over 24 hours later no random black screens...
That's what I thought initially too - however they still have appeared afterwards on rare occasions. (only a fraction of what it was before)
PS I've got the same display on my Mac mini's (2010 server and 2011 base)
When your monitor is connected by Mini-Display-Port, does your Thunderbolt bus Port Status (in Systems Information) show a device connected?
Is this what you are talking about? -
Thunderbolt Bus:
Vendor Name: Apple Inc.
Device Name: Mac mini
UID: 0x0001000D1961D3E0
Firmware Version: 14.1
Port:
Status: No device connected
Link Status: 0x101
Link Controller Firmware Version: 0.11.9
If so, negative.