Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Garden4Gods

macrumors member
Oct 12, 2012
47
0
This is an Intel issue with the HD4000 MRC code (Memory Resource Code) and is happening across PCs and Macs. I'm not hopeful Intel will figure this out after reading the Intel Support forum, which looks like a cluster-F attempt to obfuscate the problem. Intel released MRC1.5 claiming it was fixed, only to have users reporting the same issue with that release and later with v1.6. I'm returning my Mini when it comes in next week, and wait for the next round of Intel chips, maybe they'll get the GPU right in Haswell, but I'm not keeping my fingers crossed. This is why Apple needed a discrete GPU solution for the Mini, and instead of trying to save a few bucks and depending on Intel to finally get it right.
http://communities.intel.com/message/171244

That settles it for me. I won't be buying a Mini.

It's either an iMac, a windows based HTPC, a Mini PC or a laptop now for me.

I don't know what to do.
 

Orcelo

macrumors newbie
Nov 8, 2012
3
2
That settles it for me. I won't be buying a Mini.

It's either an iMac, a windows based HTPC, a Mini PC or a laptop now for me.

I don't know what to do.


Lol

After browsing through some of your posts it seems you've been quite negative on the mini from the get go. Posting in the majority of the mac mini problem threads.

Don't need to down other people who are interested in the mini, bro.
 

DitteVilladsen

macrumors member
Nov 1, 2012
43
0
This is an Intel issue with the HD4000 MRC code (Memory Resource Code) and is happening across PCs and Macs. I'm not hopeful Intel will figure this out after reading the Intel Support forum, which looks like a cluster-F attempt to obfuscate the problem. Intel released MRC1.5 claiming it was fixed, only to have users reporting the same issue with that release and later with v1.6. I'm returning my Mini when it comes in next week, and wait for the next round of Intel chips, maybe they'll get the GPU right in Haswell, but I'm not keeping my fingers crossed. This is why Apple needed a discrete GPU solution for the Mini, and instead of trying to save a few bucks and depending on Intel to finally get it right.
http://communities.intel.com/message/171244

I'm not having this problem of flickering.

----------

Intel HD Graphics 4000:

Chipset Model: Intel HD Graphics 4000
Type: GPU
Bus: Built-In
VRAM (Total): 768 MB
Vendor: Intel (0x8086)
Device ID: 0x0166
Revision ID: 0x0009
 

noteple

macrumors 68000
Aug 30, 2011
1,505
523
More problems for the Mac Mini 2012:

3rd issue (OSX makes harddisks dead by default)
https://forums.macrumors.com/showthread.php?p=16252659#post16252659

http://joernhees.de/blog/2011/09/16/mac-os-x-harddisk-high-load-cycle-counts/

&

4rd issue (Mac Mini cpu gets 103c (217f) and become simpy unstable!)
https://forums.macrumors.com/showthread.php?p=16252549&posted=1#post16252549

If Apple just make the Mac Mini 0.5 Inch larger it was all oke. But no, they had to make it thin. :mad:

Even the HDD problem shows how incompetent they are at Apple.

HDMI stinks
Disk stinks
CPU stinks
And you think their all incompetent.

So other than that Mrs Lincoln, how did you like the play?
 

Darby67

macrumors 6502
No, just mad. Ofcourse i am mad. The options of Macs are limited and if they have flaws i am certainly not happy.

What would you be mad about? You stated that "OSX makes harddisks dead by default" with links that show no such thing. You also stated, "Mac Mini cpu gets 103c (217f) and become simpy [sic] unstable!" with a link that again show no such thing.

While there may or may not be something deeper here, certainly nothing to be angry about and make conclusions about as no flaws have been concluded.
 

Snowcake

macrumors regular
May 18, 2010
187
0
What would you be mad about? You stated that "OSX makes harddisks dead by default" with links that show no such thing. You also stated, "Mac Mini cpu gets 103c (217f) and become simpy [sic] unstable!" with a link that again show no such thing.

While there may or may not be something deeper here, certainly nothing to be angry about and make conclusions about as no flaws have been concluded.

I am "mad", and disappointed. Steve, for example, could get mad too if something wasn't right. Wel these flaws are not right. I need a Mac!

The downside of parking the heads is that HDs are usually not designed to do this every few seconds. Typical limits range from 300,000 to 600,000 (link) load cycle counts.

After he fixed the problem:

Now, after four months since discovering the problem went by, my load cycle count only increased by about 500 (which is about the number of times i sent the mac to standby). Isn’t that a figure compared to the 36500 within the first month?

Roughly said, on a standard/non-fixed Mac Mini, 1.5 year later, the limit of 600,000 load cycle has already been exceeded and is more likely to fail. Especially after 3 years...

Apple is doing something what shouldn't be done. Link about it below:
http://joernhees.de/blog/2011/09/16/mac-os-x-harddisk-high-load-cycle-counts/


I think you are blind. The CPU where i was referring to showed errors in CPUTest when the CPU was on high temperatures. And you are right, not at 103C but at 99c-99c. Proof below.



Screen Shot 2012-11-07 at 6.32.31 PM.png
 
Last edited:

Heynonny

macrumors member
Jan 4, 2012
30
4
I'm feeling very much left out since I'm not having any problems, with the HDMI or otherwise. Should I hook up an oscilloscope or some other test equipment to try to get with the program? What does white clipping look like, to the eye?

A week or so I ago I got some snow and black screens for a few seconds as I was switching my HDMI input to the monitor back and forth between the mini and some other devices. Should I be getting all upset about HDMI handshaking glitches every once in awhile?

I am a little pissed about external DVD drives not working right with remote management, but I now understand the issue of DRM with HDMI and I'm going to work around it.

DAMN these intolerable problems!!!!!!!

I'm RETURNING this POS, yelling and screaming all the way.

Well, I tried but I just can't work up a head of steam. Sorry.
 

dazey

macrumors 6502
Dec 9, 2005
327
55
Mine is running fine. 2.3 i7 with an ssd and hard disk (fitted with owc kit) and upgraded with 10gb ram. No problems with displays (although other than testing, it is run mainly headless). I did need to trick it into thinking it has a display attached though! I do see the high reported temps but the unit remains cool to the touch and I have had no stability issues under load tests (CPU test)
 

Snowcake

macrumors regular
May 18, 2010
187
0
Mine is running fine. 2.3 i7 with an ssd and hard disk (fitted with owc kit) and upgraded with 10gb ram. No problems with displays (although other than testing, it is run mainly headless). I did need to trick it into thinking it has a display attached though! I do see the high reported temps but the unit remains cool to the touch and I have had no stability issues under load tests (CPU test)

10GB? ram? :rolleyes:
 

calvol

macrumors 6502a
Feb 3, 2011
995
4
Got a 2.6/4/1 Mini today, BTO Nov-week-1, trying to reproduce HD4000 issues. So far:

- HDMI->HDMI ok with Samsung 40" LCD TV, played 2 hours of A-team, color and detail appear normal, no flicker.

- miniDP->HDMI ok to same Samsung TV, no problem as above (use this setup with a 2010 MBA).

- HDMI->DVI with Dell 21.5" monitor (U2211H) at 1920x1080, no issues for first 30 minutes. As I typed this message, got a 2-second blank, dark screen, then back to normal. This happened as I was concurrently running memtest.

Going to test DVI with my other Dell monitor later this evening. (Dell 2412M)

Hard drive is not making noise reported in other threads.

Will add 16GB RAM upgrade and keep testing this weekend.
 

knight1

macrumors newbie
Nov 6, 2012
9
0
London,UK
I've posted on the Apple forums and I will be returning the Mini for a refund. I've looked into it further and my conclusion is it could be a hardware related issue which Intel is unable to fix. They are using delay tactics until the next Processor upgrade.
They have released three fixes so far, MRC 1.4.0.0, MRC 1.5.0.0, and MRC 1.6.0.0, all of which they claim to fix the issue, but users using Windows and Linux based PCs are still reporting problems with HDMI connection after applying an updated BIOS from their motherboard manufacturer.
But if it turns out to be hardware related, no amount of firmware/bios/driver patches will fix the underline hardware problem.

It's a shame though, I really like the Mini when its working properly, it's perfect for my needs.

On a side note, did anyone else notice how little they showed the new Mac Mini at the launch event, compared to the imac and ipad. Could apple have known about the issue before release and were nervous?
 
Last edited:

limesmoothie

macrumors 6502a
Apr 20, 2009
917
697
Edinburgh, Scotland
On a side note, did anyone else notice how little they showed the new Mac Mini at the launch event, compared to the imac and ipad. Could apple have known about the issue before release and were nervous?

No, its because the Mini was a minor spec bump, whereas the iMac was a reasonably big redesign and the iPad Mini was a big deal. Also, both those products are bigger sellers than the Mac Mini by quite some distance.
 

dlewis23

macrumors 65816
Oct 23, 2007
1,149
1,827
This is awful from Apple.

I said it before, but God knows what problems the iMacs could have.

Time to start looking at PC's I think :(

I have no issues at all with my 2012 Mac mini. I upgraded the hard drive to a SSD and added 16GB of ram. The thing has been on since the day I got it and has been perfect.
 

koyeung

macrumors newbie
Original poster
Nov 5, 2012
17
2
Unfortunately, unlike mba or mbp, most people rely on external display on mini (other may use remote desktop).

This revision removes discrete graphics and make proportion of using intel HD + HDMI becoming higher.

It may be same issue on mba/mbp using intel graphics+hdmi (via mini-displayport). (see https://discussions.apple.com/message/20247181#20247181 ) Perhaps Apple has investigated on that too and didn't think it would occur on mac mini when they released mini few months later.

On the other hand, there are reports in this forum that no issue on using HDMI on the Mac mini-late2012, i.e. they could see all gray levels on page: http://www.photofriday.com/calibrate.php . That may imply replacing the unit may be helpful.

I've posted on the Apple forums and I will be returning the Mini for a refund. I've looked into it further and my conclusion is it could be a hardware related issue which Intel is unable to fix. They are using delay tactics until the next Processor upgrade.
They have released three fixes so far, MRC 1.4.0.0, MRC 1.5.0.0, and MRC 1.6.0.0, all of which they claim to fix the issue, but users using Windows and Linux based PCs are still reporting problems with HDMI connection after applying an updated BIOS from their motherboard manufacturer.
But if it turns out to be hardware related, no amount of firmware/bios/driver patches will fix the underline hardware problem.

It's a shame though, I really like the Mini when its working properly, it's perfect for my needs.

On a side note, did anyone else notice how little they showed the new Mac Mini at the launch event, compared to the imac and ipad. Could apple have known about the issue before release and were nervous?
 

philipma1957

macrumors 603
Apr 13, 2010
6,367
251
Howell, New Jersey
Unfortunately, unlike mba or mbp, most people rely on external display on mini (other may use remote desktop).

This revision removes discrete graphics and make proportion of using intel HD + HDMI becoming higher.

It may be same issue on mba/mbp using intel graphics+hdmi (via mini-displayport). (see https://discussions.apple.com/message/20247181#20247181 ) Perhaps Apple has investigated on that too and didn't think it would occur on mac mini when they released mini few months later.

On the other hand, there are reports in this forum that no issue on using HDMI on the Mac mini-late2012, i.e. they could see all gray levels on page: http://www.photofriday.com/calibrate.php . That may imply replacing the unit may be helpful.

my quad core is going to be rma'd.

1} black screens

2} usb3 beach balls often

3} failed cpu test with oem ram and aftermarket ram

4} failed prime95 test with both 4gb and 8gb ram

5} lacie t-bolt stutters in raid0 and raid1 but not jbod
 

Snowcake

macrumors regular
May 18, 2010
187
0
my quad core is going to be rma'd.

1} black screens

2} usb3 beach balls often

3} failed cpu test with oem ram and aftermarket ram

4} failed prime95 test with both 4gb and 8gb ram

5} lacie t-bolt stutters in raid0 and raid1 but not jbod


You used the text-based prime95 Max osx version?
You said in another thread that it had a possible fail in prime95 test.
Why possible? A error is a error.

Which program you were using to test your ram with and how much passes and how much time did it take?
 

philipma1957

macrumors 603
Apr 13, 2010
6,367
251
Howell, New Jersey
You used the text-based prime95 Max osx version?
You said in another thread that it had a possible fail in prime95 test.
Why possible? A error is a error.

Which program you were using to test your ram with and how much passes and how much time did it take?

you are correct it is an error I listed as possible because i could not burn the cd for memtest86 +. I was only able to run a 1 hour apple test and a 15 pass rember test. So i knew i could not be certain if the cpu or ram was bad. Since then I put in 3 sets of 1600 ram 2 oem sets of 2x 2gb sticks of ram and the after market set of 2x 4gb sticks of ram.


all failed the aftermarket ram failed on 3 of 8 cores in under 15 minutes with cooling in place ie under 180f . the two different sets of oem mac mini ram failed on 4 of 8 cores in under 10 minutes at temps under 180f.


my 2010 mac mini passes this test with flying colors

my 2011 mac mini passes it

as does my 2 core 2012 mac mini

my 2 core 2012 mac mini passes with all 3 sets of 1600 ram.

so the sticks are good.
my 2012 quad core has an issue
maybe the mobo
maybe the ram slots
maybe the cpu/gpu

at this point it goes to the local apple store to be looked at.

I do not want it back so we will see what they tell me.

Well at least I only did one upgrade ram and it is back to stock.

glad i did the testing before I put in a ssd.


this is one reason I no longer like to put in ssds right away.

in defense of mac minis this is my third or fourth out of the box bad mini since 2006.

over 450 upgraded and 4 bad right away is very good stat.

I hope these quads are not lemons due to heating issues. I hope I just got a bad one. here are my 2010 core 2 duo marks after 7 hours in a 77 f room. all good and it runs cool since it is a low watt c 2 d processor. the 2.4 c2d cpu uses 15 watts while the 2.3 i7 quad uses 35 to 37 watts.

this is a lot more heat to dump. personally my gut tells me if you own a quad it is going to want a lot of space to keep cool and consider 3 year apple care as a must.
 

Attachments

  • Screen shot 2012-11-10 at 8.53.19 PM.png
    Screen shot 2012-11-10 at 8.53.19 PM.png
    422.1 KB · Views: 93
Last edited:

MrXiro

macrumors 68040
Nov 2, 2007
3,850
599
Los Angeles
two issues:
1. snowy screens, black screen
2. poor colour on HDMI display

Given macbook air (using HD graphics) still have the colour problem (see thread:
https://discussions.apple.com/message/18825794#18825794
(first post: Jul 4, 2012) ) and no official words from Apple, I don't think they are able to fix similar issue for MacMini.

I got a "black screen" a few times... it's like the computer goes into sleep mode.

But I don't see any color issue going back and forth from a MDP>HDMI VS HDMI built in...

----------

Both of my Mini's were perfect out of the box. One was purchased last week, one this week. Both are connected to LCD TVs via HDMI, both work perfectly with the displays. Reading through these forums gives people the impression rhat every device is inherently flawed. While statistically some will be, it's not a general rule.

Yeah... I'm still on day 4 with this Mac Mini... but I haven't really had much of an issue.
 

Snowcake

macrumors regular
May 18, 2010
187
0
you are correct it is an error I listed as possible because i could not burn the cd for memtest86 +. I was only able to run a 1 hour apple test and a 15 pass rember test. So i knew i could not be certain if the cpu or ram was bad. Since then I put in 3 sets of 1600 ram 2 oem sets of 2x 2gb sticks of ram and the after market set of 2x 4gb sticks of ram.


all failed the aftermarket ram failed on 3 of 8 cores in under 15 minutes with cooling in place ie under 180f . the two different sets of oem mac mini ram failed on 4 of 8 cores in under 10 minutes at temps under 180f.


my 2010 mac mini passes this test with flying colors

my 2011 mac mini passes it

as does my 2 core 2012 mac mini

my 2 core 2012 mac mini passes with all 3 sets of 1600 ram.

so the sticks are good.
my 2012 quad core has an issue
maybe the mobo
maybe the ram slots
maybe the cpu/gpu

at this point it goes to the local apple store to be looked at.

I do not want it back so we will see what they tell me.

Well at least I only did one upgrade ram and it is back to stock.

glad i did the testing before I put in a ssd.


this is one reason I no longer like to put in ssds right away.

in defense of mac minis this is my third or fourth out of the box bad mini since 2006.

over 450 upgraded and 4 bad right away is very good stat.

I hope these quads are not lemons due to heating issues. I hope I just got a bad one. here are my 2010 core 2 duo marks after 7 hours in a 77 f room. all good and it runs cool since it is a low watt c 2 d processor. the 2.4 c2d cpu uses 15 watts while the 2.3 i7 quad uses 35 to 37 watts.

this is a lot more heat to dump. personally my gut tells me if you own a quad it is going to want a lot of space to keep cool and consider 3 year apple care as a must.


I don't see errors in your printscreen on prime95. (I don't know if the osx version is the same as windows, but below here i quote some website)

Before You Start with prime95, Enable Round Off Checking

However, before you do anything with Prime95, you should enable Advanced -> Round Off Checking (which is shown in the picture below). This makes Prime95 check for errors after each round, so you’ll be notified if there’s an error.

enable-prime95-round-off-checking.png



Once you do that, you can get to the torture test screen that you see below by going to Options -> Torture Test.

prime95-torture-test-screen.png



The Small FFT test is for testing the processor only, but I’d recommend just using Intel Burn Test for that, and the Blend test will test some of everything (cpu, ram, and chipset).

However, the Custom test will let you manually enter whatever settings you want. If you know what you’re doing, the custom test can help you to zero in on a specific part of the system, so you’ll be able to fail faster if there’s a problem with the component you’re currently overclocking.

If you’re wanting to test fewer than all of your cores (or threads), just change the Number of torture test threads to run box (shown in the middle of the picture above).


Quickly Testing the Overall System

The custom torture test settings (that you see in the picture above) work really well for quickly testing how the overall system is working by sending large amounts of data between the cpu and ram.

The blend test usually works really well for thoroughly testing the overall system.

I’d also recommend customizing this test to use 90% of your unused memory. You can do this by first clicking blend and then clicking custom (which will let you modify the normal blend settings). You can see the amount of memory used in activity monitor.

In this case 4358 MB is available, so we’d want to run the custom blend test with 4358 * 0.90 = 3922 MB of ram.

The amount of time you should actually run this test differs depending on who you ask. Some recommend a few hours is enough, and others say 24 hours is required to ensure stability under heavy loads (like 24/7 folding). However, most of the stable overclocking clubs require 12 hours to prove stability, so that seems like a happy median. I personally just run it overnight for about 8 hours, but run it for as long as you need to feel stable.

The source for the quotes above: http://www.masterslair.com/ram-cpu-stress-testing-software-tools-prime95-intel-burn-test/

You can also put Memtest86+ on a usb-stick.
You can also boot from ubuntu an use the option: Memtest86+
The Ubuntu image can also run from a usb-stick. Link below:
http://www.ubuntu.com/download/help/create-a-usb-stick-on-mac-osx

If you can find the memtest86+ option during live boot. Read the link below:
http://ubuntuforums.org/showthread.php?t=1542941

But the question is if it is useful to still test memory. Since you also tested other memory. But if you are unsure you could always test it again. (I would then choose to test the 2x2gb because it don't take long testing in comparison to 2x4gb.
 
Last edited:

calvol

macrumors 6502a
Feb 3, 2011
995
4
I've posted on the Apple forums and I will be returning the Mini for a refund. I've looked into it further and my conclusion is it could be a hardware related issue which Intel is unable to fix. They are using delay tactics until the next Processor upgrade.
They have released three fixes so far, MRC 1.4.0.0, MRC 1.5.0.0, and MRC 1.6.0.0, all of which they claim to fix the issue, but users using Windows and Linux based PCs are still reporting problems with HDMI connection after applying an updated BIOS from their motherboard manufacturer.
But if it turns out to be hardware related, no amount of firmware/bios/driver patches will fix the underline hardware problem.

It's a shame though, I really like the Mini when its working properly, it's perfect for my needs.

On a side note, did anyone else notice how little they showed the new Mac Mini at the launch event, compared to the imac and ipad. Could apple have known about the issue before release and were nervous?

+1... I think Apple had to know about this issue at the launch event because it was on the Intel forum 5 months ago, and before that Intel probably saw it in internal testing. Not much doubt Apple shipped the Mini knowing the bug is random and unusual, hoping the user base would not be affected too much. If it's a hardware issue, it will never be fixed, because Intel will not spin new silicon for Ivy this late in the cycle when Haswell is due in a few months. It's looking more like that's the case since several MRC code releases have not fixed the issue.

My Mini is going back Monday for full refund. I'm 1 for 3 with Macs now. My 2010 13 MBA is a keeper, but had a 2011 MacBook die in the first week due to bad logic board and took it back for a refund. Hit or miss, buyer beware!
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.