Ram is Samsung 1333 (DDR3).
We got 21 servers with this motherboard, all got the same problem. It stuck at boot, it passes the Intel Management screen and after that we got a black screen. No option, we must shutdown the server manually and switch it on again.
We have no time to see the code in the down right corner of the screen.
I have similar problem o this board. But in my case, it is Windows 7. The compuer often gets stuck after POST (The monitor shows "0_" code at the lower right corner) or right after starting to load OS (The monitor shows Starting Windows, but Windows flag animation is not shown yet), after a soft reboot. . I have to power cycle the pc to get it to work. I've tried WD Blue and Seagate 7200.9. The memory is 2 x 1GB Kingston Value DDR3-1333. I found that the board had been resealed by either the vendor or Intel because there are double yellow stickers on the package bag. I figured the board might be a defective returned item, but passed the re-examination. Now I am waiting for a replacement and see how it goes.
Thanks for the update, now I can see that I'm not the only one.
All our motherboard are brand new, didn't noticed any yellow sticker. Strange thing all our motherboard got the same behavior, I don't know if they are from the same production serie. It can be a serie problem I don't know.
Sure thing, it can't be an OS problem now.
Is there someone else with this problem ?
I got the replacement board. It is a big disappointment. The board has also been opened before. This time, the computer would also stuck at the "0" screen after power on. I have to press the reset button to let the computer restart and boot into OS. I guess I would not take chance for another replacement. Hopefully the vendor can accept the RMA of this replacement and let me order Gigabyte H55 ITX.
brice_capelle, any progress at your side?
No progress on my side. All our servers (25) with this motherboard are in production state so we can't remove them from service or do more tests. We are waiting another group of 25 motherboards like this one for our servers. We can't break the deal
We'll do more tests on this but I'm pretty sure there is a BIOS bug. I hope Intel is aware of the problem.
Yesterday, I brought the pc to the work for test this morning. It booted fine. No single failure. What I did were that I tried another pair of 2 x 2G (instead of 2 x 1G) memory, and the board was connected to a different monitor, which has single link DVI. The one at home is with dual-link DVI. Another difference was at work, the board was connected to a standard mouse, but at home, it was connected to a wacom tablet simulated as a mouse. I understand this intel board supports single-link DVI only. I thought all I got from using dual-link was 1280x800 resolution max. The htpc will go with TV eventually. I am not sure if this different types of DVI really affect the booting process. After returning home, I tried the dual-link DVI again. But booting is mostly fine. It only failed once in warm reboot. No failure when it was connected to the TV. I guess probably you used the same monitor to test all the boards since they were used as servers. Could that be the reason? Or mouse? So I will keep the Intel for now, and I will put the previous mb back to do further verification.
I've finished testing. The observation, in my case, is that both dual-link dvi and wacom tablet may contribute to the random freeze duing booting process. It seems that once the board has been connected to a single-link dvi monitor, then the chance of freeze is greatly reduced regardless if it is dual-link or single-link in the future.
The board may still occasionally freeze duing boot up (before entering OS) because of the wacom tablet. This can be verified by when the booting is freeze, unplugging the tablet resumes the boot process. Enabling USB backward compatability in BIOS may solve this issue (I did not test it though since the frequency of freeze is already very low and the htpc is primarily used with tv and remote).
After I installed the htpc in the home theature, there is no failure during booting up.
It's really strange that the DVI port cause problem. But our testing bench is old, I mean the hardware like the monitor is an old CRT monitor and we must use an adaptator.
I'll have the next motherboards in 2 weeks, I'll can test that on a real new monitor. I'll do the same for the Keyboard.
Thank you for the info, I'll keep you informed of my tests
It is great that the cause of the issue was found. How will you solve it? Fast hard drive may be essential for the server. Has Intel be notified about this? The next thing I can think of is to use SSD. But if the application require huge number of writes, that could be a problem of reliability.
It looks like the board is indeed "targetted" to the htpc application. The board does not like stuff that shouldn't be in that application. So far, after I keep the "advanced" peripherals away from the board, no single boot failure occured. I also tested, sleep to S3, wakeup from S3 by UIRT and WOL from S5/S3. All are working great.
After new tests on our last servers we found out the problem.
In fact the problem was the RAM modules. To solve it we had to change to Samsung RAM Modules.
Older ones were PQI DDR3-1333U 2GB and Apacer 4GB UNB PC3-10600 CL9 .
If someone knows how to solve this or knows if it is the BIOS which got a problem with those specific RAM modules, I'll be glad to hear it !