Archives Discussions

[maxreason]

phenomIIx4 965 BE hypertransport must be reduced to 1600MHz with winxp64 ???

I built a new system with phenom II x4 965 BE and gigabyte motherboard GA-MA790FXST-UD5P running windows xp 64-bit edition, but the system would freeze every 1 to 20 minutes at random times (including when no task was running and nothing was happening).

I'll skip the 3 days of careful and thorough tests I ran, but it included running memtest86+ for hours from boot CD generated (zero errors), removing everything except video card from the motherboard slots, trying other memory modules, underclocking the CPU from 3.4GHz to 2.4GHz, underclocking the 1600MHz memory modules to 800MHz, and so forth.

For some odd reason the BIOS sets the northbridge/hypertransport speeds to 2000MHz instead of the 2600MHz that this CPU and chipset should support, so that should have been very conservative.  Nonetheless after I had tried everything else I downshifted via BIOS the northbridge/hypertransport speeds from 2000MHz to 1600MHz --- and the system became completely reliable, stable, solid thereafter (for 2+ days so far).

Thinking the motherboard and/or northbridge chipset must be defective (slower than specification) I ordered a new motherboard.  I figured this would surely solve the problem.  Today I got the new motherboard and found it has ***exactly*** the same problem.

My question is --- where is the bug?

I am certain AMD chip testing tests the hypertransport bus on their CPUs at the full 5200MT/s hypertransport v3.0 speeds, so I tend to believe the CPU is not defective.

That leaves two possibilities that I can think of:

#1: Something is fundamentally wrong with this motherboard.

#2: Some code in Windows XP 64-bit edition malfunctions when the northbridge/hypertransport speeds exceed 1600MHz.

To me, #1 seems quite unlikely for the following reason.  I chose this motherboard because most reviews of the Phenom II x4 965 BE CPU were performed on this motherboard --- and none of them reported any problems with flakiness in general or the northbridge/hypertransport in particular.  In fact, many of these reviews overclocked the various parts of the system to see what kind of performance they could achieve.

That leaves #2, which seems somewhat unlikely too --- but what other possibilities are there?  Am I missing some more likely possibility?  Obviously there is one other possibility:

#3: The CPU chip I have was not properly tested, or malfunctioned within the first 15 minutes of running in my system.  That seems highly unlikely to me, but it is possible.

Q1: Does anyone else run this combination of <CPU, motherboard, OS> with northbridge/hypertransport at 2000MHz to 2600MHz (or higher)?

Q2: Does anyone see any other possibilities that did not occur to me?

PS: The CPU temperature never exceeds 41C, the system temperature never exceeds 41C, the GPU temperature never exceeds 51C, and the system has lots of fans and excellent airflow (and the fans always run at full speed).  The problem is no more likely to occur with lots of CPU activity versus minimal activity.

#####  system specifications  #####

motherboard: one gigabyte GA-MA790FXT-UD5P
CPU: one 3.4GHz AMD phenom II x4 965 BE (black edition)
DRAM: four mushkin 996601 modules (4 x 2GB DDR3-1600MHz @ 7-7-6-18)
video: one evga nvidia GTX 285 with 2GB VRAM
disks: two seagate 1TB SATAII drives (ST31000528AS)
disks: one western-digital 1TB SATAII drive (WD10EADS-00M2B0)
CDDVD: one NED CD/DVD burner (ND-3540A)
power: one coolmax greenpower 950W (CUG-950B)
CPU-cooler: zalman CNPS9700
fans: eight 80mm ball-bearing fans running constantly at full speed

NOTE:  the DDR3 memory is running at 800MHz to 1333MHz @ 9-9-9-24
NOTE:  often the tests were run with only one hard disk drive connected

PS:  Reply here or send reply to <maxreason at gmail dot com>.

[avk]

Please try another AMD forum and post your problem in AMD chipsets/motherboards.

[maxreason]

Here is some new - and more confusing - information.

I purchased another phenom II x4 965 BE and tried that in both motherboards.  To my amazement, the problem persists in all four motherboard / CPU combinations!  Now what?  Is the problem with all of these motherboards?  All of these CPUs?  All of these 790FX northbridges?

Or somehow winxp64 has a race condition or timing problem that doesn't happen until northbridge reaches 2000MHz?  It is true that memtest86+ would run for hours without a freeze... but then again, it probably doesn't perform any operation that requires a single northbridge access.  No, that's not true either - to update the display requires GPU access.  Hmmm.

Any ideas?

PS:  Thanks for the idea about the other forum.  I just posted an updated version of this message on that forum.

[avk]

You see, WinXP x64 is a very rare OS, and, perhaps, not all BIOSes and/or drivers are 100% compatible with it because of this. Maybe should you post your problem at the motherboard vendor' site too?

P.S. BTW, all the current AM3-Phenom II has the NB/HT standard frequency for 2.0 GHz, not 2.6 GHz as you've mentioned.

[maxreason]

Yes, of course.  That's how I started my quest for answers, actually.  Unfortunately, they have no helpful answers so far.  I also submitted my issue to Microsoft, and answered their first round of comeback questions.  After that, no response... yet.

After many more hours of tests, I do notice that one of the CPU chips freezes less often.  I'm not sure whether that's a hint or just due to expected variations from chip to chip.  But whatever the difference, it does seem to indicate the CPU chip is somehow relevant, and maybe the cause.

[avk]

Four DIMMs... Are they two-sided? I heard that some models of CPUs and/or motherboards don't work correctly with four two-sided DIMMs. Did you try to reduce them or replace them just in case? What about WinXP x86 (I mean 32-bit) version, does it works?

[maxreason]

I had not heard about a problem with 2-sided DIMMs.  However, I do recall reading messages or articles that claimed 2 DIMMs worked reliably while 4 DIMMs did not [unless slowed down].  However, I did try running with only 2 DIMMs in the motherboard, but that didn't seem to change anything.

Are my DDR3 DIMMs 2-sided?  They certainly appear to have ICs on both sides of the PCB, and the heatsink certainly covers both sides.  You can see for yourself at:

http://www.newegg.com/Product/Product.aspx?Item=N82E16820146777&Tpk=996601

or

http://www.mushkin.com/doc/products/memory_detail.asp?id=651

I did try two DIMMs someone lent me for a few hours, and they were another brand, but I'm pretty sure they were also 2-sided.

By the way, in your mind doesn't all the evidence point to the northbridge/hypertransport bus and not memory?  I mean, if the memory has a problem, why would slowing ONLY the northbridge/hypertransport bus from 2000MHz to 1600MHz eliminate the problem?  These are not Intel CPUs that have memory controllers in the northbridge chips, these are AMD CPUs that have separate memory controller and bus in the CPU chip.  Therefore, I don't see how the evidence suggests memory problems.

[avk]

I was just trying to understand your problem .

Ok then. So what about 32-bit Windows?

[maxreason]

Yes, I understand and appreciate that.  Sorry if the tone of my text doesn't match the appreciation in my brain for your attempts to help.

I don't have any 32-bit McWindows, and I am not inclined to buy one just to test this out.  I do have 32-bit and 64-bit ubuntu linux DVDs, as well as 64-bit ubuntu computers to develop and test cross-platform code.

I am exchanging emails and phone calls with microsoft to see what they think about this problem.  No results yet, except that everyone is too busy to invest 10% as much time as I have to figure this out.  Unfortunately I haven't been able to give AMD this opportunity, because I can't figure out who to call to get this ball rolling.

[avk]

Well, there is another option you could try: to replace your graphics card with another one and/or to install different driver.

[maxreason]

Yup, that's a good idea too.  But I already replaced my GTX-285 video card with an older GT 8600 card, without seeing a change.  Just to try everything, I also found a three or four month old version of the driver and tried that too.  Again, no change.

I just ordered my third phenom II x4 965 BE CPU tonight.  This will be my last attempt to fix the problem with a new CPU.  If they have this many of these CPUs in the supply lines with underspeed hypertransport interfaces, I will wait for some future CPU model before I try again.  I don't know what else to try.

[avk]

Hmm... your case is a definitely remarkable one. Perhaps you've really managed to find a serious problem. Did you try to post somewhere on AMD Developers Blog? Maybe you'll manage to get response that way.