Issues with an amd 6000+

[Faatmole]

Hey, not sure if this is the right place to post this, or if anyone has had a similar experience // is able to identify any problems that I'm having... anyway

I purchased an amd am2 6000+ last Thursday (Upgraded from my am2 3800+) and it was running fine for the first day or two. After a while, I was getting random BSOD's, however they only occurred while running Firefox (games ran fine, had no issues with the hellgate demo). Yesterday, while playing the hellgate demo some of the sounds where stuttering, and I was getting a slight delay when performing certain actions (the primary issues of my 3800, which was under the minimum requirements for the game). Left to go to work, and when I came back I kept getting BSOD's when I tried to boot the PC. I've had a few different messages, none of which seem to occur any more often than the others. I read on another forum that this was more than likely due to an issue with ram, so I tried swapping out the sticks but it didn't change. the BSOD's where occurring up until the vista account login page was, however it usually restarted before then. I swapped out the 6000 for the 3800 (still using the heatsink//fan from the 6000) and it booted fine. put the 6000 back in, had a quick browse through the BIOS settings and didn't see anything wrong, and future attempts at booting the PC have led to it getting stuck on the screen with the bar scrolling across... I've left it for around 10-15 minutes so far and it hasn't changed. Swapped the CPU's back over, and I've now found myself posting this message on various sites to see if others may be able to help me >.>

If it helps at all, here are the specs of my PC:
-320gb Seagate barracuda, 7200rpm, SATA 2
-2x 512mb Geil ram, 800Mhz (5, 5, 5, 12)
-2x 1gb Geil ram, 800Mhz (5, 5, 5, 12)
-Antec TruePower Trio 550w PSU
-Asus M2N SLI deluxe mobo
-8800GTS (640mb) - eVGA

Would be greatly appreciated if I could get a response to this message

~Faatmole

[AMDave]

There are known bios issues with a few of the M2N-SLI-D and the M2N-E mobos. Looks like you could be one of the unlucky few.

If you contact Asus, they will probably tell you to RMA it back to the point of purchase.
If you go the RMA path make sure you don't get a 'repaired' RMA'd board like some have reported:
http://vip.asus.com/forum/view.aspx?id=20071017072834531&board_id=1&model=M2N-SLI+Deluxe&page=1&SLanguage=en-us

The BIOS reflash may or may not work. If it doesn't, a replacement chip will cost about $5-$10 and take a couple of days to arrive.

I don't have that board. I'm just reading from various google searches.

NB - Try booting it with less than 4GB RAM - just a 2GB RAM pair.
It will probably work fine (see notes at ASUSTEK website via first linky below)

eg
http://vip.asus.com/forum/topic.aspx?board_id=1&model=M2N-SLI+Deluxe&SLanguage=en-us
http://www.tomshardware.com/forum/195746-30-asus-deluxe-updating-bios
http://www.tomshardware.com/forum/195541-30-asus-start-issue
http://vip.asus.com/forum/topic.aspx?board_id=1&model=M2N-E&SLanguage=en-us
http://www.theinquirer.net/en/inquirer/news/2006/05/22/socket-am2-screws-up-the-ddr-ii-clock

[Faatmole]

I'll try swapping out my current mobo for my old one (an MSI K9N Neo-F), although I had some issues with it randomly crashing (crashed a lot more often playing warcraft 3 than it did in Oblivion) and wouldn't support the 3gb of ram I currently have...

If it can support the 6000, I guess I'll be sending my current mobo back...

[AMDave]

You have done all the right things by unit-testing the RAM and CPU.
Given some of the reports on the Flash results on boards with problems, I would not proceed in that direction.
It really does sound like the mobo you have is defective.
If it was mine I would be RMA'ing it immediately.
It is a shame that you have had this experience as it really is a good capable mobo.

[Faatmole]

Ok, "Tomorrow" was pushed back slightly, got a bit lazy with school & work...

Anyway, after hacking at my computer the entire afternoon, running the K9N Neo with the 6000+ (only running with 2x 512mb sticks of ram) I recieved the following message on startup:

******WARNING: UNKNOWN PROCESSOR REVISION******
THE PROCESSOR(S) INSTALLED IN YOUR SYSTEM ARE OF AN UNKNOWN REVISION. PLEASE CONTACT YOUR BIOS VENDOR FOR APPROPRIATE UPDATES.
FATAL ERROR... SYSTEM HALTED.

Not sure if I should bother updating the BIOS on my old motherboard and trying again, or trying to update the BIOS on the M2N... Or just sending the CPU back by itself

[AMDave]

The bios on your old mobo doesn't know how to handle the new CPU and therefore shuts itself down so no damage is done.

The new CPU is most likely to be fine.

The new mobo is most likely to be the source of the issue.
I'd RMA the new board. (see above)

[BobCat13]

Which 6000+ do you have? The 125W or 89W?

The M2N SLI Deluxe requires a newer BIOS version for the 89W processor. This page lets you choose the motherboard and see the BIOS versions necessary for each processor:
http://support.asus.com/cpusupport/cpusupport.aspx?SLanguage=en-us

[Faatmole]

Bobcat:
I'm not sure.. It's not on the website I ordered it from (the url remains the same while on the website, so I'm unable to link it - www.auspcmarket.com.au though if you follow the links) and I can't find it on the CPU box... Any other ideas on locations where it's shown? If it's printed on the actual CPU I'd be needing some more thermal paste before it goes back into the computer to check it..

The main thing that confuses me is that it was working flawlessly at first, and then stopped.

[AMDave]

It would be nice if CPUZ gave that extra bit of information.
However, it should give you the chip series and you should be able to look that up to find out which you have.

Great info from BobCat13. I didn't pick that up.

[BobCat13]

Any other ideas on locations where it's shown? If it's printed on the actual CPU I'd be needing some more thermal paste before it goes back into the computer to check it..

You can try using the Central Brain Identifier program available from MajorGeeks

http://majorgeeks.com/downloads14.html

It lists the OPN number, unlike most other CPU ID programs I have tried. I believe AMD uses the following:

ADA6000IAA6CZ = 89W
ADX6000IAA6CZ = 125W

[Faatmole]

I managed to boot the PC in safe mode after 4 attempts, to run CBid, to be told it's a 125W... Adding a screenshot of the rest of the results in case there's something wrong that I can't see...
http://content.imagesocket.com/images/CBId00917.jpg (http://imagesocket.com/view/CBId00917.jpg)

EDIT:
I've got no idea what HTL speed is, but comparing the screenshot to the CBid I've got running at the moment (It may have just been that the 6000+ was running in safe mode, I've got no idea) shows the 3800+ running at 1004.63Mhz, and the 6000+ is running at 94.18...

[Frederic Brillouet]

I managed to boot the PC in safe mode after 4 attempts, to run CBid, to be told it's a 125W... Adding a screenshot of the rest of the results in case there's something wrong that I can't see...
http://content.imagesocket.com/images/CBId00917.jpg (http://imagesocket.com/view/CBId00917.jpg)

EDIT:
I've got no idea what HTL speed is, but comparing the screenshot to the CBid I've got running at the moment (It may have just been that the 6000+ was running in safe mode, I've got no idea) shows the 3800+ running at 1004.63Mhz, and the 6000+ is running at 94.18...
HTL is your Hyper Transport Link (aka hypertransport bus). Its weird its only running at those speeds, in fact the reading cant be true, unless they've got hypertransport multipliers that use half increments. The normal speed at stock processor speed should be 1000 MHz at a x5 multiplier (reference clock [normally 200 at stock] * multiplier) your processor voltage seems rather awkward as well. the clock ratio is off as well, because i thought the 6000+ was supposed to have a x15 multiplier and here it says x16. also the fact that your reference clock is 188 MHz instead of 200 or more seems wried, but that might be because the pprocessor has coolnquiet enable.

now anyone that can explain this? :icon_lol:

[BobCat13]

HTL is your Hyper Transport Link (aka hypertransport bus). Its weird its only running at those speeds, in fact the reading cant be true, unless they've got hypertransport multipliers that use half increments. The normal speed at stock processor speed should be 1000 MHz at a x5 multiplier (reference clock [normally 200 at stock] * multiplier) your processor voltage seems rather awkward as well. the clock ratio is off as well, because i thought the 6000+ was supposed to have a x15 multiplier and here it says x16. also the fact that your reference clock is 188 MHz instead of 200 or more seems wried, but that might be because the pprocessor has coolnquiet enable.

now anyone that can explain this? :icon_lol:

Frederic,

Those numbers come from booting into Safe Mode. I just tried it and everything looked like Faatmole's except the voltage was higher.


Faatmole,

Restart the system and get into the BIOS. Goto Advanced and choose JumperFree Configuration.
Set AI Tuning to Manual so you can change the following settings.
Overclock Options Disabled
N.O.S. Option Disabled
CPU Frequency 200.0
DDR2 Voltage 1.90v (see note below)
CPU Voltage 1.35v
CPU Multiplier 15Note: One the DDR2 Voltage, you should check your memory manufacturer's info to see what to set this at. I have 2 different sets of DDR2 RAM; one runs at 1.80v, the other at 1.90v.

Judging by the voltage you showed in Safe Mode, it may be set too low. Mine was reporting 1.150 in Safe Mode, 1.200 in Normal Mode even though it is set at 1.3500 in BIOS. If yours is reporting 1.050 in Safe Mode, then it may not be enough for the 125W 6000+ as they are some hungry beasts.

[Faatmole]

I tried running the forced options you listed, Bobcat, but there was no change... I let it run to the BSOD a few times and noted down some error messages (I hadn't noticed "Driver over-run by a stack based buffer" before) and the most common message that I got was an error in NTFS.sys...
Due to that, is there any chance that the issue could be software based? vista not being compatible with some part of it (or having had something changed in an update, and that's why it stopped working after 2 days) or Norton anti-virus which is constantly creating rules for every new application I run?

[Lagu]

This is a tip from Vaughan to me. Check if there your disc for damage following this:

Close all programs.

From a CMD prompt (DOS box) i.e. Start -> Run -> in the Open: box type in cmd and press Enter.

Then type in chkdsk c: /f then press Enter
(replace the c: with whatever hard disks or partitions you have.

If you get a warning message about the drive being locked that's OK just do what it says and a chkdsk will be performed when you re-start your computer.

Now re-start your computer and you should see chkdsk run a series of tests, usually three are run (sometimes more, I think it can be 3) and the second one is quite slow (have a cup of coffee during this one :icon_wink: ) if you have a large capacity hard drive.

Lagu:icon_wink:

[BobCat13]

Try the chkdsk that Lagu suggested first, but I don't believe you have any problem with the HDD. I'm inclined to agree with AMDave and say you have a faulty mainboard, at this point.

I would ask if you have tried ASUS Probe to check temps on the 6000+, but since you can't get it to run that long it would be difficult to check temps.

Another thing to try would be the BIOS. Check your version (put the 3800 back in if necessary) and see if that one has any problems on the links AMDave provided. If so, you could try flashing another version.

[Faatmole]

Bobcat, I can check the temps in the BIOS settings, which I will do when I get back from work tonight. I'll post it if anything odd comes up, and if not I'll try updating the BIOS.
Also, the Chkdsk didn't show any issues.

[Faatmole]

Oh the frustration...
I put the 6000 in to check the temperatures on the CMOS pre-boot (shown below), and then booted it, waiting for a BSOD. It didn't happen. I managed to log on completely for the first time since it started acting up, ran CBid (screenshot also below) but as I went to post on here that I managed to log on, "Readyboost" had stopped working. When I clicked to end the program, I got a BSOD.
"ReadyBoost is a disk caching technology first included with Microsoft's Windows Vista operating system. It aims to make computers running Windows Vista more responsive by using flash memory on a USB 2.0 drive, SD card, CompactFlash, or other form of flash memory, in order to boost system performance."
-Wikipedia
I hadn't even heard of readyboost until it crashed...
Anyway, on to the CPU temperatures:
before getting to a BSOD with the 6000, it began at 35 degrees (All temperatures listed are in Celsius) and eventually rose up to stay at around 39-41. After it ran for a few minutes then crashed on me, the temperates where ranging from 43-45. If it's of any relevance, the motherboard was ranging from 38-40 at the same time
After putting in the 3800, it was running at 40-42 (mind you I'd just taken out the 6000 - CPU//motherboard// general case was still a bit warm), and according to CBid it's currently averaging at 42 degrees.
The following picture is a comparison of the 3800 and the 6000 - the 6000 was taken running normally (Not in safe mode) but the only difference it had in safe mode is .01 in the TSC clock..
http://content.imagesocket.com/images/CPUcomparison5b5.jpg (http://imagesocket.com/view/CPUcomparison5b5.jpg)

[AMDave]

Rockin!

The OPN number for your 6000+ gives it all away.
Going straight to
http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/33954.pdf

ADX6000IAA6CZ breaks down like this:

AD=Manufacturer code AMD
X=Power@125W
6000=Model: AM2 socket AMD 64 X2 Dual Core at 3000Mhz with 2MB cache
I=Socket: AM2
A=Operating voltage: Variable
A=Die temperature: Variable
6=L2 Cache Size: 2MB
CZ=Part Definition: Rev F3, 00040F33h

That takes us down to table number
2.3.9 ADX mmmsvtc CZ Thermal and Power Specifications

which says for the 3000Mhz Max-P state, the voltage should be between 1.350 V @ 55 degrees Celcius and 1.400 V @ 63 degrees Celcius.

Your report above shows the core voltage at 1.050 V. Note that whilst this is the default power-on voltage for socket AM2, it is not the operating voltage of that CPU. The power-on voltage allows the BIOS to detect the CPU and then voltage should step up/down after power on to suit the CPU.

Which brings us to the diagnosis:
Your BIOS has failed to set the correct voltage for your CPU.

A screwed up BIOS may also explain that awful HTL speed that is showing. I believe the correct clock ratio for that chip is meant to be 32, not 16.

You can try to flash the BIOS. Make sure the correct procedure is followed. (If the 3800 is stable enough in the M2N board you can install the ASUS BIOS update tool and use it to update the BIOS and if it works then put the 6000 back in. I have used that tool successfully several times with my ASUS boards. BUT there are posts by people saying it did not work for them with this board per my original post so be prepared. If the BIOS update fails you should be able to get a fresh (and up-to-date one from ASUS for a few buck within a couple of days (according to other posts).
OR
Take it back to the place of purchase and have them give you one that works (eg RMA the mother board). After all, that is what you were paying for.

If anyone can add more to or contradict my explanation please post as I can always learn some more.

Best of luck.

[Faatmole]

><
I flashed the BIOS without any problems at all, put in the 6000, and still had the same issue...
however, it did prove to show another oddity. When I went to check the temperature in safe mode while running the 6000 (for some reason the field was blank), and I noticed that, while the power limit was 125W, it was using 129.20W... (entire page screen capped below). The 3800 currently is running at 54.26W with a 62W cap, so there seems to be a bit of an issue there.. The first page with the core info was running identical to the previous one's I've posted, however it's probably due to it running in safe mode.
http://content.imagesocket.com/images/CBId02dbe.jpg (http://imagesocket.com/view/CBId02dbe.jpg)

[AMDave]

only 3 things left you can try.
1 - try the 6000+ cpu in another board that is known to be working well with a 6000+ cpu in it to rule that out
2 - try a different power supply to rule it out
3 - RMA the motherboard to rule that out (if your RMA number has not expired yet)

[BobCat13]

I don't think you can trust the numbers CBID is giving under Safe Mode. Here are two screenshots from my M2N-SLI Deluxe w/ 6000+ box:

http://i135.photobucket.com/albums/q133/rahicks13/Safe_Mode.png http://i135.photobucket.com/albums/q133/rahicks13/Normal_Mode.png

[Faatmole]

Sorry about the lack of response...
I tried swapping out the power supply, but the problem persisted. I had a friend come over on friday with a half built PC (motherboard, power supply & CPU) which had a gigabyte AM2 board in it, however we had issues removing the heatsink (the faceplate was stuck on the board, so it wasn't possible to unscrew that and go from there) and given that it was relatively new hardware he didn't really want to cause any damage to it so I couldn't test that..

I think I'll just RMA the board on Wednesday (Monday / Tuesday I've got major exams - supposed to be studying now)
Thanks to everyone that's posted here, with all the suggestions / assistance you've provided, and I'll post back on how the RMA goes :D