Phenom 9500 w/ MSI K9A2 Platinum

[madfaze]

hmm i updated to v1.3 how come i cant find those auto express option?!

[Jaivan]

@KTE
Well the only difference with the DFI and the Sapphire is that the Sapphire has an odd color scheme lol.

[KTE]

Thats really strange... Must have been the USB flash procedure then, have you ever Flashed a BIOS like that KTE?

Never, not BETA's. Known to give random issues and be a problem causer IME.

Now the even stranger thing is, I swapped out the PSU and the VC. Same results and they both work fine in other machines. I went and got a new board today, same results... I'll put the 9600Std back in tommorow, about the only thing left would be the 9600BE went bad...

It won't have become naft but it looks exactly like the problem I've had many times after a CMOS clear or flash. The CMOS doesn't clear and keeps a mix of old and new values but throws the VCore too low and that results in no boot, maybe for many days.. until you go subzero with the chip.

Thanks Much for all your help Man...
And sorry about muddying your thread with my problems...

You're welcome. No need to apologize, I'm still alive.

hmm i updated to v1.3 how come i cant find those auto express option?!

It doesn't have it, P0H does.

[aGeoM]

JESTER: Just curious if you flashed the same version of "P0H" that aGeoM linked back on page 50? Thats the one I used and I'm having similar problems. Although mine never even posted after the flash...

aGeoM: Did you get your version from that link, and did it work OK? I'm certainly not accusing you of anything malicious, but If it's a bad file I'd hate to see someone else flash a bad bios.

I've looked all over and still can't figure out what the jumper right next to CMOS reset does (JCI1), but I'm assuming it's not the backup bios I was hopeing for...

Not sure what I'm going to do now....

Hi

The link it's just the upload from the bios posted in MSI beta bios. I had used no problems found worked with same settings used in 1.131 beta.

[aGeoM]

Thanks Jester, I thought that may have been the one you used too.

I've done everything I know to do and followed the instructions per Informal & KTE... Mobo is just bricked...

KTE: Did you get a chance to DL that version of P0H? Personally, I wouldn't flash it just in case! But you may be able to run a checksum comparison with your known good copy.

The only other thing I can think of is that we both used a USB drive to flash the bios. Thats the first time I've ever used that method, but everything seemed to go as it should, ie: ran all the steps and completed as expected.

I use bootable USB pen, and run afud4184 BIOS.NAME normally.

I only use commands /n /pbnc for reverting to previous version eg. 1.P0H to 1.P0D.

Sorry if the file I uploaded make your boards brick, but it was to help you guys to have it without register in the German site just to download.

@Informal I play AA mainly in Interdiction Map, but not playing ATM.

[Daveburt714]

Hi

The link it's just the upload from the bios posted in MSI beta bios. I had used no problems found worked with same settings used in 1.131 beta.

Thanks aGeoM, no offense intended, it just seemed odd that Jester and I both had similar problems..

I put the 9600Std back in last night (on the new board) and did finally get it to post, but it's still acting strange... Looks like it may be a PSU problem after all.

BTW, aGeoM: I do appreciate you putting the file up for us!!

[linhvndiy]

Got tired with phenom.
I just got new 9600BE with main giga 790FX DQ6, can not stable with 2.5ghz.
Changed to Asus M3A32 MVP, can not pass 2.5ghz too.
I tried near all options about volt. Sad about my phenom.

[KTE]

Apologies to all but I'm on holiday and so been too busy to get back with all I wanted. Brought the stuff with me here (car, not far) so I can speak details and reply to PM's later on but things will be very slow until after March 8.

Quickly speaking=>

aGeoM: we don't doubt you. As I said, I tested it and it's fine, it's the original file. The problem could only be with the method and functions used.

Jaivan: will get it as soon as I have office access. Need permission from uncle first and he's out of country yet.

linhvndiy: The 790Fx board you wanted to maximize Phenom is either the DFI or the Sapphire.

OK. Well been testing Phenom for many days now, +25 days rigorous. Here are my findings and trends.

Firstly, I went from 2.55GHz all the way up to 2.832GHz testing stability. My purpose wasn't to bench or take screenshots of Prime but to avoid the freeze/lockup after 2-3 and then 5 days.

*Moving up stability test/benchmark pass I reached 2.754GHz @ 1.3572V Prime 95/EVEREST/OCCT/Memtest fully stable. 20 hours for P95.
*No matter what cooling (air), ratio, volts was used, not a single MHz more was stable with stability testers.
*But when I tested real long term stability incl. freezing, the best +5 day stable was 2.7GHz @ 1.352V idle and 1.336V loaded. Fully perfect for over 7 days.
*I then moved to oc on various settings and didn't find any better again. So I reverted back after 5 days to trying to find stability with my most stable known, Last Known Good Conf in other words.
*However, now, 2.691GHz @ 1.512/1.506V was the max I could get stable. Yep, it really did require this voltage, 1.32V at 2.69GHz would reboot on load. More stability was possible below 2704MHz but yet to be explored.
*I then went ahead and tried higher than 2704MHz. All of them rebooted at any volts whilst I was idling, randomly and different periods ranging from 10 minutes to 11 hours. I could boot 2.832GHz maximum and benchmark it.
*However, when I now tried oc again to a lower spot in between 2704 and 2691 I found another phenomena. 2691MHz all the way to 2639MHz was not stable and was now rebooting randomly with an Ati VGA driver BSoD. Typical BSoD of instability (I analyzed the files). I dropped down to 1.456V and after 33 hours it froze again randomly while starting a YetiSport game (one of my tests).
*Trying anything from 2.618GHz to 2.639GHz with 1.56V to 1.3V was repeatedly calling the freeze.
*So for the past 11 hours since I got here I set it up and booted 2.613GHz at 1.408V and have been seeing if it froze. It still hasn't shown the freeze yet and is looking like the most stable setting so far... but for how long?
*In each case I tried all the voltage ranges, CPU/NB/RAM dividers, HT ref., and even with and without the TLB patch and with and without any videoing, stress, gaming and Firefox usage.

This is what I mean by not made to oc, look at the volts as well. Stability over 2.6GHz is just plain out random and hardly exists. ATM I'm at 2613MHz 1.392V idle/1.376V load, I had this setting stable for over 14 days at only 1.252V before and am not happy with the amount of MHz I'm losing day by day even with higher voltage.

Has anyone else experienced similar happenings?
All my chips have reacted this way at top after one solid month.

BTW I'll have the Sapphire and the Abit boards soon to test but this MSI is also giving me a "USB device not recognized" random error very consistently where my mouse/keyboard stop working. Really does wind me up!

[Daveburt714]

Hey KTE,

I've only had this board for a couple days (Gigabyte DS5), but if you get a chance you should check it out!

Flash the F3/F4f bios. Like I said, I'm still a noob on it, but the bios features work (pretty much) the way you would expect them to work, the one bad thing is that you can't control the NB v's but it actually seems like their TLB disable works!

I like MSI, and I've been a fan for a few years, but it seems like they've given up on the K9A2 bios...

So far, I like this board MUCH better than the MSI offering...

[KTE]

I have previously checked it out Dave.
TBH I'm not fond if it, it clocks worse than the MSI and has less oc options than the 113 BETA.
The only one of GBT's lineup that is better in oc/stability is the DQ6.

BTW the MSI board has dropped in value a lot around places here I'm seeing while the DQ6 retains it's value.

[linhvndiy]

Thanks a lot KTE.
I will try DFI. I think AMD need a lot of tweak for phenom.

[madfaze]

BTW the MSI board has dropped in value a lot around places here I'm seeing while the DQ6 retains it's value.

hmm that's odd, does this mean that MSI is loosing its reputation as a good mobo?

[KTE]

Hehe... I want to show you all what ERRATA actually means.








DON'T BE FOOLED! That's just the clock generator going whacko, actual speed does not change from what we know and the core performance will show it clearly. The actual NB clock there is same as HT clock (1809/1809). That's not the maximum either, I've seen 4650x35 (162,750MHz NB) and 4650x2.66 (12,369MHz RAM) in those apps stock NB/RAM volts

AMD has some very serious clock generator problems.

[Oc-Ghost]

[justapost]

Hehe... I want to show you all what ERRATA actually
DON'T BE FOOLED! That's just the clock generator going whacko, actual speed does not change from what we know and the core performance will show it clearly. The actual NB clock there is same as HT clock (1809/1809). That's not the maximum either, I've seen 4650x35 (162,750MHz NB) and 4650x2.66 (12,369MHz RAM) in those apps stock NB/RAM volts
AMD has some very serious clock generator problems.

You did all modifications with bar/wpcredit on a running system without a reboot i assume.

Hmm changing the nb multi requires a warm start says the bkdg. Can be the clock generator might need some sort of refresh command to display correct values if the multi is changed on a runnin system.
Polygon's bios-mod modifies MSR registers via the lan-boot module. This one should be loaded after the prozessor and stuff got initialized so the nb-multi change only confuses the clockgenerator.
If the nb multi is locked upward on the cpu the symptom is more obvious as the ref HT changes, with no nb-multi lock things look different.

[LIKMARK]

If the nb multi is locked upward on the cpu the symptom is more obvious as the ref HT changes, with no nb-multi lock things look different.

not neccesary so. See here :

9500 nb multi x10 wich is not supposed to be possible. edited with BAR EDIT on the fly.

[KTE]

I already know that, I've been doing it since December. No Achim, it booted like that and it can get worse. It's like that since I booted for over 8 hours now.

If I modify NB registers on the BE and try to boot, it won't boot because I need NB voltage which the P0H BIOS doesn't allow (stock voltage, see above). And my NB is downward locked, no POST with anything below 9x.

CPUZ Dump attached [worthless for this since this is a CPU error].

[tictac]

You did all modifications with bar/wpcredit on a running system without a reboot i assume.

Hmm changing the nb multi requires a warm start says the bkdg. Can be the clock generator might need some sort of refresh command to display correct values if the multi is changed on a runnin system.
Polygon's bios-mod modifies MSR registers via the lan-boot module. This one should be loaded after the prozessor and stuff got initialized so the nb-multi change only confuses the clockgenerator.
If the nb multi is locked upward on the cpu the symptom is more obvious as the ref HT changes, with no nb-multi lock things look different.

I am with justapost... I already state it back in barcelona
http://www.xtremesystems.org/forums/...8&postcount=83

[Mathos]

These processors can't be taking electron path dmg that quickly already can they? (forgot the technical term) Maybe that would be why Intel said native quad at 65nm wasn't practical... Granted AMD chips have never been spectacular OCers. But.... I can't see how that could happen if you're not running above stock volts, I was running 2.4ghz NB/IMC at stock nb voltage without issues, on 1.1b3. I've been considering going to P0H but don't want to if I can't do the nb thing. My luck with OCing nb on my phenom has either been stable at a voltage or no boot. For example I can do 2ghz nb at 1.1v VID stable, probably a bit lower too, never tried lowering it below that.

I was running 2.6ghz Core at 1.262VID 1.248v Actual for a long time, with the NB running at 2.4ghz 1.25v VID (according to AMD Power monitor). Had to lower back to stock recently to eliminate all possible issues trying to figure out if I was having a video card problem. Ended up having to RMA the 3870, wouldn't run graphics intensive games even with everything else at stock. Though I run perfectly stable with the 2600xt installed, granted it doesn't root out a possible PSU issue.

[KTE]

Don't really know, I can only speak on what we experience.
I have a problem where high MHz and overclock makes the VGA unstable, not load up of give BSoD's too. Even though the oc be Prime95 stable.

[Mathos]

Don't really know, I can only speak on what we experience.
I have a problem where high MHz and overclock makes the VGA unstable, not load up of give BSoD's too. Even though the oc be Prime95 stable.

See thats the thing I don't get with mine. I was completely stable at 2.7ghz 1.262v VID 1.248v actual on the cores 2.0ghz 1.1v VID on the NB. Then after upgrading from the 2600XT to the HD3870, suddenly I'd get a machine check BSOD in games like Civ IV with the graphics set to high on 1680x1050 res. I'd get graphic corruption, then either that machine check bsod, or I'd lose video and have to reset the computer twice. Some times when it would come back up after the first reset, the windows splash screen would be way dark and not load beyond that, but after a second reset it would be fine, until putting a heavy graphics load on it, and that continued to happen even with everything at stock. Now I'm back on the 2600XT with no problems at all, had to RMA the 3870.

The bad thing about the machine checks was it never mentioned the video card. It always listed Processor, Power supply, Memory, or system board, even though I knew for a fact that the bsod's didn't start till after putting the 3870 in. I probably could of got a dud vid card, but like I said, it still doesn't rule out a weak psu. Won't be able to fix that part till either my quarterly bonus from work, or the $110 in rebates comes back from Raidmax.

[KTE]

You know something, my BSoD's relating to the GPU I only figured out after lengthy troubleshooting, the stacktrace and analysis logs didn't show anything but the CPU or RAM at error. Typically I had to uninstall and reinstall GPU driver many times to figure this out.
I was stable 1.28V 2.7GHz at the beginning as I've shown, which was then moved up to 1.344V for 2.754GHz. But now that is impossible without those dreaded freezes. The max I've had without freezes was 2613MHz. It's now at 1.28V idle at 2613MHz stable for many hours even benchmarking, priming and gaming without the random freezes. Still far off from 1.232V 2.6GHz stable as beginning.

I don't want to spread unsubstantiated FUD but I will tell you the honest facts to keep in mind. The TLB errata doesn't have to crash your system alone.. it can make you lose hard drive data, cause drive fragmentations, loss of files and corrupt files especially those held in system memory i.e. VGA SYS and DLL files included.

But you really shouldn't rely on that PSU. It's not even a PSU, honestly advising you, it's a total rip off for what it's doing to your components.

[Daveburt714]

Whats the stepping on that 60Ghz Phenom KTE!!

That is wierd man! You think it's a software problem? Seems like if it was an actual problem with the clocks it would have crashed long before that....

[KTE]

It's the CPU MSR's and/or PLL feedbacks and any software which depends on them. Actual clocks do not change with these bugs, core limitations are well known by now.

Once I opened up Crystal CPUID to check my MSR values and a lot of them were totally blank (0's) where they shouldn't be at all.
It's the same with BarEdit/WPCREDIT access of NB registers, they totally whack out many times. The last time I checked that also was reading 0b for everything in the NB FID D4 register, whereas it should not be and was still running 9x multi as it showed. At those instances L3 performance and access becomes very bad and slower than stock.

If I could capture it, I'd show you some much higher MHz to show you the scale of the problem. Dilemma is, when the [fake] high MHz show up in all software the system freezes for 2-3 seconds at a time and so the PrntScrn is never saved until after it restores from freezing. The way to test the bug is to test the system on multiple boards really.

BTW what config are you now running on the GBT? Any more oc on HT, NB or CPU?

[Daveburt714]

BTW what config are you now running on the GBT? Any more oc on HT, NB or CPU?

The best I've gotten so far with any stability (not really stressed) was 2.76Ghz (240x11.5x8x8) using AOD, those setting would not boot windows when set in Bios though... Looks like 2.7 may wind up being doable, while my best stable on the MSI board was 2.5...

So far it looks much better than K9A2, I can boot at 13x multi now.
To be fair to the MSI board, I never really got a chance to play with it much using XP, most of my testing was done using Vista 64....

A short story of how I wound up with the DS5:
I have a 9600std and a 9600be, when my K9A2 died I thought I'd be shady and go buy another one switch them out, return it, then I'd have 2 boards right?
Well, the board I bought actually WAS bad!! When I went back to the store they were out of K9A2's and all they had were DS5/DQ6 and the Asus boards, so it wound up costing me an extra $45!!

Poetic Justice I guess..
I'm just not good at being bad..