• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD FX9590 + GA-990FXA-UD7 woes

Neuter

Neuter

Neuter

Associate
Joined
20 Dec 2015
Posts
10
Hi all, and Merry Crimbo :)

EDIT: For clarification, it's the rev 3 NOT the rev 1 motherboard.

Prior to doing all of the below, I did update my BIOS to the latest BETA version (and then rolled it back 1 version to check for differences).

I decided to treat myself to a RAM and CPU upgrade this weekend, bunged it all in, booted windows, loaded up 3dMark... First freeze. Hrm, strange that. Shut the PC down, tried again. This time on the login screen. Getting slightly concerned it was temps, I quickly loaded up the BIOS and checked the temps. Bit on the warm side, but 35-40C is reasonable.

Left it in the BIOS screen for 10 minutes to see if it'd lock up again. Nothing. Shut it down, took the heatsink off and put my old CPU back in. No problems, then shut down and put the new CPU back in.

So, did a little digging around and someone said to disable a bunch of things in the BIOS relating to CPU, as follows:


Core performance boost
CPU Unlock
Cool & Quiet
C1E Support
Core C6 State
HPC Mode
APM

Disabled that lot, and the CPU was "stable" again. By stable, I mean, I could now load into windows, no dramas, no freezes... or so I thought. Loaded up 3dmark, and the minute it ran a physics test, locked up again.

So, back to the drawing board, I went and downloaded BOINC and ran seti@home. As soon as it peaked to 100% CPU usage, instant lockup again. Temperatures were 48-50C according to CPUID, so it wasn't running hot. I then tweaked BOINC to only use 90% CPU max, no problem.

I then rebooted and decided to underclock the CPU to 4.4Ghz (now this thing is meant to run at 5gz turbo... so I'm a bit miffed at this point).

Ran BOINC overnight at 100% and PC was happy as can be getting back to it this morning.

So, I rang up OC and they basically said there is only 1 motherboard which will work with this CPU. I asked if there were any BIOS settings I could do to get it to work. "No, sorry, we get a lot of these CPU's RMA'd, so I suggest you do the same within 14 days etc". Was the reply.

I have a Megahalems cooler, 650W Antec Platinum PSU, motherboard as per title.

Can anyone who has got this combination to actually work let me know what you did? I'm at my wits end having forked out £175 to only get 200Mhz over my old trusty 8320. I am not planning on overclocking the CPU at all, I'm well aware it's a heavily overclocked 8350. I just want it to run at the stock speed it comes shipped with.

If I have to go down the watercooling route, then so be it. I'd just like to know if there is anything else I can try prior to forking out for additional hardware.

Thanks for reading my wall of text :)
 
Last edited:
Thanks for the reply theaterix. I suppose I didn't explain myself well enough. My old 8320 got up to 4.2Ghz overclock, no dramas at all. I was "hoping" for 5 with this new one. But right now, that's why I'm saying I forked out £175 for only 200Mhz "upgrade - read underclock" (when I'd prefer it to have been 800 boost as advertised when buying it). I hope that someone can come along at some point and say "oh, do xyz in the bios, problem solved".

I don't think it's a temperature issue, as the lockup would instantly happen the very second the CPU went to 100%. CPUID would freeze on screen saying it only got to 48-50C. Now, it could very well be that it did over-temp extrememly fast, the software might not catch it quick enough before it locks up. Still have my reservations getting a watercooler yet, and as you say, I might just send it back and go for Intel (been an AMD fan since the good old Athlon XP days). Anyway, I'll lurk around for a few days to see if anyone managed to get them to play nice, if not, it's going back.
 
The PC is mainly for gaming, however I do network router simulations on it and initially ran into capping out my 16Gb RAM and couldn't add more routers to the network, so the extra 16Gb should solve that problem. (GNS3, running JunOS).

When I loaded "optimised defaults" in the BIOS, it set the memory to 10600 (I think), so it was already unstable. I've since bumped it up to 1833 today and it's ok at 4.4.

I'll look into going down the Intel route then. Thanks for the help thus far all :)
 
Thanks Dave, I'll give this a go, but I can guarantee that if I set it back to defaults, windows will stop working. I'll get freezes at the login screen, and windows gets really upset wanting to do a recovery (as it thinks the problem is with itself).

I honestly had massive stability problems prior to disabling those features. Didn't even need to run anything to get lockups.

Currently (in its crippled 4.4Ghz version) Phenom MSR has the following P-states:

P0:
Core multi 25
CPU VID 1.5
NB VID 1.55

P1:
Core multi 22
CPU VID 1.5
NB VID 1.55

P2:
Same as P1

P3:
Core Multi: 20
CPU VID: 1.3875
NB VID: 1.55

P4:
Core Multi: 17.0
CPU VID: 1.2750
NB VID: 1.55

Apologies if this isn't the information you're looking for, let me know what else I should have looked at?

Idle: http://i.imgur.com/7mTBi1a.jpg

100% load: http://i.imgur.com/uFbd9Pd.jpg

The maximums have been over the last 24 hours while I've had seti@home running 100%

I'm going to try reboot and reset to optimised defaults and see if I can get a reading for you.
 
Hi Dave,

I reset it to defaults and sure enough, I had endless problems. I had to take photo's of the screen with my phone (as I obviously can't do a screen grab when it's all locked up!).

Anyway, here's a few screenshots of what was going on.

Before resetting to default (current stable)

http://imgur.com/a/WV2QE

After reset values and lockup photos:

http://imgur.com/a/nlNa9

The last crash photo was me lowering the clock multi to 22 (and it still crashed), had to go back and disable all the CPU features mentioned previously, and I'm back up and running again.

After the last crash, Windows tried to do a recovery again, then the PC completely shut off, had to perform a bios recovery.

I have noticed however since this bios recovery my VCORE is much lower.

http://imgur.com/fOZqtmt

Anyway, any help is appreciated. I'm no expert when it comes to things in the BIOS, I just know at the moment how to get it stable.
 
Hi Dave,

I need a tiny bit of handholding here if I may.

I went into the BIOS and went looking for LLC and found one option to adjust it.

Vcore Loadline Calibration. My options are:

"Auto" - currently set to this
Normal
Extreme
Ultra High
High
Medium
Regular

So yeah, I have no clue as to what each one does, nor does clicking on them show how much it'll adjust it percentage wise.

Thanks for the help thus far from everyone.
 
Hi Dave.

I've canned it mate. I just can't get this CPU to be happy for love nor money. Got a BSOD after your suggestion. I'm just going to send it back, it kept locking up and as others have said, a poor choice for an upgrade.

http://i.imgur.com/mzoEqL4.jpg

Stuck my old faithful back in there (OC'd from 3.5 to 4.2 quite happily) and I'll wait off for a while before I upgrade.

I appreciate all the help you've all given in this thread.

This old CPU is perfectly happy running at 100% load and 42C. Maybe in the capable hands of someone like yourself, this CPU could have run just fine. Unfortunately I just didn't have the patience to get it to work.

Take care.
 
You must log in or register to reply here.
Back
Top Bottom