Trouble overclocking Athlon II X4 630 on Asrock 880GMH/U3S3

[doveman]

I'm having major problems trying to overlclock with this setup.

Previously, on a Biostar TA790GXBE I was able to set the FSB to 230 to get 3.22Ghz @1.456v, with the DDR2 RAM at 460/920Mhz and it managed 5 passes of IntelBurnTest OK. I might have pushed it further but that's all the notes I have.

Now I've got 8GB of Crucial 1600Mhz DDR3. I'm using a Xigmatek HDT-D1284 top-down cooler and temps are all under 42c idle at default clocks

Asrock pointed me to this table which shows that officially Athlon II X4 630 on a 4-slot board with 2-slots populated only officially supports 1066Mhz RAM http://support.amd.com/us/kbarticles/Pages/ddr3memoryfrequencyguide.aspx

So I've tried setting the RAM to Auto instead of 1333Mhz, which with a FSB of 230Mhz it sets to 960Mhz and I've also tried setting the timings manually to match the table for 1066Mhz, 6-6-6-18, (otherwise it seems to use the timings for 800Mhz, 5-5-5-15 which I guess might be unsuitable at 960Mhz).

I can POST at FSB 230Mhz but Windows BSOD on boot. Even at 220Mhz with manual timings it doesn't POST reliably and the fans runaway sometimes. It actually seems more unreliable with the manual timings, which doesn't seem to make much sense.

The BIOS has a setting "Load-Line Calibration - Auto" and says when manually overclocking it should be set to 1/2 for AM3+ and Disabled for AM3, so I've been setting that to Disabled. Spread Spectrum I've left on Auto. I'm using the onboard HD4250 but the PCI-E is set to 100Mhz so I assume that the GPU shouldn't be overclocked. Default clocks for the GPU are 500 or 560Mhz (stats for HD4250 show 560Mhz but GPU-Z showed around 500Mhz but maybe it was just downclocking at idle).

I've had some success using OC Mode - CPU Mode, which lets me select a percentage to overclock by. 5% tested stable with IntelBurnTest, so I tried 20% for 3360Mhz. That bumps up the CPU-NB to 1.25v despite underclocking it to 1920Mhz. It set the HT to 2100Mhz so I reduced that to 1920Mhz manually.

It set the CPU to 1.35v and the NB Core to 1.19v (default 1.10v) which seems strange as what I've read says to leave this alone. It set the RAM to 960Mhz. At those settings, IBT failed (well Linpack actually) on the first run, so I bumped up the CPU to 1.40v and that failed on the fourth run.

Temps were CPU 62, Core 62, MB 42. I believe the Athlon II X4 630 is rated for 71c but that's getting a bit close for my liking, so I'm not sure if I should increase the voltage to 1.45v. My CPU fan (on Xigmatek HDT-D1248 top-down cooler) ramps up from 1000 to 1400RPM at 50c and my front fan ramps up from 600 to 750RPM at 60c. My exhaust fan runs at a constant 750RPM.

Do I need to disable Spread Spectrum and Cool'n'Quiet? If I have to disable the latter, I'll have to use PhenomMSRTweaker to downclock at idle and I'd prefer to avoid having yet another utility running but if it's necessary so be it.

Any assistance would be most welcome. I've got photos of the BIOS screens I can post if that would help.

 

[RGone]

Core 62 << we have found that the Phenom2 cpus become unstable many times when the core temp is over ~55c.

 

[doveman]

Core 62 << we have found that the Phenom2 cpus become unstable many times when the core temp is over ~55c.

Yeah, this is an Athlon II not a Phenom II. The Phenom2 is only specced to 61c whereas the Athlon II is specced to 71c

 

[ShrimpBrime]

Temps were CPU 62, Core 62, MB 42. I believe the Athlon II X4 630 is rated for 71c

This is a max TDP for a stock processor. The goal is to lower temps to be able to raise voltage. I'd try and keep it in the 55c range at max temp while OCed.

Yes, turn off cool and quiet and C1E support.

Gonna need to see some screen shots of Cpu-z and such. Need a little visual to see where your at.

 

[doveman]

Of course that's the max but surely if the Athlon II is rated 10c higher than the Phenom II then the former can safely go a bit higher and I've been told to keep my Phenom II below 55c?

Anyway, I disabled Spread Spectrum and Cool'n'Quiet in the BIOS (I'd already disabled C1E) but left Load Line Calibration on Auto and it passed five runs of IntelBurnTest Standard at Auto 20% overclock (FSB 240, CPU 3360 @ 1.40v (auto was 1.35v), HT and NB 1920, NB 1.25v, RAM 800Mhz@5-5-5-15-2T (it autoset 960Mhz before so I'm not sure why it choose 800Mhz this time, 800Mhz isnt even presented as an option most of the time, with 960Mhz being the lowest, so if that is the best setting I can't even select it!).

Even though IntelBurnTest passed, I have DVDFab Passkey set to load with Windows and that crashed. This is only the second time this has happened and both times were when overclocking, which makes me think it's not as stable as IBT might suggest.

It's quite cool here today (indoor thermo reads 25c) and the PC is right next to an open window. At idle, I've currently got my rear exhaust fan running at 930 RPM, my front intake fan at 590 RPM and my CPU fan at 1054 RPM.

At load, the exhaust (Power) fan stays the same as it's not a variable speed motherboard connector, the CPU fan hits 1490 RPM and the front (Chassis) fan 1000 RPM. For that, I get CPU 62c, Core 62c, MB 37c. I guess I could reduce the target temps and see if that helps keep it lower at load (currently CPU 50c, Chassis 60c).

This shot of HwInfoMonitor shows the currently idle temps with the peak temps reached when running IBT on the right (as you can see, the MB temp has actually increased a bit at idle compared to when running IBT and now sits at 40c). Not sure why HwInfo shows Core VID 1.40v and CPU-Z shows 1.376v.



HwInfo is also wrong about the NB Clock which is set to 1920Mhz, as this image shows



This is the SPD tab from CPU-Z

 

[doveman]

Well it BSOD within seconds of posting my last message, so it obviously wasn't stable.

Worringly, after resetting the BIOS to default clocks (I had a saved profile for that) and rebooting, it BSOD again within minutes.

I went into the BIOS again and checked and everything was as it should be but I reset to UEFI defaults and adjusted the settings as needed and rebooted and it seems OK again now (fingers crossed).

I've uploaded some photos I took earlier of the BIOS in case they're helpful. The first two are the overclocking page, the third is the RAM timings page and the rest are the Advanced settings.

http://s10.postimg.org/sehh6o955/SAM_0311.png

http://s10.postimg.org/rjqc6wbwp/SAM_0312.png

http://s10.postimg.org/mrl4976mh/SAM_0313.png

http://s10.postimg.org/urutjxy6h/SAM_0314.png

http://s10.postimg.org/n1smsjl8p/SAM_0315.png

http://s10.postimg.org/vb4i6yeyx/SAM_0316.png

http://s10.postimg.org/v0wznm0cp/SAM_0317.png

http://s10.postimg.org/gyb28muyx/SAM_0318.png

http://s10.postimg.org/7rsrlcpqh/SAM_0319.png

http://s10.postimg.org/lzn1nqrt5/SAM_0320.png

 

[doveman]

I've attached the minidumps from the two BSOD in case they show anything useful.

 

[dostov]

That CPU OC is very close to what you can expect from a Atlhon II X4 on air in my experience. I'm thinking you may need more CPU voltage to get it stable at your current 3.35Ghz. My 635 needs 1.4v to do that, I see yours is getting only 1.375v.

 

[Mr.Scott]

Needs more vcore. Unfortunately, your current cooling solution won't allow it.

 

[doveman]

That CPU OC is very close to what you can expect from a Atlhon II X4 on air in my experience. I'm thinking you may need more CPU voltage to get it stable at your current 3.35Ghz. My 635 needs 1.4v to do that, I see yours is getting only 1.375v.

That's disappointing. I've seen quite a few reports of people achieving 3.5Ghz quite easily. One person is getting 3.5Ghz on his Athlon II X4 620 at 1.325v (1.3695v in the BIOS) http://www.pix01.com/gallery/FCECA5...I_O_C/bigpage.html?image=1001637130_orig0.jpg

I've got 1.4v set in the BIOS, so I'm not sure why it's only using 1.376v according to CPU-Z

 

[doveman]

Needs more vcore. Unfortunately, your current cooling solution won't allow it.

I'd have thought my Xigmatek HDT-D1284 would be sufficient for a fairly mild 500Mhz overclock. Looking at these results, it seems it works better on 65mm Kentsfield CPUs than 45mm Yorkfield CPUs but I'm not sure which the Athlon II X4 is closer to http://www.xbitlabs.com/articles/coolers/display/scythe-zipang-xigmatek-hdtd1284_12.html.

On my old Biostar board I was using a TRUE rev.C to run it at 230 FSB/[email protected], HT 1600Mhz and the temps were Idle/Load 24/40c (not sure if that was CPU or Core), so it's surprising it's running so much hotter with the Xigmatek. I'm using the TRUE on my Phenom II X4 955 now with an Arctic F12 PWM fan http://www.arctic.ac/en/p/cooling/case-fans/73/arctic-f-pwm.html and it doesn't seem a particularly great cooler anyway (around 50c when under load (i.e. gaming, not IntelBurnTest) at 3.5Ghz)

Idling/Mild activity (5-20%@800Mhz), I currently get CPU 37c, Core 27c, MB 41c.

If the Xigmatek just can't handle overclocking I guess I'll just have to settle for a quiet machine at default clocks but I'm wondering if it's just not working as efficiently as it should be now.

 

[doveman]

I've attached the minidumps from the two BSOD in case they show anything useful.

Looking at these myself, it seems the one that occurred whilst overclocking says "Probably caused by memory corruption" which makes sense if the overclock was unstable. The second, which happened after disabling the overclock, refers to Sbiedrv.sys (a driver for Sandboxie).

Sandboxie works fine normally, in fact I'm using it now, so it seems a strange coincidence that it should only cause a BSOD just after testing the overclock. At default clocks, I haven't had any BSOD in 2-3 weeks and since rebooting for a second time it's been stable for the last 5 hours, so I'm rather perturbed about that.

EDIT: Maybe I need to set the RAM to 1066Mhz? Despite that table that Asrock linked me to, showing that officially the Athlon II X4 only supports 1066Mhz when using 2 sticks in a 4 slot board, the BIOS on Auto sets it to 1333Mhz, so maybe that could cause the odd intermittent crash, as happened with Sandboxie after disabling the overclock. Having had any more BSOD since but Iron Portable running in Sandboxie did crash once (ironically when I was trying to upload the minidump to the Sandboxie forum and clicked on Browse to select the zip on the desktop!)

 

[dostov]

That's disappointing. I've seen quite a few reports of people achieving 3.5Ghz quite easily. One person is getting 3.5Ghz on his Athlon II X4 620 at 1.325v (1.3695v in the BIOS) http://www.pix01.com/gallery/FCECA5...I_O_C/bigpage.html?image=1001637130_orig0.jpg

I find that hard to believe, especially the "quite easily" part. Mine is a rev. 3 chip and it does 3.4Ghz at 1.4v.

I wouldn't expect your cooler to be a very good cooler. It's not bad, but I would use the TRUE instead. Tower style heatsinks are always better than comparable top-down style IMO.

 

[doveman]

I find that hard to believe, especially the "quite easily" part. Mine is a rev. 3 chip and it does 3.4Ghz at 1.4v.

I wouldn't expect your cooler to be a very good cooler. It's not bad, but I would use the TRUE instead. Tower style heatsinks are always better than comparable top-down style IMO.

Well I'm not sure just how easily but certainly I came across a fair few reports of running at 3.5Ghz. No way of knowing if they're true of course but I have reason to doubt the person who posted his Athlon II X4 620 result.

Having said that, mine needed 1.456v to get to 3.22Ghz on the Biostar board with the TRUE, so I can equally believe that some chips can't do much better than that.

As for the Xigmatek, it might not be as good as the TRUE but nonetheless the difference in temps seems remarkable (I'm being generous and assuming the recorded figure I have for the TRUE is Core and not CPU, I'll see if I've got any fuller records somewhere).

Idle: Xigmatek CPU 37, Core 27 @2.8Ghz(800Mhz)@1.36v(1.10v) vs TRUE Core 24
Load: Xigmatek CPU 62, Core [email protected]@1.36v vs TRUE Core [email protected]@1.456v

 

[doveman]

Hmm, seems I need to re-orient the Xigmatek so that the heatpipes are at the bottom (they're at the top at the moment).

http://www.xbitlabs.com/articles/coolers/display/scythe-zipang-xigmatek-hdtd1284_5.html

In this review they only tested with them over the RAM slots and found it was much worse than having them facing up, so I can't be sure there's going to be the same result from moving them from the top to the bottom but it's worth a shot (even if it is a major pain in the *** to disconnect everything and remove the motherboard from the case to get to the nuts!)

 

[dostov]

There's is a significant difference in overclocking potential between different chips. Some will OC much better than others. Yours is a stepping 2 chip, while others that have had it to 3.5Ghz "easily" might have a stepping 3 chip.

Just find your best stable OC for everyday use and be happy with it. There's nothing more you can do.

Good luck with repositioning the Xigmatek, maybe it will help!

 

[doveman]

Well I took the heatsink off to re-orient it and as you can see from these photos, the TIM coverage was awful, which I don't understand as I filled in the channels and applied two short lines, as shown at the bottom of this page http://archive.benchmarkreviews.com...ask=view&id=170&Itemid=1&limit=1&limitstart=5

Any advice on how I can get better coverage when I re-apply it this time? I'm using MX-2.

 

[Mandrake4565]

Doveman I have a CM 212 + with similar exposed heat pipes. I put a thin line of TIM approximately 10mm long on the middle two pipes and I get a good spread from it. You definitely applied too much TIM the last application. You may want to try it and then pull it off to see how it spread.

 

[doveman]

Thanks Mandrake4565, I guess I put it on the dividers rather than the actual heat pipes before as I copied that picture, so I'll try it on the heatpipes as you suggest this time.

I'll probably need to remove the motherboard from the case to get the nuts back on the heatsink anyway, so that'll make it easier for me to remove it again and check the spread this time. I just bought a new tube of MX-2 so I've got plenty to spare.

 

[Mandrake4565]

While you have it off, if you have a jigsaw available cut a hole in the motherboard tray so you don't have to remove it next time to mount an aftermarket heatsink. In addition it will help reduce the socket temps. if you do it just make sure you don't have any of the edges sticking up, a file will also come in handy to remove them.