FIRED UP! X79-UD7 + 3960X @ 5480 1.67V + 4way 580 ...

[massman]

So, some people seem to be convinced this type of config wouldn't work on a gigabyte X79 board. So, figured I'd give this a test.

Fired up the rig at 5480MHz 1.67V vcore, ran 3Dmark11 and WPrime 1024M

and then ...

suddenly ...

...

A benchmark score appeared!




No problems with stability at all. So, that's that then; let's move on to something else now ...

More info (like bios screens) in the hwbot thread.

[sin0822]

Very nice job mate!

[l0ud_sil3nc3]

nice to see proper results

and nothing caught on fire jk

wondering what vtt vccsa did you use for that run?

[massman]

1.2V for VTT and IMC. I posted ALL bios settings (incl PWM settings) in the hwbot thread. They are pictures from the BIOS itself (you know, the F12 function).

[OC Nub]

Very nice results!

Looks like they've got it working well now, glad to see it.

[l0ud_sil3nc3]

1.2V for VTT and IMC. I posted ALL bios settings (incl PWM settings) in the hwbot thread. They are pictures from the BIOS itself (you know, the F12 function).

very nice, you think a lot of the problems that people were having were user error?

seems like you had no problem producing good results, I think I may have to grab one of these to play with. I have only built one system for a friend with a UD5 / 3930K and it was pretty hard to get 4.8 stable for dailey use as he is not a bencher, could have just been a crappy cpu.

and the thread title rocks btw, very

[sin0822]

very nice, you think a lot of the problems that people were having were user error?

seems like you had no problem producing good results, I think I may have to grab one of these to play with. I have only built one system for a friend with a UD5 / 3930K and it was pretty hard to get 4.8 stable for dailey use as he is not a bencher, could have just been a crappy cpu.

and the thread title rocks btw, very

What people had users error? lol, yea. this board just needed some time to come out with a good BIOS capable of benching.

[massman]

I can't really speak for the overclocking capability on air cooling, since don't really my systems on air cooling. In fact, I don't even use mounting for the CPU heatsink; just use the thermal paste and gravity. It kinda limits the overclock by heat, but I got the same results on the UD7 as I get on the R4E with 6c12t (~4500 3D11 with 1.4V). I do recall there were some OC limitations (around 4.8G) with the earlier bioses (pre-F7), but I should verify if that's still the case. I think I wrote about it in the thread at hwbot forums (link).

Even back then, though, I never had any issue clocking up with subzero cooling. Same goes for the GD65 and Sabertooth. When you take those boards subzero, you should be hitting the same CPU OC all the time.

This board is not as easy to work with as the R4E. It's more of a challenge to get where you want. Well ... once you figured out how it responds, you'll know how to work the bios and get your result. There are a couple (ehem) things I'd like to see fixed ASAP, though. Anyways, have fun with that F5 board .

//edit: concerning the "user error" part. Last week, I ran this exact same chip on the R4E for the PCMark7 stage of the Country Cup and the board did a hard shutdown at 5.2GHz. Two days later, it got to 5400 no problem. Just to say ... user errors happen ALL the time

[l0ud_sil3nc3]

What people had users error? lol, yea. this board just needed some time to come out with a good BIOS capable of benching.

I have had plenty of user errors while learning this platform

I can't really speak for the overclocking capability on air cooling, since don't really my systems on air cooling. In fact, I don't even use mounting for the CPU heatsink; just use the thermal paste and gravity. It kinda limits the overclock by heat, but I got the same results on the UD7 as I get on the R4E with 6c12t (~4500 3D11 with 1.4V). I do recall there were some OC limitations (around 4.8G) with the earlier bioses (pre-F7), but I should verify if that's still the case. I think I wrote about it in the thread at hwbot forums (link).

Even back then, though, I never had any issue clocking up with subzero cooling. Same goes for the GD65 and Sabertooth. When you take those boards subzero, you should be hitting the same CPU OC all the time.

This board is not as easy to work with as the R4E. It's more of a challenge to get where you want. Well ... once you figured out how it responds, you'll know how to work the bios and get your result. There are a couple (ehem) things I'd like to see fixed ASAP, though. Anyways, have fun with that F5 board .

//edit: concerning the "user error" part. Last week, I ran this exact same chip on the R4E for the PCMark7 stage of the Country Cup and the board did a hard shutdown at 5.2GHz. Two days later, it got to 5400 no problem. Just to say ... user errors happen ALL the time

I know exactly what you are saying, I was having issues getting past 5391 for the PCMark7 stage of the country cup and I realized that the low current extreme oc profile had vccsa a bit high on default (1.25v) for my chip, reducing this gave me the extra mhz I was looking for, less is more it seems

F5 board?

[massman]

Oops, I meant UD5 .

I reckon OCers should go for the UD3 or UD7. UD5 is targeted at workstation I think.

[Charles Wirth]

Thanks update, now to test a few processors.

[Schnuppl]

Good job massman!

[AMD_Freak]

really nice.... good job