knopflerbruce

OK looks like I misunderstood what API really is:D

A socket specific list would be cool:)

Ticket ID: 932 Priority: Low http://en.wikipedia.org/wiki/List_of_AMD_Sempron_microprocessors#.22Georgetown.22_.28Socket_754.2C_D0.2C_90_nm.2C_Desktop_replacement.29\r\n\r\nOnly one of these missing I think, and I have it on my desk now

I bet you'll come back someday Nothing beats the feeling of getting awesome scores:D

I'm not 100% sure, but I think it the FX series won't work on the /B-boards.

I have the non-B board. I'm thinking about replacing the BIOS with the B version on at least one board, don't need two for benching

Go ahead, I was just browsing my valid.canardpc-profile looking for reference clock scores Not sure what the limit is of my board, as it wasn't what I aimed for back then.

Ticket ID: 931 Priority: Low http://valid.canardpc.com/show_oc.php?id=1145383 - if you need more information about the board, let me know.\r\n\r\nPS: I will post some scores with an FX64 and an AMD Le-1640 Lima core soon, FX64 is like FX62 but 3GHz stock, and LE-1640 is like LE-1660, but runs at 2.7GHz stock and has 512kb cache. You may want to relabel the old one to lE-1640 Orleans while you\'re at it, and call this one LE-1640 Lima

If you have an orb link, post it. Massman can prove his results are OK by having screenshots that have no errors. Very simple.

Maybe I should just kill it to piss off everyone I'll give it a spin when I've got the REX tweaked properly:) But I want to finish some AMD stuff first. Some guys were messing with me lately

CSX Duablo + REX arrived a few days ago:D

If you find some (more than one) similar results from other userS, or find a way to reproduce the graphical errors, then that will be discussed, I guess. I don't mind of there are other mods who will take part in this thread, but why do you need EVERYONE to take a look at this?

The CPUZ artifact is very coomon, and the 3DMark01 artifact that's being discussed here is unique. If this happens at relatively low clocks like in this case, then there should be plenty more of the same type out there.

If you find some other submissions that have this artifact, drop me (or some other mod) a PM or post it here. If it's a bug that happens when you push the cards, then there should be more of these.

Ty:D Not completely finished with this category, though... Maybe I can up the multiplier one notch:)

Since they are server chips, it's reasonable to assume they're more carefully selected - which IMO is good enough to make a separate category.

I always run on DFI Expert or Venus (usually Expert unless I really need to squeeze out the last few MHz). They suck for TCCD, though, nearly killed my board because of some maxxmem-testing

I think it's alot more, as I don't do PCMark, and I still haven't benched X2 3800+ manchester, which is good for alot of boints. I'll calculate it later tonight Grabbed 4 more boints today (was missing 2 3500+ San Diego gold cups ). We'll see what I do later, maybe I'll give my Opteron 165 a last spin before I sell it (I got a feeling it has degraded a bit, after I ran wprime at 1.81v and ~0c a couple of years ago )

I have 1000 within my reach, still lots of superpi 32m and wprime 1024m to run. Boring, but every boint counts:p

Maybe you can add some box for the sockets this applies to - like "socket xxx to yyy adapter used".

Allright... I thought it was some info that was accidentally "erased" by the edge of another window or something. This is quite different, I'll send him a PM.

Common sense applies for minor mistakes like these. If the score looks fair enough for the clocks, then it's probably legal. However, it it looks a bit weird, a detail like that may be enough to take it down.

CAn't both be implemented? I mean... you go by motherboard first, and if there's no mobo info, you go by the CPU?

Uhh, I have am2+ as well? LOL. Ahh, I forgot I still have some work to do on 754 Will do it next month:D

I just blocked it You're right, I don't think this can be a proper run.