cinebench 11.5

[chew*]

Was hoping Zambezi would do better.

Current Top 12 (Excluding LN2 & DICE)

You might as well include DI if your including cascade that can go colder than di

[El Gappo]

CPU @ 5.39Ghz, enabled C6 to tame the heat @ idle

Nicely done

After a recent BD death, really not to keen on over 1.5 above zero....Wont be doing it again.

[demonkevy666]

those thuban up above 7+ have north bridge speeds well over 3.0ghz.

[ReKcOlNu]

You might as well include DI if your including cascade that can go colder than di

I wasn't aware of that, shows how much I know about this stuff.

OK, here's current top 15, see my sig link for all results.

[AMDnord]

You might as well include DI if your including cascade that can go colder than di

this cascade can hold 1090T @ ~ -98°C @ load.
i was just testing to match a Sandybridge of a friend to see score diff :p


so yeah, DI should be included.

[equinox1]

[QUOTE=Dumo;4975415]CPU @ 5.39Ghz, enabled C6 to tame the heat @ idle

nice run Dumo, not sure i can top that. but i will try

[wez]

Aw man, the more results I see, the more I want a BD

[Manicdan]

why has no one bought some ice and give us a 7ghz run of BD yet?

[Zeus]

Like this? http://www.youtube.com/watch?v=-Yzdk...el_video_title

[Manicdan]

Like this? http://www.youtube.com/watch?v=-Yzdk...el_video_title

if hes not on ReKcOlNu's chart, he dosnt count

[Dumo]

Bios 9913

[Daveburt714]

Nice improvement with the new bios Dumo.

Now where's the new bios for Sabertooth...

I assume that's still on phase?

[eRazorzEDGE]

Raja said over on the ROG forums that the 9913 BIOS was just an improvement for 3Dmark 2001 SE. Is that an understatement? Does the chip actually get higher clocks on the new BIOS?

[Dumo]

Nicely done

After a recent BD death, really not to keen on over 1.5 above zero....Wont be doing it again.

Thanks El, wow sorry for the loss

I tried air for about half an hour and realized that it wasn't gonna go nowhere near 5Ghz+ full cores benched

Nice improvement with the new bios Dumo. Now where's the new bios for Sabertooth...

I assume that's still on phase?

Thanks Dave. Yep with SS. It seems bios 9913 has a problem with oc settings, it reverted back to default in windows most of the time

[hyc]

Has anyone done any analysis of scaling on these results? (Sorry, I've only read thru 5 pages of this thread.) It seems pretty inconsistent. E.g., X3@3900MHz: 3.02, X4@3900MHZ: 4.65. 4.65/3.02 = 1.539, 54% higher score with only 33% more cores.

On the other hand, X1@3375: 0.82, X2@3392: 1.58, 93% higher with 100% more cores is closer to what you'd normally expect.

X6@3360: 6.00 vs X2@3392 :1.58, 3.8x higher with 3x as many cores. Again, higher than expected scaling. (Obviously there's many other system factors here, memory speed etc. so numbers aren't really comparable in many cases, but I wouldn't expect memory speeds and such to make more than 10% difference.)

X4@3510: 4.17 vs X6@3500: 6.25, 50% higher score with 50% more cores - just about perfect scaling there. Kind of implies that the rest of the systems were comparably configured.

X4@4400: 5.22 vs X6@4402: 7.89, 51% higher score with 50% more cores - still about what you'd expect.

Do we have FX scores with fewer than 8 cores to compare?

[Manicdan]

is that with thuban or FX your talking about there hyc?

[hyc]

is that with thuban or FX your talking about there hyc?

I was thinking FX-81xx but it would be interesting to see one or two tests for Thuban too. I'd figure Thuban results would be perfectly uniform; for FX it will obviously depend on which cores you use (same module or different module).

[Manicdan]

I was thinking FX-81xx but it would be interesting to see one or two tests for Thuban too. I'd figure Thuban results would be perfectly uniform; for FX it will obviously depend on which cores you use (same module or different module).

im still confused,
the numbers you have in your post with the odd scaling, are they from a thbuan or fx chip?

[hyc]

im still confused,
the numbers you have in your post with the odd scaling, are they from a thbuan or fx chip?

Those numbers were pulled from the Cinebench results table. They're Athlon X2s and X4s, (and yes, X6 Thubans). http://www.xtremesystems.org/forums/...d.php?t=255323

[Manicdan]

Those numbers were pulled from the Cinebench results table. They're Athlon X2s and X4s, (and yes, X6 Thubans). http://www.xtremesystems.org/forums/...d.php?t=255323

then what your seeing is a wide variety of tweaking and some results missing L3.
going from Athlon X3 to Phenom X6 is going to have better scaling from L3.
other results, like 2-4 cores, could be because of ram or NB speed differences. some guys are experts with CB11.5 and able to get nearly half a point better than i can with the same clocks as me. and i wasnt using stock ram/NB timings either.
if everything stayed the same but the number of cores, scaling would be near perfect

EDIT: thanks for making me look at the chart again, im the second WORST thuban on the list. god i remember those days of using DDR2 and a corsair H50, lol. i think its time i post a score of 7pts, lol

[hyc]

OK, I guess that makes sense. So really the Phenom X4 and X6 scores are most comparable.

DG Lee's thread gives a data point for FX with only 4 cores, too bad his CB11.5 slide's scores are obscured by his watermark.

[Dumo]

OK, I guess that makes sense. So really the Phenom X4 and X6 scores are most comparable.

DG Lee's thread gives a data point for FX with only 4 cores, too bad his CB11.5 slide's scores are obscured by his watermark.

Heres hyc..

[hyc]

Thanks Dumo. But just to be certain, what if you ran your 8C and 4C tests at the exact same clock speeds? Then we should see 8C being exactly 2x 4C result, right? Or 4C result will be higher than proportional, due to no sharing overhead? What about 6C?

[Lightman]

Thanks Dumo. But just to be certain, what if you ran your 8C and 4C tests at the exact same clock speeds? Then we should see 8C being exactly 2x 4C result, right? Or 4C result will be higher than proportional, due to no sharing overhead? What about 6C?

Never look at Cinebench reported clocks, just a hint

Dumo ran both 8C and 4C at the same 5418MHz clock shown in CPU-Z, it just happens that one of the runs he booted @4.49GHz from BIOS and then tweaked clocks to 5418MHz.

[hyc]

Doh.... I see, thanks for the heads up.

So, 8.85/5.25 = 1.6857: the 4 extra cores are only worth 2.7 real cores.