*NEW UPDATED* Post Your Cinebench R11.5 Score

[taserbro]

5960x at 4.5ghz:

 

[crashtech]

Man, if you could get another hundered megahertz more out of that, you'd take the pole! I'll get your results in later tonight or tomorrow morning..

 

[taserbro]

Man, if you could get another hundered megahertz more out of that, you'd take the pole! I'll get your results in later tonight or tomorrow morning..

Actually I did but I'm not going to submit that since I can't pass realtest's stresstest at this frequency.

I could probably go even higher but the problem is, I can't control the temperature enough for my lack of gonads to run the full stress test through. When I see the temps suddenly spike through the mid 90 degrees after just 10 minutes, my heart skips a beat and I panic shutdown the program...

I guess this is the limit of the amount of heat the swiftech h220-x can deal with unless I can add another radiator to it or even go full custom loop but that'll be a project for much later.

 

[crashtech]

I won't enter into the list if you don't want me to. It's within the rules, though...

 

[Flapdrol1337]

Cpu doesnt want to do 4.9 unfortunately, I tried

 

[crashtech]

I'm thinking about splitting the quads into two categories, 4C/4T and 4C/8T. It may also be more fair to put the 8T FX CPUs into the 8T category and rename it "4T/8T and 4M/8T," but I would welcome input from FX owners before implementing this change. I may also separate the 2T/2C CPUs from the 2T/4C ones at some point.

 

[biostud]

Personally I think it's fine the way it is.

 

[SlowSpyder]

I'm thinking about splitting the quads into two categories, 4C/4T and 4C/8T. It may also be more fair to put the 8T FX CPUs into the 8T category and rename it "4T/8T and 4M/8T," but I would welcome input from FX owners before implementing this change. I may also separate the 2T/2C CPUs from the 2T/4C ones at some point.

AMD calls them cores, Cinebench doesn't. I personally don't care where I am in the list, the score is what it is regardless of how it is categorized. I don't mind staying in the 6+ because if nothing else, AMD tells us these are cores, so that's how they're positioned to compete, at least by how they're labeled.

But then again Cinebench doesn't see them as cores and this is the test we're running. < shrug > Don't care personally, if it's less work for you leave it as is.

 

[crashtech]

When the lists get too long I find them cumbersome in this format. Maybe someday I will get them all into a spreadsheet.

 

[guskline]

Results from my 3930k @4.6Ghz
The CPU-Z post is

 

[crashtech]

 

[rvborgh]

Latest results for my quad opty home project PC:

Its the baseline after updating to 4 Opteron "Magny Cours" 61xx "Extra Spicy's" (4x12 K10 cores). These are running at stock 2.1 GHz.

26.83

video of the run:
https://www.youtube.com/watch?v=pvXkhovF8OI

 

[crashtech]

I'm not yet certain whether I will put an engineering sample in the list. But while I am considering it, you might want to manually set the number of threads in the File>Preferences menu to 48 and see what that does to your score.

 

[rvborgh]

Hi Crash,

I decided to play back my video and actually count the # of cores (squares) and it did come to 48... so there seems to be a display bug with CB 11.5. i did make sure it was set to 48 in the preferences before the run.

AFAIK, all the Magny Cours (Opteron 61xx) chips are pretty much the same... just binned differently and named differently with different max freqs and voltages set in. The ES just have unlocked mult and voltage. The stock GHz on these is 2.1 GHz so effectively these would be equivalent to an unlocked Opteron 6172. Hope that helps with your consideration.

Oh... i should add. The interesting thing about these quad Opteron setups... you can mix and match Opteron chips so long as they are in the same series (ie 61xx, 62xx, 63xx)... and on initialization... all 4 will run at the speed of the lowest common denominator. So its possible to mix and match ES chips with the stockers...

I'm not yet certain whether I will put an engineering sample in the list. But while I am considering it, you might want to manually set the number of threads in the File>Preferences menu to 48 and see what that does to your score.

 

[Mydog]

Almost got to 21 points

 

[rvborgh]

Had a bit of time to see what these new Opteron chips would do at stock 1.1v... 2.6 GHz... its completely stable... didn't really want to take it much higher or increase voltage because of the two dynatron coolers on the top two processors (they are more of a stocker type of cooler, and don't work nearly as well as the Noctuas).

power consumption was just over 660w during this Cinebench run:

32.78

 

[SlowSpyder]

Had a bit of time to see what these new Opteron chips would do at stock 1.1v... 2.6 GHz... its completely stable... didn't really want to take it much higher or increase voltage because of the two dynatron coolers on the top two processors (they are more of a stocker type of cooler, and don't work nearly as well as the Noctuas).

power consumption was just over 660w during this Cinebench run:

32.78

:thumbsup:

 

[rvborgh]

Latest results for Quad Opty at 3.0 Ghz (slight voltage bump):

37.21

power draw was around 940w during this run. Had to bump voltage to 1.175 for proper operation at 3.0 GHz.

video for folks that like boxes moving around

https://www.youtube.com/watch?v=p47cJijwfmc

 

[crashtech]

Almost got to 21 points

Awesome! What is your cooling setup?

 

[crashtech]

Tweaked some settings and added an H110.

 

[jason166]

I love the video... That's a mean setup..

Power draw @ 940w? I think you need a better cooling setup

 

[rvborgh]

Hi Jason, glad you enjoyed the video yeah it drew that at the wall... but surprisingly the chips themselves don't really get much past 50C. When i was running them at 2.6... i don't think they even broke 50 and that is with the top dynatrons (which aren't really a cooler you want for a non stock setup)... the lower two processors with Noctua coolers ran around 47C... very strange to me but i'll take it. At some point i think i'll try for 3.2 GHz... but i need another two Noctua UD9, some heat sinks on the VRMs and extra fans blowing on them (especially the one up in the upper left corner).

I love the video... That's a mean setup..

Power draw @ 940w? I think you need a better cooling setup

 

[SlowSpyder]

Latest results for Quad Opty at 3.0 Ghz (slight voltage bump):

37.21


power draw was around 940w during this run. Had to bump voltage to 1.175 for proper operation at 3.0 GHz.

video for folks that like boxes moving around

https://www.youtube.com/watch?v=p47cJijwfmc

Gotta say, hearing those fans spin up and all those boxes chew through the benchmark is pretty cool!

 

[Mydog]

Awesome! What is your cooling setup?

High-end water cooling, two 360x60 mm radds and two 280x60 radds, 16 2k rpm fans and a Hailea water cooler/chiller. All components are hooke into the loop with QDC's
Atm I'm waiting on the 980 Classified so I got no GPU on water cooling but as soon as the classies and blocks arrive I split this loop into two. Thanks to the QDC's that only takes 15 mins and no leak-testing needed.

This is what it looked like with the R4BE and two 780 Ti Classies

 

[rvborgh]

hi, i kind of like the sound as well although its not intentional i just didn't have anything but the smaller Dynatron A6's to slap on there. The Noctuas are almost dead quiet... i'll be replacing the Dynatrons with Noctua's soon and replacing 4 of the fans on the Noctuas with Noctua's latest PWM fans... as they have a higher potential flow than the constant speed fans on there now. So there should still be a bit of a whir.

btw, if you like that sound you'll also like the sound of the twin screw supercharger kits i used to make... check out the dyno run on my channel (customer of mine dynoing his kit back in the day) :biggrin:

Gotta say, hearing those fans spin up and all those boxes chew through the benchmark is pretty cool!