Cinebench R15 Benchmark Thread

[CHADBOGA]

Seems this is a relatively friendly AMD benchmark.

 

[Durp]

Might want to fix that chart, Makaveli has a 970 not a 920. BIG difference.

 

[Makaveli]

Might want to fix that chart, Makaveli has a 970 not a 920. BIG difference.

about two cores worth of a difference

 

[Edgemeal]

OK, lets try this again,

EDIT
Scores updated and posted here,
http://forums.anandtech.com/showpost.php?p=35610521&postcount=80

 

[inf64]

I forgot to attach the image in my "report".
Here it is:


750K 4T@4.3Ghz, 1600MHz DDr3 Cas8

 

[Makaveli]

OK, lets try this again,

looks great edgemeal thanks for putting in the work as always

 

[Burpo]

Yes, thanks edgemeal.. The work is appreciated.. nice graph!

 

[Durvelle27]

FX-8350 @4.8GHz

 

[Engineer]

Xeon E3-1230V2 with 16GB of DDR3-1333 (4 x 4GB)

Intel 3770 with 32GB of DDR3-1333 (4 x 8GB)

All stock speeds.

 

[bononos]

We are not in 2008 or before anymore. Its time to move on.

And since you must have skipped it:
http://www.realworldtech.com/forum/?threadid=135978&curpostid=136051

So using ICC with Cinebench is an advantage for AMD CPUs vs Intel CPUs. AMD will not be any better without, it would even perform worse in relation.

Simply dismissing the issue by saying its from 2008 doesn't make it go away. Intel was crippling AMD in its compiler, got caught by anti trust investigators and had to pay a large fine. But icc still hobbles AMD according to Agner Fog.

The RWT link does not give any evidence, only one Behne asserting that the dispatcher does not discriminate which directly contradicts Agner Fog's (and others) findings.

"To be more specific: With the (SSE2) compiler setting used in CINEMA 4D and CineBench 11.5, the speed advantage of ICC over MSVC (roughly 15-20%) has been slightly bigger on AMD cpus than it was on Intel cpus."

From what I read from the above, the icc for AMD cpus is faster than msvc, but all it really means is that msvc is bad for AMD. Not that icc is equally good for AMD and Intel cpus.

 

[YBS1]

Ok, so we know you think Intel are a bunch of dirty cheaters. Yet still they hold a performance advantage (sometimes massive) in virtually all real world software. So how does this make me want to buy an AMD cpu again? Go ahead, dig deep and cherry pick...I'm sure you'll come back with some obscure piece of software no one has heard of where AMD has a slight lead.

 

[TerryMathews]

Ok, so we know you think Intel are a bunch of dirty cheaters. Yet still they hold a performance advantage (sometimes massive) in virtually all real world software. So how does this make me want to buy an AMD cpu again? Go ahead, dig deep and cherry pick...I'm sure you'll come back with some obscure piece of software no one has heard of where AMD has a slight lead.

You realize those aren't mutually exclusive points right?

Intel can be the better CPU maker and a cheater. Which, I will point out Intel distributes a disclaimer with ICC that it may not fully optimize for non-Intel CPUs and may not take advantage of features that are supported by the processor.

The fact that Intel is faster makes this especially sad.

 

[YBS1]

You realize those aren't mutually exclusive points right?

Intel can be the better CPU maker and a cheater. Which, I will point out Intel distributes a disclaimer with ICC that it may not fully optimize for non-Intel CPUs and may not take advantage of features that are supported by the processor.

The fact that Intel is faster makes this especially sad.

Good point, and I'm aware of that, but doesn't that sound like AMD has some work to do? In the end though, the Intel end user still has the better experience/performance.

 

[TerryMathews]

Good point, and I'm aware of that, but doesn't that sound like AMD has some work to do? In the end though, the Intel end user still has the better experience/performance.

It sounds like Intel should do the ethical thing and honor its cross-licensing agreement in spirit especially since the market is shifting so that they are benefiting from the technology that AMD provided more so than vice versa (AMD64).

With the desktop PC market withering, they should each spend some more time and money working together and less trying to slash each others throats before there is no desktop PC market yet.

And, in the interest of fairness, I will say that AMD's Mantle on the surface seems to be a load of hogwash as well. So by no means is AMD innocent. They both act like spoiled children.

 

[Edgemeal]

Yes, thanks edgemeal.. The work is appreciated.. nice graph!

You guys are welcome, but I don't see the point in updating it anymore, what started out as a benchmarking thread is now anything but and will just get closed anyway.

 

[Engineer]

You guys are welcome, but I don't see the point in updating it anymore, what started out as a benchmarking thread is now anything but and will just get closed anyway.

Regardless, thanks anyway!

I just rebooted my 3770 to see if I could get to 700+ (because I really wanted that extra bit) and made exactly 700.

No other changes other than a reboot (it had been about a week between reboots):

Much appreciated.

 

[YBS1]

5.1/1866

 

[BallaTheFeared]

 

[BallaTheFeared]

4,875Mhz with 1.25 bootstrap (thus 2500 mem)

 

[Shellls]

Athlon x4 651k @3.6ghz 4GB 1333
MT 352 cb

 

[SPBHM]

i3 2100 (3.1GHz) with 1333 memory
MT 266
ST 106

 

[Durvelle27]

FX-8350 @5.02GHz

 

[Makaveli]

FX-8350 @5.02GHz

Great job on hitting 5ghz but man that is a scary looking vcore.

what are you temps and is this a 24/7 stable overclock or just benchmark stable?

 

[Durvelle27]

Great job on hitting 5ghz but man that is a scary looking vcore.

what are you temps and is this a 24/7 stable overclock or just benchmark stable?

actually with the vcore i have its pretty good for a 8350. Most need 1.56-1.6v just to hit that. Temps hit 65c and this use to be a 24/7 OC but since adding my 7970 into the loop 4.8GHz is now my daily clock

 

[BallaTheFeared]

What does your kill-a-watt tell you, red alert?

But seriously, what is your draw do you know for 4.8/5?