X4 PHENOM taken to 3GHz

[muzz]

Cleaned up a few posts.

Thanks Boss

m

[macci]

Here is a bit better benchmark (ran under Vista32bit without tweaking) -- single card 3DMark05!



I have a reference score with 3GHz Yorkfield using the same GFX clocks and its 300 points less (ran under WinXP32bit though)...

Here is another Yorkfield comparison score at 3GHZ (this one is with OVCKed FSB & RAM) (by Sampsa);


Pretty close match, eh?

[PhilDoc]

Saw this in the 3dmark section. Very nice and on a slower OS!

[madwelshboy]

That is just amazing!,
congrats to you sir!

[nemrod]

Pretty close match, eh?

Sorry for my probably stupid question but what is it supposed to show?
Is there a huge difference between a yorkfield at 4GHz and a Yorkfield at 3GHz for the same setup 1 HD3870 and the same bench?

[informal]

I don't know if that makes any difference, but there are other tests (Feature tests) ran on the Phenom system, but not on the Yorkfield. Why don't you run both systems on same OS using same GPU drivers and same settings. A 3D mark 2005 validation would be nice. BTW, why MemTest shows DDR2 at 444MHz while CPUz shows DDR 571MHz?

Oh boy here he comes again
The second SS in maccis's post is taken from Sampsa's OC test of Yorkfield with an 3870.You do know who Sampsa is,do you?Somehow i think you don't..
If you wish to question Sampsa's integrity,by all means do it .I know how it will end,for you at least
The Yorkfield in question had OCed FSB and Ram,so it matches pretty closely the situation macci made with his Phenom and RAM.

The Phenom in macci's last post was OCed with 11.5 multi,which represents the 9600 model and is done with Black edition 9600, with a multi set to that value(11.5).This is what i understand from his last few posts.

The point is that Phenom@3Ghz is faster or close to Yorkfield at the same clock speed in this particular test.

[gallag]

The point is that Phenom@3Ghz is faster or close to Yorkfield at the same clock speed in this particular test.

[[STOP THE PRESSES]] Are you trying to tell me that you finally found a bench were a maxed out phenom comes close to a stock yorkfield. Have you alerted reuters?

Just kidding witsha, but you did seem to cream your pants at this rather below average score lol

[Sparky]

[[STOP THE PRESSES]] Are you trying to tell me that you finally found a bench were a maxed out phenom comes close to a stock yorkfield. Have you alerted reuters?

Just kidding witsha, but you did seem to cream your pants at this rather below average score lol

3GHz Phenom = 3GHz york in this test. What's wrong with that? Theoretically a Phenom at 3.5GHz then could equal a 3.5GHz york. If so, it would come down more to how much you want to spend (and whether you are a fanboy of one or the other).

Don't know what is "below average" about the score either.

[gallag]

3GHz Phenom = 3GHz york in this test. What's wrong with that? Theoretically a Phenom at 3.5GHz then could equal a 3.5GHz york. If so, it would come down more to how much you want to spend (and whether you are a fanboy of one or the other).

Don't know what is "below average" about the score either.

YOU ARE RIGHT. I have never looked at it that way before, Just cant wait till the wife gets home so I can tell her to cancel that q6600 and order me a phenom from santa. Yes yes I know that th q6600 will hit 3.5ghz with ease but at least I know that my unstable power sucking buggy phenom would be just as fast it it could have got to 3.5.

I have seen the light

[informal]

3GHz Phenom = 3GHz york in this test. What's wrong with that? Theoretically a Phenom at 3.5GHz then could equal a 3.5GHz york. If so, it would come down more to how much you want to spend (and whether you are a fanboy of one or the other).

Don't know what is "below average" about the score either.

Ah,gallag has just found a way to post in here ,let us leave him with a struggle to comprehend clock Vs clock comparison in this test

[nemrod]

Is it usefull to go back to flaming?

[gallag]

o.k guys. I am sorry for taking the p1ss a bit. Ill refrain from posting in this section until there is something positive to say, because it seems that saying something bad about AMD is worse than naming a teddy bear after the prophet Mohamed (or drawing a picture of him)

[Sparky]

YOU ARE RIGHT. I have never looked at it that way before, Just cant wait till the wife gets home so I can tell her to cancel that q6600 and order me a phenom from santa. Yes yes I know that th q6600 will hit 3.5ghz with ease but at least I know that my unstable power sucking buggy phenom would be just as fast it it could have got to 3.5.

I have seen the light

You either didn't read what I said, or you are just looking at things through those blue tinted glasses

Is it usefull to go back to flaming?

No, and I hope it doesn't come down to that again.

[BeardyMan]

[[STOP THE PRESSES]] Are you trying to tell me that you finally found a bench were a maxed out phenom comes close to a stock yorkfield. Have you alerted reuters?

Just kidding witsha, but you did seem to cream your pants at this rather below average score lol

since when does an air cooled chip equal to maxed out? LOL i say LOL

[macci]

Sorry for my probably stupid question but what is it supposed to show?
Is there a huge difference between a yorkfield at 4GHz and a Yorkfield at 3GHz for the same setup 1 HD3870 and the same bench?

5GHz Yorkfield (with exact same GFX setup) scores 26k. So yes its a benchmark that scales VERY well with CPU power (even with 1GFX card).

[macci]

I don't know if that makes any difference, but there are other tests (Feature tests) ran on the Phenom system but not on the Yorkfield.

doesnt make a difference

Why don't you run both systems on same OS using same GPU drivers and same settings.

I dont have both systems. the Yorkfield result is from sampsa. Its the same GFX driver version and exactly the same clocks though.
I'll run another loop under WinXP later. Or we could run the Yorkfield under Vista (which would result in lower score)...

A 3D mark 2005 validation would be nice.

coming as soon as FM adds Phenom CPU support to their ORB (right now I can upload the result but cant publish it).

[nemrod]

5GHz Yorkfield (with exact same GFX setup) scores 26k. So yes its a benchmark that scales VERY well with CPU power (even with 1GFX card).

thanks.

[Lightman]

@macci: do you know when Futuremark will add Phenom to ORB??
I want to post my scores with CF as soon as possible

[D4rkn3ss]

i don't understand this.. really.. everyone can see that the intel cpus, for now, outperforms the phenons.. why keep posting this over and over again? is like some people feel smart posting the obvious? i suggest keeping this thread as a phenom performance discussion thread. thanks.

[b1tterman]

and the results are like the ones i used to pump out of my fx-55 back in the days :roll

[D4rkn3ss]

Darkness, look again in the pictures provided by macci. The Phenom OC-ed at 3GHz slightly outperforms Yokrfield QX9650 in 3D Mark 2005.

yes it does, but given the brutal overclockability (?) of the intel competitor i don't consider this much an advantage. i use amd systems since the duron ages and the way the things are standing now, i rather buy an intel rig.
but.. i'm curious, and don't plan to upgrade right now. this thread is being quite interesting, and i don't refer to anyone especificaly on my last post, it's just i think it adds nothing usefull to flame poor amd on this thread.

[muzz]

@macci: do you know when Futuremark will add Phenom to ORB??
I want to post my scores with CF as soon as possible

Was wondering that myself, I've seen the Quad Opterons on there already, and their CPU score (per clock) is fine.

Edit- It seems they are Phenoms, under Opteron (just opteron).

[Sparky]

yes it does, but given the brutal overclockability (?) of the intel competitor i don't consider this much an advantage. i use amd systems since the duron ages and the way the things are standing now, i rather buy an intel rig.
but.. i'm curious, and don't plan to upgrade right now. this thread is being quite interesting, and i don't refer to anyone especificaly on my last post, it's just i think it adds nothing usefull to flame poor amd on this thread.

I bolded an important part.

I'm hoping that once the errata is taken care of at a hardware level we'll start seeing stock chips towards 3GHz with perhaps the same kind of overclockability we are seeing out of the BE 5000+ and such (and even better would be nice of course ). I don't expect that to happen all at once, I have a feeling it'll be a rather drawn out process, but with time I hope AMD can get there.

[malvindo]

nice OC

hopefully with 2008 batch, can go near 3.6ghz

TS, what weeks ur CPU?

[VulgarHandle]

this is truly looking like the AXP vs. P4 days, when AXP won clock per clock, but the P4's had the o/c to outperform

let's hope K10 is to the next tweak as clawhammer was to san diego