K10 at 2.5 GHz beats QX6800 in multimedia-33 to 66% faster in Sandra Multimedia

[informal]

Fuad gets to look at early benchmark scores in Tunisia it seems

http://www.fudzilla.com/index.php?op...d=663&Itemid=1

A 33 to 66 percent faster

In Sandra multimedia benchmark Agena / Barcelona K10 2.5 GHz scores over 400.000 it/s (instructions / second) in integer test and about 300.000 it/s in floating point test. Yeah there are some Agena / Barcelona parts runing at 2.5 GHz at press time and we got the chance to share the scores with the world.


Intel's fastest QX6800 Quad core at 3 GHz scores about ~300.000 it/s in integer and ~180.000 it/s in floating point test.

So at least in multimedia stuff and this synthetic benchmark AMD's native quad core is 33.3 percent faster than Intel's fastest Quad core, while in the floating calculations it is a massive 66 percent faster.

The QX6850 with FSB 1333 might improve the situation a bit, but it won't be enough save the day. Intel needs to go much higher than 3 GHZ to match the 2.5 GHz K10 part.

Yeah it's synthetic bench,but still impressive(if we know by what margin C2D/C2Q beats K8,it's a good indicative how faster K10 may end up to be when compared to Conroe/Penryn)

[Shintai]

Just a shame even the Sandra bench is fake.

http://shintai.ambition.cz/pics/cloverton1.png

Thats a 8 core 2.66Ghz (ES) with only 1066FSB on a 533Mhz dualchannel and not quadchannel FBDIMM.

Removing 4 cores is down to over 200000 in floating. And thats not even accounting for scaling issue etc.

So..looks like more fud from fuad.

[milkcafe]

Kentsfiled is not a real 4 core CPU

8 core Yorkfield XE 3.73G is waiting for the battle

[informal]

I think that Shintai here wants K10 to fail(and wants that badly ).
Or he wants to see AMD dead.
Neither is good for the consumers.

[dinos22]

I'm sorry but Sisoft Sandra means horse to me in benchmarks.............

why not post some REAL figures instead of teasing like this..........................................we'v e been here before when Intel had Sisoft figures and AMD was smashing them with S939 in real performance >>> it means all

[XSAlliN]

Did he used Glasses when he looked at those charts, cause he has some eyes problems, I got to this conclusion from his past news.

[Shintai]

I think that Shintai here wants K10 to fail(and wants that badly ).
Or he wants to see AMD dead.
Neither is good for the consumers.

LOL, did you miss the part where i PROVED it was wrong?

But I guess you couldn´t see that for the K10 wishes. There is something named reality. And Fuds QX6800 scores aint valid there.

I never even said anything about the K10 scores.

[Getttosmurf]

no source , no screenies , no news

[informal]

LOL, did you miss the part where i PROVED it was wrong?

But I guess you couldn´t see that for the K10 wishes. There is something named reality. And Fuds QX6800 scores aint valid there.

I never even said anything about the K10 scores.

You are the one posting FUD here.The Q6800 scores are perfectly valid!Proof:http://www.hothardware.com/articles/...QX6800/?page=3
integer:299063it/s
float:162836it/s
So Fuads numbers are even a bit higher!

[MAS]

cool

even i didn't expect such sandra score for K10

[ahmad]

Just a shame even the Sandra bench is fake.

http://shintai.ambition.cz/pics/cloverton1.png

Thats a 8 core 2.66Ghz (ES) with only 1066FSB on a 533Mhz dualchannel and not quadchannel FBDIMM.

Removing 4 cores is down to over 200000 in floating. And thats not even accounting for scaling issue etc.

So..looks like more fud from fuad.

I can't even understand what you are saying. Here is a qx6800 sandra shot:

http://www.hothardware.com/articleim...big_san_mm.jpg

299000
163000

And they were using PC6400 4-4-4-12 so I don't know how you can call any benchmarks fake. These are the scores of the Intel quad core, and Fuad says AMD Barcelona is 33%-66% faster.

We take his word until someone has something else to say. For now, thats all we have.

EDIT: LOL 7 minutes late

[Shintai]

You are the one posting FUD here.The Q6800 scores are perfectly valid!Proof:http://www.hothardware.com/articles/...QX6800/?page=3
integer:299063it/s
float:162836it/s
So Fuads numbers are even a bit higher!

http://www.techreport.com/reviews/20.../index.x?pg=12
314217 int.
242087 fp

And add my test above with it.

[ahmad]

http://www.techreport.com/reviews/20.../index.x?pg=12
314217 int.
242087 fp

And add my test above with it.

Nope. Thats x16 Int and x8 float. The other numbers are x8 and x4.

The sandra test in windows 32 is x8/x4, in 64bit its x16,x8. Therefore, that is the more likely result. So you were wrong as you admitted here.

Regardless, barcelona numbers are still higher. At 2.5GHz....

[ted3]

The Kentsfield scores seems about right to me, in tomshardwares interactive chart QX6700 scores ~293k integer and ~159k floating point.
Integer: http://www23.tomshardware.com/cpu.ht...=694&chart=160

E6400 scores more than twice as much as X2 6000+ in Sandra MM Integer, hard to say how important this bench is in real world performance. E6400 also outperforms dual FX-74's in this bench so 33% more from Barcelona over Kentsfield isnt revolutionary, but 66% MM FP would be impressive if true, which its most likely not. If Fuddy-boy had access to Barcelona he wouldnt have posted that fake screenshot would he?

[accord99]

Probably the difference comes from running 32-bit vs 64-bit. Techreport used 64-bit Vista while Hothardware used XP SP2.

[informal]

http://www.techreport.com/reviews/20.../index.x?pg=12
314217 int.
242087 fp

And add my test above with it.

Those are numbers in X86-64 version of the test!

[Vapor]

The way this thread has gone so far....I don't see it being open for long.

Keep it clean guys....

[Brother Esau]

How does anyone make a statement like 33-66% faster? It's either one or the other or a figure in between!

[informal]

How does anyone make a statement like 33-66% faster? It's either one or the other or a figure in between!

Well since those are two different tests within Multimedia benchmark.One is integer and the other is floating point bench.Hence the 33 - 66 %.

[Anemone]

Real scores, real software...

If they have something to show, then show us, with real software. I very much want AMD to survive but they're playing a shell game and hoping people are willing to wait 9 months to get their chips. It's not like Intel doesn't know how these things perform. It's your potential customers who want more solid info.

[perkam]

Post and discuss Fud news in the k10 sticky pls.

CLOSED

Perkam