PCPER.com: 5 GHZ+ Phenom II Overclock on Dry Ice, 6Ghz on LN2

[Ghostbuster]

wPrime 32M http://enthusiast.hardocp.com/articl...hlbnRodXNpYXN0

[informal]

Guys don´t you think you are a little overhyping deneb? i mean, i would really really really LOVE to see i7 being destroyed by deneb, but there is like 1% possibilities of that happening *snif*

Why don´t we let it be and wait instead of creating higher hopes and in the end deneb will be a flop because it didn´t deliver what it promised?

Nope,nobody is talking about core i7 being destroyed.Justapost just posted( )a link to a planet3dnow thread where a guys who was at teh AMD event saw some numbers from both 3Ghz deneb and i965.He didn't gave any more details apart from saying very competitive (perf. of the two).Like justapost already said,it could very well be a HT off/turbo off situation ,which in my view would be rather lame to do(since both of these features are out of the box for core i7).
Anyway,a month or so to go before we know the real results.Or someone gets a Deneb on 16th of December and breaks the silence

wPrime 32M http://enthusiast.hardocp.com/articl...hlbnRodXNpYXN0

Here's what Yorkfield and Agena at 3Ghz achieve in wprime(source:madshrimps.be):

[justapost]

Thx, so round 9,7s with HT off. Means deneb must be ~29% on IPC to be onpar. Dunno if they ran 32m.
I assume they ran the i965 system with turbo and ht of to give an IPC comparison and here deneb was able to compete (can mean was not destroyed). Sry about that cache crap. Sad to see amd lost that benchmark.

EDIT:
i965 3.2GHz HT off 9.7s / 9950BE 3.2GHz 11.7s / PII 3.2GHz => ~11s (assuming 6% improvement here) => i7 ~10% faster on same clock with ht off in wprime.

[Zucker2k]

IF AMD brought the i965 to the demo then they're on to something. Though, most of the little info there is seem to concentrate on the overclockability of P2 (which does make sense) but if it was a head to head demo, then the NDA has been very effective. At this point, I'm confused as to where exactly AMD plans to target its new chips, that would be the biggest indicator of their performance compared to the competition.

[Ghostbuster]

Here's what Yorkfield and Agena at 3Ghz achieve in wprime(source:madshrimps.be):

This benchmark seems to favor memory bandwidth and cache. At 32M Phenom takes the lead but at 1M the Yorkfields took it back.

Edit: Certainly that big bandwidth on Core i7 made some difference even with HT off.

[informal]

IF AMD brought the i965 to the demo then they're on to something. Though, most of the little info there is seem to concentrate on the overclockability of P2 (which does make sense) but if it was a head to head demo, then the NDA has been very effective. At this point, I'm confused as to where exactly AMD plans to target its new chips, that would be the biggest indicator of their performance compared to the competition.

Early pricing seems to be on par with i920(a920 vs i920).At least thta's what this article suggests.

[justapost]

This benchmark seems to favor memory bandwidth and cache. At 32M Phenom takes the lead but at 1M the Yorkfields took it back.

From hwbot

WPrime is a simple, easy to use multithreaded benchmark application that can quickly test your processor performance. In contrary to most other simple benchmark applications, wPrime is written to take full benefit of processors with multiple cores, like the new Intel Core 2 Duo or AMD Athlon 64 X2.

Source is optimized for C2D and AMD 64 X2.

[informal]

This benchmark seems to favor memory bandwidth and cache. At 32M Phenom takes the lead but at 1M the Yorkfields took it back.

Nope,wprime is a specific multithreaded benchmark that uses a recursive call of Newton's method for estimating functions while super pi is an old singlethreaded test(a poor one while we are at it,a lot slower than other methods for pi number calculations)which likes big L2(dataset) and uses specific old x87 instructions that have very low latency in Yorkfield(Merom/Nehalem) case.That's why you see totally different numbers from these two tests.

edit: justapost was faster

[Ghostbuster]

Over here the Q9550 is a bit more expensive than the Core i7 920.

EDIT: Here's another one for comparison


Source: http://www.custompc.co.uk/reviews/60...l-core-i7.html

[Ghostbuster]

From Ex-INQ http://theovalich.wordpress.com/2008...less-than-400/

After dropping the bombshell last week, more details about AMD’s exciting fight back are slowly coming to life. Several e-tailers decided to get under the limelight and listed pricing of upcoming Phenom II processors.
As well all know, the Phenom II will be launched on the first day of CES 2009 (January 9, 2009), and the launch models are 920 (2.8 GHz) and 940 Black Edition (3.0 GHz). According to Australian website Austincomputers, the price of Phenom II 940 Black Edition is set at 532.40 Australian dollars (€264/US$342). It does not stop there - TDshop.it , Italian e-tailer who listed prices for both Phenom II 920 and 940. According to Italian store, price of 920 is set at 251 Euro, while 940 will set you back for 299 Euro.
As you probably know, Sales Tax/VAT in EUrozone is extremely high, in the range of 15%+. For instance, Croatia has 22% and that isn’t the highest on the continent. Thus, if we deduct VAT from that Italian store, we get following:

* AMD Phenom II 940 Black = €245 = US$318
* AMD Phenom II 920 = €205 = US$266

There you have it. Now, we can get clearer picture at the cost of upcoming 45nm processors from AMD.

[Macadamia]

This pricing is much better, LOL.

[Zucker2k]

This pricing is much better, LOL.

It should come down even further by launch as I see an Intel response in the near future.

[duploxxx]

Just in... AMD Phenom II Bring Us No Exciting Improvements



The 3DMark CPU score does tell us a bit, but why compare to Q9400 and Q9300 only? And are those integrated graphics gaming scores?

people make so much fuss about these stupid pics, where it is even not confirmed that these are from AMD, let alone the information from the site is big fanboism and wrong.

And yes if they are comparing a 790gx against a G45 with any nice blue cpu you want i am sure you get such results but they have 0 added value to the market.

[largon]

^
Obviously, with a fan.

¦¬

[Rammsteiner]

^
Obviously, with a fan.

¦¬

Any chance it was like -30C in Finland lately though Then it might have been fanless.

But more serious though, think indeed a fan is needed. That's what making such heat sinks good, to exchange heat quick to the air. If it would have been fanless it would have exchanged it way to slow, but actually I dont think Ive to explain much about that though

[kompulsive]

Lord I hope it's true. Back to the days of K6-2 glory!

[Calmatory]

Lord I hope it's true. Back to the days of K6-2 glory!

...indeed, which was beaten by Intel counterparts. Now waiting for K7 and K8...

[kompulsive]

...indeed, which was beaten by Intel counterparts. Now waiting for K7 and K8...

Pffff...there was a solid period of time there when the K6-2 was the pound-for-pound, dollar-for-dollar champion.

[Donnie27]

Pffff...there was a solid period of time there when the K6-2 was the pound-for-pound, dollar-for-dollar champion.

No it wasn't! Then it had crappy VIA motherboards as well. Wayyyyyy too many folks blamed Data corruption on Win95 and Win98. Thank goodness AMD isn't that same boat.

[Rammsteiner]

No it wasn't! Then it had crappy VIA motherboards as well. Wayyyyyy too many folks blamed Data corruption on Win95 and Win98. Thank goodness AMD isn't that same boat.

Not like Windows 95/98 were that solid and stable anyway

[SAE]

I never had data corrupion with my several via boards... My first AMD was a 80386, then I skipped the 486-geneation with a P133, afer that the great K6-2 times began... K6-2 300 was replaced later with my beloved K6-III 400 beating every higher clocked celeron around (compared Q3Arena and such, using a Riva TNT) - dunno if that gem was able to beat these great P2-400, but it performed similarly well nevertheless.

Win98 SE was very stable for me BTW, but everything went crap w/ WinME.

[Calmatory]

At least my K6-2 @ 350 MHz has no chance against Celeron(Mendocino core) @ 333 MHz. Thanks to integrated, full-speed L2 on-die in the Celeron. Though, P2's are different thing with their half-speed L2 on the PCB, let alone K6 with on-motherboard L2

[Ghostbuster]

Had a K6-III 450MHz before, and I would say the Pentium-II 300MHz smokes it.... of course my Pentium-II was running at 450MHz (its was a SL2YK, the 450 in a 300 disguise).

[Zucker2k]

Not like Windows 95/98 were that solid and stable anyway

What kinda argument is that? I thought we were talking about hardware? Did it dawn on you Intel based chipset boards and cpus had to run that OS too?

Win98 SE was very stable for me BTW, but everything went crap w/ WinME.

Win98 SE was awesome.

[mongoled]

Win98 SE was very stable for me BTW, but everything went crap w/ WinME.

ahhhh winme, yeah winme was my first ever Windows OS, what a great start to using computers! I remember the blue screens I used to get when using the cdrom, a very steep learning curve I can tell you, LOL