AMD Phenom II 940 "Xtremely" Benchmarked

Page 14 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
I've heard that one won't be able to actually buy a Phenom II from a retailer in the U.S.A.(like for example newegg) until the middle of February of 2009.So that makes it nearly 2 months away.
 
The DDR2 Denebs at 3.0 unlocked and 2.8 locked were supposed to be out January 8th.

I can't see a European or Asian launch a month before North America. The DDR3 Denebs won't be out till February, and the new Ph2 Athlon's afterwards. We'll see if they're delayed.

 
http://www.ocxtreme.org/forumenus/showthread.php?t=4302

Phenom II
p2.JPG


Benchmark Setup
AMD Phenom II X4 940
MSI DKA790GX Platinum
ASUS 4870 TOP 512MB
Kingston PC2-8500
Gigabyte 800w Odin Pro
Patriot Warp V2 32GB SSD

Stock settings.

SuperPi1m
spi1m.PNG


SuperPi32m
spi32m.PNG


Cinebench 10 (32-bit, like all tests, unfortunately)
cinebench.PNG


WPrime2.0
wprime2.PNG


Aquamark 3
aq3.PNG


3Dmark06
3d06.PNG



Special regards to 'coolice'.

 
AMD has been pushing the platform and fusion concept for a couple years now, and with Dragon it looks like they are taking it to the next level. As they've said, fusion isn't just about cooking the gpu and cpu side by side on the same piece of silicon. So one would expect their are benefits from pairing an ATI gpu with an AMD cpu on the same board. Maybe a future catalyst driver will take advantage of this by opening up the sideport, which would really push the platform concept to the forefront. There are some very credible sources stating how impressed they are with Deneb.

Now when Little Dragon hits, imagine the platform advantage AMD has with an Athlon X4 Neo at ~140mm^2, a Little Dragon and an excellent low priced motherboard. Or how about an Athlon X2 Neo at under 100mm^2, with Little Dragon.

It's not too hard to envision the direction AMD is heading. To me it looks like they want to redefine the performance of the value and mainstream segments, and with die sizes and performance like that it looks like they can do it very easily.

It seems to me that many people are still thinking of AMD and ATI as two seperate entities. Just because the process isn't small enough yet to marry the gpu and cpu on the same silicon doesn't mean they can't leverage the massive advantage they have in owning ATI.
 
looking at the 3dmark06 scores above it has me scratching my head a little. both the Deneb and the q6600 score that dattimir linked. i think those scores seem a little skimpy on both for quad cores. i was hitting higher with a single 260 and a dual core. i know 3dmark06 is a piss poor indicator but generally quads do better than duals since it is so CPU bound.
 
Im thinking pie has hit it square on the head. A 40nm gpu cheap, with 4830-50 capabilities, on a smaller new Athlon quad with low cache, high stock clocks, cool, cheap, nice performance, no fancy mobos, itll be a great mid system
 
Dont know what kind of motherboard/memory/ambient conditions they are running in, but my Q6600 when I had it at 3.6ghz with a 4870 scored P9100 and some change.
 
Kinda looks like 4 should be a piece-o-cake on air. Still no core voltages? Wonder what gives with that. Ya know the worst part of those screen-shots? Those boys are getting a check to play with those new II's. You and I would pay to get that chance.LOL
 
OK, I may be way off here, but when the 4870 was released, gpuz recognized it as having 480 shaders, or 320 shaders. W1zzard had hinted all along about 800 shaders, but very obscurely, and his gpuz read different from fact. Just wondering if this is another case of this or not? It is possible
 
http://www.youtube.com/watch?v=b248iTbplag

The Phenom II Air 3.9Ghz_Liquid 6.3Ghz

Top overclockers Sampsa and "SF3D" from Finland take on Team AMD in overclocking competition at event in Chicago, Il. Results include air-cooled, overclocked AMD Phenom II running Crysis Time Demo at 3.9GHz and up to 6.3GHz using liquid nitrogen cooling. 12/9/08.
 
Phenom's 9950 clock is equivalent to 86.6% of that from Q9650, but scores roughly 73% of what Q9650 does, so, it's more like a 12% difference, I suppose?
 
Im going 600Mhz difference. Then dividing that difference into 2.6Ghz, and adding that % to 73.4%

Something like 4 is baseline. If you have 3, then 33% slower, adding the 33% makes it whole, or 4. Being theres a difference to begin with, as in 2.6 vs 3.2, you have to use the base, right?
 
By the way, looking a little closer at those scores, most of the ones who help Phenom's 9950 average scores are synthetic benchmarks or some games. So, it's more like it usually scores close to 65% of Q9650 in what 'really matters'.
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom