'Nehalem' 2.93 GHz Benches Revealed

[kidswithguns]

[citation][nom]jimmysmitty[/nom]Right now on a clock per clock Q6600 vs a Phenom 9750BE, the Q6600 gets a higher 3DMark CPU score. So whats that tell ya?[/citation]

This is a plain fact, and it can not be denied, AMD fans please accecpt it.
Quad 6600 > Phenom 9750 clock vs clock
And Nehalem > QX6800 clock vs clock
=> Nehalem > Phenom clock vs clock.
Dude, this is so easy to understand.

 

[xenol]

I don't know if these benchmarks are good or not. I looked over at my 3DMark 06 and PCMark 05 benchmarks for my new PC (E8400, P35 motherboard, 8800GT 512MB, 2GB DDR2-1066) and the Nehalem system beats mine by like 10%-15%. Sure, I'm not using the same OS, but unless Vista cripples the 4850 to 8800GT levels, then wow.

 

[gto127]

Based on the phenom score of 3.5% increase for 4% more clock you could predict a score of 8695 at 3ghz for phenom. Still a good choice for the money. Intel has a very good product now but I feel that if AMD could rework the core to get rid of L3 and double the L2 cache they would be competitive again due to lower latencies and a wider memory bus than they have now.

 

[silicondoc]

I know it must be tough, but the 14% 24% and 31% doesn't do it for me. I'm just so underthrilled. As usual, someone could just take the E8400, overclock it to 3.8 or 4.4, and be ahead anyway.
It's like the videaocard jumps... underwhelming... the G80 768 ultra still kicks in some cases. It just takes forever, while they piddle out barely above 2 digit increases...
I guess the idea of a breakthrough is moot.
I note on the videocrads, they keep the crappy 128 bit bus, or make it 256, when they've aleady released 512bit , and have 448 bit... etc.
At least ATI jumped to ddr5.
Now it will be 5 more years of ddr3 anyway, while they still try to sell ddr2 512 megs for 150 bucks to fools.
Sorry not impressed.
Nehalem - bllpppbpbpllpbpbpoppppoooooooooooop - who cares.
I thought those who had to brag about folding - poor fellers - since nothing much uses quadcore ... what a ding dang gimmic... 2 cores actually did help... now it's whackin along the looney line on fluffy hype...
( Yes, some of this is sourpuss, I'd be singing a different tune if I had 5 grand to blow on this stuff evey 6 months )- LOL take that into account but yeah I meant every word.

 

[silicondoc]

AS far as the arguments on what does this test and article really mean, heck they had little time, just look up the scores you are interested in they're on this site. No I'm not doing it, I'm underwhelmed.
I think the manufacturers design, research, and produce for the big boy servers markets, where all these cores can be used in meggawhomper multi cpu setups, then we get the steam and urine leftover off their silicon presses. LOL
Ya know it's true, noone wants to say it but that's the way things are.
LOL
Yes, I'll take the prozac now or slug down a shot of whiskey. rofl
What crap.
I still say some of us get together and make a laptop with the latest vidcore, and a harddrive sound and ram and USB ports only(anything attatches for cheap). Then CRUSH the googleheadsaleswonks with all their crap unvalue added junk on every laptop, selling em for 600 bucks each - 300 million sold in a quarter...
Come on, where are the motherboards that do that too. How about some REAL VALUE for the enduser ? ( Don't say e-pc or whatever 12 inch tinbox they called it.)
3 grand for a decent gaming laptop ? Oh really... the videocard setup costs 2 grand ?
Same with this nehalem - it's just more buffoonery. Overclocked E7200 probably smokes it in gaming and everything but "folding" and "stealing dvd videos".

 

[Guest]

Their setup must be screwy, the QX6800 is much better than that. Look at the old cpu reviews, it gets a 9458 with lesser components.

 

[Guest]

OK, if Nehalem is 11% faster than an X6800, and Penryn is (supposedly, read it here on Tom's,) about 10% faster than Conroe, does that mean that Nehalem is 1% faster than Penryn?

AMD will increase instructions per clock and overall clock speed with the forthcoming AM3 and 45nm, and it looks like won't take that much of an increase to catch Nehalem... and contrary to what some are saying, there won't be any massive increase in performance before Nehalem is released, if the mobo manufacturers are testing it, it's more or less a finished product.

 

[wingless]

How about a comparison of a Phenom 9950 BE @~2.93Ghz against the 2.93Ghz Nehalem, or a 2.6Ghz Nehalem against a stock 9950 BE? I would like to see the exact clock for clock difference and make a comparison there. It will be interesting what AMD can do with the K10.5 at this point. If they can pull off the performance they said, then there should be less than 10% difference between K10.5 and Nehalem clock for clock (roughly).

 

[wingless]

[citation][nom]wingless[/nom]How about a comparison of a Phenom 9950 BE @~2.93Ghz against the 2.93Ghz Nehalem, or a 2.6Ghz Nehalem against a stock 9950 BE? I would like to see the exact clock for clock difference and make a comparison there. It will be interesting what AMD can do with the K10.5 at this point. If they can pull off the performance they said, then there should be less than 10% difference between K10.5 and Nehalem clock for clock (roughly). [/citation]

Well, less than 10% until we really get into some nice multi-threaded programs. The gaming performance gap will surely be much bigger.

 

[the last resort]

let the flames begin.

 

[Guest]

"
Well, less than 10% until we really get into some nice multi-threaded programs. The gaming performance gap will surely be much bigger."

Games won't benefit from Nehalem any more than the apps that support upto 4 cores.And even then Nehalem will not be as compelling when compared to same clock Shanghai(AMD's 45nm chip).Shanghai will bring upto 3Ghz speeds in Q4 and will be roughly 10-15% on average faster than today's phenom,per clock.
So Nehalem better OC like a mofo or it will be noticeably bigger than Shaghai while performing marginally better in large number of cases and beating it by a good margin in a very few...

And just a side note on "Nehalem silicon" state:it is mosty(95%) finished product and performance is what you see.What is not finished is OCing functions and that is what will be done in the area of BIOS optimizations next couple of months.

 

[onearmedscissorb]

[citation][nom]gxsolace[/nom]onearmedscissorb, i don't think that's the case here.[/citation]

Doh, you and some other people talking about them comparing to a dual core threw me off. I retract my entire former statement, at least as it pertains to this particular article. Sorry TH.

That is actually a QX6800 quad core, and their test makes perfect sense.

 

[onearmedscissorb]

Now I see what the problem is. They need to take out that part where they talk about comparing it to the X6800. It's pointless to begin with, and just downright confusing with the update including the QX6800 quad core.

 

[master9716]

they better make boards that support ddr2 though bc if not im going to B stuck with 6gigs of ddr2. Well once prices drop to around $300 a chip

 

[wingless]

[citation][nom]master9716[/nom]they better make boards that support ddr2 though bc if not im going to B stuck with 6gigs of ddr2. Well once prices drop to around $300 a chip[/citation]

Looks like you'll want to build an AMD Deneb system. Nehalems will be DDR3 only...

 

[randomizer]

The core voltage read by CPU-Z is wrong, the CPU needs at least 1V to operate.

 

[dmacfour]

I'm underwhelmed by these benchmarks, and excited to see if AMD can catch up.

 

[ZootyGray]

quote: onearmedscissorb 07/10/2008 2:22 AM
"This site absolutely baffles me lately. Not that I'm ungrateful that they do tests and take the time to write about it, but on a very consistent basis, there have been glaringly obvious things wrong with the articles that totally defeat the purpose of bothering."


Agree - and this is a crock of caca.

Also funny - Anandtech site did a review on 9950 Black Edition(!!!!) 2 weeks ago - this is the first mention of it here on thg - BIASED MUCH??
And yes they could have oclokd to make a comparison.

Funny = no review - and suddenly 9950BE appears HERE in a comparison w a non-releaseD propaganda schmozzell. - BIASED MUCH??

thg is really starting to suk bad - the forums are more informative than the whole site.

This is really a waste of time and an insult. (spit)

Also the 9800 PHENOM is going to be phased out in favour of the 9950 BLACK EDITION

Nehalem is obviously crawp at that clock and for what lame price $1500??

BANG FOR THE BUCK = AMD OWNS ntel = hahahahahahahahahahahaha!!!!

 

[mdillenbeck]

Article is okay for a "teaser"

Things I am always interested in but rarely see in reviews: energy consumption per hour under no load, consumption per hour under full load, cost of system (cpu + motherboard + ram at least for "core" cost), and dB at 0/1/3 meters.

Then, if one system is cheaper than another while the other uses more energy, I'd love to see the break even point after 3 hrs full load/3 hrs no load per day (assuming an average of say, 10 cents per kWh).

Myself, I've generally found that AMD platforms are fairly cost efficient when building - but I've never measured my energy usage. Guess I'll have to pick up a meter...

 

[justjc]

@kidswithguns I don't know if you notished the fact that there is a update written both before the introduction of other quads and the Phenom x4, so you might owe an appology to asdasd123123 and spuddyt.

About the performance of Nehalem I'm actually a bit dissapointed. After all the QX6800 that have been on the market since april 2007 is only 11% slower clock for clock(PCMark05) not much considering all the innovations of Nehalem such as the integrated memory controller, the QuickPath Interconnect, Hyper-threading and general optimations. Honestly I expected more. I doubt AMD would be seen as making a huge step forward if they don't get more improvement, than 11%, when they present their next generation Phenom processors.

 

[ZootyGray]

Yes, but it's not even a teaser. It's a lamer. More misleading hype for innocent noobeez with too much money to burn. biased much??

Play that nice jingle music, Nihilum inside (lol = nothing), and show the happy families at home all surfing the net at the same time on one pc - such a happy scene!!!

Wow - I love gettin all disturbed over total BS.

LMAO at overpriced junk - of course the price??? $50 right - o yeh.
And then we'll see bozo the clown bragging about his $3K new pooter - and he'll be whining for help in the forums.

"plz Help!!! my nEW NIHILum cra$hed. nEED hELP!!!" - signed Bozo, ntelz latest fanboy.

 

[randomizer]

You know you guys don't actually have to buy this if it sounds that crap to you

 

[theblackbird]

I'm interested in 3 benchmarks:
* A 1920x1200 render of a heavy 3D-scene in 3DStudio Max done by a Nehalem Quad vs 'Old' Quad.
* Average fps in Crysis of a Nehalem Quad vs 'Old' Quad.
* Averge fps in Supreme Commander of a Nehalem Quad vs 'Old' Quad

Until then... *silence*

 

[Guest]

There is much craziness in these comments

'I expected more with QPI...' in a 1 socket system, what do you think QPI is going to do? Why are people expecting huge benefits on a single socket system from QPI and IMC? (Now you can talk about expectations for HT or general core improvements...) Please tell me you are not one of those fanboys that think IMC and HT (hyper transport) is what made K8 so good on desktop... it was the core and the IPC! These things help obviously in server.

'Should be compared clock for clock...' This would be good if both suppliers were releasing products at similar clocks! As AMD is significantly slower, doing clock for clock is merely an academic exercise... now if you want to argue compare similar price points then you might have a valid argument.

'You should OC a Phenom to 3.0'... so OC the AMD chip to near its max, but leave the Intel chip at stock? Again nice academic exercise!

'I doubt AMD would be seen as making a huge step forward if they don't get more improvement, than 11%...'

And how much better is K10 than K8? When you are not comparing 4 K10 cores to 2 K8 cores? It is so great, apparently AMD has decided to ABANDON 65nm K10 dual cores... if K10 was vastly better wouldn't they be making dual cores by now? Please apply your logic to both suppliers!

THG - thanks for the early numbers! It gives a high level view, and given this product is probably another 5-6 months out, I'm not sure what people are expecting. I look forward to the complete review when it is ready (though you will get the haters coming out in full force again)

 

[martel80]

How can you people argue that much about some CPU LOLbenchmark? You can't be serious!
They should have picked something like superPI for single-core clock for clock comparison and some H.264 encoder or cinematic renderer for multithreaded performance.
This article says next to nothing.