Anandtech benches Nehalem

[CBONE]

A guy says "I think even the most optimistic person would say AMD is in deep trouble."

and scientia cleverly points out "I would say that anyone who doesn't care about proper testing would say that AMD is in deep trouble."

[SNIP]

At the end of the day though this is all moot because the_ghost suggests that anandtech is pretty much owned by Intel so all this actually means nothing.

Thank the gods for rational folk http://www.amdzone.com/phpbb3/viewto...?f=52&t=135197

Where is the drying-my-eyes-because-I'm-laughing-to-tears-'cause-these-guys-are-delusional emoticon?

"Proper testing"... heh... That's good stuff... Oh, man, here it comes again! Proper testing!

[gallag]

Regardless of how good or bad Nehalem is he's right that the testing done by Anandtech is incredibly bad.

before I looked down at your system specs I guessed you had a AMD cpu. Amazing. Only messing with ya man but how far do you think the river in Egypt will run this time?

No one is saying that this was a full review, not even anand said that but it does give us some numbers that pretty much make it certain than nehlam will be a monster, No?

[JumpingJack]

Regardless of how good or bad Nehalem is, he's right that the testing done by Anandtech is incredibly bad.

Well, not to argue that a carefully pefected study is not completely available... how much data do we have on Deneb given that these proc's are supposedly to come out roughly the same time?

Point is... there are some tests Anand did that are so straight forward even Scientia could not screw them up ... Valve RAD maps, cinebench for example, and considering it was hands on for a day... the amount of info pretty remarkable.

What Anand showed is more than what initially broke at March 06, IDF when Intel let loose some Conroe numbers... Scientia said the same things, and what turned out to be the truth?

Nonetheless, we do not know how Deneb will fair ... so it is in appropriate to call 'AMD dead meat' or 'no more competition'... etc etc. Final numbers will tell the tale.

Jack

[eleeter]

This makes me wonder, now that they finally implemented everything AMD invented so far what will they do next?

By far the biggest benefit Intel has gotten from AMD is from x64. Scary to think Intel told us all that 64bit on the desktop is not needed.

K8 copied the IMC from 386SL and 486SL.

K8 didn't copy anything except the concept. AMD took the memory controller and made it mainstream, right down to the budget desktop.

About a 30% perf increase is what I would have expcted. Nothing exceptional.

30% is an excellent improvement.

This test would be commonly used, if you were only allowed to use one cylinder on 90% of roads.

HAHA good point.

AMD did it to itself. AMD bought ATI without any short return investment.

AMD would be in trouble today if they didn't own ATI. K10 would have little or no platform to run on. Look how long it took Nvidia to field a K10 chipset.

Actually guys, If you nip over to AMDzone scientia and the_ghost have some pretty compelling theories as to why we should not get to excited about this.

I started that thread over at AMDzone.com. I have been a very strong AMD supporter for years, but I see the writing on the wall. Scientia is a very, very knowledgeable person and has excellent technical understanding. But he also tends to find the silver lining for AMD more often than not. I agree with a lot of what he says yes, but in this instance I think he is grasping a bit. AMD IS in trouble. Their last bastion, the server CPU with IMC and multi-socket scaling due to HT is coming to an end.

I really think AMD's biggest problem right now is their fab tech. If they had their ducks in a row, K10 would probably be running at above 3Ghz with decent power consumption. It is damn near impossible to make up for a 500+ mhz deficit, which is what AMD is dealing with. Never mind that Nehalem is bring improved IPC into the mix.

[ak_47_boy]

HOLY

[BrowncoatGR]

Well, not to argue that a carefully pefected study is not completely available... how much data do we have on Deneb given that these proc's are supposedly to come out roughly the same time?

Point is... there are some tests Anand did that are so straight forward even Scientia could not screw them up ... Valve RAD maps, cinebench for example, and considering it was hands on for a day... the amount of info pretty remarkable.

What Anand showed is more than what initially broke at March 06, IDF when Intel let loose some Conroe numbers... Scientia said the same things, and what turned out to be the truth?

Nonetheless, we do not know how Deneb will fair ... so it is in appropriate to call 'AMD dead meat' or 'no more competition'... etc etc. Final numbers will tell the tale.

Jack

Actually its the way they chose to present the data that bothers me. I guess that might have been to keep Intel happy(Obfuscate the actual numbers as much as possible) but really the only usefull conclusion from these tests is that SMT scales very well and that Nehalem performs as expected in multithreaded applications. No doubt it's going to be a really good chip though.
[OT]BTW if AMD copied the IMC from any company that would be DEC actually[/OT] and that's the last time i'll mention AMD in this thread.

No one is saying that this was a full review, not even anand said that but it does give us some numbers that pretty much make it certain than nehlam will be a monster, No?

Stone cold killer( ) in apps that scale well with multiple threads certainly and Nehalem seems to actually deserve that comment

[gallag]

Actually its the way they chose to present the data that bothers me. I guess that might have been to keep Intel happy(Obfuscate the actual numbers as much as possible) but really the only usefull conclusion from these tests is that SMT scales very well and that Nehalem performs as expected in multithreaded applications. No doubt it's going to be a really good chip though.
[OT]BTW if AMD copied the IMC from any company that would be DEC actually[/OT] and that's the last time i'll mention AMD in this thread.

Please explane what you mean, i.e how the numbers have been Obfuscated to keep Intel happy and also how do these tests have no use except proving smt performance?

[BrowncoatGR]

Please explane what you mean, i.e how the numbers have been Obfuscated to keep Intel happy and also how do these tests have no use except proving smt performance?

Well if someone was to publish a full featured review of Nehalem at this point in time with detailed numbers Intel would be very very pissed.

[Donnie27]

Actually its the way they chose to present the data that bothers me. I guess that might have been to keep Intel happy(Obfuscate the actual numbers as much as possible) but really the only usefull conclusion from these tests is that SMT scales very well and that Nehalem performs as expected in multithreaded applications. No doubt it's going to be a really good chip though.
[OT]BTW if AMD copied the IMC from any company that would be DEC actually[/OT] and that's the last time i'll mention AMD in this thread.

Intel owns DEC IP Intel had to agree to license EV6 to AMD or the buyout wouldn't have been approved. I don't think any cares about AMD being mentioned in proper context LOL!

RE: No single threaded benchies? by Anand Lal Shimpi, 3 hours ago
Interestingly enough, none of our standard CPU benchmarks are single threaded at all - even the most benign ones are multithreaded (including the games). I did run some single thread Cinebench numbers though:

Nehalem - 3015
Q9450 - 2396

[Periander6]

Well if someone was to publish a full featured review of Nehalem at this point in time with detailed numbers Intel would be very very pissed.

Same tired old FUD from the AMD fanboys we saw the last time with nothing to back it up.

Nehalem is and will be another C2 style home run.

[Shintai]

Well if someone was to publish a full featured review of Nehalem at this point in time with detailed numbers Intel would be very very pissed.

Oh please drop the conspiracy crap. We heard this over and over again. Here we have numbers that usually match what we got. Its the same rerun all over again from 2006. Intel wont be pissed, why? This is cheap and free marketing of the best kind. Just like with Core 2. Intel gives them the option to test what they want. AMD is lightyears away in terms of competition so there is nothing to hide.

[Nedjo]

Nehalem doesnt equal a BMW M3, perhaps some of the top bins. But Nehalem strecthes all the way down to the lowest CPU.

Absolutely not true!

Nehalem is intended to single digit % of quad's market, and for mission to cutt off AMD from 4-way x86 server market!

It's FUDiculous(TM by me :p) to think that Nehalem will go down to E7000/E8000 or even Celeron/Pentium market segments!

[jas420221]

Just found this and added a nice little quote to my sig...

Hopefully this is motivation for AMD to step up after Phenom!

[GoThr3k]

@ GoThr3k

Did you skip this part or did Anand revise it? Yet, even if performance was the same or slightly less per core, Nehalem ends up being faster due to other improvements. Most apps are Dual Core and WinXP and up uses Dual Core as well. At a minimum, one core runs the OS while another would run a single thread.

To the other guy, this does sound or read like the same comments during 2006 when C2D was tested by this same person, WOW! After over two years, you'd think they'd learn something by Now.

i dont know, maybe i missed it, i think singlethread nehalem wil have a 10-15% advantage, but not more, on the multithreaded side however

[BrowncoatGR]

Same tired old FUD from the AMD fanboys we saw the last time with nothing to back it up.

Nehalem is and will be another C2 style home run.

Oh please drop the conspiracy crap. We heard this over and over again. Here we have numbers that usually match what we got. Its the same rerun all over again from 2006. Intel wont be pissed, why? This is cheap and free marketing of the best kind. Just like with Core 2. Intel gives them the option to test what they want. AMD is lightyears away in terms of competition so there is nothing to hide.

FUD? What the hell did i say besides the obvious? Why the hell would Intel want a full fledged review months ahead from the launch. Sure the rumors and leaked info help keep the buzz alive but a review now would be really bad marketing and anticlimatic because by the time Nehalem will get out it will already be old news...

[jas420221]

FUD? What the hell did i say besides the obvious? Why the hell would Intel want a full fledged review months ahead from the launch. Sure the rumors and leaked info help keep the buzz alive but a review now would be really bad marketing and anticlimatic because by the time Nehalem will get out it will already be old news...

-OR-

It could live up to the hype and make everyone want it??? Marketing 101, Business Management 101, simple stuff here guy.

I mean would you really not want to buy a product b/c its performance specs were shown well before it was available (saying it doesnt change for the worse)? Seriously?

[Nedjo]

Just found this and added a nice little quote to my sig...

Hopefully this is motivation for AMD to step up after Phenom!

Anand has proven to be less, and less informed about what's going on in this industry (I still remember his ridiculous CeBIT "diaries")

This sentence: "2009 should hold a new architecture for AMD, which is the only thing that could possibly come close to achieving competition here" further prove that! As far as I'm aware (maybe I've missed some announcement) but AMD for 2009 only plans to further shrink/refine K10 architecture...

[BrowncoatGR]

-OR-

It could live up to the hype and make everyone want it??? Marketing 101, Business Management 101, simple stuff here guy.

I mean would you really not want to buy a product b/c its performance specs were shown well before it was available (saying it doesnt change for the worse)? Seriously?

You want the WOW effect to happen on or close to release to help maximise sales. Besides whats the point of an NDA on reviews if you wanted all the info out there.

[Shintai]

Absolutely not true!

Nehalem is intended to single digit % of quad's market, and for mission to cutt off AMD from 4-way x86 server market!

It's FUDiculous(TM by me :p) to think that Nehalem will go down to E7000/E8000 or even Celeron/Pentium market segments!

So Lynnsfield and Auburndale will cost what?

[Nedjo]

FUD? What the hell did i say besides the obvious? Why the hell would Intel want a full fledged review months ahead from the launch. Sure the rumors and leaked info help keep the buzz alive but a review now would be really bad marketing and anticlimatic because by the time Nehalem will get out it will already be old news...

c’mone Browncoat, do you honestly think that Intel isn’t behind this!?
It’s clear as a day that sole purpose of this “review” is to make anyone considering buying AMD’s 4-way Barcelona, or making plan for 4-way Shanghai, to rethink idea once more, and hopefully go in idle ‘till Nehalem based Xeon is out!

Intel purposely went with Anand and not some specialized server oriented site ‘cos of the chance to make bigger impact… these Intel marketing guys do know really well what and how they are doing what they are payed for!

On the other side this doesn’t do harm to Intel’s current lineup on desktop market, ‘cos desktop market will receive these CPU’s later, much later…

[terrace215]

i dont know, maybe i missed it, i think singlethread nehalem wil have a 10-15% advantage, but not more, on the multithreaded side however

Do note that on the one single-threaded test Anand ran, a Nehalem alpha-silicon part running on a crippled mobo outperformed an identically clocked Penryn by 26%

[bowman]

Absolutely not true!

Nehalem is intended to single digit % of quad's market, and for mission to cutt off AMD from 4-way x86 server market!

It's FUDiculous(TM by me :p) to think that Nehalem will go down to E7000/E8000 or even Celeron/Pentium market segments!

Uh, you do realise that 'Nehalem' isn't just Bloomfield? It's an architecture, a collection of modular CPU bits and bobs which they can toss into any market segment. That's the whole point, it's a modular architecture. They'll make low-end dual-core laptop CPUs with integrated Intel graphics next year with this same architecture.

To say that Intel's official roadmap and product plans are FUD would definitely be stretching it.

[eleeter]

...do you honestly think that Intel isn’t behind this!?

I certainly do. And it's an effective strategy for Intel.

[kuhla]

Awesome. Can't wait to get one.

[jaredpace]

clock for clock appears to be ~ 36% increase over Penyrn in general.

heres to hoping everything goes multithreaded in 2009!