Next Gen VS. Current Gen CPU: Crysis Benchmark

[Hornet331]

what does that have to do with it?

go read nub:
http://en.wikipedia.org/wiki/Errors_..._in_statistics
http://en.wikipedia.org/wiki/Margin_of_error

[mstp2009]

According to you maybe....
we shall see what Q1 brings,3ghz at 25W would be a real accomplishment if you ask me,would be some nice notebook chip
i still haven't seen any proof of a 3ghz 25W part

I haven't seen any announcement as well, but it is WELL WITHIN their capabilities.

Especially according to an engineer I know at D1D at Intel's Oregon facility.

[GoThr3k]

ANY 1-2% variance is within the level of statistical error.


Basic math.


You can't get a significant p value between two data sets that vary by only 1-2% unless the number of measurements within EACH data set is HUGE. In this case we would need about 2-500 INDIVIDUAL measurements from EACH (Phenom and Barcelona) to say with 95% certainty there is a STATISTICALLY SIGNIFICANT difference between the two (i.e. a P <= 0.05, which is universally recognized as the MINIMUM cutoff to say something is significantly different or not).

again,what does this have to do with it?
cant admit you are wrong for once?

I haven't seen any announcement as well, but it is WELL WITHIN their capabilities.

Especially according to an engineer I know at D1D at Intel's Oregon facility.

K10 @ 3GHz is also possible,but it doesnt mean we will see it anytime soon

[mstp2009]

go read nub:
http://en.wikipedia.org/wiki/Errors_..._in_statistics
http://en.wikipedia.org/wiki/Margin_of_error

again,what does this have to do with it?
cant admit you are wrong for once?

I have no problem admitting IF I am wrong. Problem is I am not.

What is your problem?

[mstp2009]

K10 @ 3GHz is also possible,but it doesnt mean we will see it anytime soon

The ONLY reason we won't see it is b/c there is nothing from AMD that can compete with a 3GHz (or even a 2.4GHz) laptop C2D. B/c of that, Intel is in ZERO rush to push the chips out.

I know for a FACT they could be made YESTERDAY.

[GoThr3k]

go read nub:
http://en.wikipedia.org/wiki/Errors_..._in_statistics
http://en.wikipedia.org/wiki/Margin_of_error

i do know what it meens nub
thats not the point of discussion here

I have no problem admitting IF I am wrong. Problem is I am not.

What is your problem?

if you are calling in margin of error for your 1-2&#37; improvement for other rams,you are just searching for excuses and you cannot admit you are wrong...

The ONLY reason we won't see it is b/c there is nothing from AMD that can compete with a 3GHz (or even a 2.4GHz) laptop C2D. B/c of that, Intel is in ZERO rush to push the chips out.

I know for a FACT they could be made YESTERDAY.

you know..... you can say whatever you want,still no proof
and it still doesnt mean we will see 25W 3ghz in Q1.....

[mstp2009]

if you are calling in margin of error for your 1-2&#37; improvement for other rams,you are just searching for excuses and you cannot admit you are wrong...

No one is talking about other RAM here. Just you n00b.

you know..... you can say whatever you want,still no proof
and it still doesnt mean we will see 25W 3ghz in Q1.....

I never CLAIMED we would see it in Q1.

Read what you quoted - I SAID 45nm laptops chips in Q1, but never said 3GHz was PART of that, just that is was POSSIBLE. Fine line, but it illustrates your DIMINISHED READING COMPREHENSION.

[GoThr3k]

No one is talking about other RAM here. Just you n00b.

What people are comparing is IPC, and for that Phenom and Barcelona are identical in architecture and memory timings will make AT BEST only a 1-2% improvement. So no, Phenom = Barcelona for these comparisons.

oh right,that improvement is statistical error

I never CLAIMED we would see it in Q1.

Read what you quoted - I SAID 45nm laptops chips in Q1, but never said 3GHz was PART of that.

you clearly say,25W at 3GHz....
just read my sig or the post of yours

[mstp2009]

oh right,that improvement is statistical error

Sigh.

Simple - it cannot be CLAIMED as an improvement simply BECAUSE it is within the statistical error. You would run the same analysis again and it could be 1-2% slower, not faster. Hence the term VARIANCE.

you clearly say,25W at 3GHz....
just read my sig or the post of yours

READ the lines separately genius. FIRST line - statement of fact. SECOND line - statement of what I would like to see (that is within the REALM of possibility). I didn't CLAIM we would see it in Q1, but just that we would see in on 45nm.

I understand English is not your first language, but if you are going to get into an argument in another language, you BETTER understand the nuances of that language.




I'm done with you. You have a lot more education needed before you can converse on this level.

[GoThr3k]

reread the thread....
everyone was laughing at your stupid statement.....

we will see 15th november when phenom is launched...
according you,it will perform the same as barcelona

and yes,english is my third language....
i understand nuances in english,i am not that stupid...

funny thing is,when you are losing an argument,you say things like,proof me god doesnt exist and you start flaming and throwing in statistical error

really sad if you ask me.....

[mstp2009]

we will see 15th november when phenom is launched...
according you,it will perform the same as barcelona

Care to put your money where your MOUTH is.

I will bet you ONE 2.4GHz Phenom that Phenom is within 3&#37; (i.e. the limit of statistical error) on a CLOCK for CLOCK basis with BARCELONA (ALL other settings of the system MUST be the same - HDD, RAM speed and timings, RAM size, Video card, OS, etc.).


So, chicken little, you GAME? I'm calling YOU OUT.

[mstp2009]

I'll also bet you A PAIR of 2.4GHz Phenoms that it is slower, clock for clock, than Yorkfield.

[GoThr3k]

Care to put your money where your MOUTH is.

I will bet you ONE 2.4GHz Phenom that Phenom is within 3% (i.e. the limit of statistical error) on a CLOCK for CLOCK basis with BARCELONA (ALL other settings of the system MUST be the same - HDD, RAM speed and timings, RAM size, Video card, OS, etc.).


So, chicken little, you GAME? I'm calling YOU OUT.

who said anything about same ram,according to your statement,ram doesnt improve anything
Barcelona uses ECC DDR2-667 by default
Phenome uses DDR2-1066 by default

according to you,they will perform the same even with different rams
do you dare to put your money where your mouth is??

[LowRun]

[savantu]

...
i still haven't seen any proof of a 3ghz 25W part

What about 3.16GHz DC at 40w ?

Xeon L5250 3.16 GHz 40W 1333 MHz 6MB

Or QC at 50w ?

L5430 2.66 GHz 50W 1333 MHz 12MB

[GoThr3k]

What about 3.16GHz DC at 40w ?

Xeon L5250 3.16 GHz 40W 1333 MHz 6MB

Or QC at 50w ?

L5430 2.66 GHz 50W 1333 MHz 12MB

i know intel has LowPower parts savantu,i don't doubt that
but the 25W 3ghz chip is part of an old argrument...

[Ubermann]

So even after reading 5 pages of denial the Intel is still faster, as in the first post.

[Periander6]

So even after reading 5 pages of denial the Intel is still faster, as in the first post.

Faster clock for clock. Add in the differences in clockspeeds, and the gap between Intel and what AMD put out is going to be huge.

Of course none of this should be a surprise to anyone. The presentation AMD gave to Sun early this year that was leaked and posted here several times clearly noted that K10 was going to offer "up to 15% IPC improvement" over K8. That still would leave it behind Conroe's IPC much less Penryn's. Why would AMD lie to it's own partners?

Anyone not in denial should have known for months that K10 was not going to be able to match Conroe's IPC. The barcelona results suggested the same. Now the first Phenom benches are leaked, and the results are still the same. The denial is just getting silly at this point. No one can post facts anymore without AMD fans coming in and trashing them.

[Hornet331]

1 1/2 week till the madness stops.

[happychappy]

1 1/2 week till the madness stops.

And it'll still carry on after that for a while because they will all moan the reviews aren't fair

[Shintai]

I see GoThr3k is still strong in the denial skills. Oh well...I guess reality was a too bitter pill to swallow for him.

Be nice to him..its a rough time...

[Ubermann]

And it'll still carry on after that for a while because they will all moan the reviews aren't fair

"WTF did they test that pro Intel game for ??"
"OMG look at the timings!"
"Bios will mature and you get another +30fps"
"They used the wrong RAM slots, the traces have longer distance FFS!"
"We all know that AMD sux at that game!"

Did i miss anything ?

Ohh i did, R2 V1.02 B02 A03 will fix it!
Not to mention compilers!

[mstp2009]

I see GoThr3k is still strong in the denial skills. Oh well...I guess reality was a too bitter pill to swallow for him.

Be nice to him..its a rough time...

Yeah, I'm done w/ him.

You can only have so much fun with the mentally disabled before it falls under the definition of child abuse.


I also see he didn't have the BALLS to take me up on my bet(s). Guess I'll just have to go out and buy a pair of Phenoms. I was hoping he would "donate" some to me.

[arisythila]

Oh, and GoThr3k, thanks for quoting my comment on 3GHz C2D Penryn chips.

Based on preliminary wattage info here, a Penryn Quad at 3GHz consumes an actual 73W at load (even though TDP is listed as 125W IIRC):

http://www.tomshardware.com/2007/10/...ng/page14.html


72.98W on the QX9650 at full load is almost 1/2 the wattage we saw on the 65nm Quads.

That would put an UNBINNED DUAL CORE at 38.5W at full load.

25W at 3GHz for a DUAL CORE based on Penryn, with proper binning, is WELL WITHIN the possibilities for Intel's 45nm HMGK process.


Now, how dumb do you look and feel?

Pwned!1!!

[GoThr3k]

Yeah, I'm done w/ him.

You can only have so much fun with the mentally disabled before it falls under the definition of child abuse.


I also see he didn't have the BALLS to take me up on my bet(s). Guess I'll just have to go out and buy a pair of Phenoms. I was hoping he would "donate" some to me.

maybe you should try to read,because i did reply....

@ Shintai,you really made me laugh
really

It's not that i am an AMD fanboy,i like AMD more then Intel,that is true
but i will buy what is best

I'm just trying to correct what some morrons are stating that phenom wil only be 1-2% faster even with othe rams
and then started blamin it statistical error because they cant win the argument
I am sure K10 will lose/be on par with Kentsfield though,i dont need reviews to tell me that