K10 Scores starting to surface

[zaarath]

Again from coolaler's site

http://forum.coolaler.com/showthread.php?t=161127

http://img.coolaler.com.tw/images/mawngo04gmhmwdiyjtze.jpg
http://img.coolaler.com.tw/images/uvjzyzolmj5d13ttimmz.jpg

The setup is Dual Opteron 2332 @ 2G with total 8 cores.

From what I can gather from dual clovertown scores

http://www.xtremesystems.org/forums/attachment.php?attachmentid=63418&stc=1&d=1187979867

1X: Clovertown 22% advantage clock for clock
8X: K10 speedup 7X, Clovertown speedup 6X

So even K10 scales better, it still may lose at dual socket, at least for Cinebench 10.

Don't shoot me down, I'm just a messanger.

[xlink]

we still need to know how well it scales.

I want 3.3Ghz on water

[msimax]

ooooh baby cant wait

[guess2098]

how about /2 still 4000 extra then 4.7G QX6700... hehe

[kl0012]

For comparision:
http://www.3dfluff.com/mash/cinebench/top.php
Core 2 Duo 2GHz = 2016

[zaarath]

For comparision:
http://www.3dfluff.com/mash/cinebench/top.php
Core 2 Duo 2GHz = 2016

You have to realize the 2016 score is run under windows XP 32bit
The score is higher under 64bit

[zanzabar]

has the core htt clock been released yet im curios on weather amd has upped it from 200mhz

[Movieman]

I got to go look that up. I think I ran cine10 at 2000mhz for one of the guys..might have been 9.5..

[accord99]

Xbitlabs has a Fritz score for an 8-core X5365 system:

http://www.xbitlabs.com/articles/cpu/display/intel-v8_13.html

[tictac]

Xbitlabs has a Fritz score for an 8-core X5365 system:

http://www.xbitlabs.com/articles/cpu/display/intel-v8_13.html

http://img255.imageshack.us/img255/2520/cinebench3258fc8.jpg

[kl0012]

Xbitlabs has a Fritz score for an 8-core X5365 system:

http://www.xbitlabs.com/articles/cpu/display/intel-v8_13.html

Xeon faster by 60%. Clock difference is only 50%.

[savantu]

This is consistent with previous information; K10 is slower clock/clock than Core in rendering ( I expect similar results in Povray).No magic dust inside...

A 3.33Ghz Yorkfiled scores 13068 , almost the same as 8 2Ghz K10 cores.

[Movieman]

Running it now at 2013mhz(closest I can get to 2000) on 32 bit win2K3.
Will take a bit..Machine isn't used to running under 3000mhz:D

[zaarath]

Running it now at 2013mhz(closest I can get to 2000) on 32 bit win2K3.
Will take a bit..Machine isn't used to running under 3000mhz:D

Can you please run it under 64bit windows?

Core 2 has 15% advantage in 64bit against 32bit

[n91htmare]

Can you please run it under 64bit windows?

Core 2 has 15% advantage in 64bit against 32bit

well if the score is higher or same in 32bit then just assume that it's 15% faster in 64bit.

Wouldn't wanna make AMD look THAT bad in cinebench

[Slay0r]

Better speed up in cinebench when compared to core 2, but that's no news is it .

Would help to know what kinda rams speeds were used for the tests, also i'd love to take a peak at pi calculations even though I have a feeling they will be real bad .

Awaiting your results MM :) I'm very curious .

[kemo]

So is that quad or 2X quad
http://img.coolaler.com.tw/images/mn1vj2zk32nmzfzgtjuh.jpg


same score of a 4GHZ quad Qx

[zaarath]

wPrime score
Edit: Kemo6600 beat me to it.

Can someone post a dual clover wprime score?

[savantu]

2x quad ; fairly obvious when you see processor #8

[Movieman]

Can you please run it under 64bit windows?

Core 2 has 15% advantage in 64bit against 32bit

I can, thats the other machine, that runs on win2K3 Ent 64 bit.
I've already run cine10 at 3000 and at 3240 on that machine and posted in the first cimebench10 thread here..9% slower than Intels posted benchmarks at the same cpu speed..
At my max of 3240 I was app 400 points behind the penryn setup at 3000

[kemo]

So simply you are saying that Intel dual socket is faster and indeed it looks faster in wprime

[Slay0r]

So simply you are saying that Intel dual socket is faster and indeed it looks faster in wprime

The multi core efficiency sees amd as the winner . That said, the performance seems still higher on Intel's side clock per clock .

we still don't know how k10 clocks/scales though...

[Movieman]

2013mhz..win2K3 32 bit:
http://img241.imageshack.us/img241/7969/cine10zs8.jpg (http://imageshack.us)

[kl0012]

http://images.people.overclockers.ru/120410.png

[nn_step]

2013mhz..win2K3 32 bit:

now to make sure that the OS isn't effecting it, 64bit may be required.

[Movieman]

Now you got me going..Going to crank down the other machine(64 bit) and see what happens..I got a feeling that there's going to be some surprised and maybe unhappy people here..back in a few..this takes a few minutes to do..server hardware..:D

[savantu]

http://images.people.overclockers.ru/120410.png

Core totally owns it in wPrime.

Let's wait for some Penryn scores.

[Slay0r]

now to make sure that the OS isn't effecting it, 64bit may be required.

If anything, the score is going to be higher on x64 os .

[n91htmare]

oh boy....... I can see why AMD kept silent.

[zaarath]

Now you got me going..Going to crank down the other machine(64 bit) and see what happens..I got a feeling that there's going to be some surprised and maybe unhappy people here..back in a few..this takes a few minutes to do..server hardware..:D

Yeah, I'm eager to see dual clover @2G cinebench R10 under 64bit windows

[kl0012]

Core totally owns it in wPrime.

Let's wait for some Penryn scores.
Not realy.
It is 2.66GHz 2xXeon.
From the russian site.
http://forums.overclockers.ru/viewtopic.php?t=174447&postdays=0&postorder=asc&start=0

[zaarath]

Assuming 100% clock scaling
dual K10 @ 2.66G will yield 8.00 sec and 246.17 sec in wprime.

[generics_user]

unless every revision posts the SAME info on CPU-Z this has to be a B0

= EVT sample which are supposed to have bad performance....

i wait for final benches with final revisions :rolleyes:

[Movieman]

From the other clover: Win2K3 ent 64 bit at 2000mhz:
http://img208.imageshack.us/img208/1466/cine102000eg0.jpg (http://imageshack.us)

[zaarath]

SPI 1M
http://img.coolaler.com.tw/images/zmhjxx0wenmlwemyzmzk.jpg

[Slay0r]

PI1M @ 2G

Edit: Damn you, zaarath :P

[n91htmare]

Ouch!!!!

Bad News : So far it's down in Cinebench, wPrime and Superpi

Good News: All are synthetic and not real world benches

[zaarath]

unless every revision posts the SAME info on CPU-Z this has to be a B0

= EVT sample which are supposed to have bad performance....

i wait for final benches with final revisions :rolleyes:

Well, it's certainly a valid concern, but aren't all EVT samples clocked @ 1.6G?

[nn_step]

From the other clover: Win2K3 ent 64 bit at 2000mhz:


Thank you

[Movieman]

If it's of any matter my best on that 64 bit machine on SP1M is 15.562s at 3330mhz..
That's an untouched OS and stock volt chips with DDR2-667FBDimms

[generics_user]

Well, it's certainly a valid concern, but aren't all EVT samples clocked @ 1.6G?

vr-zone got one @ 1.8ghz

but is this one really an ES, if it is supposed to be one then i smell BS because amd ES never show processor numbers...

EDIT: second scrrenshot shows full information, well we'll see :)

[R101]

I sure hope those are the sucky ones. If this is retail, run for your Q6600's while they are still below 300$!!

[Movieman]

Thank you

My friend, are you sure you want to say that to me after seeing this?
I fully expected to be on the losing end of this not the other way around.
Then add in that Penryn is 9% faster than my Clovers at the same clock.
Honestly, I am sorry to see this and I think you know me well enough to know I'm dead serious.

[n91htmare]

I don't think this is an ES.......

[n91htmare]

My friend, are you sure you want to say that to me after seeing this?
I fully expected to be on the losing end of this not the other way around.
Then add in that Penryn is 9% faster than my Clovers at the same clock.
Honestly, I am sorry to see this and I think you know me well enough to know I'm dead serious.

If that came from anyone else, I would have called you a Fanboy but its not the case, it's the truth. I truly am disappointed at the initial results.
I know these are preliminary results and synthetic, but in your experience movie man, how have these synthetic results transfered over to real work performance?

[generics_user]

Second scrrenshot shows single socket F system with 1 cpu :confused:

EDIT: ok i pass out, this is just as much validated as Theos 30k in 06, no info on the system, the RAM or anything else, i'll pass :shakes: , much smooke around no info, intel ES benches had mor system specs and better screens to back it up :rolleyes:

[zaarath]

Second scrrenshot shows single socket F system with 1 cpu :confused:

It shows "Processor #8"...

[Slay0r]

My friend, are you sure you want to say that to me after seeing this?
I fully expected to be on the losing end of this not the other way around.
Then add in that Penryn is 9% faster than my Clovers at the same clock.
Honestly, I am sorry to see this and I think you know me well enough to know I'm dead serious.

I too expected amd to take back the crown, but it seems it's not the case . Perhaps we're too quick to come to conclusions, afterall we don't know how k10 speed scales as the clock increases .

I don't think anybody expected anything great in SPI tbh, afterall 512kb vs 4096kb L2 says it all... what I'm eager to see now is 3d06 cpu scores . Maybe we're in for a surprise ? :shrug:

[Movieman]

If that came from anyone else, I would have called you a Fanboy but its not the case, it's the truth. I truly am disappointed at the initial results.
I know these are preliminary results and synthetic, but in your experience movie man, how have these synthetic results transfered over to real work performance?

I have been saying for the last 8 months that these clovers are the damnest things I've ever seen. I can load all 8 cores at 100% and it still doesn't slow down. nn_step can tell you, I run the 2 of these machines at 3157/8mb/1402 at 100% load 24/7 doing DC work..Thats nonstop 100% load except for a few benches and updates since last December..In March this machine had a 99.8% uptime, I remember that figure. I'd done 4 reboots for updates..Thats it..Yea, I could go to work selling these for Intel they are so good.

[alayashu]

if that CPU exists, it's certainly not an ES, but does AMD give nonES's to
people? never heard of this.

smells fishy
and the performance gains are very low compared to what people having
this CPU in hands said.

[Mikesta]

WOW K10 has failed.

I thought that with all the Microarchitecture enhancements that it would have an advantage over 65nm Core2's and be at least even with the 45nm ones.

How wrong many of us were!

[generics_user]

It shows "Processor #8"...

i was talking about the SPI screen which shows 4 cores and 4 threads but greyed out the processor so not the same system

[n91htmare]

if that CPU exists, it's certainly not an ES, but does AMD give nonES's to
people? never heard of this.

smells fishy
and the performance gains are very low compared to what people having
this CPU in hands said.

coolaler has MANY connections. Chances are it's a retail version.

[zaarath]

coolaler has MANY connections. Chances are it's a retail version.

Corrrection: "People on Coolaler's site" have MANY connections.

[Slay0r]

i was talking about the SPI screen which shows 4 cores and 4 threads but greyed out the processor so not the same system

Different version of cpu-z .

[duploxxx]

If anything, the score is going to be higher on x64 os .

maybe, maybe not. remember 64bit disables the inteligent caching from the core architecture


if that CPU exists, it's certainly not an ES, but does AMD give nonES's to
people? never heard of this.

smells fishy
and the performance gains are very low compared to what people having
this CPU in hands said.

this is no longer an ES those were shipped with 1.6-1.8GHZ speed. It is a DVT or one of the "early birds" as some OEM like to call these

and for the performance... wprime/pi etc says nothing to real performance.
on INT it's not faster than micro on FPU it's a huge difference same with memory bandwith, power consumption is to early to tell.


cpu-z, doesnt 1.40.5 show the voltage on K10 CPUs?

it look likes cpu-z is able to read vcore of the phenom, but it can't on barcelona


WOW K10 has failed.

I thought that with all the Microarchitecture enhancements that it would have an advantage over 65nm Core2's and be at least even with the 45nm ones.

How wrong many of us were!

funny how people think that 45nm penryn will be so much better.

I'll give some hint's:
yes it will have advantages because of the higher speed, don't expect miracles, they already had troubles fitting 3.16ghz wolfdale in the 120W tdp package
if the program has advantages due to additional cache yes it will benefit
if it supports sse4 yes it gives huge performance increase.
all the rest clock/clock it is the same performance

[alayashu]

cpu-z, doesnt 1.40.5 show the voltage on K10 CPUs?

[zaarath]

At higher frequencies it can only be slower(clock/clock) due to less memory bandwidth per MHz.

It is not up to the cache. Run a SPi test on a 2GHz Core2 with 512kB L2 and you'll get a very similar score to the one of the 4MB L2 Core2. Also the K10 has 2MB L3 dedicated to only one core when needed and two independent integrated memory controllers providing twice the memory bandwidth of Core2.


Actually 2MB C2D already has 10% deficit in Spi against 4MB, 512kb L2 Celeron 4xx is even worse. SPI is very cache sensitive.

Also, I don't think you can get 2X memory bandwidth against core2 in a single threaded application, more like 1.2X

[savantu]

WOW K10 has failed.

I thought that with all the Microarchitecture enhancements that it would have an advantage over 65nm Core2's and be at least even with the 45nm ones.

How wrong many of us were!

That should have been obvious from 5 reasons :

-constant delays
-far slower speed than expected
-changing tune from "Intel killer" to "perf/w over current Opteron" in AMD's PR
-AMD stressing Spec_rate benchmarks ; that was a clear indication that single threaded performance sucks
-early benchmark figures ( some from AMD itself )

The whole thing boils down to : K10 has nothing over Core to suggest it will have better IPC; it is simply a K8 upped to Core-like level.I say Core-like because IMO Intel still has a few advantages such as better prefetch and vastly better L2s.

Cinebench is a test where K8 kept up with Core , even exceeded it clock/clock.From these results , it is obvious that K10 is slower clock/clock than K8 at least for this benchmark.I attribute this to the cache hierarchy , the cpu simply needs to search in more places for data and this takes time.

[Movieman]

I would guess it's an ES as AMD does give them out, just sparingly..
I'm not usually surprised, most things in this game are somewhat predictable.
I can't beleive that AMD would release a chip with lesser capibilities than the current Intel offering KNOWING that Intel has even faster waiting in the wings with Penryn. After almost 40 years in the business world that flyies in the face of logic.
SO if that is wrong, what are the alternatives?
AMD is sandbagging Intel to get Intel to relax and not release Penryn a weel before AMD releases these?
That is sound business logic and It wouldn't surprise me.
That is the only option that makes sense to me.
They control their chips much better than Intel does in terms of who gets ES and might they have intentionally sent out "slow" cpu's?
Sound loudricous? Yes, but when the obvious fails then you look at the not so obvious..

[savantu]

maybe, maybe not. remember 64bit disables the inteligent caching from the core architecture

...

Huh ?!

[Slay0r]

At higher frequencies it can only be slower(clock/clock) due to less memory bandwidth per MHz.How do you work this out ? bandwidth is something Amd never lacked, on the other hand it's also something that had minimal influence on the overall performance . I understand the scaling is not going to be linear, but we can't know for sure that it's going to be slower c4c untill it gets tested.


It is not up to the cache. Run a SPi test on a 2GHz Core2 with 512kB L2 and you'll get a very similar score to the one of the 4MB L2 Core2. Also the K10 has 2MB L3 which are being used and dedicated to only one core when needed and two independent integrated memory controllers providing twice the memory bandwidth of Core2.It's not entirely up to the cache, on that I would agree . That said, the newer celerons already showed that fewer cache handicappes the performance on core2s (in superpi at least), the scores are not very similar at all .

[savantu]

I would guess it's an ES as AMD does give them out, just sparingly..
I'm not usually surprised, most things in this game are somewhat predictable.
I can't beleive that AMD would release a chip with lesser capibilities than the current Intel offering KNOWING that Intel has even faster waiting in the wings with Penryn. After almost 40 years in the business world that flyies in the face of logic.
SO if that is wrong, what are the alternatives?
AMD is sandbagging Intel to get Intel to relax and not release Penryn a weel before AMD releases these?
That is sound business logic and It wouldn't surprise me.
That is the only option that makes sense to me.
They control their chips much better than Intel does in terms of who gets ES and might they have intentionally sent out "slow" cpu's?
Sound loudricous? Yes, but when the obvious fails then you look at the not so obvious..

I think you are giving AMD too much credit ; aren't they allowed to fail after all ?

[n91htmare]

I would guess it's an ES as AMD does give them out, just sparingly..
I'm not usually surprised, most things in this game are somewhat predictable.
I can't beleive that AMD would release a chip with lesser capibilities than the current Intel offering KNOWING that Intel has even faster waiting in the wings with Penryn. After almost 40 years in the business world that flyies in the face of logic.
SO if that is wrong, what are the alternatives?
AMD is sandbagging Intel to get Intel to relax and not release Penryn a weel before AMD releases these?
That is sound business logic and It wouldn't surprise me.
That is the only option that makes sense to me.
They control their chips much better than Intel does in terms of who gets ES and might they have intentionally sent out "slow" cpu's?
Sound loudricous? Yes, but when the obvious fails then you look at the not so obvious..

you crazy silly man.....get some sleep:D

[Hornet331]

hmm doesnt look that good for barcelone.

But ill wait till more benches pop up.

[nn_step]

I think you are giving AMD too much credit ; aren't they allowed to fail after all ?
If we lose AMD, then we will all be :banana::banana::banana::banana:ed


you crazy silly man.....get some sleep:D

I believe you mean reasonable man :rolleyes:

[Movieman]

I think you are giving AMD too much credit ; aren't they allowed to fail after all ?

At this level of business you've got the company bet on it..
The answer is no, your not allowed to fail and if I had $$ I'd buy AMD stock tomorrow!

you crazy silly man.....get some sleep:D
We will see how nuts I am..:D

Good night Gentlemen..,Glad to help out! Sleep calls.

[Nedjo]

I think you are giving AMD too much credit

But we have you who doesn't have single post on this forum for giving AMD and those benches at least little benefit of doubt!

And just like nice Intel fan boy Mikesta you go ahead giving us ultimate sentence for K10: failed

way to go Sav :up:

[Movieman]

Gentlemen: I can see whats about to happen and I'll ask EVERYONE to think before they type.
No flaming please..No Gloating, tonight only raises questions, it provides no difinitive answers..
Please, remember, here at Xs, friends first please.
The last thing I want to do tonight is play Mod and ban people so please respect each other and each others views..
Thank you and goodnight..
(Yes, I get all these posts into my email);)

[savantu]

But we have you who doesn't have single post on this forum for giving AMD and those benches at least little benefit of doubt!

And just like nice Intel fan boy Mikesta you go ahead giving us ultimate sentence for K10: failed

way to go Sav :up:

I accept the reality of cold facts instead of replacing it with one of my own where K10 is the best thing since sliced bread...

[nn_step]

I accept the reality of cold facts instead of replacing it with one of my own where K10 is the best thing since sliced bread...

Oh yes, because God knows a single post is 100% accurate in reflecting any possible future for a processor :rolleyes:

[savantu]

If we lose AMD, then we will all be :banana::banana::banana::banana:ed

I doubt it ; Intel lost a lot of its power over the industry ; SW is becoming platform agnostic and the cellphone will ultimately replace the PC...

AMD will fail ; spending all their money on ATI instead of FABs was a turning point.

Secondly SOI is running out of steam ; AMD 65nm problems are part of ti and things will get far worse on 45nm.




I believe you mean reasonable man :rolleyes:

He is reasonable when he tries to replace facts ( even deny his own tests ) with a conspiracy theory ? That's a new one.


Oh yes, because God knows a single post is 100% accurate in reflecting any possible future for a processor :rolleyes:

So independent Cinebech , wPrime , SuperPI , Fritz , Pov-ray scores together with AMD released Cinebench scores ( which support the independent ones ) all point to K10 being slower clock/clock than Core.

I smell spring 06 Conroe-denial-syndrome once again...

[nn_step]

I doubt it ; Intel lost a lot of its power over the industry ; SW is becoming platform agnostic and the cellphone will ultimately replace the PC...

AMD will fail ; spending all their money on ATI instead of FABs was a turning point.

Secondly SOI is running out of steam ; AMD 65nm problems are part of ti and things will get far worse on 45nm.


He is reasonable when he tries to replace facts ( even deny his own tests ) with a conspiracy theory ? That's a new one.
well until I can play a DX11 game on my Cellphone

[RaZz!]

I doubt it ; Intel lost a lot of its power over the industry ; SW is becoming platform agnostic and the cellphone will ultimately replace the PC..[...]

lmao!

[Slay0r]

I doubt it ; Intel lost a lot of its power over the industry ; SW is becoming platform agnostic and the cellphone will ultimately replace the PC...

AMD will fail ; spending all their money on ATI instead of FABs was a turning point.

Secondly SOI is running out of steam ; AMD 65nm problems are part of ti and things will get far worse on 45nm.




He is reasonable when he tries to replace facts ( even deny his own tests ) with a conspiracy theory ? That's a new one.

You on the other hand were graced with foresight on birth ?

How about we tone it down a tad and wait for more results to come in ? you know, cinebench/superprime/superpi produce really fashionable numbers, ay . But what do they mean in "real world" (if it even has a meaning of its own) performance ? not a great deal .

If amd has failed, then so be it . "Better luck next time" .

[savantu]

At this level of business you've got the company bet on it..
The answer is no, your not allowed to fail and if I had $$ I'd buy AMD stock tomorrow!

...

That's blind faith territory ; Intel failed with Prescott altough it was supposed to be a killer.

You can have a superb product ( Alpha ) and still go down under , far larger companies crumbled in this space ; why whould AMD be exempt ?

K10 will offer sizeable benefits over K8 in the server arena , altough this will be obvious only when it ramps up the clock.But that's about it ; it won't obliterate Core , not to mention Penryn.

[Hornet331]

i hope later today we see some gaming benches, im curious about the performance of K10 in games.

[kemo]

well until I can play a DX11 game on my Cellphone

well you need to buy a cellphone first my friend :rofl: :rofl:

[savantu]

You on the other hand were graced with foresight on birth ?

How about we tone it down a tad and wait for more results to come in ? you know, cinebench/superprime/superpi produce really fashionable numbers, ay . But what do they mean in "real world" (if it even has a meaning of its own) performance ? not a great deal .

If amd has failed, then so be it . "Better luck next time" .

Cinebench is portrayed as representative of commercial 3D rendering SW ; that is pretty real world.

After all , what's real world ? 3Dmark ? Games ?

If anything , tests like the above show which core has more oomph , even if the results are semi-worthless for real world.

[oohms]

Just because AMD can't release anything that beats a conroe/kentsfield does not mean they will just "suddenly die"...
I guess if worst comes to worst they can go back to the days of the K6, where they were vastly inferior to intel in terms of performance.

[nn_step]

well you need to buy a cellphone first my friend :rofl: :rofl:

already taken care of that one (I use it as a modem when I travel)

[Motiv]

Cinebench is portrayed as representative of commercial 3D rendering SW ; that is pretty real world.

After all , what's real world ? 3Dmark ? Games ?

If anything , tests like the above show which core has more oomph , even if the results are semi-worthless for real world.

If the results are correct, then AMD looks like it will be without a paddle in the home market. If this is the case, they better had hope that when put in typical servers, it performs exceptionally well.

I'm just hoping these chips are "broken" revisions and the 3Dmark Phenom is the actual speed. I wont be holding my breath though.

[RaZz!]

Cinebench is portrayed as representative of commercial 3D rendering SW ; that is pretty real world.

After all , what's real world ? 3Dmark ? Games ?

If anything , tests like the above show which core has more oomph , even if the results are semi-worthless for real world.

ye, but you have to consider that:
- only cinebench and wprime were used to measure the performance. this is just a very small amount of benchmarks at all
- this is just ONE source that shows us its benchmark-scores. if more and different sources report the same performance numbers than we can say that the results are representative (almost).

we won't know for sure until barcelona is officially launched and credible review-sites put it to the acid test. ;)

[savantu]

lmao!

What's to laugh about ?

Have you heard of Fusion , Silverthorne ?

The purpose of such devices is to move the PC as we know it on the cellphone. Current cellphones are slowly going uphill into PC territory ; this will advance at a very rapid pace once system-on-a-chip devices appear next year.The whole point is to move x86 down , into cellphone territory and this is the purpose of Silverthorne early next year. ( Iphone 2 will use it according to rumours )

This is real and is happening now ; both Intel and AMD are preparing for the 2010-2012 generation , by that time , cell phones will be more powerfull than current PCs.Add 3D glasses to them and all my multimedia needs are satisfied.

/End off topic.

[RaZz!]

What's to laugh about ?

Have you heard of Fusion , Silverthorne ?

The purpose of such devices is to move the PC as we know it on the cellphone. Current cellphones are slowly going uphill into PC territory ; this will advance at a very rapid pace once system-on-a-chip devices appear next year.The whole point is to move x86 down , into cellphone territory and this is the purpose of Silverthorne early next year. ( Iphone 2 will use it according to rumours )

This is real and is happening now ; both Intel and AMD are preparing for the 2010-2012 generation , by that time , cell phones will be more powerfull than current PCs.Add 3D glasses to them and all my multimedia needs are satisfied.

/End off topic.

yep, our current gfx-cards consume 200w+ and need a bigass cooling solution and you're talking about cellphones that are faster than our current pcs by 2010-2012?

even if this would be true, who could afford these cell phones? talking about premium deluxe highend ultra extreme prices :rolleyes:

c'mon, i can't believe you're serious about this...

[Slay0r]

Cinebench is portrayed as representative of commercial 3D rendering SW ; that is pretty real world.

After all , what's real world ? 3Dmark ? Games ?

If anything , tests like the above show which core has more oomph , even if the results are semi-worthless for real world.

imho real world is different from user to user . I'm essentially a gamer, even though my job wants me sat at a pc all day for obvious different reasons . Bearing in mind that I am a fanboy of performance, I'd pass a SPI 1M 15sec pc for one that takes twice or three times as much to complete the loop, if the latter, afterall, ran cpu intensive games at a higher frame rate when compared to the former .

Given that the above has little chances to reflect reality, that's the way I see it .

[Hornet331]

more on topic:

can anyone provide benches for Friz Chess Bench, CPU Mark 99 with xeons @ 2ghz ?

[SOLDNER-MOFO64]

woooooow hold on.....am I right in saying you guys are arguing over an early K10 benchmark submitted by an INTEL fanboi?

You can't pay attention to a bench from people who jumped ship to intel and have licked their a$$e$ ever since.

I could bench a Q6700 and show you 28second Pi 1M times......big deal.

:welcome: to common sense

[Hornet331]

woooooow hold on.....am I right in saying you guys are arguing over an early K10 benchmark submitted by an INTEL fanboi?

You can't pay attention to a bench from people who jumped ship to intel and have licked their a$$e$ ever since.

I could bench a Q6700 and show you 28second Pi 1M times......big deal.

:welcome: to common sense

i didnt know you know that 84036980 dude, who posted the benchmarks... so everyone who posts something "negative" about amd is a intel fanboy... :rofl:

btw. nice ninja edit on the 18 secs to 28 secs. :p:

[SOLDNER-MOFO64]

i didnt know you know that 84036980 dude, who posted the benchmarks... so everyone who posts something "negative" about amd is a intel fanboy... :rofl:

btw. nice ninja edit on the 18 secs to 28 secs. :p:

lol...no no not everyone.....just the INTEL fanatics who submit K10 scores :)

yeah, I'm ninja-like alright ;)

[Slay0r]

woooooow hold on.....am I right in saying you guys are arguing over an early K10 benchmark submitted by an INTEL fanboi?

You can't pay attention to a bench from people who jumped ship to intel and have licked their a$$e$ ever since.

I could bench a Q6700 and show you 28second Pi 1M times......big deal.

:welcome: to common sense

my common sense fails to understand how one could be "fan boy" of either company, never mind "jumping ship", unless there were lucrative reasons in the background .

it's like me turning down a ford mustang for a fiat panda (given that money wasn't an issue) because fiat is an italian brand . what the :banana::banana::banana::banana: . :shrug:

[STaRGaZeR]

AMD ppl appears! Have been a long time without them in this thread. Only informal remains in silent :rofl:

Now we have some real numbers, more to come in the upcoming days. As spected, K10 seems to be a disappointment. And that really sucks. No competition to Intel = Intel wont innovate as fast as they do with the C2D.

AMD BAD!!!

[SOLDNER-MOFO64]

my common sense fails to understand how one could be "fan boy" of either company, never mind "jumping ship", unless there were lucrative reasons in the background .

it's like me turning down a ford mustang for a fiat panda (given that money wasn't an issue) because fiat is an italian brand . what the :banana::banana::banana::banana: . :shrug:

Put it this way, when I see an AMD user post K10 benches I'll believe them, or atleast someone with some credibility.

[Motiv]

Put it this way, when I see an AMD user post K10 benches I'll believe them, or atleast someone with some credibility.

And if the benchies are true, then AMD could well be in very big trouble.

[kemo]

my common sense fails to understand how one could be "fan boy" of either company, never mind "jumping ship", unless there were lucrative reasons in the background .

it's like me turning down a ford mustang for a fiat panda (given that money wasn't an issue) because fiat is an italian brand . what the :banana::banana::banana::banana: . :shrug:


Italian is Italian , how can you turn down a fiat :shakes: ,,,




some people will favor a product over another one due to a good past experience ,, the other type of fan boys those who pays more to get less either stupid people or retarded ,,, you should only buy whatever will give you the best for your budget and always hope for better competition and better performance with every advancement

[Slay0r]

Put it this way, when I see an AMD user post K10 benches I'll believe them, or atleast someone with some credibility.

fair enough, i still fail to see how one can only either be amd or intel. I have both in my house, does that make me an untrustable bad e-person ? :confused:


Italian is Italian , how can you turn down a fiat :shakes: well, if i was to choose between italian brands, i'd be more prone to buy off alfa romeo or ferrari(i wish). :D


you should only buy whatever will give you the best for your budget and always hope for better competition and better performance with every advancement:up: :up: :up:

[Thesavage]

Already got my Alfa :)

[mAJORD]

Soo anyway... now you've all stopped being b1tches at each other (yeah right!)

..here are some some:

1Mb K8 (Windsor) Super pi @ 2ghz , with DDR2 667, to serve as some sort of comparison, since the opty's must be running reg'd 667.

http://img509.imageshack.us/img509/3645/pi2gmy8.jpg (http://imageshack.us)


And a 32bit cinebench 10 single CPU score..

http://img170.imageshack.us/img170/504/cine2gpi2.jpg (http://imageshack.us)

So this should give a rough idea of the single core performance delta if these benchmarks are a true representation:

5% faster at super pi

somewhere between 0 ann 7% faster @ cinebench, depending on how much K8 scales in 64bit version.. (anyone?)

If someone with a socket F opty system lurks out there on 64bit would like to do the same.. would be good.

[Ubermann]

Well this was just sad!
They wont earn 1 buck on these..i bet price will be really really cheap cuz thats the only reason for going with these ones over Intel.

At least the AMD vs Intel fight is over and now we have to wait for 2009 or so when they fired the whole crew and make a good product.

[[XC] riptide]

Somebody somewhere has to start posting up a full suite of K10 benchmarks for 3 reasons.

1) To get a clear picture of whats what.

2) To get an idea of overclock ability.

3) Cos I'll go banannas waiting and wondering.

[KTE]

Wow crazy. :D

Who knows whats happening. This is all a game, can be played by Intel and AMD. R600... not too much different in the buildup. Pure lunatic fights and lies passed on as fact everywhere, minute day and night. Most guys I saw speak affirmatively lost their credibility as quack "prophets" because of it, and hopefully learned how to refrain from gross exaggerations which can destroy a company if the product doesn't match the hyped expectations.

Was it any decent? Not when it was released which most so die hard expected, drivers were poor, early DX10 was poor, massive delays, hyped up grandiosely but hype was just that, pure vacuous hype. But you can find versions of it in retail now (i.e. extreme-pc) which beat the G80GTS and GTX - get a good enough MB for it first, which I think comes in the form of RD790 with the appropriate PCIe 2.0 16x bandwidth (remember, it was originally made for PCie 2.0), CF scales excellently and it does hold the benchers 2k6/2k5 records, so it 'aint a failure at all. Still, people tend to regard it as failure due to not leaving a shock attack on them when first released. :D

One things damn obvious: K8 was awesome for its time and Core 2 was and still is, in its time.

Barcelona-Penryn? That's a new phase with different competition.

Barcelona was not originally made or planned to compete against Penryn or Nehalem at all, but Core 2. The 6 month delays meant it now faces a more renewed and optimized opponent. This is a fact anyone at AMD will tell you. That in itself is a major loss and stress for them.

-Barcelona server or desktop: different ball game.

How good? We have The Inq and Rahul Sood saying one thing, and Coolaler forums saying another. I have more reason to believe none, both parties are out on jury until further evidence, but Rahul speaking about this in his position seems a little strange. They do have the retail and sample CPUs and they would've tested them quite well by now, so he has a better position to speak than both the previous sources or anyone I know online. This makes the perfect setting for confusion in the face of Coolaler results, and thus, I step back to see who pans out closest to the truth.

Phenom has higher clocks attainable than Barcelona Opteron. I think highest AMD might hit late this year is 2.4-5GHz. I doubt higher, as an estimation, but they have headroom is what everyone so far is stating and might ramp it up sooner. I don't expect a quad core Agena/Agena FX will get beat by a QX6700 clock for clock. I find that hard to believe after R600 they would carry on if that be the case. I doubt business works like this at all, at the least, it will outperform a current Kentsfield but not attain high clock speeds.

-Penryn wouldn't be released if Intel believed Barcelona was worse than a current Xeon/Kentsfield, simply put, and current prices would be that of Tigerton, much higher.

Penryn any good clock-clock? Apart from what SSE4 and smaller fab brings, highly doubt it. In multimedia I believe Intel has hit home. Do they have supply numbers to release late November like they've been leaking? If they did, they wouldn't be releasing Tigerton now but Penryn instead. I doubt they have the supply to release it yet, and are having trouble with meeting even those speeds within the given TDP packages. I reckon its a race; they want to release it in mid-late October, but numbers, bugs and clock speeds is the concern.

Any firm can make one such chip and show it around, AMD or Intel. Intel Core 2 was touting 4GHz quad cores in July 2006, but we won't even have those with Penryn 18 months after AFAWK.

What ends up in retail, the supply numbers/clock-per-clock over previous generation and competition/performance/watt/cost efficiency is what matters most.

So far, I'm undecided and refuse to play the guessing game and give sides which'll guesstimatingly be better. I just know the architectural details and whats around online.

[freeloader]

The sky is falling, the sky is falling....:D

[BeardyMan]

Great results :)

[Sparky]

Guess what... Screw intel, I still won't buy their CPUs. I don't like their business practices, and I tend to go for the underdog anyway (my wallet kinda helps me do that too sometimes) so I still would get an AMD. Just because it doesn't perform "best" doesn't mean it is a bad CPU. AMD sure has a nice price point.

[Hornet331]

Guess what... Screw intel, I still won't buy their CPUs. I don't like their business practices, and I tend to go for the underdog anyway (my wallet kinda helps me do that too sometimes) so I still would get an AMD. Just because it doesn't perform "best" doesn't mean it is a bad CPU. AMD sure has a nice price point.

everyone taste and motivation to buy products is different, and thats a good thing. :)

for myself i give a :banana::banana::banana::banana: about names or or business practices, performance is the thing what counts. :)

[BeardyMan]

Guess what... Screw intel, I still won't buy their CPUs. I don't like their business practices, and I tend to go for the underdog anyway (my wallet kinda helps me do that too sometimes) so I still would get an AMD. Just because it doesn't perform "best" doesn't mean it is a bad CPU. AMD sure has a nice price point.

If we look at the 3dmark results, they wil own Intel ver very hard in 3d and gaming...who cares about pi then :D
and in PI they did improved wel i must say..
A64 @ 2.2ghz = 42-43secs i believe
39secs at 2ghz ain't that shabby..

[Motiv]

If we look at the 3dmark results, they wil own Intel ver very hard in 3d and gaming...who cares about pi then :D
and in PI they did improved wel i must say..
A64 @ 2.2ghz = 42-43secs i believe
39secs at 2ghz ain't that shabby..

http://img509.imageshack.us/img509/3645/pi2gmy8.jpg

SPI seems 5% quicker.

Anyway, these all seem the B0 stepping. Are they the ones that rumour said were bugged?

It is ominous but if AMDs B0 samples do perform under par and the release stepping are better, then maybe all is not lost.

The positives are for the server market, it's a simple upgrade path to double the performance. The home market is a completely different kettle of fish. Without sounding like a doom sayer, there is no reason for anyone to "upgrade" onto them unless the next steppings are in fact Core killers.

[Sparky]

http://img509.imageshack.us/img509/3645/pi2gmy8.jpg

SPI seems 5% quicker.

Anyway, these all seem the B0 stepping. Are they the ones that rumour said were bugged?

It is ominous but if AMDs B0 samples do perform under par and the release stepping are better, then maybe all is not lost.

Hmm maybe, not sure, maybe I'll dig around and find out.


The positives are for the server market, it's a simple upgrade path to double the performance. The home market is a completely different kettle of fish. Without sounding like a doom sayer, there is no reason for anyone to "upgrade" onto them unless the next steppings are in fact Core killers.
Yes there is plenty of reason to upgrade. All those who have K8 have reason to, or those who have a P4. Doesn't have to be a core killer to be a good upgrade from something you have already. I know for sure that K10 will outperform my opteron 165, heck it had better or AMD just wasted their time lol.

[Morais]

I really dont want to trust in this benchmarks...Damn Sept 10 never comes!

[Aussie FX]

it's been noted here already.

Why are people trusting these benchmarks?

They are from a known intel fan.

Funny that no one wants to believe the 3dmark score.

I remember many months ago S7 had a heavily patched ES and without saying much he let on that his chip was ~ 10% faster than Kentsfield clock for clock.

I'm not taking any notice until something official comes out.

It seems like a fair few here are wanting/hoping for AMD to fail.

[Motiv]

it's been noted here already.

Why are people trusting these benchmarks?

They are from a known intel fan.

Funny that no one wants to believe the 3dmark score.

I remember many months ago S7 had a heavily patched ES and without saying much he let on that his chip was ~ 10% faster than Kentsfield clock for clock.

I'm not taking any notice until something official comes out.

It seems like a fair few here are wanting/hoping for AMD to fail.

I don't think it matters that he is an Intel fan. If you had the opportunity to be the first person with the Newest Intel chips Benchies, before anyone else. Would you mess them up.

It would only make you look bad and not the chip (especially as it seems before release people on XS will have this chip in their hands and be benching).

To me, the benchies are correct but is the chip bugged?

[Aussie FX]

Because are not coming from AMD.

Uh...fanpois. Are there any other benchmarks?
Independant review sites would be helpful.


What score? :eek: Link please?
~ 30,000 3Dmark 06
It's on the xs news page..... I can't believe you haven't seen it. (No offense meant)


Link please? It never happened before ES to be faster than retail CPU.
It is also on this site. Pg 14 of the K8L and beyond thread IIRC

[/QUOTE]

[Hornet331]

it's been noted here already.

Why are people trusting these benchmarks?

They are from a known intel fan.

Funny that no one wants to believe the 3dmark score.

I remember many months ago S7 had a heavily patched ES and without saying much he let on that his chip was ~ 10% faster than Kentsfield clock for clock.

I'm not taking any notice until something official comes out.

It seems like a fair few here are wanting/hoping for AMD to fail.

did you even have visited the link to the coolaler forum... the benches are NOT from coolaler...

they are a community just like xs.org is, a users posted this benches. If this users can be trusted or not.... i can't say, time will tell.

And why most people belive this rather than the 30k 3dmarks ?
Well at least this user provides screens of the scores rather than just making things up and giving no prove (not even a cpuz screenshot) like theinq.
Sure you can say this screens are fake, but at least the super pi score seems to valid.

[PetNorth]

Not realy.
It is 2.66GHz 2xXeon.
From the russian site.
http://forums.overclockers.ru/viewtopic.php?t=174447&postdays=0&postorder=asc&start=0


It is a dual Xeon 2.66GHz, running @2GHz. So K10, totally owned in CineBench, wPrime, SuperPi. :( I was expecting clock for clock K10 to be at least as fast as Core2. :down:

No.

it's dual Xeon Quad at 2.66 (2.0 = idle = speedstep)

look the source: http://forums.overclockers.ru/viewtopic.php?p=3788796#3788796

[Hornet331]

well if we go after this link scores are like this:

8.203 for 2x Intel Xeon X5355 2.66GHz 4*1GB DDR2 SDRAM FBD Memory @ 667MHz
9.110 for 2x Intel Xeon E5345 2,33 GHz 4 GB (4x1 GB) Kingston FB-DIMM @ 667 MHz
11.720 for 2x Intel Xeon E5320 1.86Mhz 8Gb (4x2Gb) Hynix PC5300

the score for a 2ghz clovertown should be between the 11.720s and the 9.110s, barceelona score was 10.641s.

Imho they are right neck on neck

[Carfax]

I think we should be very skeptical of these benchmarks.

More reputable sites have had opportunities to do benches with the K10, but have declined due to rumored chipset issues which degraded performance.

I think I remember Anandtech saying that before.

It's possible something could be wrong with the motherboard or BIOS which is not making the K10 perform up to speed!

I mean, come on. It's barely faster than the K8 in some benches!

[Blasar]

IF these benches are accurate, i truly hope that this is not the top performance that a system based upon

a K10 and a RD790 can achieve, or else....

:slapass:

[STaRGaZeR]

And we are only comparing clock per clock perfomance, and the fact is C2D and Penryn clocks like hell, no problems to get 4.0Ghz.

[BeardyMan]

And we are only comparing clock per clock perfomance, and the fact is C2D and Penryn clocks like hell, no problems to get 4.0Ghz.

The clocks like hell doesn't matter in the OEM or other vendro bussiness..

When K10 and it wil be most likely, beats c2d/quad, at stock then amd has succeeded in their goal.

[R101]

The clocks like hell doesn't matter in the OEM or other vendro bussiness..

When K10 and it wil be most likely, beats c2d/quad, at stock then amd has succeeded in their goal.

It all depends on the price.. Enthusiasts are buying Q6600 like there's no tomorrow, and that is where K10 must strike, but we will not see K10 at 2.2GHz for 260$. :(

[BeardyMan]

It all depends on the price.. Enthusiasts are buying Q6600 like there's no tomorrow, and that is where K10 must strike, but we will not see K10 at 2.2GHz for 260$. :(

If the performance gap is big, wich is possible, AMD ha sthe right to ask that premium.And they will flip sides by forcing intel to lower prices.
And at the end price won't matter, people are still buying the more expensive C2D over the cheap but less performing old K8.

[Dainas]

Its too soon to muse over the failures of K10s, Capitan Jumduconclusions.... Still it would not be the first architecture to have a rough start to blossom for whatever reason down the road.



The clocks like hell doesn't matter in the OEM or other vendro bussiness..

When K10 and it wil be most likely, beats c2d/quad, at stock then amd has succeeded in their goal.

I concur, people do love the Athlon 6000+ for good reason today. Allot of people don't give two licks about what something can do overclocked and buy it for the stock performance. Case in point the X1950pro 512mb AGP I got my sister months ago, so she could has no hassle great performance(well for $170) in her fav game Oblivion. Like it so much I got another one for old system.

[gallag]

It still clocks like hell though at stock, If peny gets to 3.3ghz at launch and AMD is 2.0-2.4ghz then it will just be a slaughter

[alayashu]

do you know how K10 will perform to tell this?
reputed guys that tried the CPU told it will be a great performer, AMD officials said that too.
i preffer to trust them instead of suspicious posts on various sites.

[kl0012]

reputed guys that tried the CPU told it will be a great performer, AMD officials said that too.
i preffer to trust them instead of suspicious posts on various sites.
Actualy we doesn't seen any benchmark from those "reputed guys". Still I believe them :). (For example I believe to SUN about 15% IPC improvement of K10 over K8).

[zabomb4163]

Coolar is reliable...

[Enoc]

zaarath since you are in that thread over a coolalers...can post that that he can download cpuz 1.41..since it detects the X4 correctly..and then we can see the revision that he has...

http://www.vr-zone.com/articles/CPUZ_1.41_Detects_Phenom_X4_Correctly/5208.html

[Think]

Price estimate?

[zaarath]

delete

[zaarath]

zaarath since you are in that thread over a coolalers...can post that that he can download cpuz 1.41..since it detects the X4 correctly..and then we can see the revision that he has...

http://www.vr-zone.com/articles/CPUZ_1.41_Detects_Phenom_X4_Correctly/5208.html

Enoc, the cpu-Z in the 1m-spi test screenshot is already 1.41, which shows info consistent with vr-zone. Check it again.

[Think]

Price estimate?

Well, anyone??

[Enoc]

Enoc, the cpu-Z in the 1m-spi test screenshot is already 1.41, which shows info consistent with vr-zone. Check it again.


did not see that one...my bad...wonder why does not show the revision?

[Lightman]

did not see that one...my bad...wonder why does not show the revision?

Either it is A1 revision and CPU-Z can't recognize it or it is B1/C1 and same apply!

Pics from VR-Zone are showing Stepping 0, these pics are showing Stepping 1 so definitely both CPUs are different.
I'm afraid guy on Coolalers forum has final B1 rev, but that only time will tell...

[informal]

The CPUs @ coolaler's forum are not DVT...
They are EVT B0's just like VR-zones...

This Agena processor is clocked at 1.8GHz and is B-0 revision which is an EVT sample for SVID testing only

All the intel fans jumped to early.

[FghtinIrshNvrDi]

here's to hoping it's broken...

Ryan

[Shadowmage]

Hopefully coolaler is reading this thread. If not, zaarath:

If I can get TWO (yes just two) data points at different clock speeds with constant FSB, I can do an Amdahl's Law curve fit and predict performance at 3GHz (or any clock speed) within 1-2%, depending on the quality of the two points (ie variation between runs).

Please ask coolaler for 1.8GHz and 2.0GHz, same FSB and mem, for a benchmark (eg Cinebench 10R).

Also, core scaling obeys Amdahl's Law as well, so this may be interesting too for 4core and 8 core data points.

[Shadowmage]

Also, I'm wondering the reasons why K10 loses, when it meets or exceeds C2D except in a few areas (cache size, cache latency, issue/retire width, slightly stronger branch predictor).

How sensitive are these benchmarks to cache size? Does anyone have C2D 2M vs 4M comparisons?

EDIT:
http://www.xbitlabs.com/articles/cpu/display/core2duo-e6420_5.html

Looks like Cinebench 10R is NOT sensitive to cache size or memory bandwidth.

[FlawleZ]

Soo anyway... now you've all stopped being b1tches at each other (yeah right!)

..here are some some:

1Mb K8 (Windsor) Super pi @ 2ghz , with DDR2 667, to serve as some sort of comparison, since the opty's must be running reg'd 667.

http://img509.imageshack.us/img509/3645/pi2gmy8.jpg (http://imageshack.us)





So this should give a rough idea of the single core performance delta if these benchmarks are a true representation:

5% faster at super pi

somewhere between 0 ann 7% faster @ cinebench, depending on how much K8 scales in 64bit version.. (anyone?)

If someone with a socket F opty system lurks out there on 64bit would like to do the same.. would be good.

Wouldn't it be optimal to have this comparison in 1T? Nothing else just to rule out any other variables...

[terrace215]

The CPUs @ coolaler's forum are not DVT...
They are EVT B0's just like VR-zones...


B0's show up as "Model 2, Stepping 0"

These results are "Model 2, Stepping 1", i.e., they are B1 results.

[Donnie27]

Wow, known Spinner vs. Coolater + Members and the Spinner trusted more, just frackin' wild:D Coolater and folks posting there proved they know their stuff. Subsequent C2D tests and owners backed up just what they said with early samples.

I disagree with at least 3 posters here. No matter if Barkie sucked, there is a hardcore 12 to 13% (across all markets) that would buy AMD no matter what it did performance wise. No threat of death to the Core of AMD's company either BTW. If they didn't die as bad as K6 sucked (processor and platform) they'd not die now. They didn't die when the PR-150 was slower than a Pentium 133 and so on.

IF not this round and that's still not truly NOT certain yet, AMD will have another K7 sooner or later. The other poster is right, even Intel Fans wants AMD to ship something that keeps Intel working to get better and keeps prices down. Can't have either if AMD is stinking up the place.

Just as the excellent A64-X2 didn't stop folks from buying P4, C2D didn't stop folks from buying X2, it will make no to little difference how Barkie and PhenomX4 perform, they'll still get loved. I'd still bet there'll be apps k10 will run faster and others it will run slower. SSE4 will still offer Intel a big time advantage.

[StealthyFish]

You on the other hand were graced with foresight on birth ?

How about we tone it down a tad and wait for more results to come in ? you know, cinebench/superprime/superpi produce really fashionable numbers, ay . But what do they mean in "real world" (if it even has a meaning of its own) performance ? not a great deal .

If amd has failed, then so be it . "Better luck next time" .

AMD execs had gloated about these benchmarks and if these benchmarks are true and really benched on a K10, AMD looks to have failed the enthusiastic outlook people held for them. But I'm going to wait for the retail chips before I solidify a conclusion about K10.

[[XC] hipno650]

SPI 1M
http://img.coolaler.com.tw/images/zmhjxx0wenmlwemyzmzk.jpg

i would hope thats not true. cuz form my personal experience a 2.3ghz core 2 is 20sec faster and my P4 will do 30sec

[informal]

B0's show up as "Model 2, Stepping 0"

These results are "Model 2, Stepping 1", i.e., they are B1 results.

So,all pre-B2 were buggy.
Here you go:
http://www.vr-zone.com/articles/BA-Step_Barcelona_For_Launch_In_Sept/5196.html


We heard that AMD partners has gotten B1-step Barcelona for compatibility testing with better and newer AGESA code compared to the B0-step Barcelona. There is also B2-step Barcelona which is bug free and then BA-step Barcelona for mass production

[BeardyMan]

Superpi is L2 cache loving...

[terrace215]

So,all pre-B2 were buggy.
Here you go:
http://www.vr-zone.com/articles/BA-Step_Barcelona_For_Launch_In_Sept/5196.html

AMD is LAUNCHING the K10 on the B1 stepping, per everyone else. Then B2 in Q4. At any rate, it is unlikely that B1 has any huge performance bugs. These results fit precisely with what AMD themselves claimed.... up to 10-15% better IPC than the K8. About in the ballpark of Conroe/Kentsfield per clock. Better on some bandwidth-heavy server stuff & HPC, worse on desktop apps.

[Enoc]

there are details that are still missing here...what motherboard(chipset), what ddr2 ecc single or dual channel? etc...
a lot of people giving opinions but what's the platform that this was tested...

[informal]

I still say these results are not the representative of final scores.

[Yoxxy]

I still say these results are not the representative of final scores.

I would love to believe so, but in the end I believe this chip is going to be a retail sample.

[AliG]

god I seriously hope those super pi times aren't real, if they are no wonder amd didn't allow anyone to bench the 3ghz systems

[terrace215]

I still say these results are not the representative of final scores.

Based on what? Hope?

[informal]

Based on what? Hope?

Hmm you are one intel only dude,aren't you?

Here you go,knock yourself out :):
http://forums.anandtech.com/messageview.aspx?catid=28&threadid=2090639#27132633



Originally posted by: Viditor
There appears to be some problems in those reports...

1. It reports as an Opty 2332, but also as a Phenom/Agena (the 2332 is a socket 1207, the Agena is a socket AM2+)
2. It's only a stepping 1 chip...


The latest Barcelona chips are B02 steppings with one more to go. Believe me, the reason we did not post any numbers at Computex or since then is the simple fact that the CPU/boards/BIOS have undergone dramatic changes over the course of the summer. If you have an earlier stepping there is a very good chance that HT and the secondary cache is disabled, this will affect the benchmarks dramatically. We expect to see final stepping chips and board revisions early next week, until then, it is all speculation for the most part.

The one caveat that I will add, this chip really does not get into a groove until you get over 2.4GHz and then it scales incredibly well. Also, the first RD790 boards we have will undergo another spin so any Phenom results with those boards are subject to interpretation depending on whether you like AMD or not.

[Periander6]

I still say these results are not the representative of final scores.

I say they will be. Slide 16 here was a dead giveaway:

http://www.sunmicrosystems.se/virtualisering/pdf/AMD_Quad_Core-Leif_Nordlund.pdf

That's not "up to 15% IPC gains" against C2D, it's 15% against K8.

Add to that keeping a tight lid on benchmarks when there was no reason to do so if they were good, the change in emphasis from performance to performance/watt, the Henri Richards resignation, low launch clockspeeds, the focus on Specfp_rate (which is really little more than a memory bandwidth bench), the rumors out of Computex, and the "simulated" benchmark scores.

Really, the writing has been on the wall for months, just like it was for the R600 and for many of the same reasons. The only thing keeping the hype alive were the AMD forum boosters and their relentless efforts to shout down anyone who pointed out inconvenient facts.

[[XC] gomeler]

I just have 2 hopes, these K10 results are somehow skewed a bit due to a buggy processor and that Opteron prices won't be through the roof so us normal people can buy one to play with.

[AliG]

as long as the actual phenom cpus won't be that buggy when they debut and produce better scores, I'll be happy, but the servers are going to be gipped, unless if the super pi and cenebench results don't display the true processing power of the k10 cpus

[informal]

I say they will be. Slide 16 here was a dead giveaway:

http://www.sunmicrosystems.se/virtualisering/pdf/AMD_Quad_Core-Leif_Nordlund.pdf

That's not "up to 15% IPC gains" against C2D, it's 15% against K8.

Add to that keeping a tight lid on benchmarks when there was no reason to do so if they were good, the change in emphasis from performance to performance/watt, the Henri Richards resignation, low launch clockspeeds, the focus on Specfp_rate (which is really little more than a memory bandwidth bench), the rumors out of Computex, and the "simulated" benchmark scores.

Really, the writing has been on the wall for months, just like it was for the R600 and for many of the same reasons. The only thing keeping the hype alive were the AMD forum boosters and their relentless efforts to shout down anyone who pointed out inconvenient facts.

Read mu previous post..

[Carfax]

Haha Informal, I was just about to post that :D

[informal]

Beat you by 1 minute :D.Still you did post some very useful info:up:

[kl0012]

Read mu previous post..

there is a very good chance that HT and the secondary cache is disabled
That is stupid explanation. Barcelona can't work with "disabled HT". And try to disable L2 - you will see slide show instead of your windows. Any way, you can see in CPU-Z that it wasn't disabled.

[railer]

Well if these SPI scores are real it will mean one thing that K10 will suck at real games. Most games are single threaded now and those 8 cores are useless in anything but in 3d06 which is really far from actual game performance. I'm really disappointed with K10. I hopped for a way better. AMD should build a good performing single CPU and then add those cores. And Collaer is a reputable source, 99% it wont be any different on the sept 10. Let hope in 2009 AMD strike back.
Intel pwn's right now in prices and performance:welcome:

[w0mbat]

u compare superpi and games? lol!

[XeRo]

u compare superpi and games? lol!

Superpi is a game. It's just like to some people that 3dmark is a game. :D

[railer]

u compare superpi and games? lol!
Il give you a hint for every 1 sek decrease in PI my fps in FEAR increase 3%.
I don't compare its just an indication that at single threaded apps i don't seee much difference between K8 and K10.

[oldblue]

... Specfp_rate (which is really little more than a memory bandwidth bench)

I've seen this belief pop up a few other times on this forum, and it's a little strange. Specfp_rate tests how quickly a system runs 17 different real-world floating-point codes under full load. (Specint_rate is similar, but for integer codes). Of course it becomes a memory bandwidth benchmark if you don't have enough memory bandwidth. That's true about every benchmark that doesn't fit in the cache.

If you look at the results (www.spec.org), you'd see that systems with the same memory bandwidth but faster processors tend to do better.

It's probably better to think of it as a scalability benchmark. If your system scales well, you want to show off specfp_rate and specint_rate. If it doesn't, you want to stick with single-threaded benchmarks.

[PetNorth]

As kl0012 says, I don't think HT or cache are disabled in these B0/B1 stepings. But I think that the new (or part of them) features are disabled or something.

I say this because when I see to these results, I conclude that they are like if it would be a K8 (for example, I've run Cinebench 10 x64 with my X2 at 2.0 and 1cpu gives me 1905 -this K10, 1896-. Equal result. The same thing I can say about those pov-ray benchs). And with new K10 features (like 128 SSE etc) this isn't possible at all.

So certainly, I think that Gary Key form anadtech, is on the way, but he fails in the reasons. Of course no HT or Cache disabled (without cache, it would be an absolute turtle), but like I said I think new K10 features (or part of them) are disabled in these steppings and thats the reason why we don't see basically improvements in relation to K8, in heavy optimized SSE apps, like Cinebench.

IMHO, we'll have to wait till commercial step (it seems to be B2 or BA) to know how this thing really perform.

[awdrifter]

Damn, I might as well build a quadcore now, I was expecting K10 to be at least as fast as Kenfield, but I guess this is another R600. If these benches are true, AMD might as well just keep the K8 arch and try to get it to 3.5+ghz on air. It would yield about the same performance as the 3ghz K10.

[Mad_Man]

talking about "disabled HT"... noone noticed that in every cpu-z it says bus speed 200mhz, rated fsb speed 200mhz ? should not HT link be much, much more? at 200mhz it will be a real bottleneck.
just a though...

[Sparky]

Bus speed is the HTT. HT is different than the bus speed/FSB/HTT.

[n91htmare]

I just don't think it's an early sample this late in the game :(.

IF anything, I think we can attribute the low performance to the mobo or something.

[duploxxx]

Well if these SPI scores are real it will mean one thing that K10 will suck at real games. Most games are single threaded now and those 8 cores are useless in anything but in 3d06 which is really far from actual game performance. I'm really disappointed with K10. I hopped for a way better. AMD should build a good performing single CPU and then add those cores. And Collaer is a reputable source, 99% it wont be any different on the sept 10. Let hope in 2009 AMD strike back.
Intel pwn's right now in prices and performance:welcome:

i think its about time school is going to start again.... so you guys can learn something.

the benches are from a 2 socket opteron socket and as additional info the memory is running 667 speeds cas5. what superpi perfromance do you expect.
the performance can't be compared with the phenom. try to bench a k8 with memory 667 cas5 and then change it to 1066 cas5 like the demo system AMD is showing. even at the same speed it will be a huge performance difference.

looking and comparing the cpu-z from Coolaler to mine it is a fact that his sample is a dvt or early bird from oem. so it will be the chip that launches within a few days. All features are enabled, K10 is no INT miracle it will be as fast as core and no penryn is not faster then conroe, it's just when the program is affected by cache or not it will be faster clock/clock and the magic sse4. k10 will increase the gap on fpu and IO even more not to mention virtualization. How well it scales will be depending on the new stepping coming in oktober and you'll see how well it does when people get phenom es in there hands.

informal's information about the bios is correct, it is changing alot, systems with b0 had no CPU options in BIOS available except changing HT speed.

[Flambo]

http://img.coolaler.com.tw/images/zmhjxx0wenmlwemyzmzk.jpg

If this score is truly representative of K10, then AMD must be going backward because even my 2.67 GHz 3800+ X2 can do SuperPi in 32 seconds! Of course,my 3.124 GHz Opteron 175 can do it in about 28 seconds. I cannot really believe that this will be what the shipping chips will be like.

[Mad_Man]

If this score is truly representative of K10, then AMD must be going backward because even my 2.67 GHz 3800+ X2 can do SuperPi in 32 seconds! Of course,my 3.124 GHz Opteron 175 can do it in about 28 seconds. I cannot really believe that this will be what the shipping chips will be like.
and my windsor at 3.17Ghz can do it in 24.8s
whats the purpose of your post?
OCed cpu&ram against stock cpu with ddr2 @667 cl5 ....

sPi aint real world...

[terrace215]

and no penryn is not faster then conroe, it's just when the program is affected by cache or not it will be faster clock/clock and the magic sse4.

You're forgetting some other Penryn improvements. Take a look at throughput and latency improvements to various SSEx operations, for instance. (Super-shuffle engine)

But in most apps, most of what Penryn brings will be higher speeds & lower power. Assuming AMD provides any pressure at all for the higher-speeds part...

In terms of a 2-socket comparison, one near-term boost for Clovertown (and Yorkfield) will be the transition from Blackford --> Stoakley. Much better snoop-filter & throughput up 25%... should give the spec_rate benchmarks and performance-related apps a nice boost.

[savantu]

I say they will be. Slide 16 here was a dead giveaway:

http://www.sunmicrosystems.se/virtualisering/pdf/AMD_Quad_Core-Leif_Nordlund.pdf

That's not "up to 15% IPC gains" against C2D, it's 15% against K8.

Add to that keeping a tight lid on benchmarks when there was no reason to do so if they were good, the change in emphasis from performance to performance/watt, the Henri Richards resignation, low launch clockspeeds, the focus on Specfp_rate (which is really little more than a memory bandwidth bench), the rumors out of Computex, and the "simulated" benchmark scores.

Really, the writing has been on the wall for months, just like it was for the R600 and for many of the same reasons. The only thing keeping the hype alive were the AMD forum boosters and their relentless efforts to shout down anyone who pointed out inconvenient facts.

Excellent post . :up:

[[XC] flat-four]

Bus speed is the HTT. HT is different than the bus speed/FSB/HTT.

Shouldn't CPUZ show "HT Link" link instead of "Rated FSB" like usual AMD cpus? If the HT link is actually running at 200 instead of 1000+, this would be a huge bottleneck as Mad Man said.

[informal]

Isn't it very convinient this info comes from the intel exclusive place(coolaler forums),and that it is coming only a day after Theo's post??It could be that Theo's results were in some way inflated(since the score is really hard to believe),but this avalanche of "poor K10 results" comes after he wrote this:

This statement warrants at least three hatemails from Intel's R&D lads, but all that we will disclose here are results we have in our possession

[Mad_Man]

Bus speed is the HTT. HT is different than the bus speed/FSB/HTT.

i was talking about this:
cpuz 1.41 should be barcelona compatible, if i can call it that.
so cpu-z should show bus speed as 200mhz, and HT link 1000mhz for stock system??

[Sparky]

flat-four;2402277']Shouldn't CPUZ show "HT Link" link instead of "Rated FSB" like usual AMD cpus? If the HT link is actually running at 200 instead of 1000+, this would be a huge bottleneck as Mad Man said.

Should, don't know why it says FSB since AMD doesn't use a FSB. Should have the bus speed (the HTT, has been 200MHz on K8) and the HT link (1000MHz on K8). Tells me that CPU-Z still doesn't detect K10 properly.
Unless that is a fake screenshot, with an intel chip in there and things mspainted or something like that ;) :p:

[terrace215]

Isn't it very convinient this info comes from the intel exclusive place(coolaler forums),and that it is coming only a day after Theo's post??It could be that Theo's results were in some way inflated(since the score is really hard to believe),but this avalanche of "poor K10 results" comes after he wrote this:

Conspiracy theories? The results fit with AMD's OWN DOCUMENTS on K10 performance vs. K8. Why can't you accept that AMD knows how the K10 performs? :shrug:

Don't tell me you listened to the non-technical hype from Pat Patla and the now-fleeing Henri Richard?

[SEA]

...
And try to disable L2 - you will see slide show instead of your windows

If L2 disabled but L3 is enabled - this will change things from slideshow significantly.

[informal]

Conspiracy theories? The results fit with AMD's OWN DOCUMENTS on K10 performance vs. K8. Why can't you accept that AMD knows how the K10 performs? :shrug:

Don't tell me you listened to the non-technical hype from Pat Patla and the now-fleeing Henri Richard?

Lmao.With all the NDAs around,somehow Sun's got leaked and spilled out the all hush hush about K10 performance and summed it up to poor "up to 15% IPC improvement":rolleyes:

Can you post anything positive about AMD,ever?

[savantu]

If L2 disabled but L3 is enabled - this will change things from slideshow significantly.

Oh sure , the magic 38 cycle latency L3 will come to the rescue.

Sorry , but only an idiot could believe any of the caches or execution units to be disabled when the chip performs +/- 10% of C2D.

Reality is this : the chip is perfectly fine, but it's "bugs" prevent it from scaling , that is reaching higher clockspeed while maintaining data integrity.In other words ; you're fine at 2GHz , but at 2.3GHz* due to speedpath problems you get silent data corruption or other nasty surprises.

* Example.

[savantu]

Lmao.With all the NDAs around,somehow Sun's got leaked and spilled out the all hush hush about K10 performance and summed it up to poor "up to 15% IPC improvement":rolleyes:

Can you post anything positive about AMD,ever?

I have the same number from a Cray presentation ; 15% better IPC ( that is for HPC code , which K8/10 likes most ) + extra % due to higher clock ( didn't happen ).

Can you post something anything true about AMD , ever ?

[Sparky]

Oh sure , the magic 38 cycle latency L3 will come to the rescue.

Sorry , but only an idiot could believe any of the caches or execution units to be disabled when the chip performs +/- 10% of C2D.

Reality is this : the chip is perfectly fine, but it's "bugs" prevent it from scaling , that is reaching higher clockspeed while maintaining data integrity.In other words ; you're fine at 2GHz , but at 2.3GHz* due to speedpath problems you get silent data corruption or other nasty surprises.

* Example.

Reality is this:
You, nor I, nor most people here know enough to say for certain how the chip is going to perform. We have conflicting benchmarks from different sources. Just wait until after Sept. 10th and then we'll know for sure.

And no reason to be calling people idiots :rolleyes: just gonna raise tempers and not do any good. Geeez why can't people be respectful enough of each other to refrain from name-calling.... :rolleyes:

[SEA]

Oh sure , the magic 38 cycle latency L3 will come to the rescue.

Sorry , but only an idiot could believe any of the caches or execution units to be disabled when the chip performs +/- 10% of C2D.

Reality is this : the chip is perfectly fine, but it's "bugs" prevent it from scaling , that is reaching higher clockspeed while maintaining data integrity.In other words ; you're fine at 2GHz , but at 2.3GHz* due to speedpath problems you get silent data corruption or other nasty surprises.

* Example.

L2 is still twice faster than memory.
Also it has some smart prefetch.

Is there any such test showing how much performance would degrade with L1 and L3 enabled for other chips that already on market?

And finally, why K10 having lots of improvments against K8 shows exactly same performance per core? Where is 15% over K8???

[kl0012]

Oh sure , the magic 38 cycle latency L3 will come to the rescue.

Sorry , but only an idiot could believe any of the caches or execution units to be disabled when the chip performs +/- 10% of C2D.

Reality is this : the chip is perfectly fine, but it's "bugs" prevent it from scaling , that is reaching higher clockspeed while maintaining data integrity.In other words ; you're fine at 2GHz , but at 2.3GHz* due to speedpath problems you get silent data corruption or other nasty surprises.

* Example.
+1
I don't understand why ppl thinks that it must be faster then C2D/C2Q in non-bandwidth dependent aplications.

[red]

If L2 disabled but L3 is enabled - this will change things from slideshow significantly.

http://www.xtremesystems.org/forums/showpost.php?p=2402255&postcount=173



the benches are from a 2 socket opteron socket and as additional info the memory is running 667 speeds cas5. what superpi perfromance do you expect.
the performance can't be compared with the phenom. try to bench a k8 with memory 667 cas5 and then change it to 1066 cas5 like the demo system AMD is showing. even at the same speed it will be a huge performance difference.

looking and comparing the cpu-z from Coolaler to mine it is a fact that his sample is a dvt or early bird from oem. so it will be the chip that launches within a few days. All features are enabled, K10 is no INT miracle it will be as fast as core and no penryn is not faster then conroe, it's just when the program is affected by cache or not it will be faster clock/clock and the magic sse4. k10 will increase the gap on fpu and IO even more not to mention virtualization. How well it scales will be depending on the new stepping coming in oktober and you'll see how well it does when people get phenom es in there hands.

informal's information about the bios is correct, it is changing alot, systems with b0 had no CPU options in BIOS available except changing HT speed.

The chips aren't bugged??? it's the BIOS at worst. Or maybe this is K10 performing at its best lol

[kl0012]

And finally, why K10 having lots of improvments against K8 shows exactly same performance per core?
Actualy it is faster than K8 per core (if these tests are true).

[informal]

Actualy it is faster than K8 per core (if these tests are true).

Yeah the whole 3-8%.

[SEA]

Actualy it is faster than K8 per core (if these tests are true).

Actually it is within 4% difference for me.
Need screeny?

[Donnie27]

there are details that are still missing here...what motherboard(chipset), what ddr2 ecc single or dual channel? etc...
a lot of people giving opinions but what's the platform that this was tested...

So why were folks being flamed for suggesting old boards would suck with new processors? So, why bother with all of the BS about Socket Continuity? Guys, you can't have it both ways. It's either great with old boards and it's all good or its only so so and a new board is needed anyway. Ibet a new board is needed.

Rumor! A guy in Germany told me one tester is running Penryn on an i865, I told him I'd believe it when I see it. Yet, I wouldn't touch that setup with AGP, SATA1 and all of yester-year's tech. In fact, I'm already wanting to upgrade my GA-965 DS3.

[kl0012]

Yeah the whole 3-8%.

I doubt we can expect more in non-SSE2-vectorized aplications. As was said before it is only up to 15% IPC improvements.

[Hornet331]

Rumor! A guy in Germany told me one tester is running Penryn on an i865, I told him I'd believe it when I see it. Yet, I wouldn't touch that setup with AGP, SATA1 and all of yester-year's tech. In fact, I'm already wanting to upgrade my GA-965 DS3.

Same here. :D

whatever is out in Nov-Dec and is the fastest, I'll buy. :)

[PetNorth]

Actualy it is faster than K8 per core (if these tests are true).

not really. In Cinebench is exactly the same, what doesn't make any sense, keeping in mine K10 improvements.

[jabway]

not really. In Cinebench is exactly the same, what doesn't make any sense, keeping in mine K10 improvements.

Somehow the massive core improvements amount to a 3-4% increase against K8.:rolleyes:

Doesn't make any sense at all.

[Nedjo]

What's to laugh about ?

Have you heard of Fusion , Silverthorne ?

The purpose of such devices is to move the PC as we know it on the cellphone. Current cellphones are slowly going uphill into PC territory ; this will advance at a very rapid pace once system-on-a-chip devices appear next year.The whole point is to move x86 down , into cellphone territory and this is the purpose of Silverthorne early next year. ( Iphone 2 will use it according to rumours )

This is real and is happening now ; both Intel and AMD are preparing for the 2010-2012 generation , by that time , cell phones will be more powerfull than current PCs.Add 3D glasses to them and all my multimedia needs are satisfied.

/End off topic.

so you're SUN guy:
http://www.dailytech.com/article.aspx?newsid=8658
:)

[uOpt]

not really. In Cinebench is exactly the same, what doesn't make any sense, keeping in mine K10 improvements.

The numbers are about 7% better per-core with the clockspeed normalized.

[PetNorth]

No, like I said before, I've run my X2 at 2.0ghz (with cinebench 10 x64), and 1cpu result was 1905 (1896 this K10).

[Donnie27]

Same here. :D

whatever is out in Nov-Dec and is the fastest, I'll buy. :)

Doesn't matter, I can be hungrey for and buy the Phenom X4. According to some in this thread, I'd still be called an Intel Fanboy LOL!

[JVguest]

I have the same number from a Cray presentation ; 15% better IPC ( that is for HPC code , which K8/10 likes most ) + extra % due to higher clock ( didn't happen ).

Can you post something anything true about AMD , ever ?

I can believe K10 would run rings around K8 at same clock. The mere doubling of SSE throughput means that a lot of HPC apps will run near twice as fast. Core 2 is in many cases near twice as fast as P4 clock for clock
in HPC is some cases up to 4 times as fast.
The 32 byte fetch means K10 should be a floating point monster (HPC).The load forwarding capabilities of K8 are quite deficient (none!) compared to Core 2 ( load forwarding already in pentium pro) which means that their inclusion in K10 will give an even bigger boost than Core 2 got from it. Too bad the clock rate is low and the cache is relatively small.

[kl0012]

The 32 byte fetch means K10 should be a floating point monster (HPC).
32 byte fetch can help with decoding long instructions, but K10 still limited by 3 x pipeline (Core has 4 x pipeline). It can't help in legacy code with short instructions.
But Core(tm) feature 64-byte fetch buffer wich can help short loops run faster (on any code).


The load forwarding capabilities of K8 are quite deficient (none!) compared to Core 2 ( load forwarding already in pentium pro) which means that their inclusion in K10 will give an even bigger boost than Core 2 got from it. Too bad the clock rate is low and the cache is relatively small.
Core(tm) is still better in almost all which is related to the memory subsytem.

http://www.xbitlabs.com/articles/cpu/display/amd-k10.html#sect0

As a result, we see that the memory subsystem of K10 processors has undergone some positive improvements. But we still have to say that it still potentially yields to the memory subsystem in Intel processors in some characteristics. Among these features are: the absence of speculative loading at unknown address past the write operations, lower L1D cache associativity, narrower bus between L1 and L2 caches (in terms of data transfer rate), smaller L2 cache and simpler prefetch. Despite all the improvements, Core 2 prefetch is potentially more powerful than K10 prefetch. For example, K10 has no prefetch at instruction addresses so that we could keeps track of individual instructions, as well as no prefetch from L2 to L1 that could hide L2 latency efficiently enough. These factors can have different effects on various applications, but in most cases they will determine higher performance of Intel processors.

[EvlUndrWareNome]

Until i see a reputable site trusted by XS post a full review, its all speculation. Lets wait and see mature results, enough with the speculation.

[STaRGaZeR]

Until i see a reputable site trusted by XS post a full review, its all speculation. Lets wait and see mature results, enough with the speculation.

This is speculation, 30k in 06 is a fact. :clap: :rolleyes:

[jabway]

This is speculation, 30k in 06 is a fact. :clap: :rolleyes:
Inferring what isn't there?:rolleyes:
He's talking about all the speculation, not just this.

[STaRGaZeR]

Inferring what isn't there?:rolleyes:
He's talking about all the speculation, not just this.

Seems that you have not read the other thread :)

[JVguest]

32 byte fetch can help with decoding long instructions, but K10 still limited by 3 x pipeline (Core has 4 x pipeline). It can't help in legacy code with short instructions.
But Core(tm) feature 64-byte fetch buffer wich can help short loops run faster (on any code).


Core(tm) is still better in almost all which is related to the memory subsytem.

http://www.xbitlabs.com/articles/cpu/display/amd-k10.html#sect0

As a result, we see that the memory subsystem of K10 processors has undergone some positive improvements. But we still have to say that it still potentially yields to the memory subsystem in Intel processors in some characteristics. Among these features are: the absence of speculative loading at unknown address past the write operations, lower L1D cache associativity, narrower bus between L1 and L2 caches (in terms of data transfer rate), smaller L2 cache and simpler prefetch. Despite all the improvements, Core 2 prefetch is potentially more powerful than K10 prefetch. For example, K10 has no prefetch at instruction addresses so that we could keeps track of individual instructions, as well as no prefetch from L2 to L1 that could hide L2 latency efficiently enough. These factors can have different effects on various applications, but in most cases they will determine higher performance of Intel processors.

1st point. That's a loop detector. I'm interested in pure SIMD FP, where 32 byte fetch should help.

2nd point. AMD has implemented a write burst buffer and real RAM prefetcher into the IMC. Intel probably has better prefetchers. The real hurt is the low launch clockspeed if true.

[informal]

People seem to forget how one early K7 sucked (simply put)
Link:http://firingsquad.com/hardware/k7550preview/page7.asp
Even loses to K6-3 at the same clock.Has abysmal FPU performance,also goes for integer too.

[w0mbat]

People seem to forget how one early K7 sucked (simply put)
Link:http://firingsquad.com/hardware/k7550preview/page7.asp
Even loses to K6-3 at the same clock.Has abysmal FPU performance,also goes for integer too.

Do u think that 11 days before launch these CPUs are early K10s?

[Donnie27]

Do u think that 11 days before launch these CPUs are early K10s?

I do:) So few good ones are out there that leaks of the so called bad ones would likely be seen first.

Yet, I'll say what I've said from day one. If AMD had something to show, they or someone friendly to them would have shown it by now. I hope like hell I'm wrong on this one.

[KTE]

Oh sure , the magic 38 cycle latency L3 will come to the rescue.The AMD stated working latency is "less than 38 cycles and depends on the clock speed of the southbridge". Higher clock speeds offset the latency as in all processors. L3 cache is just the shared victim cache for the L2 cache, nothing more. It operates to reduce latency very well between RAM<->CPU for the K10 as the larger L2 does in Core 2.

K8 had a 12 stage pipeline, Barcelona a 12 stage, and Core 2 a 14 stage.

K8 L2 latency is 12 clock cycles, Core 2 is 14 and Barcelona is 12.

K8 L2 cache bus width is 128-bit, Core 2 is 256-bit, Barcelona is 128-bit.

SSE engine width of K8 was 64-bit (2 per clock), Core 2 was 128-bit (3 per clock) and Barcelona is 128-bit (2 per clock).

L1+L2 cache latency is 15 cycles for the K8, 17 cycles for Core 2, and 15 for Barcelona IIRC.

Correction: L1+L2 cache access combined latency is median 13 cycles for Core 2 and Barcelona. That's twice as much data in the same time frame accessed by Core 2 due to the double bus width between L2<->Core.

There's much more improvements with larger stack load and reordering of load/store of the many which have the potential to make the most difference. Many of the improvements are identical to what was done with Yonah -> Core 2. Many more specific, and even some more advanced.

Based on the technicalities, the improvement seems like this:

K8 > K10 as with Yonah > Core 2. Like I forementioned, I think its a retail clock speed yield race, nothing more. We'll wait and see how it pans out in reality.

[Theli]

Until i see a reputable site trusted by XS post a full review, its all speculation. Lets wait and see mature results, enough with the speculation.
Coolaler isn't trusted by XS?
From what I recall, the early Kentsfield benchmarks found on that forum were quite accurate, and so were the Core 2 overclocks. Faking a benchmark like this would be really stupid, seeing how easy it would be to prove the forgery once the CPU's went retail.

The results are disappointing taking Intel's current and future (45nm) offerings into consideration, but they are not unreasonable. They show a few percentages performance increase over the K8.
There will probably be benchmarks where the K10 scores like the K8, and there will probably be benchmarks where the performance delta is bigger than in Cinebench (say circa 20-30&#37;).

[TigeriS]

http://firingsquad.com/hardware/k7550preview/

Take a good look at that. Those were pretty bad K7 pre-launch benchmarks. And we all know the K7's story don't we?

Better wait a few more days. You never know with AMD, I remember T-Bred A and B. Well that was a surprise too, many thought AMD couldn't get the K7 over 2GHz properly when T-Bred A came out.

I'd say don't worry too much, AMD has the reputation to exceed everyone's expectations. It happened so many times before.

Edit, wow informal, I for sure take tooo long to write a reply.
Anyway, that's the same point I wanted to present here.

[Donnie27]

Coolaler isn't trusted by XS?
From what I recall, the early Kentsfield benchmarks found on that forum were quite accurate, and so were the Core 2 overclocks. Faking a benchmark like this would be really stupid, seeing how easy it would be to prove the forgery once the CPU's went retail.

The results are disappointing taking Intel's current and future (45nm) offerings into consideration, but they are not unreasonable. They show a few percentages performance increase over the K8.
There will probably be benchmarks where the K10 scores like the K8, and there will probably be benchmarks where the performance delta is bigger than in Cinebench (say circa 20-30%).

Abso-Frackin'-lutely QFT!

Nothing really wrong with questioning the the Platform or even the revision of the Processor. But I do trust Coolater and many of his cohorts!

[Donnie27]

http://firingsquad.com/hardware/k7550preview/

Take a good look at that. Those were pretty bad K7 pre-launch benchmarks. And we all know the K7's story don't we?

Better wait a few more days. You never know with AMD, I remember T-Bred A and B. Well that was a surprise too, many thought AMD couldn't get the K7 over 2GHz properly when T-Bred A came out.

I'd say don't worry too much, AMD has the reputation to exceed everyone's expectations. It happened so many times before.

I remember them very well and hopes like Hell Barkie doesn't follow in its footsteps. All too often the Platform those first K7's shipped with sucked big ones:D Cu-Mine put Intel right back in the Game until Tbird Launched.

[Dainas]

I remember them very well and hopes like Hell Barkie doesn't follow in its footsteps. All too often the Platform those first K7's shipped with sucked big ones:D Cu-Mine put Intel right back in the Game until Tbird Launched.


Kinda funny to note that both Pentium III and K7 did not reach their pinnacle until they were well in the shade of anticipation for the newer tech. Barton and Tualatin, too few remember ye.

K8 has gone through allot too, but besides the X2... it hardly feels like an mutated alien in the face of its original self.

[hurleybird]

Do u think that 11 days before launch these CPUs are early K10s?

Hmm, lets see.... :rolleyes:


It has also been brought to our attention that there has been question as to how old the particular K7 that we looked at was, as some people (and websites) have speculated that it is a very old "debug" model. Quite frankly, it's not. While it is an "Engineering Sample" the K7 we have is a very recent build, and according to our sources, is as close to release as AMD will be able to accomplish by June. It is highly unlikely that any architectural changes will be made to the K7 from the one that we were able to see, if the timeline for its release is still being followed. We are talking about a product that is supposed to be released within a month's time!

Does that answer things for you? :D

[xPliziT]

Guy's

i would like to mention that AMD said that software using the FPU need a re- compile to take full advantage of the new Barcelona FPU right ?
So, running current benchi's like cinebench or super pi with apparently 'old' code doesnt show the full speed up the K10 would get.

Dunno if super PI uses FPU but basically alot of scientific programs use it.
Please correct me if i am wrong here.

[Clairvoyant129]

I love how everyone is coming up with conspiracy theories. I totally trust Coolaler.

[accord99]

Guy's

i would like to mention that AMD said that software using the FPU need a re- compile to take full advantage of the new Barcelona FPU right ?
So, running current benchi's like cinebench or super pi with apparently 'old' code doesnt show the full speed up the K10 would get.
Conroe showed it strength immediately in the same old software. Plus, Barcelona's FPU didn't get upgraded much, more effort went into doubling the SSE capabilities in order to match Conroe.

[hurleybird]

I love how everyone is coming up with conspiracy theories. I totally trust Coolaler.

I love how discussions like these bring out the stupid in stupid people.

*sigh*

First off, it's not Coolaler, it's some guy on the site's forum. Secondly, remember the original Athlon previews? Even if somebody credible were to bench the chip, the results would still be bogus it it were bugged. Thirdly, Rahul Sood has stated that the 3GHz Phenom is considerably faster than all of AMD and Intel's current offerings. These benchmarks fly against Rahul's claims, and, quite frankly, I trust Rahul Sood just a bit more than some guy on some forum who's possibly using bugged chip/mainboard/bios to boot!

[accord99]

I love how discussions like these bring out the stupid in stupid people.

*sigh*

First off, it's not Coolaler, it's some guy on the site's forum. Secondly, remember the original Athlon previews? Even if somebody credible were to bench the chip, the results would still be bogus it it were bugged.
How about the AM2 previews, they turned out to be correct. Performance decrease vs DDR-400 unless paired with DDR2-800 memory.



Thirdly, Rahul Sood has stated that the 3GHz Phenom is considerably faster than all of AMD and Intel's current offerings. These benchmarks fly against Rahul's claims, and, quite frankly, I trust Rahul Sood just a bit more than some guy on some forum who's possibly using bugged chip/mainboard/bios to boot!
Sood also questioned the initial Conroe benchmarks so I doubt he really knows much more than anybody else.

[Movieman]

I love how discussions like these bring out the stupid in stupid people.

*sigh*

First off, it's not Coolaler, it's some guy on the site's forum. Secondly, remember the original Athlon previews? Even if somebody credible were to bench the chip, the results would still be bogus it it were bugged. Thirdly, Rahul Sood has stated that the 3GHz Phenom is considerably faster than all of AMD and Intel's current offerings. These benchmarks fly against Rahul's claims, and, quite frankly, I trust Rahul Sood just a bit more than some guy on some forum who's possibly using bugged chip/mainboard/bios to boot!

Put's on Mod hat:
And how would you feel if I refered to you in that way?
Make your point without the name calling or go elsewhere.
I've said it nice, now I'll say it not nice..Stop the flaming!

[Clairvoyant129]

I love how discussions like these bring out the stupid in stupid people.

*sigh*

First off, it's not Coolaler, it's some guy on the site's forum. Secondly, remember the original Athlon previews? Even if somebody credible were to bench the chip, the results would still be bogus it it were bugged. Thirdly, Rahul Sood has stated that the 3GHz Phenom is considerably faster than all of AMD and Intel's current offerings. These benchmarks fly against Rahul's claims, and, quite frankly, I trust Rahul Sood just a bit more than some guy on some forum who's possibly using bugged chip/mainboard/bios to boot!

Stupid people huh? Like the person before said, everyone thought AM2 was going to be competitive with Core 2 clock for clock, well well look what happened.

Atleast this seems more crediable than the 30K in 3DMark06 :rolleyes:
And everyone else who believes in that 30K isn't "stupid," right?

Stupid.

[informal]

How about the AM2 previews, they turned out to be correct. Performance decrease vs DDR-400 unless paired with DDR2-800 memory.


Sood also questioned the initial Conroe benchmarks so I doubt he really knows much more than anybody else.

Lol man,Sood is the head of HP gaming department (you've heard of that company HP,have you?)
If AMD wants Phenom in HP's systems,they sure have handed a few Phenoms to them and asked them to sign the stuff all others have signed.Other than breaking NDA,Sood simply said Phenom @ 3ghz is "stone cold killer" and that it is and will be considerably faster than any intel or AMD chip when it is out.

[SOLDNER-MOFO64]

Other than breaking NDA,Sood simply said Phenom @ 3ghz is "stone cold killer" and that it is and will be considerably faster than any intel or AMD chip when it is out.

and he'd be correct :)

[Movieman]

Gentlemen:
I settled the statement.
The next one who quotes that guy and keeps this going on the stupid comment takes a vacation? Is that plain enough? Stop the fighting!

[EvlUndrWareNome]

Lets see a concrete review. A review with pictures and an author. I do not trust pics from a foreign nation across the world.

The only hype or benches I will believe is when it is a true posted review, written up by an author, with pics of the packaging and cpu. I want to know what chipset it is running on as well as memory.

I dont trust a few screen shots from across the globe as a reliable performance benchmark.

Also note these are early opteron K10's not the Phenoms that willl be out later this year, nor the chipset we will run them on, and most likely different ram speeds/timings.

[JumpingJack]

Also note these are early opteron K10's not the Phenoms that willl be out later this year, nor the chipset we will run them on, and most likely different ram speeds/timings.

So, just curious, are you saying it is in AMD's best interest to release the buggy, less reliable material into servers first?

[informal]

So, just curious, are you saying it is in AMD's best interest to release the buggy, less reliable material into servers first?

No he means EVTs are not representatives of the final performance...

[JumpingJack]

No he means EVTs are not representatives of the final performance...

Thanks.... typically enterprise chips are the cream of the crop. Though there is some ambiguity in the stepping represented here, it doesn't look like CPUID has been updated with the CPUID string/look up table to designate stepping yet... so if this is B1 as some are postulating, then this is the launch stepping is it not?

This is of course proceeding under the assumption that this data is factual and not made up, back alley leaks like these deserve a healthy dose of skepticism.

[informal]

Thanks.... typically enterprise chips are the cream of the crop. Though there is some ambiguity in the stepping represented here, it doesn't look like CPUID has been updated with the CPUID string/look up table to designate stepping yet... so if this is B1 as some are postulating, then this is the launch stepping is it not?

This is of course proceeding under the assumption that this data is factual and not made up, back alley leaks like these deserve a healthy dose of skepticism.

A small number of people really knows what is the launch stepping...
There is one more possibility,apart from EVTs,the boards weren't up to the task(ie early BIOS,some features turned down just to be safe in order to test the early chips for various reasons)

[JumpingJack]

A small number of people really knows what is the launch stepping...
There is one more possibility,apart from EVTs,the boards weren't up to the task(ie early BIOS,some features turned down just to be safe in order to test the early chips for various reasons)

The BIOS argument is a decent one, but typically BIOS code does not interact to produce a computational result... it provides very low level IO code, most of which OSes simply ignore today. Though it could be that there are initializations for various functions that are not occuring correcty and such, but I have never really experienced a BIOS update, even a buggy one, to affect performance that much.

Nonetheless, there is way way too much contradictory information or conjecture. Theo seems to think it is a monster, Hector Ruiz is downplaying the significance, Raul is certain of great performance, AMD has conceded single thread perfomance to Intel in various press statements.... so it is hard to know what to make of this.

[Donnie27]

Kinda funny to note that both Pentium III and K7 did not reach their pinnacle until they were well in the shade of anticipation for the newer tech. Barton and Tualatin, too few remember ye.

K8 has gone through allot too, but besides the X2... it hardly feels like an mutated alien in the face of its original self.

Yup!

[terrace215]

,Sood simply said Phenom @ 3ghz is "stone cold killer" and that it is and will be considerably faster than any intel or AMD chip when it is out.

No, he didn't say that, actually.

This is what he said:

And for the record, if you were to benchmark Phenom at 3GHz you would see that it kicks the living crap out of any current AMD or Intel processor

It's also worth noting here that Rahul has been an AMD fan for some time, predicted good Q1 results for them, and IIRC, has mentioned that he is invested in their stock.

[terrace215]

The BIOS argument is a decent one, but typically BIOS code does not interact to produce a computation result... it provides very low level IO code, most of which OSes simply ignore today. Though it could be that there are initializations for various functions that are not occuring correcty and such, but I have never really experienced a BIOS update, even a buggy one, to affect performance that much.

Nonetheless, there is way way too much contradictory information or conjecture. Theo seems to think it is a monster, Hector Ruiz is downplaying the significance, Raul is certain of great performance, AMD has conceded single thread perfomance to Intel in various press statements.... so it is hard to know what to make of this.

Don't leave out the most important indicator: Henri Richard up and quit right before the K10 is to launch. Says it all right there.

[JumpingJack]

Don't leave out the most important indicator: Henri Richard up and quit right before the K10 is to launch. Says it all right there.

I don't read too much into this, other than to say -- top execs leave for various reasons, and Richard's departure, especially the timing, subjects the event to an enormous amount of speculation.

My underlying point is that we simply are still in the dark, this back-alley bench is the most data was have had to date, and it consisted of Cinebench, CPUMark, and SP1M... we have various people making strong but non-data backed claims of trouncing this or failing that...

Having said that, this data did originate on Coolaler's site... that name alone attaches a certain degree of credibility to the results based past experience and on a reputation some of which was built on this site. However, it is still not sanctioned, and fails some basic principles, such as clearly spelling out all the details necessary to reproduce the results -- and, most importantly, the stepping is ambigous at best ... it is none-the-less-fun to debate.... but this is such a hot topic, I approach this with a high degree of trepidation. Do not underestimate the degree of importance this product is to AMD, they are betting the farm on it and, frankly, I prefer to give them the benefit of the doubt.

[LordEC911]

Don't leave out the most important indicator: Henri Richard up and quit right before the K10 is to launch. Says it all right there.

You don't think he was forced out???

[Zytek_Fan]

You don't think he was forced out???

I think he was getting bored. Being the chief of marketing at a company that doesn't market very much makes little sense. I believe Henri resigned because he wanted to do the best thing for the company, and seeing he had little to offer at the moment, he resigned. Takes a lot of balls to do something like that. :up:

[JumpingJack]

AMD is LAUNCHING the K10 on the B1 stepping, per everyone else. Then B2 in Q4. At any rate, it is unlikely that B1 has any huge performance bugs. These results fit precisely with what AMD themselves claimed.... up to 10-15% better IPC than the K8. About in the ballpark of Conroe/Kentsfield per clock. Better on some bandwidth-heavy server stuff & HPC, worse on desktop apps.

It is hard to determine what stepping this site is showing, if it is real (my little disclaimer :) ).... but there is a hint.

The CPUID string that CPUID read from BIOS does not read "AMD Opteron processor unknown" or "Engineering Sample (ES)".... AMD uses a special registers, which the BIOS reads and constructs the processor name and model string, and information.... see for example, page 345 http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/32559.pdf

The CPUID report from Coolaler's site correctly identifies the processor family as 10, and also does not read the CPUID string as anything other than the branded processor name....

Assuming this is not faked (which is well could be), it appears to be the release product.

Jack

[Donnie27]

Put's on Mod hat:
And how would you feel if I refered to you in that way?
Make your point without the name calling or go elsewhere.
I've said it nice, now I'll say it not nice..Stop the flaming!

Oh and I PM-ed you:) Before I forget, Rahul Sood was one of the folks calling those early Conroe test done here by *that guy, fugger, fcg, victor and etc bogus. Hinted that some of the folks here was being paid. That was the cause that misunderstanding in that other post. At least 4 Webmasters weren't as rude but said similar things.

Saying he is bias is contrary to what he says, his own words.


Many people have been asking for my latest perspective on AMD, and I just finished writing an article on them. I wrote it because I love the company, but I am utterly disappointed in what has transpired with them lately. I haven't lost hope though, but I think there are changes necessary in order to turn around. You'll see it on the blog in a few days.

Love blinds!

[freeloader]

I'll wait for the real release from AMD before I comment on the performance; not that my opinion means anything, anyhow. :D I'd like to see how these chips run on updated BIOS and motherboards (shipping) at speeds of 2.5ghz plus.

[nn_step]

Love blinds!

So do M23 NFDDs (Noise and Flash Diversionary Devices aka flashbangs)
:rolleyes:
but eventually the truth is reveled

[Zytek_Fan]

Wait and the truth will be revealed.

[fakhrain]

Stupid people huh? Like the person before said, everyone thought AM2 was going to be competitive with Core 2 clock for clock, well well look what happened.
well i don't. in fact there was still a lot of people who realized that AM2 is just the same old S939 but with DDR2 support.

[Kenetixx]

Man these Phenom threads are getting messy :|

[Zytek_Fan]

Man these Phenom threads are getting messy :|

Because the people posting positive stuff (including me) are excited :up: :yepp: :D :shocked:

[kl0012]

Because the people posting positive stuff (including me) are excited :up: :yepp: :D :shocked:
I suppose the stuff must be objective, not positive/negative.

[jmilcher]

Put's on Mod hat:
And how would you feel if I refered to you in that way?
Make your point without the name calling or go elsewhere.
I've said it nice, now I'll say it not nice..Stop the flaming!

I totally agree, people tend to particiapte more and enjoy the threads that dont involve flaming...

But the nature of this beast topic is hard not to show your true fanboy colors.

6 months from now there will be no arguing.

[savantu]

so you're SUN guy:
http://www.dailytech.com/article.aspx?newsid=8658
:)


Nope , it's just thinking ahead into the future ; it doesn't take a genius to notice the current trends , just look at Intel IDF's presentations about ultra mobile stuff.

[savantu]

Somehow the massive core improvements amount to a 3-4% increase against K8.:rolleyes:

Doesn't make any sense at all.

Yes they do.

The K8 core is already strong to begin with.The law of diminishing returns is in full force here.

When you already start high , the improvements you make will bring little advantages.

Compare that with the P4 core which was far more fragile ( code quality was vital while the K8 eats just about anything ). Core brought massive improvements over the P4 , the score jumped a lot.Even compared to the K8 , core has a huge number of improvements , in the end it is only 20% better on average.

There are a lot of situations where K8 core+ 1MB L2 will be as fast or faster than K10 core + 512Kb L2 + 2MB L3 , especially where latency counts.In multithreaded apps , this will be more pronounced as the L3 will get trashed by different threads.

[biohead]

Put's on Mod hat:
And how would you feel if I refered to you in that way?
Make your point without the name calling or go elsewhere.
I've said it nice, now I'll say it not nice..Stop the flaming!

not trying to fuel a fire here, but calling stupid people stupid isn't flaming, it's just calling them exactly what they are.

[ >> continue ]