Shintai, why not wait barcelona availibity. You will not be able to convince those guys before.
Printable View
Shintai, why not wait barcelona availibity. You will not be able to convince those guys before.
Shintai you are really taking that article seriously? Anyway K10 has double the SSE throughput of K8. What are we even discussing here?
You chose beetwen what Allen has said, you don't care about what allen said June 29 2007 but you keep old allen mistakes? But probably, you've tested K10, and you even better know what is true beetwen allen x 3,6 and allen 40-50% better...Quote:
Originally Posted by informal
"Old Allen mistakes?"Quote:
Originally Posted by nemrod
Are you serious man? The guy who was behind K9(dual core K8) has made such a dumb mistake?Wow,just wow.
The June quote was probably the reference to low clocked 2 Ghz part(and we don't know what that number really represents,do we?)
Yes wow, just wow if not him the writer of the paper has not understand and fail to tell what Allen said.Quote:
Originally Posted by informal
That is a transcript man.Transcript.And the per core reference has been repeated by Allen a couple of times.Still not clear?Quote:
Originally Posted by nemrod
Jesus, could the mods here do something about this thread crapping?...what was once a nice thread has turned into a flame fest... :(
could you read that february
http://www.amd.com/us-en/Corporate/V...118193,00.html
Quote:
Optimum Performance Levels
While special attention was paid to the power-saving design features in native Quad-Core AMD Opteron processors, equal emphasis was placed on delivering industry-leading performance. Quad-Core AMD Opteron processors are packed with core and cache enhancements designed to improve performance on a range of server and workstation applications. Cache-sensitive transactional applications such as Web, database and email servers can benefit from the addition of a 2MB shared L3 cache. Simulations conducted in AMD laboratories indicate that certain database applications will see performance improvements up to 70 percent and certain floating point applications will experience performance gains of up to 40 percent over platforms powered by current dual-core AMD Opteron processors. High-performance computing (HPC) applications can benefit tremendously from a doubling of Barcelona’s floating-point execution pipeline to 128-bit width, which includes an AMD-only doubling of instruction and data delivery capabilities. Finally, through enhancements to AMD Virtualization™ (AMD-V™), including Nested Paging, virtualization customers can enjoy additional application performance improvements over non AMD-V driven applications.
or june
http://www.amd.com/us-en/Corporate/V...118193,00.html
Is it a transcript?Quote:
With planned availability at launch in a range of frequencies up to 2.0 Ghz, AMD expects its native quad-core processors to scale to higher frequencies in Q407 in both standard and SE (Special Edition) versions. Designed to operate within the same thermal envelopes as current generation AMD Opteron processors, AMD estimates that the new processors can provide a performance increase up to 70 percent on certain database applications and up to 40 percent on certain floating point applications, with subsequent higher frequency processors expected to significantly add to this performance advantage.
“More than ever before, customers are expecting energy-efficiency and performance-per-watt leadership as much as absolute performance. With this new reality of computing, greater performance at the expense of greater power consumption is no longer an option,” said Randy Allen, corporate vice president, Server and Workstation Division at AMD. “AMD has prioritized production of our low power and standard power products because our customers and ecosystem demand it, and we firmly believe that the introduction of our native Quad-Core AMD Opteron processor will deliver on the promise of the highest levels of performance-per-watt the industry has ever seen.”
here's a roadmap from DT
http://www.dailytech.com/AMD+SecondG...rticle7913.htm
nemrod,both articles are mentioning the average per. figures(ie 2GHz K10 Vs the current K8s).
Those ZDnet quotes are especially for FP/SSE code!
Read this line please :Quote:
Originally Posted by nemrod
Designed to operate within the same thermal envelopes as current generation AMD Opteron processors, AMD estimates that the new processors can provide a performance increase up to 70 percent on certain database applications and up to 40 percent on certain floating point applications, with subsequent higher frequency processors expected to significantly add to this performance advantage.
Their talking about a 70% increase with the exact same power consumption.
I would say that is very Good:up:
That should apply to you also;)Quote:
Originally Posted by nemrod
not agree "up to" is not average. But it doesn't matter we will see that in a month or two and I'd like to be wrong.Quote:
Originally Posted by informal
yes they put 2x more cores with the same power consumption. but as said we will see laterQuote:
Originally Posted by [XC] Teroedni
Well nobody knows for sure until K10 finally lands in reviewers hands.
It's incredible how lame Theinq can be : http://theinquirer.net/default.aspx?article=40749
1.In the Spec tests they compare the estimated perf of a 2.6GHz K10 vs. 2.66 Clovertown
2."Recently, we managed to sse the short of performance figures that AMD is promising for Barcelona."
Too bad the first 2 slides are form december 06 and the last 2 from february 07.
It's 2.6GHz, clear as day on the footnotes http://multicore.amd.com/GetFile.asp...trate_040207v2Quote:
But in the Integer test, a Barcelona 2.3GHz yields 21% higher score than Clovertown 2.66 GHz, but Floating Point test leaves a staggering 50% performance deficit for Clovertown
http://www.spec.org/cpu2006/results/...528-01175.htmlQuote:
and this is not something 45 nanometre Penryn can solve overnight
106 vs 102, 106 wins. On 65nm.
Wait a minute...Quote:
Originally Posted by savantu
I see this:
and this:Quote:
These performance numbers were based on systems using tge Opteron 2356, or known to the world of us regular folk as AMD QuadCore at 2.3GHz
Where do you see 2.6GHz K10? :confused:Quote:
Barcelona 2.3GHz yields 21% higher score than Clovertown 2.66 GHz
And where are the slides?
Look on the picture. They estimated the performance at 2.6Ghz for barcelona. And gotta love memory intensive synthetic benches. (Last line on the text on the picture)
I probably need to wear eyeglasses from now on...
I don't see any picture there on inquirer! :confused:
Can you please explain why you link AMD slides with link to inquirer? I might miss something...
I'm 100% certain that their story is entirely based off this presentation which has been circling around, http://vd.verysell.ru/files/ie/252_1...ion_PUBLIC.ppt around slide 70. 2.6GHz clear as day.
http://www.xbitlabs.com/news/cpu/dis...702235635.html
X-bit has them up too.
Basically, the SPECrate benches were run at 2.6GHz, the others were at 2.3GHz
I like people that this certain.Quote:
Originally Posted by red
But i am still not - Inquirer did not provide any link, but they stated 2.3GHz for Barcelona (even twice!) -Quote:
Barcelona 2.3GHz yields 21% higher score than Clovertown 2.66 GHz
It is the INQ afterall. Did you check the powerpoint? It says 2.6GHz K10 21% faster than X5355. That would just be a big coincidence that they improved a 2.3GHz K10 to also be 21% faster. Where do you think the TPC-C, SAP-SD, SPECweb benches surfaced from? No other place than the same place he's reporting these old specrate scores, that powerpoint.
The most important, isn't 2.3 or 2.6 in the text butQuote:
Originally Posted by SEA
These figures were from a few months ago,
http://www.xbitlabs.com/news/cpu/dis...702235635.html
its 2.6 vs 2.66.
Once again it confirms that K10 will be a rendering (and not only) monster,
the lowest end K10 should match/surpass the top performance Core2 chips
with ease here.
20% INT clock/clock advantage over C2 isn't negligible either IMO.
this is still the same FUD and even the paper's writer tell it
Quote:
These figures were from a few months ago,
/.../
The only real question that now remains is whther AMD can execute. Sadly, the company's recent track record does not bode well, with constant delays of products and events. And we'll also wait to see real tests of the Barcelonas rather than paper promises
:rolleyes: I've only said this many times... AMD compares their 2.6GHz K10 score of 102 to stuff before 4.16.07. This is the result AMD is referencing for Intel's INT score. http://www.spec.org/cpu2006/results/...219-00526.html The best X5355 has since gone up to 101 http://www.spec.org/cpu2006/results/...612-01275.html and a 3GHz score of http://www.spec.org/cpu2006/results/...528-01175.html
clock/clock does not matter. What matters is wattage, performance, cost.
BTW, compare the percentages of http://www.xbitlabs.com/news/cpu/dis...702235635.html to http://tweakers.net/reviews/661 and you might not be so impressed.
Umh...no.Quote:
Originally Posted by red
They were "simulated".You take a lower clocked part and extrapolate.
Anyway , Intel submitted new scores for Clovertown which beat even the simulated 2.6GHz K10.
Spec_Int_rate
5355 : 99.9
5365 : 106
K10 2.6 : 102*
*simulation
Please keep in mind that Spec_rate is the best case scenario for K10 because it makes full use of the superior interconnect technology ( HT vs. FSB ).All other tests will show smaller gains.
You do know K8 already today beats Core 2 with about 20% in that bench?Quote:
Originally Posted by alayashu
And specfp_rate/specint_rate is heavily memory intensive. Why do you think there is no dual dualcore K8 in the benchresults? Its because it would own the K10 due to more memory bandwidth.
It would be like showing DivX encoding with a penryn..and only that.
this has been said many times, the _rate benches do depend on memory BW,Quote:
And specfp_rate/specint_rate is heavily memory intensive. Why do you think there is no dual dualcore K8 in the benchresults? Its because it would own the K10 due to more memory bandwidth.
as they use all the cores of the system (thus the more cores the more it
depends on the BW, just like in real world). But saying it is a pure memory
benchmark is naive from you. The _rate are the right benches for multicore
systems, if i need to say it again...
they should be about equal clock/clock in the fp_rate, maybe you should redoQuote:
You do know K8 already today beats Core 2 with about 20% in that bench?
the bench again, as it looks like some chips get more power with time, as seen above :P
Anyway, why do i have the feeling we discuss the same things over and over? :)
one more month to go, maybe at a certain booth on Siggraph we'll learn more;)
http://www.generation-3d.com/UserImg...big-family.jpgQuote:
Originally Posted by Dailytech
:up:
Well to be able to see what really the CPU can achieve we have many kind of benchmarks, some CPU's can achieve higher scores on a determined test but the overall performance is quite different.
Do you want to, say, read the title instead of cheerleading AMD?
red, it's just powerpoint presentation under NDA
read (if you can) it carefully ))
What does it say?Quote:
Originally Posted by MAS
1. Why do you post the K8 vs Woodcrest graphs?Quote:
Originally Posted by MAS
2. I read it, and there is only 3 (One a reprint of an old one) interresting parts where barcelona is. There is nowhere else in the slides its there with performance numbers.
Inquirer says "systems using tge Opteron 2356"
There is no word about Opteron 2356 in that presentation.
So why you guys still link them together?
It's really simple. http://vd.verysell.ru/files/ie/252_1...ion_PUBLIC.ppt this has been just recently circulating. INQ copied. In the INQ's article, it's clear they're talking about slide 71,72. Theo also babbled about slide 70 even though we've already seen it and even though we've already known it's 2.6GHz.
http://i8.tinypic.com/52wg50p.png
It would be a big coincidence if they somehow improved 2.3GHz to perform like old 2.6GHz at the same 21% rate. But I digress, believe what you want, merry July 4th.
And the 2356 is nowhere present but it's obvious they put 1 (recent surfaced model numbers) and 1 (the 2.3GHz on slide 71,72) and called it the 2356.
Another one on the specint_rate2006
http://blogs.zdnet.com/Ou/?p=567
Sad we have so much FUD today :(
http://blogs.zdnet.com/Ou/images/int...-quad-core.png
sad we have such lame comparisons everywhere....Quote:
Originally Posted by Shintai
its getting really boring, time for AMD to release K10
tired of everyone claiming this and that, time for some hard numbers, i don't care if it are good or bad numbers
http://www.spec.org/cpu2006/results/rint2006.htmlQuote:
Originally Posted by Shintai
If you consult the table you will see that that score of 101 is achived by a Dual xeon X5355 machine.
I´am not understand nothing. Starting with AMD graphic that is very confusing :lol:
Quote:
Originally Posted by v_rr
If you read the footnote of the AMD figure, you will see dual-cpu quadcore :shrug:
http://multicore.amd.com/GetFile.asp...trate_040207v2
Oh, please. Go to AMD's own page:Quote:
Originally Posted by v_rr
http://multicore.amd.com/us-en/AMD-M...rformance.aspx
and you will see that AMD has clarified it's own comparison. Read the fine print at the bottom of the slides.
Of course, the idea that this was anything but a dual socket against dual socket comparison is ridiculous, but I guess there is no limit on the ridiculous statements some fanboys will make if you don't spell everything out.Quote:
Dual-CPU Quad-Core AMD Opteron processor estimates based on internal AMD simulations at 2.6GHz
Lack of brainpower is the culprit...Quote:
Originally Posted by Periander6
Have you Intel droids (and the zdnet genious) notice that those results are with different SOs and of course with different compiler?
-AMD page results are with: WINDOWS SERVER 2003 64bits -32bits spec version- Intel Compiler v9.1 IA32. Here is 84.8 dual 5355 spec page http://www.spec.org/cpu2006/results/...219-00524.html
-Zdnet genious and venerable IT picked results are with (btw not to mention they are from june, AMD page is from april), OK with: SUSE LINUX Enterprise Server 10 64bits, Intel compiler IA32/EM64T v10.0 Here is 101 dual 5355 2.66ghz spec page http://www.spec.org/cpu2006/results/...612-01275.html and here is 106 dual 5365 3.0ghz spec page http://www.spec.org/cpu2006/results/...528-01175.html
How somebody can compare benchs with different SOs and compiler? I know, nonsense fanboys.
Do you even know how the configuration for the other systems AMD used in the comparision?Quote:
Originally Posted by PetNorth
http://www.amd.com/us-en/BenchmarkPl...117239,00.html
I guess its ok for AMD to use 64bit on the K8 vs 32bit on the Woodcrest?
And whats this, now the compiler cant be used? Thats like saying all 2900XT benchmarks with newer better performing drivers is unfair.
And its not like they used the best one for Intel is it, even with the 9.1 compiler and 32bit.
http://www.spec.org/cpu2006/results/...402-00728.html (just example, not sure its the best)
i'm not going to waste time with zealots. But again, Linux. Not windows. And of course compiler do differents, and more with different OS.
Who is the zealots, you don't even click on shintai link. So as you're too lazy.Quote:
Originally Posted by PetNorth
AMD benchmark configuration information for such kind of comparison:
# 2 Xeon 5355 processors in FSC Primergy RX300 S3, 16GB memory, 73GB SAS disk drive, Microsoft Windows®Server 2003 x64 Enterprise http://www.spec.org/cpu2006/results/...219-00524.html
# 2 AMD Opteron™ processors Model 2222 SE in Tyan Thunder n4250QE motherboard, 8GB memory, 80GB SATA disk drive, SuSE Linux® Enterprise Server 9 SP3 http://www.spec.org/cpu2006/results/...319-00684.html
# 2 Xeon processors 5160 in FSC PRIMERGY TX300 S3, 16GB memory, 73GB SAS disk drive, Windows® 2003 Server Enterprise Edition SP1 http://www.spec.org/cpu2006/results/...122-00249.html
# 2 Xeon 5355 processors in FSC Primergy RX300 S3, 16GB memory, 73GB SAS disk drive, 64-Bit SUSE LINUX® Enterprise Server 10 http://www.spec.org/cpu2006/results/...219-00523.html
# 2 AMD Opteron™ processors Model 2222 SE in Tyan Thunder n4250QE motherboard, 8GB memory, 80GB SATA disk drive, SuSE Linux Enterprise Server 9 SP3 http://www.spec.org/cpu2006/results/...319-00685.html
# 2 Xeon processors 5160 in FSC PRIMERGY TX300 S3, 16GB memory, 73GB SAS disk drive, 64-bit SUSE Linux Enterprise Server 10
And again in the barcelona simulated comparison, if you take about 3s to read the footnote:
http://multicore.amd.com/GetFile.asp...trate_040207v2
you read:
"competitive benchmark results stated above reflect results published on www.spec.org as of april 16. 2007. The comparison presented above is based on the best performing blabla... "
Best performing... Nowhere best performing using this and that blabla, just best performing, they don't care of os or compiler, they only take the "best performing"
What part of "best result" is too hard for you to comprehend ?Quote:
Originally Posted by PetNorth
Maybe you should worry more about your problems than calling others zealots.
what is a zealot?
Religious fanatics. Google is your friend.
Someone who is zealous, filled with or motivated by zeal; fervent, enthusiastically mad.
http://www.dailytech.com/Welcome+to+...rticle7927.htm
DT rips on INQ, incorrectly stating 2.3GHz specrate benches compared to 2.6GHz... and
Barcelona...when it does launch on August 27.
Seems this thread just revealed another Intel drone - this time in the form of George Ou.
"Blatantly", "Deceptive", "Ethics" - nice word choice George.
I'm sorry, but for an Intel fanboy to cry about ethics is simply laughable. So is George implying that Intel is the model citizen for ethics? Looks like we found Tom's HW 2.0 in the form of George Ou.
Right! 6 weeks to go and no one here with final silicon to leak some numbers :confused: I can't believe that... Hector if you reading XS then create account for yourself and post some numbers :D .Quote:
Originally Posted by red
i prefer some pictures :DQuote:
Originally Posted by Lightman
and what's supposed it means?Quote:
Originally Posted by nemrod
that AMD guy or guys who made that nonsense table or graph, is stupid?
I knew that, Big deal.
But I'm sorry, it doesn't change zdnet genious stupidity or Intel zealots stupidity embracing George Ou crap.
Fine, so a gift for you:Quote:
Originally Posted by madcho
http://www.lsi.upc.es/~aofa05/pictures/barcelona1.jpg :)
no world record here :down:Quote:
Originally Posted by nemrod
You, G. OU or De Vries have just miss a very very small detail...Quote:
Originally Posted by PetNorth
SPEC: Standard Performance Evaluation Corporation
s a non-profit corporation formed to establish, maintain and endorse a standardized set of relevant benchmarks that can be applied to the newest generation of high-performance computers.
X86 or not doesn't matter
linux or windows or unix doesn't matter...
Who is behind spec? ALL the world involve in server business:
SPEC Members:
Acer Inc. * Advanced Micro Devices * Apple Inc. * ATI Research * Azul Systems, Inc. * BEA Systems * BlueArc * Bull S.A. * CommuniGate Systems * Dell * EMC * Fabric7 Systems, Inc. * Fujitsu Limited * Fujitsu Siemens * Hewlett-Packard * Hitachi Data Systems * Hitachi Ltd. * IBM * Intel * ION Computer Systems * Itautec S/A * Microsoft * NEC - Japan * NetEffect * Network Appliance * NVIDIA * Openwave Systems * Oracle * Panasas * Principled Technologies * QLogic Corporation * The Portland Group * Rackable Systems * Red Hat * S3 Graphics Co., Ltd. * SAP AG * SGI * Sun Microsystems * Super Micro Computer, Inc. * SWsoft * Sybase * Symantec Corporation * Trigence * Unisys * Zeus Technology
etc...
And your comment about zealots embrassing G. OU crap... Just :
G. OU crap: July 4th, 2007
July 3 here:
Quote:
Originally Posted by savantu
http://www.sunmicrosystems.se/virtua...f_Nordlund.pdf
Slide 16
Projected IPC gains up to 15%
or enough to make a 2Ghz Proc run like a 2.3Ghz ProcQuote:
Originally Posted by red
Which Ironically is more than what conroe does over k8 in many many many places.
So what exactly is your point?
Is it a problem to obtain new data in the barcelona thread?Quote:
Originally Posted by nn_step
I never knew that reporting data without analysis means that I'm trying to make a point :rolleyes: Sorry that you don't like the data.
Great find. Thanks.Quote:
Originally Posted by red
He did kinda skip Slide 42, which effectively means double Virtualization performance. Which would relatively put a single 1Ghz Barcelona on even performance with a Quad core Kentsfield running at 8Ghz (virtualization applications and Operating systems only)Quote:
Originally Posted by Periander6
Which means AMD is going to make a killing in the ISP and high end server market.
This is misleading since FP/SSE performance gains are going to be much much bigger than 15%...
(vs. K8)Quote:
Originally Posted by red
That puts the K10 core IPC right about at Core2 IPC, as expected.
Throw in Intel's 50% clock advantage at Barc launch, and both a speed bump *and* a 5-10% IPC bump with 45nm, and it's...
Game over.
AMD may scrape out a win in 4-socket HPC apps, but that's too small a market to survive in. And Nehalem is coming.
Phenom is going to be an absolute dog vs. Penryn. Desktop disaster.
And mobile? Oh that's right, there is no K10 mobile, just a warmed over K8 dual core Turion 1.01 served up as "Griffin".
lmao,game over..
terrace215 you win the "extreme intel fanboy award of the year".Competition has been tough,we have a few good candidates,but you have shown what it takes to take the "prize".
Are you paid by Intel or do you know absolutely nothing about electronics?Quote:
Originally Posted by terrace215
And Integer gains are going to be much smaller.Quote:
Originally Posted by informal
It's an average. What isn't to understand? (besides simple math)
A 15% IPC improvement over k8 would indicate that they need a 2.5-2.6GHz Phenom, which they are nowhere near atm, to match a Q6600 that is going to be selling for $266 in 2 weeks. Nevermind Penryn coming out in the same timeframe as Phenom. 60% of AMD's CPU revenues come from the desktop segment, and without a miracle stepping they sure look like they are going to be toast in this segment.Quote:
Originally Posted by nn_step
Servers are nowhere near a large enough segement to carry AMD, and anyway AMD is and will be facing much much tougher competition with Seaburg/Harpertown and Tigerton, not to mention Gainestown on the horizon.
So, please explain your position as to why all this is just rubbish instead of hurling personal insults, or are insults all that you've got?
umm I would hate to burst your bubble but the performance difference between Conroe and K8 isn't that large, infact in a few places K8 does out perform conroe. So logically a 2.3Ghz Barcelona should directly compete against a 2.66Ghz Kentsfield.Quote:
Originally Posted by Periander6
On a side note the server market has carried Sun for the majority of their history and I do doubt that AMD could possibly be forced off the desktop market, where a 386 is sufficient for some people to browse the web and read their email.
This thread is the last bastion for AMD fanboi's. Why on earth would they want to embrace logic after all this time?Quote:
Originally Posted by Periander6
I really hope Phenom turns into a true competitor, if solely for sake of having a choice besides Intel.
But as others have pointed out here - if you have the goods, and your competition is doing nothing but repeatedly packing your lunch in terms of sales and revenue, you don't keep tight-lipped about it.
Unless, of course, you are sitting on a dud.
BS. On average, C2D is 20% faster than K8 in 32-bit code and 14% faster in 64-bit code.Quote:
Originally Posted by nn_step
And the corollary is also true. In some things C2D is far more than 20% faster than K8. I know that in Folding@Home it is about 70-80% faster because of the SSE optimizations to that program.
I don't know about you but AMD is telling a few things but is keeping the details close. As for people with samples talking... The Current AMD NDA involves ritual suicide in event of breach. So those that know don't talk much if at all about it.Quote:
Originally Posted by mstp2009
You aren't bursting any bubbles.Quote:
Originally Posted by nn_step
You want to pretend that C2 doesn't have a 20% IPC advantage over K8, suit youself. The rest of the world gets results like the ones Xbit, Techreport, Anand and every other competent testing outfit gets, 20% or so, if anything that is being generous to the K8.
Sun is a seller of complete systems, and are still struggling. If they tried to make it just as maker of Sparc chips, they would be out of business in weeks. Absurd to compare them to AMD.
Yes, people can browse the web and read e-mail with inferior CPUs. The problem is that AMD has to make enough of a profit to make payments on their 5+ billions in debt, fund chip development, equip their fabs ect.. Slower, inferior chips don't fetch much of a price. Not nearly enough to keep AMD from sinking.
Slow(er) and inferior?K10? Are you joking?
Well at least they went for a nice round number.Quote:
Originally Posted by mstp2009
i'm with nn_step on this one, yes C2D got an 20% IPC advantage but they still WIN in SOME apps (I DO NOT TALK ABOUT THE MAJORITY APPS, but in some apps amd can keep up with C2D, and in 64bit the disadvantage is less significant ;) )Quote:
Originally Posted by Periander6
for gaming: i doubt that you see much of a difference when using mid range graphics cards in systems for up to 1000$ ;) , but in the higher and amd got no chance against intel :yepp:
Um, I didn't sign anything and I just found it off Google so I'm fine right :DQuote:
Originally Posted by mstp2009
No one bothered with this one? For some reason :rolleyes: , I'm inclined to think that these numbers are poo, but if K10 is double performance, a 1 core 1GHz K10 matches a 4 core 8GHz Kent, by deduction, a 1 core 2GHz K8 matches a 4 core 8GHz Kent, therefore, increasing AMD's advantage from 16x to 32x. If a current 16x advantage isn't compelling enough to save AMD, what makes you think a 32x one will :rofl:Quote:
Originally Posted by nn_step
Based on what ?Quote:
Originally Posted by mstp2009
Have you been under a rock for the last 15 months?Quote:
Originally Posted by savantu
http://www.xbitlabs.com/articles/cpu...w_9.html#sect0
http://www.xbitlabs.com/articles/cpu...review_11.html
http://setiathome.berkeley.edu/forum...d.php?id=39379
http://www.tomshardware.com/2006/07/...out_athlon_64/
http://www.extremetech.com/article2/...2014685,00.asp
http://blogs.msdn.com/rickbrew/archi...13/664890.aspx
http://www.techreport.com/reviews/20...s/index.x?pg=1
EDIT - 64-bit analysis here:
http://www.xbitlabs.com/articles/cpu...duo-64bit.html
Google 101: Core 2 Duo vs. AMD Athlon X2
The 64bit part interested me.I knew you would point out the Xbit test.Then why did I bother to ask ?Quote:
Originally Posted by mstp2009
Because the conclusion is flawed IMO.
When you average results , you need to eliminate anomalies , like the one provided by ScienceMark , because they skew your results.
ScienceMark is a test where K8 does abnormaly well ; in the same way SuperPI is a test where C2D does abnormaly well.Including them has a profound effect on the average.
Once you eliminate ScienceMark , the results are damn close.The graphic at the end shows this best.
Savantu,then you need to check the techreport x64 testing,where 6000+ gets really close on many occasions to e6700.
The big advantage of the C2D is the SSE. K10 have better SSE ( full 128bits, C2D not all ).
if you disable the SSE on a C2D and a K8, K8 would be faster.
64bits mode use SSE, AMD don't loos too much here with the lot of GPRs in the K8. K10 will perform more fast in 64bits mode, with 8 more GPRs and double SSE.
K10 is a bit more fast in integer, but not a big improvement. about 15%.
Phenom will be a bit faster than barcelona clock for clock 8-10% i think, with 1066 non registered memory ( barcelona 800 registered ), and HT3 could be usefull for 4 cores.
Wrong. C2D is fully 128bit SSE (on all 3 SSE decoders).Quote:
Originally Posted by madcho
Wrong. The integer unit in C2D is MUCH better than that in K8.Quote:
Originally Posted by madcho
Do you have a clue what you are talking about? Phenom IS BARCELONA (same core, just different sockets - desktop vs. server).Quote:
Originally Posted by madcho
While i dont think a 10% IPC for Phenom over Barcelona is reasonable these are not the same processors for different sockets. Barcelona is a HT2 cpu using up to DDR2 5300 registered memory whereas Phenom is a HT3 cpu using up to DDR2 8500, which gives Phenoma huge advantage in Memory bandwidth. Whether Phenom can use the increased memory bandwidth to get improved performance remains anyone's guess right now but id say it will at least give it a slight advantage clock for clock.Quote:
Originally Posted by mstp2009
It's all a moot point for the desktop.Quote:
Originally Posted by BrowncoatGR
What can fully take up that memory bandwith NOW on a desktop app?
Nothing.
It isn't until you get to 4P systems in a server environment with HEAVY database transaction loads that you see any difference.
This is exactly why we see no performance gain going from Socket 939 (DDR1) to Socket AM2. The CPU core didn't change except for the memory controller. Bench identically clocked systems and you will see that all that extra memory bandwidth is going to waste.
It will be the same thing in Phenom w/ HT2 vs. HT3. This is pointless for desktop apps.
there won't be any surprise. it will be about equal to the advantage core 2 gets when the fsb is upped from 1066mhz->1333mhz.. 2-3%, maybe 5% (core 2 isn't really memory starved at all)Quote:
Originally Posted by BrowncoatGR
Quote:
Originally Posted by madcho
please stop the FUD and write decent englishQuote:
Originally Posted by madcho
Jacky well id say that is both application dependent and really difficult to predict right now. K8 benefits quite a bit from PC6400 memory in some applications and my guess is so will K10 from both PC8500 memory and the increased HT speed. Although in some applications the advantage will be close to zero. I expect the biggest impact to be in games where consumer apps are concerned. BTW imo Core2 could use additional memory bandwidth, in particular quadcore parts, but increasing the fsb does not have that much of an impact because of the limitations of the FSB bus. Im also guessing Core2 prefetching and cache structure was also optimized with a 1066 FSB bus in mind. Penryn will probably change this though.
Madcho C2D clearly hasa faster integer unit than K8, there is really no discussion here. But K8 is multitudes faster at x87 fp code but that is not really relevant for desktop apps. What allows K8 to somewhat keep up is its memory bandwidth mostly. And in 64bit macro-ops fusion does not work.
K10 won't be slower or inferior to K8. It seems a lot like Core, so it should at least be on par. The extra bandwidth should help AMD a bit.
If that was at me, I was referring at the impact increased memory bandwidth could have for K10. I was not comparing K8 to K10Quote:
Originally Posted by Zytek_Fan
And the many times where it loses to the E6600.Quote:
Originally Posted by informal
Xbitlabs made a mistake in their simple average. Going to 64-bit, the X6800 ends up having the same % advantage as in 32-bit.Quote:
Originally Posted by mstp2009
http://img85.imageshack.us/img85/32/...4bitrunol4.jpg
No, not at you.Quote:
Originally Posted by BrowncoatGR
Since I just love throwing wrenches into the works
http://www.silentpcreview.com/article735-page1.html
What in tarnation does that have to do with Barcelona?Quote:
Originally Posted by nn_step
not true. the units are 128bits but some functions are not available in 128bitsQuote:
Originally Posted by mstp2009
yes.Quote:
Wrong. The integer unit in C2D is MUCH better than that in K8.
The new IMC of the K10 will allow to use more bandwith, not like the K8 on AM2. The K8 on AM2 is not able to use the full bandwith of the DDR2, and cache speed is too slow.Quote:
Do you have a clue what you are talking about? Phenom IS BARCELONA (same core, just different sockets - desktop vs. server).
Registered memory is slow, and have bad timings, Barcelona will not able to show all. Phenom will be a bit faster. HT3 will improve a bit for 4 cores. I'have seen with 1ghz HT with dual core a bottleneck. ;)
We were discussing how much each CPU gains from 32bit -> 64bit , not absolute performance.Quote:
Originally Posted by informal
Secondly , too bad Xbit didn't include how much P4 gained from 32bit to 64bit.I wouldn't be surprised too see it exceeds both C2D and K8 according to their tests.I'll calculate this tomorrow , it's 2 am now and I need to sleep.
Really ? Which ones aren't available for C2D but are for K10 ?Quote:
Originally Posted by madcho
Well , K10 introduces another cache in the equation , an L3 with unknown latencies thus far.Quote:
The new IMC of the K10 will allow to use more bandwith, not like the K8 on AM2. The K8 on AM2 is not able to use the full bandwith of the DDR2, and cache speed is too slow.
In what tests did you notice a performance penalty due to the 1GHz HT link ?Quote:
Registered memory is slow, and have bad timings, Barcelona will not able to show all. Phenom will be a bit faster. HT3 will improve a bit for 4 cores. I'have seen with 1ghz HT with dual core a bottleneck. ;)
Quick search on your posts reveals,well nothing interesting(except extreme intel bias).But that's nothing new these days.Quote:
Originally Posted by mstp2009
On topic:
Article from tgdaily sums it up nicely:
http://www.tgdaily.com/content/view/32802/135/
Quote:
AMD feels “confident” about Barcelona performance estimates PDF Print E-mail
Hardware
By Wolfgang Gruener
Friday, July 06, 2007 13:04
Recommend article:
Slashdot
Digg
Delicious
Technorati
YahooMyWeb
Chicago (IL) – Two Barcelona benchmark charts published on AMD’s website caused quite a stir yesterday: Intel wasn’t happy about outdated performance numbers causing AMD to remove the charts.
In Pictures: AMD Quad-Core "Barcelona" technical details ...
AMD today in fact acknowledged that its performance charts were outdated and that new Intel numbers were not included in the comparison. Spokesman Phil Hughes conceded in a conversation with TG Daily that AMD “failed to update the numbers” and has taken down the charts as a result.
In the meantime, Intel had stressed that Xeon performance numbers, which far exceed the numbers used by AMD, are available. And according to Intel, the 2.66 GHz Xeon 5355 does not trail the estimated integer performance of a 2.6 GHz quad-core Opteron by about 20%, but by only about 2% (99.9 vs. 102 points in the Specint_rate2006 benchmark). The 3.0 GHz Xeon 5365, which is shipping in limited numbers at this time, is actually slightly faster than the 2.6 GHz Opteron (106 vs. 102 points) – which is expected to be AMD’s fastest quad-core to be available by the end of this year.
From that perspective, the future does not look favorable for AMD, given the fact that Intel’s 45 nm server chip will be available around the turn of the year and Intel already said that it does expect substantial performance gains in the double-digit range across the board. But the truth of the matter is that we are talking about unreleased products on both ends, with Intel holding the performance lead in products that you can buy today.
So, if AMD has taken down the benchmark numbers, does that mean that Barcelona will not be as fast as AMD claims? Hughes told us that AMD still expects the performance of Barcelona at 2.6 GHz to come in at what the estimate has shown – or “even a bit higher”. However, there is no doubt that the initially available Barcelonas (with up to 2.0 GHz) will not be able to match Intel’s current performance level.
Interestingly enough, plain speed isn’t really the whole story. Power consumption is a performance factor that has not received a whole lot attention. In the light of today’s server environments that try to decrease heat and use the available power as efficiently as possible, this may actually the discipline where the battle between Intel and AMD will be fought. At least according to AMD’s Barcelona website, the firm’s 68w/95w/125w chips will have an advantage over comparable Xeons (65w/80w), at least as long as 65w Xeons are pitched against 68w and 95w Opterons and the 80w Xeon against the 95w Opteron. And yes, the low-power Clovertown processors (50w) were left out of consideration in this comparison.
If industry rumors are right, then AMD will be releasing Barcelona in September and we soon will be able to verify what the processor is really capable of and what impact it may have in the market.