Barcelona 'openssl speed' benchmarks

[dmens]

not that george ou is always right or anything, but i have more respect for him that the shit-for-brains at amdzone.

 

[Phynaz]

Originally posted by: bradley
George Ou? The same guy who writes for ZDNet, and defends the dumbed-down concepts behind monopolies and oligopolies with great furvor and zeal? I didn't realize he was so respected, or that many visited ZDNet for technology information.

http://www.amdzone.com/index.p...4c6f4a72f61f0169525e20

http://www.amdzone.com/modules...&file=article&sid=8018

Wheather you agree with what he writes does not have an effect on his accomplishments and itegrity as a journalist.

BTW, ZDNet gets 22.1 million uniques a month. George Ou is the guy responsible for building that infrastruture. ZDNet properties include PC Magazine, Gamespot, Techrepublic and 1up to name a few.

 

[Viditor]

Originally posted by: dmens
not that george ou is always right or anything, but i have more respect for him that the shit-for-brains at amdzone.

Probably one of the best places to see how little George is respected is in his own "Talkback" section
Talkback
It's not even so much whether he's right or wrong, it's his complete and utter lack of objectivity! He would give even some of the most rabid Intel fanbois here a run for their money...obviously that's not journalism.

George is basically an Intel version of Sharikou...I have nothing against either one of those guys, but neither one is really a journalist.

 

[Viditor]

Originally posted by: CTho9305

Originally posted by: Viditor

Originally posted by: keysplayr2003
Why doesn't everyone just post what they "want" to see in this thread? It would cut down a lot on the beating around the bush posts. I mean, I can see why certain members arguments are what they are. But what is the point? Seriously?

To answer your question from my own POV, I think it's important to keep asking for the most information we can get...
There are obviously quite a few holes in our knowledge of Barcelona at this point, and I feel that mentioning what they might be and discussing them can only help us as we pass along our requests to review sites like AT.
At the end of the day, it's only places like AT that are going to answer some of these questions...

Given that Barcelonas are available now, couldn't you just buy one and run the tests you're interested in? With targeted tests you can determine a lot of microarchitectural details (see microarchitecture.pdf).

True, but I would also have to buy several other platforms to generate good comparisons...

 

[dmens]

Originally posted by: Viditor

Originally posted by: dmens
not that george ou is always right or anything, but i have more respect for him that the shit-for-brains at amdzone.

Probably one of the best places to see how little George is respected is in his own "Talkback" section
Talkback
It's not even so much whether he's right or wrong, it's his complete and utter lack of objectivity! He would give even some of the most rabid Intel fanbois here a run for their money...obviously that's not journalism.

George is basically an Intel version of Sharikou...I have nothing against either one of those guys, but neither one is really a journalist.

heh, at least george doesn't claim to be a PhD.

 

[Phynaz]

It's not even so much whether he's right or wrong, it's his complete and utter lack of objectivity! He would give even some of the most rabid Intel fanbois here a run for their money...obviously that's not journalism.

Nice try, but you know there is no comparison between the two of them.

Whether or not George is objective (in your opinion) in his writing does not alter his credentials and accomplishments as a journalist. How do you feel about Tom Yager? He's such a huge AMD fanboy it makes you look positivly neutral. It doesn't affect the fact that he is a top technology writer.

BTW, I assume you meant to post a link to the comments section of his blog, and not AMD. As if the comments section on a blog has anything to do with credentials of a writer.

Really, bringing Sharikou into this was lame. So lame as to make any further discussion pointless. This is my last on this subject.

 

[Viditor]

Originally posted by: dmens

Originally posted by: Viditor

Originally posted by: dmens
not that george ou is always right or anything, but i have more respect for him that the shit-for-brains at amdzone.

Probably one of the best places to see how little George is respected is in his own "Talkback" section
Talkback
It's not even so much whether he's right or wrong, it's his complete and utter lack of objectivity! He would give even some of the most rabid Intel fanbois here a run for their money...obviously that's not journalism.

George is basically an Intel version of Sharikou...I have nothing against either one of those guys, but neither one is really a journalist.

heh, at least george doesn't claim to be a PhD.

Fair enough...but you have to admit that he's still way out there.

There certainly are writers out there with an admitted strong Intel bias that I read and respect...Paul DeMone for example. This is mainly because they actually look at both sides and make very intelligent comments.
But the very best writers are people like Anand, Hans De Vries, Johan De Gelas, David Kanter, Kris Kubicki, Dean Kent, etc... They are objective, intelligent, insiteful, and they admit it when they make a mistake (though that's quite rare for them). These guys are far closer to what Edward R Murrow taught us journalists should be.

 

[AlabamaCajun]

Originally posted by: savageseb

Originally posted by: AlabamaCajun
The problem with trying to force L1 cache is that the look ahead architecture can load the data faster that the time it takes for the code that needs it to drop through the pipes. It might show if we optimize for Opteron but it seems that Intels large L2 keeps up even with the FSB bottleneck.

Personally I don't see why people think rooting for Intel is a sport like a favorite team. It's a monopolistic company for godsakes.

And i couldnt agree more, muahauhahaha

my first post at anandtech! wooooooo, lols

Hey, cool, I was starting to think all my posts were stealth. Seems there is little support for AMD here except for the OP.

 

[zpdixon42]

Originally posted by: Markfw900
$340 for a cpu and $270 for a moytherboard, just to test ? Oh, and then memory ?

Just FYI, I was looking for a cheap mobo to test the quad-core Opteron the other day and found one for $140 (on Newegg: open box Tyan S3970; single socket F; Bios update released on Sep 14th to support the quad). But I think I'll wait for cheaper AM2+ K10 processors instead.

On a side note, I initiated this thread because I (correctly) suspected K10 would outperform Core in some areas of openssl, and I am frustrated to see the number of contributors to this thread without a single person with a K10 on hand... I thought this forum was supposedly full of enthusiastic early-adopters

 

[myocardia]

Originally posted by: zpdixon42
On a side note, I initiated this thread because I (correctly) suspected K10 would outperform Core in some areas of openssl, and I am frustrated to see the number of contributors to this thread without a single person with a K10 on hand... I thought this forum was supposedly full of enthusiastic early-adopters

It is, that's why there are so many people around here who have built systems around the only desktop versions of quad-cores that are available. We don't run server farms. Shouldn't you be telling us how well the only K10's that are available so far perform in a server farm?

 

[AlabamaCajun]

Originally posted by: zpdixon42

Originally posted by: Markfw900
$340 for a cpu and $270 for a moytherboard, just to test ? Oh, and then memory ?

Just FYI, I was looking for a cheap mobo to test the quad-core Opteron the other day and found one for $140 (on Newegg: open box Tyan S3970; single socket F; Bios update released on Sep 14th to support the quad). But I think I'll wait for cheaper AM2+ K10 processors instead.

On a side note, I initiated this thread because I (correctly) suspected K10 would outperform Core in some areas of openssl, and I am frustrated to see the number of contributors to this thread without a single person with a K10 on hand... I thought this forum was supposedly full of enthusiastic early-adopters

It's still a little early with only the single 2S version on the egg at this time. I'm ready to pounce saving up all my Lincolns until that time comes.

 

[zpdixon42]

Just now I have been able to get my hands on exhaustive 32-bit "openssl speed" scores for the quad-core Opteron 2347 (slightly overclocked @ 1950 MHz) ! See this post of mine on another forum, I made a detailled clock-for-clock analysis of K8 vs K10:

http://www.xtremesystems.org/f...2494063&postcount=1097

Overall, K10 performs, clock-for-clock, 0% to 18% faster than K8 on these 32-bit (100% ALU) OpenSSL speed tests, with most of the algorithms showing a ~10% increase on K10 (+0% and +18% are really the extremes...) The guy who provided the results will try to get us 64-bit openssl scores. This is pretty good considering that (1) OpenSSL is not yet optimized for K10, and that (2) the K10 ALU unit is the one that received the fewer improvements over the K8 ALU unit (compared to, say, the SSE and FPU units).

- Z