Bookmark and Share


Gateway, a maker of servers, has accidentally revealed clock-speeds of Advanced Micro Devices' upcoming AMD Opteron chips for servers code-named Interlagos and Valencia, which are based on Bulldozer micro-architecture. As projected initially, clock-speeds of the new microprocessors are not high.

Based on the configurations of Gateway's future servers (which were noticed by Bright Side of News web-site), AMD's initial lineup of Opteron 6200-series microprocessors will consist of four models: two with sixteen cores, one with twelve cores and one with eight cores. The chips with twelve and sixteen cores are multi-chip modules with two 6- or 8-core dies, whereas the one with eight cores is likely to be powered by a single eight-core die code-named Valencia.

Clock-speeds of the chips are not really high because AMD needed to limit thermal design power to maintain compatibility with current-generation G34 socket servers. The four models that AMD seems to plan to release initially will be the following SKUs:

  • Opteron 6220: 8-core, 3.0GHz, 3.5GHz Turbo, 8MB L2, 8MB L3 cache
  • Opteron 6234: 12-core, 2.4GHz, 2.9GHz Turbo, 12MB L2, 12MB L3 cache
  • Opteron 6272: 16-core, 2.1GHz, 2.6GHz Turbo, 16MB L2, 16MB L3 cache
  • Opteron 6276: 16-core, 2.3GHz, 2.8GHz Turbo, 16MB L2, 16MB L3 cache

At present AMD's fastest twelve-core Opteron 6180 SE server chip works at 2.50GHz, so it remains to be seen how significantly faster will the new sixteen-core Opteron 6276 chip with 2.30GHz/2.80GHz clock-speeds be against its predecessor.

AMD recently promised to initiate commercial shipments of its server microprocessors based on Bulldozer micro-architecture in August, which means that actual servers running the new chips will be out in September.

AMD pins a lot of hopes onto its Bulldozer micro-architecture in general and Interlagos and Valencia products in particular. Firstly, Bulldozer is the first major micro-architectural update for AMD in almost eight years. Secondly, it will also be AMD's first chip architected for the current workloads. Thirdly, it has to be the chip that will be as successful as AMD's first breeds of Opteron microprocessors.

"We expect to begin shipping our first server platform featuring the Bulldozer this quarter. The Interlagos platform is our first server offering optimized for today's cloud datacenters. The [Bulldozer] [micro]-architecture excels at compute-intensive and HPC workloads, where it will deliver up to 35% performance improvements compared to our current offerings. Customer excitement for Interlagos is high: all of our major customers are expected to introduce servers based on the new platform this year. We are committed to the server market and are focused on returning the business to a growth trajectory," said Thomas Seifert, interim chief executive officer of AMD, during a recent conference call.

AMD did not officially confirm specifications of its upcoming chips.

Tags: AMD, Opteron, Bulldozer, Interlagos, Valencia, Gateway


Comments currently: 25
Discussion started: 07/25/11 07:15:40 PM
Latest comment: 07/29/11 03:52:45 AM
Expand all threads | Collapse all threads


If they are going to launch higher clocked bulldozer parts for home computing, can we expect to see higher clocked Opterons soon?
0 0 [Posted by: Filiprino  | Date: 07/25/11 07:15:40 PM]

Intel pays manufacturers hundreds of millions of dollars to drop AMD server products.

I guess we won't be seeing AMD servers
4 3 [Posted by: jamminjoe  | Date: 07/25/11 08:22:04 PM]
- collapse thread

Heh, even today, you see OEM sites splashed and painted with Intel everything, from laptops to servers and everything in between.
2 2 [Posted by: RtFusion  | Date: 07/25/11 11:04:00 PM]
unfortunately Intel does that, that's why i hate them! Monopolistic SOB!!!
4 4 [Posted by: 63jax  | Date: 07/26/11 12:33:16 AM]

Intel does what most Monopolistic companies do to STAY ontop!.
cheats, lies, and blackmails, pretty much anything to keep itself number 1
2 3 [Posted by: vid_ghost  | Date: 07/26/11 02:22:31 AM]

Why do you slam AMD for the clock speeds of a new product?

All current Xeons (E7) with HT have a base clock of 2.4ghz or lower.
Intel's turbo is only 400mhz while AMD's turbo is 500mhz.

This article seems a bit biased, no?
7 1 [Posted by: mamisano  | Date: 07/26/11 06:09:50 AM]
- collapse thread

It is a bit biased, there is a rumor that a few of the members of xbit are getting payed to write favorable Intel reviews and not so favorable AMD ones. At least the the hard core AMD fanboys would like to think so. But if Intel plays as dirty as they sound, I would not put it past them.
5 6 [Posted by: veli05  | Date: 07/26/11 06:43:33 AM]
Rumors spread by who? By AMD users that can't see the obvious truth that Intel has the better processors? They didn't say anything against AMD, they noticed a fact that they can't have high frequency because of the TDP. If you have no ideea about what what you are reading, you better do us a favour and stop posting nonsenses on forums.
3 4 [Posted by: cosminmcm  | Date: 07/26/11 07:09:19 AM]
... obvious truth that Intel has the better processors?

If Intel's processors are OBVIOUSLY better than AMD's then why is AMD still in business? Your statement is pure opinion with OBVIOUS Intel bias. The fanboy strikes again, this time with intelligence insults included
5 5 [Posted by: veli05  | Date: 07/26/11 07:47:59 AM]
show the post
0 3 [Posted by: cosminmcm  | Date: 07/26/11 07:53:04 AM]
This is exactly my point. When a company has enough clout to decide whether a competitor stays alive or not is a problem. It goes against all aspects of ethical competition. Monopolies are illegal in the US, I hope Intel gets what is coming to them, hopefully the government can make something stick.
2 4 [Posted by: veli05  | Date: 07/26/11 01:47:48 PM]
show the post
0 3 [Posted by: cosminmcm  | Date: 07/26/11 07:55:37 AM]
Your point being what? I'm not surprised by any of Tom's findings at all. Every review says just about the same redundant thing nowadays: Intel > AMD performance. I just refuse to support a company that likes to throw their weight around like Intel does. Its arrogant and snobbish, therefore I will support AMD in all its efforts due to the fact that the seem to have more reputable business practices.
6 4 [Posted by: veli05  | Date: 07/26/11 01:42:53 PM]
... obvious truth that Intel has the better processors?

If Intel's processors are OBVIOUSLY better than AMD's then why is AMD still in business? Your statement is pure opinion with OBVIOUS Intel bias. The fanboy strikes again, this time with intelligence insults included
4 5 [Posted by: veli05  | Date: 07/26/11 08:26:11 AM]
show the post
0 4 [Posted by: cosminmcm  | Date: 07/26/11 08:42:43 AM]
show the post
1 5 [Posted by: veli05  | Date: 07/26/11 09:16:21 AM]
show the post
0 4 [Posted by: cosminmcm  | Date: 07/26/11 09:43:26 AM]
LOLZ when did I say anything about the FACT that the reviewers were getting paid by Intel to review Intel products favorably/flame amd products. Giving Intel's supposed track record of shady business practices I would not be surprised if they were paying off reviewers. I said it was a rumor and you should treat it as such. Maybe you should read posts fully before jumping to conclusions
5 5 [Posted by: veli05  | Date: 07/26/11 09:57:45 AM]
show the post
0 3 [Posted by: cosminmcm  | Date: 07/26/11 10:05:59 AM]
The fact is K10 performs worse than K8 which is pretty damn sad
2 3 [Posted by: seronx  | Date: 07/26/11 07:14:10 PM]
c'mon Seronx you are way smarter than this, what do you want to say with that?
0 2 [Posted by: 63jax  | Date: 07/27/11 02:57:23 AM]
What are you on? Are you seriously saying that a K8 arch cpu outperforms a K10.5 arch cpu? LOL I don't know if you have been up to speed but maybe you should start reading benchmarks beyond 2008.
2 0 [Posted by: SteelCity1981  | Date: 07/27/11 03:04:44 AM]

Nice job in editing the story without mentioning it.

You took out the content slamming the clock speeds. Now all it says it that the lower speeds are due to G34 thermal design limits.

Gotta love it!
0 0 [Posted by: mamisano  | Date: 07/28/11 07:56:07 AM]

about intels practices... anybody know that intel actually payed(directly or indirectly through specialized "trainings" and information) compiler developers to use intel optimised instructions ?? Well they payed about 1 billion for this trick (but I think they made at least 100 billion because of that)... And this is not the only "trick" they played

How the hell do you justify such a stupid assumption that every chip amd released is worse then intel with a benchmark difference of <15% usually ? Don't you think the compilers do play a majour role in this?? I guess that a random instruction use (which is never done by the so called benchmarks btw!) with the same chance to hit ANY of the instructions in the intel or amd instructionset would probably give you the very basic truth... the chips are equal(or differ insignificantly). An inconsistent 15% is not sufficient a difference to call it!

When athlon came out the difference was 40-50% all across the board and now slaming amd for not being optimized for intel instruction is just plain stupid.

Make your real life tests and say this chip is good for this and that and thatone for this and that. A performance crown hasn't been given away in my opinion since the athlon. When the core2 launched things got murky with marginal improvements here and there on both camps. AMD lost the train right there they thought the athlon will hold its own for ever and that was one of the most arrogant and stupid mistakes amd has ever made. Instead of realising that thy have awaken the intel machine and that they would have to slam it at least 10 years in a row to shrink it to a manageable size they sat back and relaxed while intel threw literally billions at the problem. If they would have used the athlon era upwind they would have finished bulldozer 3 years ago and they would be allready 6-8 years into the 10... now they're back at square 1.

My take is that amd will NOT take the performance crown with the buldozer because it will not be again 40-50% over anythging from anybody on the market(like athlon was). It will probably beat intel at some things(heavy multithreaded integer) and loose at others(probably multithreaded floating point) and the difference across the board will be small.

My advice ... watch out for the real life user experience benchmarks, yes it doesn't exclude the compiler bias, but it will give you a fair idea of what you may expect from the cpu using those softwares.

I usually go for amd not because I like it (I had 2 burt down durons, had an athlon 2 crash etc...) but all in all at the end of the day somehow I allways ended up with less money spent. And that gave me the sufficient argument to never stray away. And yes I had my fair share of intel chips crashing on me too but for example in the duron era the P was much cooler and much stabler. History repeated with phenom vs c2d/q... it is time for a new Athlon... but that will not be the first incarnation of bulldozer. I personally expect it in ~2014 when they will have figured out how to properly discard all the fp operation onto the "gpu" part of the processor.

Until then!
1 0 [Posted by: tcubed  | Date: 07/29/11 03:40:09 AM]


Add your Comment

Related news

Latest News

Wednesday, November 5, 2014

10:48 pm | LG’s Unique Ultra-Wide Curved 34” Display Finally Hits the Market. LG 34UC97 Available in the U.S. and the U.K.

Wednesday, October 8, 2014

12:52 pm | Lisa Su Appointed as New CEO of Advanced Micro Devices. Rory Read Steps Down, Lisa Su Becomes New CEO of AMD

Thursday, August 28, 2014

4:22 am | AMD Has No Plans to Reconsider Recommended Prices of Radeon R9 Graphics Cards. AMD Will Not Lower Recommended Prices of Radeon R9 Graphics Solutions

Wednesday, August 27, 2014

1:09 pm | Samsung Begins to Produce 2.13GHz 64GB DDR4 Memory Modules. Samsung Uses TSV DRAMs for 64GB DDR4 RDIMMs

Tuesday, August 26, 2014

10:41 am | AMD Quietly Reveals Third Iteration of GCN Architecture with Tonga GPU. AMD Unleashes Radeon R9 285 Graphics Cards, Tonga GPU, GCN 1.2 Architecture