Print 110 comment(s) - last by eKi.. on Jun 15 at 9:03 PM

Windows 2003 R2, 64-bit identifies the "Barcelona" processor as an unknown AMD CPU. (Source: DailyTech, Anh Huynh)

AMD's benchmarking utility reveals all four cores ran at 1.6 GHz. (Source: DailyTech, Anh Huynh)
Quad-Core AMD Opteron demonstrated by MSI, Supermicro, TYAN and Uniwide

AMD took the opportunity at Computex to show server platforms running the Quad-Core AMD Opteron processors, codenamed Barcelona, with the help of MSI, Supermicro, TYAN and Uniwide

The AMD quad-core processors are designed to drop-in (following a BIOS upgrade) with all existing AMD Opteron processor-based systems using DDR2 memory, while also enabling new platform capabilities such as those being demonstrated today.

One vendor demonstrated Barcelona to DailyTech, running at 1.6GHz. According to engineers familiar with the chip technology, the current AMD Barcelona samples are not scaling core frequencies well. AMD partners confirmed the highest running, POST and OS capable, Barcelona processor is 2.0 GHz. AMD previously posted benchmarks of a simulated 2.6 GHz Barcelona.

“We commend AMD for its non-disruptive roadmaps, allowing for a smooth upgrade from dual-core to quad-core,” said Danny Hsu, president, TYAN. “Streamlining and mitigating the complexities customers often face when upgrading their IT infrastructure is especially important in today’s competitive marketplace. TYAN has collaborated with AMD since 2001, and we will continue to support and take advantage of the customer-centric innovation AMD is offering in its quad-core processors when they are made available later this year.”

AMD roadmaps show Barcelona-based Opteron processors will launch in July, but the general consensus from partners is the processor isn’t performing well enough for a July launch.

“Quad-core, codenamed Barcelona will launch later this summer, in the July, August kind of time frame -- followed by [consumer chips] on the desktop... You'll see that in the Christmas line-up,” stated Robert Rivet, AMD executive vice president and CFO, just a few months ago.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

By aurareturn on 6/6/2007 3:56:57 AM , Rating: 1
Wasn't AMD saying that the top model is around 2.5Ghz?

1.6Ghz seems kind of slow even if it's very efficient.

I can't see a 1.6Ghz Barcelona beating a 3.2Ghz Intel Quad.

RE: 1.6Ghz?
By bunnyfubbles on 6/6/2007 4:54:19 AM , Rating: 4
Intel did the same thing with C2D, and just about every chip before that. Same goes for AMD including when they first showed off the Athlon 64 and X2 and all the chips before that.

Basically this is just showing that the architecture is real and working, it isn't final silicon.

RE: 1.6Ghz?
By Thorburn on 6/6/2007 5:41:49 AM , Rating: 3
Given that Barcelona is supposed to be a month from launch its certainly worrying.

By comparison Core 2 Duo launched at up to 2.93GHz with the B-2 stepping.
Looking at the chips that made it onto XtremeSystems in the run up to the launch the A-1 silicon was clocked at up to 2.4GHz, B-0 took this to 2.66GHz and B-1 delivered the final boost to 2.93GHz.

The A-1 stepping was doing the rounds maybe 4 or 5 months before launch and saw a 500MHz increase before launch.
If AMD are demoing at almost 1GHz below there proposed launch speed they clearly have plenty work still to do.

RE: 1.6Ghz?
By Targon on 6/6/2007 8:08:30 AM , Rating: 2
It's possible that AMD is keeping their current samples in-house to keep Intel from pushing out faster chips sooner than they would if they knew the current performance that Barcelona can produce. I would be willing to bet that these current samples are 6 months old, just to throw off Intel's guestimate about the true performance that AMD can get the chips to.

RE: 1.6Ghz?
By zsdersw on 6/6/2007 8:21:32 AM , Rating: 2
Intel and AMD have things they keep from each other, obviously, but I highly doubt the "true performance" of their upcoming chips is among those things.

RE: 1.6Ghz?
By coldpower27 on 6/6/2007 9:24:33 AM , Rating: 3
Possibly but when compared to Intel demonstrating 3.33GHZ Yorkfield 6 months before launch, against AMD's Budapest/Bareclona's at 1.6GHZ 1-2 months before launch. This just doesn't inspire much confidence.

RE: 1.6Ghz?
By eKi on 6/15/2007 9:03:14 PM , Rating: 2
Cant u guys read???

"before 1-2 months from launch"... yes, its the orig date, but didnt u read that text at all?? Many of u saying same, not very good at few months before lauch, but hey guys, u know that christmas is still far away ;)

"AMD roadmaps show Barcelona-based Opteron processors will launch in July, but the general consensus from partners is the processor isn’t performing well enough for a July launch.

“Quad-core, codenamed Barcelona will launch later this summer, in the July, August kind of time frame -- followed by [consumer chips] on the desktop... You'll see that in the Christmas line-up,” stated Robert Rivet, AMD executive vice president and CFO, just a few months ago."

It says it very clearly i think, it was supposed to launch in july, but amd vice president himself said it will be seen in christmas line-up, & christmas isnt 1-2 months away, is it now? :D

RE: 1.6Ghz?
By melgross on 6/7/07, Rating: 0
RE: 1.6Ghz?
By P4blo on 6/11/2007 12:21:04 PM , Rating: 2
AMD plays dead, I love it. Then when the time comes they'll whip out some 3Ghz badboys and Intel will be like "mofos!" and slap them about with 45nm ;-)

Very curious to know which one is the most efficient though. Clock for clock.

RE: 1.6Ghz?
By INTC on 6/6/2007 5:45:11 AM , Rating: 2
This article from Spring 2006 IDF dated March 7, 2006 shows a C2D at 2.66 GHz beating an Athlon FX-60 overclocked to 2.8 GHz. That was over 4 months before the official C2D launch last July. AMD said that they didn't want to reveal clock speeds or benchmarks so as not to show all of their cards to the competition but if they don't show something better than just working architecture very soon, they are going to miss the boat. I doubt too many OEMs will be waiting around too long and risk having to make the same announcement as Cray yesterday.

RE: 1.6Ghz?
By Behlal on 6/6/2007 12:00:20 PM , Rating: 3
Yes, but who knows what NDAs OEMs are having to sign to get these chips. For all we know, AMD has already given higher speed samples out, but an NDA agreement stops them using them at trade shows. My point is that we only know that the public have been shown 1.6GHz Barcelona chips. We don't know anything else. This of course is one of our current complaints with AMD, they are playing their cards way too close to their chest.

RE: 1.6Ghz?
By melgross on 6/7/2007 1:37:06 AM , Rating: 2
But there isn't any point to it.

What are their customers thinking?

If they tell their customers that the actual chip speeds are much higher, then you don't think it will get out anyway?

RE: 1.6Ghz?
By Nil Einne on 6/11/2007 4:27:11 PM , Rating: 1
The point to it is to keep Intel guessing and perhaps also so that their end users are amazed when they actual launch the beasts.

As for it getting out, who said the info hasn't? The Inquirer has claimed at various times they're clocking well. Who's to know whether this was from a customer leaking info, AMD trying to hype things up, Intel trying to hype AMD up so they have a big fall or something The Inquirer found somewhere that was complete BS? The trouble is that even when these sort of detail leak, if the wrap is tight enough no one knows if they're true or not so everyone remains guessing...

RE: 1.6Ghz?
By Regs on 6/6/2007 7:24:30 AM , Rating: 2
I could agree. Though like what was state below this a month before launch is cutting it pretty close to now show what you have to offer. 1GHz below spec is what I call a little worrying. There is also a letter going around to AMDs partners that they wont make the deadline sending AMDs stock down. Confirmed or not, it still makes an impact and shows how AMD's tight lips is becoming more troublesome than for the better good. The only conclusion I can make is that AMD still has a lot more work to do, and I'm getting the feeling that the first spin of the K10 is going to end up like the X600. Under performing and not competitive on the high end.

RE: 1.6Ghz?
By Roy2001 on 6/6/2007 2:23:41 PM , Rating: 2
Intel did the same thing with C2D, and just about every chip before that.
Intel demoed a 2.66Ghz Conroe CPU long before C2D lunch.

RE: 1.6Ghz?
By Justin Case on 6/6/2007 12:35:27 PM , Rating: 1
First Opterons shown were running at 800 MHz, so I guess they can say they doubled the speed. ;-)

RE: 1.6Ghz?
By Screwballl on 6/7/2007 10:20:02 AM , Rating: 2
during AMDs launch of the Athlon64, they were releasing 1.6GHz versions that were stomping Intels 3.0GHz versions... at first people just rolled their eyes and kept buying the toaster oven Intels.. until the real performance news got out which put AMD in the fast lane taking a strong chunk of market share from Intel. Intel tried everything they could to take a chunk back and for sometime we didn't know if they would. Then they take the same route AMD did with a lower core speed, dual cores with more efficiency and we have the current king Core 2 Duo. AMD has been the innovator for the past decade or so with Intel taking cues but still flexing their own muscle. Some technologies were simply stolen (AMD64/EM64T anyone) or at least modified enough to fit their sales structure.
I am not sure if the Barcelona will be the next Athlon64 but if they pull it off we will see Core2Duo in the bargain bins much like the PentiumDs were.

Most interesting bit of Barcelona info, yet.
By MartinT on 6/6/2007 4:01:44 AM , Rating: 3
AMD partners confirmed the highest running, POST and OS capable, Barcelona processor is 2.0 GHz.
I guess that puts the recently rumored supposed 500 MHz gain from a supposed miracle stepping in perspective.

Sure seems like there's still quite a bit a work ahead of AMD, before they're back in the game.

RE: Most interesting bit of Barcelona info, yet.
By just4U on 6/6/2007 4:11:45 AM , Rating: 2
Do you remember the early Operton's (before launch) running at 1.4GHZ. People were not to impressed with them either when Intel was running chips 2x Faster. It was rather comical to imagine it being the powerhouse that it was touted to be. However, after it's release people had to rethink their whole mghz idea and what was classified as performance.

I am not saying that this will be anything like that but just pointing it out! One never really knows till we start seeing some real benchmarks and hands on testing. I hope that comes soon.

RE: Most interesting bit of Barcelona info, yet.
By defter on 6/6/2007 4:26:10 AM , Rating: 2
Fastest Opterons were running at 1.8GHz at launch. It's true that earlier samples were running at 1.4GHz, but that was several months before the launch.

Unfortunately, it seems that it will take several months before Barcelona will reach more competitive clockspeeds :( and by that time Penryn based server CPUs will start arriving.

By spartan014 on 6/6/2007 4:30:21 AM , Rating: 2
Even then they could have given some details regarding the performance. I mean, if they have got the systems up and running, what is stopping them from running some benchmarks?

Not thorough, by the book ones.. Something that will give us some perspective. As of now, everything is still vague regarding the performance.

By defter on 6/6/2007 4:37:36 AM , Rating: 2
I mean, if they have got the systems up and running, what is stopping them from running some benchmarks?

Probably AMD is stopping them. What's the point of running benchmarks at 1.6GHz and showing that the Barcelona is slower than existing competition?

RE: Most interesting bit of Barcelona info, yet.
By MartinT on 6/6/2007 4:45:36 AM , Rating: 2
Do you remember the early Operton's (before launch) running at 1.4GHZ.
I remember AMD not being able to demonstrate K8 above 800 MHz for the better part of a year.

But my point was not about performance at all, but rather how last week's rumor of an extra 500 MHz, when combined with this bit of information, would have AMD not even reaching the simulated 2.6 GHz they used for their SPEC_2006 estimates.

By Justin Case on 6/6/2007 12:41:23 PM , Rating: 2
The complete chip production process takes several weeks, not days. Improvements in one part of the process may not translate into faster chips for over a month, so it's possible that AMD has better silicon "in the pipeline" but not any finished CPUs.

So far AMD has delivered the speeds they said they would (I still can't buy a 10 GHz Pentium), so I wouldn't worry too much about the early samples they're sending out. The first Opteron samples ran at 800 MHz. At launch they were up to 1.6 GHz, and they reached 2 GHz not long after that.

By BladeVenom on 6/6/2007 5:09:21 AM , Rating: 5
Agreed, lets wait for the final version, and for some reviews come out before making a hasty judgment.

By DigitalFreak on 6/6/2007 8:28:04 AM , Rating: 2
Different ball game now. The Core architecture is miles ahead of Netburst in pretty much every way.

Wall Street went into panic this morning ?
By crystal clear on 6/6/2007 8:12:45 AM , Rating: 3
AMD certainly knows how to hurt itself by keeping SECRETS &
keep everybody guessing.

NO NEWS is BAD NEWS for the industry.

Read below-

"Cray started this mess yesterday by saying that component delays would cause it to miss the planned fourth quarter ship date for a revamped version of its XT4 supercomputer.

The only problem was that Cray refused to disclose which component in particular had caused the delay and failed to specify which version of AMD's upcoming four-core family – Barcelona or Budapest – it will use in the new XT4s. AMD has long said that Barcelona will ship in the Summer, while the lower-end Budapest is slated to arrive in the fourth quarter.

Wall Street went into panic this morning when the Dow Jones printed a pair of stories "confirming" that a Barcelona delay was behind Cray's issues. In one story, the newswire reported:

The part that is delayed is the four-core processor being made for the supercomputer by Advanced Micro Devices Inc. (AMD), Steve Conway, a Cray spokesman, said in an interview.

"Its postponement, at least according to everything we know right now, is a brief postponement," Conway said.

In another story, the newswire reported:

On Monday supercomputer maker Cray Inc. (CRAY) issued a revenue warning for the year, citing a delay in the availability of a component, which analysts said is likely the Barcelona server chip. Meanwhile, on Tuesday Citigroup analyst Glen Yeung said
said in a research report out of Taiwan's Computex Trade show, that "several sources," including server vendors, distributors and motherboard makers, confirmed Barcelona will be delayed to September/October from June/July.

By crystal clear on 6/6/2007 8:19:08 AM , Rating: 2
Add this to the above-

Cray has blamed delayed shipments of Opteron-based servers for a likely massive fall in 2007 revenue.

The supercomputer maker today warned that its midpoint forecast for 2007 revenue has been lowered to $200m from previous guidance last month of $230m. The revenue drop comes as a result of delayed XT4 system shipments and could result in Cray posting a loss for the year. Cray had hoped to get upgraded XT4 systems using AMD's upcoming four-core version of Opteron out the door at a quick clip.

"Due to recent information with respect to the timing of volume parts availability, the company now believes that 2007 revenue associated with quad-core Cray XT4 systems will likely be less than previously anticipated," the company said. "Cray continues to anticipate deliveries of quad-core Cray XT4 systems and upgrades to begin during the fourth quarter of 2007; however, the timing is such that most or all of the planned acceptances, and associated revenue recognition, will likely be deferred until early 2008."

Cray had been enjoying a recent return to profitability. CEO Pete Ungaro noted that he was "disappointed" with the delays that put both growth and profitability "in jeopardy."

RE: Wall Street went into panic this morning ?
By CyborgTMT on 6/6/2007 1:41:06 PM , Rating: 5
Again the copy/paste of only half the information to make the Barcelona look like the bad guy...

We're told that the XT4 boxes rely on the lower-end 1000 Series four-core chips code-named Budapest, which AMD has said will ship in the second half of 2007. The rest of the world is most excited about the higher-end four-core parts dubbed Barcelona, which are set to ship by mid-year.

"Barcelona remains on track for launch this Summer," said an AMD spokesman. "You will see platforms shipping from partners in the third quarter. Budapest is on track for the second half of the year."

In short this was Cray's fault for thinking they would have the Budapest cpu's earlier than AMD is releasing them. AMD's roadmaps have shown it's release in the later half of 2007 for a long time now.

By crystal clear on 6/6/2007 7:52:36 PM , Rating: 1
I am not taking sides nor do I represent any of them or speak on behalf of them-

AMD communications with their buyers & or prospective buyers & the industry in general is simply a disaster.

Being too secretive about their plans/roadmaps/product details/etc etc has reached its extreme.

The result of this is the Craig/AMD episode.

I always tell my clients-
"Call me up-when you want & ask what you want-I am waiting for your call"

Thats the attitude - thats how you prevent P.R. disaters & confusions.

By crystal clear on 6/7/2007 4:48:36 AM , Rating: 2
Cray had been enjoying a recent return to profitability. CEO Pete Ungaro noted that he was "disappointed" with the delays that put both growth and profitability "in jeopardy."

Cray is complaining about delays & the consequences of that delay & not the type of processor !

None of you guys out there know what AMD had promised CRAY about deliveries of CPUs.

In such cases ROADMAPS dont count rather contractual obligations between the companies.

There is no obligation to stick to your roadmaps-you can delivery to market before scheduled date or delay it further.

But when/if your are bound by contract to deliver on/by a

certain date/schedule & you fail to do so,then ofcourse you

put both growth and profitability "in jeopardy." of the

company to recv your CPUs

The SECRECY firewall of AMD is creating more problems than

protecting AMD.

By Nil Einne on 6/11/2007 4:32:42 PM , Rating: 3
Actual, Cray is simply saying that the product is coming later then they had expected. Whether AMD promised anything to Cray that they didn't deliver on is unknown. Perhaps Cray was just to optimistic, who knows? The simple fact is we have no idea what AMD promised Cray as you rightful pointed out and assuming that AMD promised something to Cray which they didn't deliver as you are suggesting is silly.

RE: Wall Street went into panic this morning ?
By CyborgTMT on 6/6/2007 1:30:42 PM , Rating: 4
Funny how you left out one part of the article....

Those of you interested in this kind of stuff will want to note that Cray won't be using the Barcelona chips in the XT4s. The systems will run on the lower-end, latter Budapest chips.

By crystal clear on 6/7/2007 3:59:23 AM , Rating: 1
I only quote a portion of the article & ofcourse provide the link/source of the story.

Its up to you to click on the link & read the whole story & draw your OWN conclusions.

I dont shape/influence opinions rather just bring the story up on the site for "strictly information purposes".

The interesting thing...
By Goty on 6/6/2007 7:46:05 AM , Rating: 3
The most interesting thing about this demonstration is that the Barcelona-equipped system supposedly draws around 50W less on average than the same system equipped with a dual-core Opteron.

(Source: )

RE: The interesting thing...
By Thorburn on 6/6/2007 8:04:28 AM , Rating: 2
You'd certainly hope so given the clock speeds though wouldn't you.

Looking at the numbers that'll be both chips at idle too, not full load.

RE: The interesting thing...
By Goty on 6/6/2007 8:08:16 AM , Rating: 4
Looking at the numbers that'll be both chips at idle too, not full load.

The quad core system drew 214W at full load, while the dual core system drew 259W.

Did you even READ the story?

RE: The interesting thing...
By Thorburn on 6/6/2007 8:30:27 AM , Rating: 1
No sir I did not :)

Shame theres no info on what the rest of the spec was, or even the dual-core parts clock speeds, as that power usage is VERY low for a Socket F system.

RE: The interesting thing...
By Brother Michigan on 6/6/2007 10:19:52 AM , Rating: 2
Why did this post get rated down? It's not fanboyish in any way, it applies to the article at hand, it contains factual information and the source the information was obtained from.

RE: The interesting thing...
By sdsdv10 on 6/6/2007 10:26:13 AM , Rating: 2
Probably because the OP linked to FUDZILLA. Rightly or wrongly, that site doesn't garner a great deal of respect in these parts, partner! ;-)

AMD System Stress Test
By McJ on 6/6/2007 7:19:09 AM , Rating: 2
Does anyone know where to find this AMD System Stress Test Tool? Couldn't find it on the AMD Site.

RE: AMD System Stress Test
By JackPack on 6/6/2007 3:32:51 PM , Rating: 2
It's probably only for partners validating future AMD processors.

RE: AMD System Stress Test
By crystal clear on 6/7/2007 2:49:49 AM , Rating: 6

(Go to forums- Register & post the querry)

If not successful contact AMD for assistance.

RE: AMD System Stress Test
By McJ on 6/7/2007 9:32:40 AM , Rating: 2

nvidia nforce 3400
By duploxxx on 6/6/2007 11:20:04 AM , Rating: 2
How can this be a benchmark of a Barcelona cpu if the nforce 3400 is an am2 socket. Do we have a Budapest cpu at teh stage?

RE: nvidia nforce 3400
By JackPack on 6/6/2007 3:38:37 PM , Rating: 2
nForce Pro 3400 works with Socket F (1207).

RE: nvidia nforce 3400
By Screwballl on 6/7/2007 10:49:27 AM , Rating: 2
another thing to remember is that some current testing may be done using AM2/1207 and not always AM2+ and 1207+. some of the reports may be coming from an AM2 board which does not support the higher HT speeds. Likely due to little to no AM2+ on the market for now

RE: nvidia nforce 3400
By Proteusza on 6/7/2007 3:00:17 PM , Rating: 2
err what do you mean little to no AM2+? there isnt a single market ready board yet.

Or is there?

New pricies announced
By crystal clear on 6/6/2007 9:45:47 AM , Rating: 2

AMD Processor Pricing

Effective June 5, 2007

RE: New pricies announced
By crystal clear on 6/6/2007 9:48:08 AM , Rating: 1
correction-Subject-should read

"New prices announced"

RE: New pricies announced
By crystal clear on 6/6/2007 10:10:30 AM , Rating: 1
AMD has sent out the following e-mail about price cuts for the desktop CPUs coming July 9th. You can search for prices on AMD desktop CPUs here.

As part of our continued commitment to you, our channel partners, we are providing this e-bulletin to communicate breaking news between issues of our monthly e-newsletter, the Tigon. We know how much email we all receive, so we will only send bulletins like this one when we have important and time-sensitive information to share with you that we feel will help you market and sell AMD-based products.
Desktop Price Move Announcement
One of our main goals is to help you manage your business more efficiently by keeping you informed of changes that could affect your purchasing decisions and sales forecast.
One such change will be announced on July 9th, when the AMD desktop processor family will have a reduction in suggested prices.

Savings for the AMD Athlon™ 64 X2 (dual-core) processor family will be about 20% and the AMD Athlon 64 (single-core) processor family will be adjusted by approximately 15%. If you are looking for a great value, look to the AMD Sempron™ processor, which will also offer savings of about 15%.
These new prices will enable you to offer your clients AMD-based desktop solutions with a true price/performance advantage at an even more competitive price. Please contact your authorized AMD distributor for specific pricing and availability details.

RE: New pricies announced
By coldpower27 on 6/6/2007 10:37:05 AM , Rating: 2
Countering Intel's July 22nd price cuts no doubt. AMD did the same thing with Intel's price cuts back in April this year.

AMD Phenom FX-xx.
By crystal clear on 6/6/2007 10:59:35 AM , Rating: 2
Watch the slides & not the female !

Leslie Sobon - Director, Desktop Business, AMD provided us some updates on Phenom today. According to her, there is no special meaning for G, S and L. G stands for Premium where Phenom X4 will compete against Kentsfield and Phenom X2 will compete against Conroe. S stands for Intermediate while Athlon 64 X2 will compete against the Intel Pentium 2000 series. As such, there will be AMD Phenom X4 GP-7xxx, Phenom X2 GS-6xxx, Athlon X2 LS-2xxx and Sempron LS-1XXX. She mentioned that there will be both Socket 1207+ as well as Socket AM2+ Phenom FX for more platform versatility and the FX series will retain back their model naming eg. AMD Phenom FX-xx

RE: AMD Phenom FX-xx.
By CyborgTMT on 6/6/2007 1:47:36 PM , Rating: 2
Watch the slides & not the female !

You really need to get out of your house more often.

RE: AMD Phenom FX-xx.
By crystal clear on 6/6/2007 8:01:11 PM , Rating: 1
That comment of mine was simply a Joke -nothing series.

My wife tells me the opposite-

"You really need to be at home more often."

Enough of the 40+ hrs a week

RE: AMD Phenom FX-xx.
By crystal clear on 6/7/2007 1:28:09 AM , Rating: 1
Correction-should read "nothing serious"

Notice something odd?
By Proteusza on 6/6/2007 5:22:28 PM , Rating: 2
The machine is using XGI Volari graphics chips. Strange for a company that has an in house graphics division that happens to be one of the largest in the world.

RE: Notice something odd?
By nerdye on 6/6/2007 11:02:45 PM , Rating: 2
good observation, my question is the same, and we also have the article here on anantech/dailytech discussing AMD's next desktop chipset, one of which mentions having a working sample using an hd2600 gpu, therefore, why is this sample not? If you haven't read it already, check out the local link, and post your thoughts on the matter.

RE: Notice something odd?
By Slaimus on 6/7/2007 6:00:51 PM , Rating: 2
Barcelona is a server chip, and as such uses PCI graphics chips to leave the PCI-E lanes to bandwith intensive devices. Remember that this was a "partner" booth, so it is not run by AMD. That partner probably uses the chip to save money compared to the standard-issue ATI Rage IIc that is on most servers.

Barcelona - AMD claimed highest performance...
By INTC on 6/6/2007 4:10:34 AM , Rating: 1
per watt and at the same clock speed - they never claimed that Barcelona will have highest overall performance.

By defter on 6/6/2007 4:27:03 AM , Rating: 4
Yes, this puts "30-50% higher performance at the same clock in SPEC" claims in perspective...

By Chadder007 on 6/6/2007 8:43:48 AM , Rating: 2
So Clock for Clock, the Barcelona will likely equal Kentsfield on final silicon at this one benchmark.?
Not much to go on, but its something at least.

RE: Soo...
By Sharky974 on 6/8/2007 9:05:43 AM , Rating: 1
SHHH!! That's not what Dailytech wanted you to take away from this!

They wanted you to think AMD IS DOOOOMED. And now you're ruining it all!

RE: Soo...
By KristopherKubicki on 6/8/2007 10:33:16 AM , Rating: 2
AMD publically announced it was demonstrating Barcelona.

All we did was tell everyone the details AMD left out of its announcement.

Dear DailyTech
By jhtrico1850 on 6/6/2007 7:16:54 PM , Rating: 2
What is meant by Current AMD Barcelona samples are not scaling too well.?

Does that mean Barcelona cannot scale to higher speeds, does not give performance linearly to clock increase, or what?

RE: Dear DailyTech
By KristopherKubicki on 6/6/2007 7:44:31 PM , Rating: 2
Specifically, this statement means that its not ramping clock frequencies as high as the roadmaps projected.

RE: Dear DailyTech
By jhtrico1850 on 6/6/2007 9:15:34 PM , Rating: 2
I think it would be nice if the article clarified that :)

RE: Dear DailyTech
By Sharky974 on 6/7/07, Rating: -1
AMD did not learn from Intel's mistakes
By arriddle on 6/7/2007 5:20:23 PM , Rating: 3
AMD's situation strongly reminds me of Intel when it attempted to change processor architecture and process node simultaneously (remember P3 to P4 transistion along with shrink to .18um in 2000). These transitions are hard enough on there own but when attempted in unison it can be crippling. That's why Intel has a tick-tock strategy in order to avoid this problem again.

By Nil Einne on 6/11/2007 4:39:25 PM , Rating: 2
Um... AMD doesn't do it either. They're already moving to 65nm on the K8 (Brisbane). Perhaps they didn't leave enough of a time gap but it's clearly not simultaneous...

AMD roadmaps
By crystal clear on 6/6/07, Rating: 0
RE: AMD roadmaps
By Martimus on 6/6/2007 12:43:50 PM , Rating: 2
Do you work for Intel? You always seem to take anything related to AMD personally. I just don't understand.

RE: AMD roadmaps
By Martimus on 6/6/2007 12:48:49 PM , Rating: 2
I didn't mean to insult you, and I appologize if I did. I just don't always understnad the motivation behind your posts.

RE: AMD roadmaps
By crystal clear on 6/7/07, Rating: 0
By sprockkets on 6/6/2007 10:55:53 AM , Rating: 2
Well, in the pic, some of the cache runs at 1.92 or at 1.5, so to me looks like it was just going up and down per the load of the core

RE: 1.6ghz
By johnsonx on 6/6/2007 5:07:10 PM , Rating: 2
You're misinterpreting what you're seeing there. The 1.92 G and 1.5 G don't refer to clockspeed. I don't know what they do refer to, but I promise it isn't clockspeed.

What about Penryn.
By ccbr01 on 6/6/2007 1:47:07 PM , Rating: 2
Just have to remember about Penryn that is coming out later this year. By the looks of this benchmark, Penryn should put Intel back on top, even if K10 is faster than the current Intel Quad.

amd partners = amd fools
By Crazyeyeskillah on 6/6/2007 4:28:35 PM , Rating: 2
Roll over already.

Windows XP 2003 Service Pack 2?
By Lord 666 on 6/6/2007 6:49:51 PM , Rating: 2
Not to be picky, but the caption of the second picture should read "Windows 2003 64-bit R2 SP 2" instead of "Windows XP 2003 SP2"

By Frags on 6/7/2007 11:50:33 PM , Rating: 2
It would be nice to see AMD succeed in this venture, but this news is as bad of news as it was months ago when they displayed 'Task Manager' to analysts.

Some have mentioned processor history, but for the Current Intel holds the crown and it looks like it will for years to come if December is the new goal date. This could set AMD up for another mass let down like R600.

Looks like the Underdog will remain.

Not bad for sample part
By CalmDown on 6/8/2007 1:48:25 PM , Rating: 2
Given the demo system is using only 1 CPU (as seen in the screen shots) this is not bad for a Budapest CPU since it is about 6 months from being available.

By innotech on 6/8/2007 12:08:18 AM , Rating: 1

Dailytech with the AMD hate
By Sharky974 on 6/7/07, Rating: -1
One never ever knows.
By jasaero on 6/6/07, Rating: -1
RE: One never ever knows.
By INTC on 6/6/2007 8:21:45 AM , Rating: 2
AMD just dropped prices by up to 50% and will drop them again in July - maybe that is when Barcelona will launch. But if that were the case, they should be filling up their partners' warehouses by now with final launch products and not development material. ?????????

RE: One never ever knows.
By crystal clear on 6/6/2007 9:59:00 AM , Rating: 1
AMD Processor Pricing

Effective June 5, 2007

RE: One never ever knows.
By DigitalFreak on 6/6/2007 8:25:47 AM , Rating: 2
OMG. Trying to read that made my head hurt.

RE: One never ever knows.
By Treckin on 6/6/2007 12:19:16 PM , Rating: 1
I dont want to pull apart your post... but I cant resist.

AMD was, and most likely will always be the underdog. (yes, Im an AMD fan, although I type from a C2D mobile :) )
The only reason people are worried about AMD is because they fail to (or are to young to) remember before 2003, when Intel smoked every AMD out there. The only competitive advantage was their prices.
In essence, Intel fucked up, sat on their laurels, and netbursted their way into second place. Intel, at least 50% larger than AMD, has FAR more resources to dump into the game. The reason AMD is falling behind, is because of the die shrinks (and the rapidity at which they are occurring... anyone remember how long we were at .90?) cost a HUGE amount of money to develop and fab. As a corporation, Intel is far better equipped to outflow that volume of resources.

I honestly dont understand why people are worried. AMD is simply defaulting back to historically what they have been good at - price cutting for a better performance/price ratio. This lets their high end stuff go up against Intel's mid/ mid-high range parts, at half the price. They hurt em where they can.

Basically the 64bit and x2 procs woke the sleeping giant that was Intel. All we can hope for now is that AMD keeps snipping at the giants heels, driving them forward.

Ive head this many times, and it pisses me off. They have the memory controller on-die. Thats it. Not the whole N-Bridge...
And I believe Intel said they were going to an integrated MC with the Penryn, although I could be mistaken

RE: One never ever knows.
By Martimus on 6/6/2007 12:40:03 PM , Rating: 4
AMD had faster 486 chips than Intel. The K6 was also faster than the Pentium and the Pentium Pro. Not many people knew these things, because Intel had a much more marketing and clout in the industry. At that time, it was always assumed that Intel had the best quality, because other chips were known to have poor quality (like Cyrix) and Intel was a safe bet. Intel did not always have better chips than AMD. I would even go so far as to say that AMD had the better processor throughout most of the 90's.

As for the memory contoller; Intel plans on incorporating that with the next architecture change, not the next die shrink. It will likely be available at the end of 2008. Nehalem is the Intel chip that should have the memory controller on the die.

RE: One never ever knows.
By Justin Case on 6/6/2007 12:45:29 PM , Rating: 3
Not to mention the Athlon MP which was, IMO, the real "turning point" for AMD in terms of going after the server / workstation market.

RE: One never ever knows.
By hans007 on 6/7/2007 7:13:37 PM , Rating: 3
Amd's faster 486 chip was the 5x86-133 which was about equaly to a pentium 66.

Sure AMD had faster 486 than Intel but by the time it came out the pentium was already out. They also had a faster 386 the 386-40. But that was actually a complete clone, just running faster.

The k6 really wasn't better than the Pentium it was maybe equiavlent and had a poorer FPU. the Pentium MMX 233 was faster than the k6-233 and at the time the k6-233 was overclocked at 3.2 volts compared to 2.9 for the k6-200 and 166 for .35 micron node. The Pentium MMX was much cooler running at the time (and at the time 35 watts TDP was a lot).

That said, it was competitive. By the time the k6-266 and 300 came out on .25 micron, the k6 was getting blown away by the pentium ii. And the Pentium II was head and shoulders above any k6-2 , the only thing really competitive was the super expensive and hard to find k6-3 chips or the few k6-2+ laptop chips.

So basically the entire k6 generation was not quite up to par / slightly behind. This was around the time that AMD was nearly bankrupt like now, and its stock fell to like $4-5. I remember I was a huge AMD fan boy and bought stock when I got out of high school.

the K7 was better than most p3s and earlier p4s until the northwood. That was around when AMD actually recovered out of their near bankruptcy around then the first german fab opened.

Also , for the reliability thing. HP was one of the first major OEMs to pick up AMD. Coincidentally I worked at an HP factory (since closed and moved to mexico btw) in california when they first got k6-2 300-350 chips.

The failure rate was easily 3 times as high for dead on arrival or chips that didnt quite perform to spec compared to the intel class equivalents (which were the celeron 300 a and 333 a in the systems we were refurbing / building).

That was a terrible summer job btw, but the AMD is not as reliable thing was definitely true back then. They were binning chips too high just to try to keep up with the Pentium III and Celeron As

RE: One never ever knows.
By nah on 6/7/2007 3:38:48 AM , Rating: 2
Intel did not smoke AMD before 2003--remember the Athlon ?

RE: One never ever knows.
By Slaimus on 6/7/2007 6:08:37 PM , Rating: 2
AMD quad cores do have the northbridge on-chip as well. If you look at the die shots from the past posts, it clearly labels the northbridge. In addition to the memory controller, the northbridge probably handles a lot of the inter-core communication.

stop worrying
By excrucio on 6/6/07, Rating: -1
RE: stop worrying
By sdsdv10 on 6/6/2007 3:24:51 PM , Rating: 3
Check C2D and X2's and FX's benchmarks. Intel have always been behind or even.

Please provide links. This is the exact opposite of everything I have read for the last twelve months (since C2D launched).

RE: stop worrying
By JackPack on 6/6/2007 3:35:03 PM , Rating: 2
He lives in a parallel universe.

You should have requested links to FUD which confirm his facts.

RE: stop worrying
By excrucio on 6/6/2007 8:16:10 PM , Rating: 2

AMD FX 74 vs EX6800 (both lastest chip from both companies. Dont campare the QX vs AMD's FX's)

As you can see they are equivalent and FX being cheaper. HELL OF ALOT cehaper by like 600 dollars



RE: stop worrying
By odiHnaD on 6/6/2007 9:00:55 PM , Rating: 2
Why don't you look at the rest of the benchmarks smart guy, if you would have you might have noticed that it's one of the only benchmarks of the bunch where the two are equal, heck in F.E.A.R. even the C2D E6400 beat it. (

FYI: Checking facts before posting helps you NOT look like a fanboy

RE: stop worrying
By odiHnaD on 6/6/2007 9:02:26 PM , Rating: 2
Sorry bad link, here's a working one:

RE: stop worrying
By utube545 on 6/6/07, Rating: -1
RE: stop worrying
By swtethan on 6/6/2007 9:54:48 PM , Rating: 2
Did he even realize that fx-74 = 4x4? TWO DUAL CORE FX's

the competitor to that is the QX6700 (or QX6800)

and we all know that 4x4 = sucks

RE: stop worrying
By Amiga500 on 6/7/2007 4:30:22 AM , Rating: 2
and we all know 4x4 = sucks

Yeap, its not great in most benches (but in some that matter to me, like MFLOPS, it still rules)... but in a couple of months time it will be 8x4, very different kettle of fish.

If its not priced stupidly, it is a pukka workstation without needing registered memory... as long as AMD give it a chipset that can handle more than 8GB.

I know I'm waiting with baited breath for one, I'm sure others are the same.

RE: stop worrying
By Accord99 on 6/7/2007 5:36:01 AM , Rating: 2
I know I'm waiting with baited breath for one, I'm sure others are the same.

I'd imagine most have already picked up faster Clovertown systems.

RE: stop worrying
By Amiga500 on 6/7/2007 8:48:09 AM , Rating: 2
I'd imagine most have already picked up faster Clovertown systems.

It is not possible to buy a 2 socket clovertown system yet is it?

(I obviously mean non-xeon that doesn't need registered RAM]

I'm waiting for a 8core job using non-registered RAM (hopefully more than the 8GB limit of the QuadFX).

RE: stop worrying
By swtethan on 6/6/2007 9:54:50 PM , Rating: 2
Did he even realize that fx-74 = 4x4? TWO DUAL CORE FX's

the competitor to that is the QX6700 (or QX6800)

and we all know that 4x4 = sucks

RE: stop worrying
By nerdye on 6/6/2007 11:10:27 PM , Rating: 3
You are right 4x4 is no logical match for any of the current Intel solutions out today, I asked the question to the VIP of Marketing of the software company I worked at last "why on earth would AMD release a chipset/platform that is by all means crippled by its current competition from Intel"?. He said "sometimes simply supporting the correct word, like "quad, or 4x" in this situation may pursuade enough of the customers that are "less in the know" to make its release worth going through with. Although in this situation AMD 4x4 sucks very badly, its easy to apply his logic to many product releases and might explain AMD's thought process despite their lack of success with the product of discussion. This is based on a moot topic though, Intel is the choice hands down, but I thought I would share the marketing logic I had learned and hope it applies.

RE: stop worrying
By CyborgTMT on 6/8/2007 6:24:55 AM , Rating: 2
In part I would agree that AMD was just trying to get the name out but part of it was also to work any bugs before the quads hit the market. With a 2x2 set up it's naturally going to lose to a chip that has all 4 cores on one processor, even if the individual cores are identical in performance. The true test of the platform is going to be the nut with too much money to spend that slaps 2 quads and a 4 X Crossfire setup on it. Maybe by the time this is all available I'll have enough saved up to throw down on one myself. Of course I'll be reduced to a Ramen noodle diet to pay the electric bill every month.....

RE: stop worrying
By Nil Einne on 6/11/2007 4:47:45 PM , Rating: 2
Exactly, I think AMD knows their the 4x4 will really shine when they get quad. 8x4 is likely crap all over anything Intel has given that Intel still lacks a 'Xeon for desktops'. Of course you'd have to be insane to want 8x4 but all AMD cares about is being able to advertise the fastest system. Intel could of course respond by sticking two of their Penryns together and making a 8 core OR by speeding up the 'Xeon for desktops' platform. But AMD will probably still have that area all to themselves for a while at least. Intel also are still stuck on the FSB and I suspect it'll start to show with a 8 core system.

"Mac OS X is like living in a farmhouse in the country with no locks, and Windows is living in a house with bars on the windows in the bad part of town." -- Charlie Miller
Related Articles

Most Popular ArticlesAre you ready for this ? HyperDrive Aircraft
September 24, 2016, 9:29 AM
Leaked – Samsung S8 is a Dream and a Dream 2
September 25, 2016, 8:00 AM
Yahoo Hacked - Change Your Passwords and Security Info ASAP!
September 23, 2016, 5:45 AM
A is for Apples
September 23, 2016, 5:32 AM
Walmart may get "Robot Shopping Carts?"
September 17, 2016, 6:01 AM

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki