ie8 fix
madison

Intel turns up the efficiency and performance heat on AMD

By | August 14, 2007, 5:16pm PDT

Summary: Intel’s latest manufacturing refinements to the 65nm Core 2 architecture has produced server CPUs that drop power consumption, boost performance, and include virtualization enhancements.  These new refinements called “Stepping G” have also given higher performance and efficiency to Intel’s desktop line of Core 2 processors allowing them to overclock past 4 GHz.  On the server side Stepping G [...]

Intel’s latest manufacturing refinements to the 65nm Core 2 architecture has produced server CPUs that drop power consumption, boost performance, and include virtualization enhancements.  These new refinements called “Stepping G” have also given higher performance and efficiency to Intel’s desktop line of Core 2 processors allowing them to overclock past 4 GHz.  On the server side Stepping G has given us the new Intel XEON X5365 and L5335 microprocessor.

The Xeon X5365 “Clovertown” quad-core CPU at 3.0 GHz with a TDP (Thermal Design Power) of 120 watts sets a new SPECint_rate_base2006 world record for two-socket servers with a score of 98.9 (peak 107).  According to TechReport’s Scott Wasson, the new stepping slashes idle power consumption of the X5365 in half from 50W idle to 25W idle power consumption which is only 1W higher than the idle power consumption of the new low voltage L5335.

Note that Apple released a 3.0 GHz “Clovertown” back in April for their 8-core workstation but that chip had a TDP of 150 watts which is too high for normal two-socket servers.

The Xeon L5335 “Clovertown” quad-core CPU at 2.0 GHz low voltage processor has a TDP rating of 50W but no SPECint_rate2006 numbers are available at this time.  We can get an idea of how well it performs by looking at the older E5335 80W part which has a SPECint_rate_base2006 score of 78.5 (peak 85) if all the other components in the server match.  The 50W Intel Xeon L5355 will make it very hard for AMD to claim the performance/watt lead on their yet-to-be-released 95W Barcelona 2.0 GHz CPU.  Even if we take in to account the extra 29 watts on Intel’s chipset because of the “North Bridge” memory controller on the motherboard and the 48 watt penalty of eight fully buffered memory DIMMs, it still gives Intel a 13W advantage in a two-socket server.

While no SPECint_rate2006 scores are published for AMD’s Barcelona CPU yet, many industry analysts are expecting an AMD “Barcelona” 2.0 quad-core to lose to an Intel Clovertown 2.0 quad-core CPU on integer performance.  Considering the fact that AMD’s Barcelona 2.0 GHz CPU is a 95W part that will likely have lower integer performance than an Intel Clovertown 2.0 GHz 50W part, it’s understandable why AMD has declined to announce any integer performance numbers yet for their 2 GHz Barcelona.  AMD has claimed that they can get a SPECfp_rate2006 peak score of 69.5 for floating point performance on the 2 GHz Barcelona which is very impressive but that only addresses a niche HPC (High Performance Computing) market.

AMD will eventually launch a low-voltage 68W variant of the Barcelona 1.9 GHz quad-core CPU but the initial version of the 1.9 which most likely launches in September will be 95W.  Even then it will only be a narrowly defined victory in a two-socket CPU 8-DIMM configuration for the absolute lowest power consumption and it may or may not be compelling on performance/watt for typical server workloads.  AMD can’t delay too much longer because they will soon be facing a much bigger monster in the form of Intel’s 45nm Penryn.

From an HPC perspective the AMD Barcelona 2.0 GHz parts will be an absolute steal because it won’t be able to command high margins due to its mediocre mainstream server computing performance.  That might be good news for scientists and researchers who need cheap and powerful HPC workstations but the modest margins won’t fix AMD’s financial woes unless they can significantly ramp up the clock speeds on Barcelona quickly.  For now it seems that all we can do is wait on news for AMD while Intel seems to be tick-tocking away on execution.

Kick off your day with ZDNet's daily e-mail newsletter. It's the freshest tech news and opinion, served hot. Get it.

Topics

Disclosure

George Ou

http://blogs.zdnet.com/Ou/?page_id=557

Biography

George Ou

George Ou, a former ZDNet blogger, is an IT consultant specializing in Servers, Microsoft, Cisco, Switches, Routers, Firewalls, IDS, VPN, Wireless LAN, Security, and IT infrastructure and architecture.

71
Comments

Join the conversation!

Just In

Q6600 quad is best bang for buck
georgeou 18th Aug 2007
E6600 is not worth it because the Stepping G based E6550 is a better buy. Even the E6750 2.66 with FSB1333 is cheaper than E6600.

If you got more budget, go for a quad core at $280 for the Q6600.
0 Votes
+ -
MAth
AMD OPTERON 14th Aug 2007
I dont know where u went to school but the total is 127 watt for the Intel and the AMD is 95 watts MAX thats a 32 watt difference in AMD's favor. You dont ever seem vary fare in your view. And now I see you can't add(FBDIM,MEMORY CONT. and Processor 127 watts)if the info you yourself gave is true
0 Votes
+ -
Check your own math!
georgeou 14th Aug 2007
2 95W barcelona chips is 190W which is 90W more than two 50W Clovertown. Now you've got a 48W FBD and 29W chipset liability for Intel so that's 77W total non-CPU liability. 90W - 77W = 13W advantage for Intel in a two CPU socket eight DIMM configuration on low voltage parts. Check your own math!
0 Votes
+ -
Snore.....
D T Schmitz 14th Aug 2007
nt
0 Votes
+ -
C'mon, that wasn't called for
georgeou 14th Aug 2007
You may not like what you're hearing, but this is an informative piece.
0 Votes
+ -
nt
0 Votes
+ -
There are plenty of other posts than this one; several new ones from this week.
0 Votes
+ -
Good article
patibulo 15th Aug 2007
Nice article. I'm now wondering though, what would this mean for average consumers. I mean, competition is good as long as long as the products are similar. But as Intel increases the gap, will this mean lower prices for AMD, will AMD go out of business, or will they simply focus on different market segments? Let's wait and see...
0 Votes
+ -
AMD won't go bankrupt
georgeou 15th Aug 2007
AMD won't go bankrupt so it Intel will continue to drop prices. They're not going to slow down because AMD is not delivering and they're not going to ease up on pricing pressures. For the foreseeable future you?re going to see better products and lower prices. AMD will keep their volume up on the bottom end by slashing prices and they?ll stay bottom feeders until they can deliver their K10 product with the kind of clock speeds they claimed.
0 Votes
+ -
It is just two dual cores slapped together.

I know, I just had to play the AMD Defender angle

Sitting and waiting is what I continue to do. Good thing I don't have any money to take advantage of Intel's market dominance.
And those glued together quad cores are smashing AMD on every benchmark until AMD can deliver some decent clocked Barcelona K10 chips. But if they wait too long to launch K10 at respectable clock speeds, those glued together Penryns with 45nm High-K is going to widen the gap again.

Call it glue, call it "not true quad core", call it whatever you like, but AMD's CEO has been on record this year saying he wished AMD had gone Intel's dual-die route if he could do it all over again. Intel has some very good "gluing" process.
0 Votes
+ -
ANOTHER FALSE STATEMENT by George Ou
TechExec2 16th Aug 2007
.
"...And those glued together quad cores are smashing AMD on every benchmark..." -- George Ou

That is a false, clearly misleading, fanboy-like statement.

An honest comparison of two dual-core Intel CPUs in one package should be made against two dual-core AMD CPUs in separate packages. And, in that comparison, Intel does NOT "smash AMD on every benchmark".

No slaughter [Updated: SLAUGHTER claim that AMD lost all tests]
http://techrepublic.com.com/5208-6230-0.html?forumID=102&threadID=205265&messageID=2131629



This blog is a comedy piece! laugh laugh laugh

There is a never-ending stream of outrageously false statements against AMD made by George Ou. It's really hilariously funny! laugh laugh laugh I read this blog for the amusement factor. I also correct the false statements so they don't go unchallenged.

This blog is for fun! It's hilariously funny! laugh laugh laugh
0 Votes
+ -
TechExec2: go back to marketing school
thetruthhurts 16th Aug 2007
"An honest comparison of two dual-core Intel CPUs in one package should be made against two dual-core AMD CPUs in separate packages"

Aw, shucks...why, maybe we should compare dual-core Intel CPUs in one package against an AMD dual-core CPU without the integrated memory controller. Oh wait...
0 Votes
+ -
Glad you're amused...
Bob.Kerns 16th Aug 2007
Two dual-core in one package would indeed be a comparison you could make.

I won't dispute the numbers with you; I haven't been following processors closely enough.

But I'll challenge your statement (or implication, at least) that this is the ONLY honest comparison to make. Sockets & density count, power consumption is often critical. It's not unreasonable or dishonest to compare apples and apples.

It seems like all you care about is raw performance (for what type of usage, I'm not sure). That's reasonable, but it's only ONE reasonable basis for comparison.

If your goal is to cram as much compute power into a laptop, or a data center, consuming as little power as possible, you'll make a different comparison.

And while George's statements may be a bit hyperbolic, I think it's fair to say that there's concern about AMD's ability to compete successfully, near-term and long-term, inside AMD, among potential customers, among investers, and in the press.

I take it your position would be that this concern isn't supported by the numbers. But I think you have to look at it market-segment-by-market-segment. I think both you and George are guilty of promoting your view, aimed at a particular market segment, as being more broadly true -- and you're probably talking about different markets.

I'm left wondering whose needs are best met by your two-socket AMD solution, vs whose needs are best met by George's single-socket Intel solution, and whether AMD's solution would translate into a successful business strategy. (Even without the FUD factor).

I've never owned an AMD CPU. I'm not even aware of having ever USED one... (though I must have). But it would certainly be a shame for Intel to end up being the sole source for CPUs.

I have this image in my mind of all the journalists rushing to one side of the boat to watch AMD's fall overboard -- only to tip the boat, causing the very disaster they rushed to report.

But the key is to not focus overmuch on benchmarks, or any single metric. Not many customers consider a single metric in their purchase decisions. Benchmarks have to be considered cautiously, in context.
0 Votes
+ -
A long reply for your long post, Bob...
TechExec2 17th Aug 2007
.
"...I won't dispute the numbers with you; I haven't been following processors closely enough..."

George's blog post here is about the numbers and what they mean. The numbers are the topic. If you don't discuss the numbers, then we're at best adjacent to the topic.



"...But I'll challenge your statement (or implication, at least) that this is the ONLY honest comparison to make. Sockets & density count, power consumption is often critical. It's not unreasonable or dishonest to compare apples and apples..."

This is an obvious straw man, Bob. I never said, nor did I imply, that there was ONLY one honest way to compare AMD vs. Intel in this case.

Number of sockets and density count are important and valid things to discuss, but they are a bit off-topic here. The topic here is efficiency and performance.

I took issue here with George Ou's false statement that "...those glued together quad cores are smashing AMD on every benchmark...". That is completely false and I proved it.



"...And while George's statements may be a bit hyperbolic, I think it's fair to say that there's concern about AMD's ability to compete successfully, near-term and long-term, inside AMD, among potential customers, among investers, and in the press..."

This is a red herring. I never discussed any of this. But, I don't completely disagree at all. AMD is only just now responding to Intel's Core 2 introduction from last year and is taking a beating financially (but they are still standing).

On the other hand, they are just about to show some of the fruits of their labor. Barcelona-based products are entering the channel and will be out within weeks. And, a few weeks ago, AMD demonstrated two computers with Barcelona-based CPUs running at 3.0 GHz. This demonstrates that AMD has still got game.

Intel claims they "shipped" their 3.0 GHz dual-dual core "quad" CPU last April. But, you could not actually buy one until August. Very very limited production, apparently only allocated to Apple for their monstrously expensive Mac Pro tower computer. Intel is twisting the truth and people like George Ou are helping them to do it (1)(2). Where's the outrage about that? Shipped in April?! That is essentially a lie.

I did not count Intel out of the game when AMD was clearly winning from 2003 to 2006. I don't count AMD out of the game now either.



"...I have this image in my mind of all the journalists rushing to one side of the boat to watch AMD's fall overboard -- only to tip the boat, causing the very disaster they rushed to report..."

Excellent! Great point!

When pro-Intel people like George Ou relentlessly attack AMD (and use false stories as well (1)(2) ), they help to bring about the demise of AMD. If AMD ever dies, we will be left with one high-speed X86 CPU supplier. Then, the price of the fast CPU you want will be $1500 (not $200). The price of the slow one you settle for will be $800 (not $80).

This is not about fanboys. I don't care about AMD. This is about competition, choices, and good pricing. So, given the current state of things with AMD losing BILLIONS of dollars, if AMD supplies the price and performance you need, EVERYONE should ALWAYS buy from AMD until they get back on their financial feet.

Doing otherwise will help to push AMD overboard, just like George Ou is doing with this blog.



"...But the key is to not focus overmuch on benchmarks, or any single metric. Not many customers consider a single metric in their purchase decisions. Benchmarks have to be considered cautiously, in context..."

Another excellent point and I completely agree.

The problem is, customers DO consider the false impressions they are left with after reading blogs like this one by George Ou. And, they DO buy based on that.

AMD had the fastest X86 CPU from 2003 to 2006 (about $1000). AMD embarrassed Intel during that period. Intel has had the fastest X86 CPU since last year (about $1000), and a dual-dual-core "quad" CPU only since December 2006 (not very long at all). But today, below that fastest CPU, AMD CPU models have EQUAL performance and BETTER pricing (they must due to mis-perceptions). In other words, AMD has SOLIDLY BETTER price-performance. If you don't need the $1000 Intel CPU, buy the $169 AMD one (3), the fastest dual-core CPU that AMD has ever produced, and the most incredible bargain.



---------------------------------------------

(1) Debunking George Ou's rabid attack on AMD
http://talkback.zdnet.com/5208-10533-0.html?forumID=1&threadID=35931&messageID=661568&start=-1

(2) Baloney! This post FURTHER PROVES George Ou is unfair to AMD!
http://talkback.zdnet.com/5208-10533-0.html?forumID=1&threadID=36193&messageID=665742&start=-1

(3) Fastest AMD dual-core ever shipped $169
AMD's then-fastest sold for about $1000 last year. Before "Core 2", it
was the fastest X86 CPU in the world.
http://www.newegg.com/Product/Product.aspx?Item=N82E16819103773
0 Votes
+ -
HPC "niche"?
Robert Crocker 15th Aug 2007
Considering the millions of dollars being spent by various companies to support ever increasing CPU requirements for simulations I'd hardly call HPC a "niche" application.

If you look at it honestly, there are vary few applications right now for a quad-core CPU. The ones that come to mind are in the virtualization market so if you're going to call HPC a niche then you'd probably best label the entire quad-core market a niche market.
0 Votes
+ -
Disagree on a number of points
DevGuy_z 15th Aug 2007
1. Virtualization is not HPC. They don't have the same goals.
2. Most major server applications (Web, Mail, Database) are designed to take advantage of multicore systems. They even have core licenses.
3. You don't really need special applications for servers, just separate processes and threads. On most servers there are many services, the OS itself and the server application or applications all of which are separate processes.

4. HPC is really an area where you expect and try to achieve 100% CPU utilization, something you definitely want to avoid with most servers. A server is like a normal auto, whereas HPC is like a race car. You are correct that there is a lot of HPC going on but it is still very small compared to the rest of the market.

5. It would be hard for a company the size of AMD to survive with a processor that only excels in a HPC environment.
0 Votes
+ -
Bashing Point 5
nucrash 15th Aug 2007
That is the bread and butter of some companies.

IBM is all about HPC. Sun, HP, Cray, and others love the idea of clustering together several high performance PC nodes and pushing the limits of computational power.
AMD has other markets, they just aren't performing as well.

Look at Microsoft. They only have two units that ever make profit, but they are enough to float the rest of the company. AMD may be able to push enough of a margin to get R&D for other project.
0 Votes
+ -
Disagree with your disagreement
Robert Crocker 15th Aug 2007
1. I did in fact point out that virtualization is not the same as HPC (I suggested it is probably as deserving of the "niche" title as HPC is.)

2. While they are designed to take advantage of multiple cores they tend to be bound by things other than CPU. Web servers are bound by their network connection. DB by hard disk throughput. Maybe a fancy spam filter could chew up CPU on a mail server...

3. The separate process routine actually will then trip over Intel's Dual-Dual weakness which is memory throughput as all of those processes are trying to work through a single memory bus and will take additional hits if the cache has to go from one set of cores to the other.
Based on that.

4. Virtualization is an attempt to achieve 100% CPU utilization as well. The whole point is that many servers don't get anywhere near 100% so they're wasting cycles.
No, you don't want 100% utilization on Virtualization. Some of the Virtualization vendors might try to sell that, but it's insane because you'll be killing usability for the users.
0 Votes
+ -
Very True
nucrash 15th Aug 2007
Granted, it would be nice to have 4 servers in one server, but at the same time, I would like to have 4 servers on one real server which serves an entirely different function. If I am replacing older NT systems, I could use the NT server as a virtual system in a box that has other systems that require a bit more stability.
0 Votes
+ -
Quite interesting
CFKane 15th Aug 2007
So now Intel's G stepping gives us the Xeon X5365 at 3 GHz? That's interesting, because more than a month ago AMD's benchmarking practices were blatantly deceptive, because:

"Intel’s XEON X5365 3.0 GHz quad-core CPU which shipped back in April was deliberately omitted from these results"

(http://blogs.zdnet.com/Ou/?p=567)
0 Votes
+ -
Don't expect a reply from George.
Bozzer 15th Aug 2007
His contract with WinTel forbids him to do so.
wink
0 Votes
+ -
You apparently don't listen.

I can think of several times where George advised on getting $300 dollar Linux appliances for firewalls and other lesser tasked systems.
0 Votes
+ -
Priceless
CFKane 15th Aug 2007
So he does the same in the other major fanboy arena? Linux is good enough for lesser tasks? That fits so well into the picture that I'm starting to think of him as a caricature and not an actual person wink.
0 Votes
+ -
My Wording not his.
nucrash 15th Aug 2007
Besides. Lesser tasked meaning more specialized. Where Linux can clearly prevail over Windows.

THIS DOES NOT MEAN THE SAME AS LESS IMPORTANT

A Lesser task to me is backups. I don't like doing then, but they are important. But they aren't my primary focus at work. Nor should they every be the primary focus. They should be important though. A firewall should be important. Not my central job, but still important.
Where I said Apple launched a pre-G version of the 3 GHz Clovertown?
0 Votes
+ -
Well done.
Bozzer 15th Aug 2007
I read CFKanes post except that I followed the links back to your Apple article. However I could not resist a little wind up post of my own.

The fact you handled it so well has impressed me and therefore I apologise for it.
0 Votes
+ -
no problem
georgeou 15th Aug 2007
nt
0 Votes
+ -
Where is it?
CFKane 15th Aug 2007
You did not mention they were only shipped to Apple in the article I refered to. There are also no SPEC results published for those machines that AMD could have used. You were accusing them of not using benchmark numbers that were not available at the time the comparison was made and which refered to systems which were not available at the time you made your accusations. The numbers you were using in your article were not for the systems that have been shipping since April. The "shipped since April"-statement I quoted was therefor blatantly deceptive.

And worse: Several commenters informed you about that mistake, but you neither corrected the post (even though you did update it in other parts), nor did you ever admit being (unintentionally, if by mistake) deceptive. That's what you always do and why your blog has nothing to do with journalism.
AMD got caught red handed only last month of passing same old bogus benchmarks even AFTER they yanked it from their own website. They were yelling at me for accusing them of leaking it to the press WHILE at the same time they were actually handing it out to the press. You, AMD, and everyone else accusing me of falsely accusing AMD of leaking it to the press have all been proven wrong. So you take your accusations and can it.

"George also points out that AMD removed the "simulated" benchmark results from their website on July 6th, and promised to post proper benchmark results of their Barcelona processor. However, we received the very same benchmark results just last week. If these results have already been officially junked by AMD, why are we still being served the same poo?"
http://www.techarp.com/showarticle.aspx?artno=424&pgno=6
0 Votes
+ -
Message has been deleted.
TechExec2 Updated - 16th Aug 2007
  • Flagged
0 Votes
+ -
That's not an answer
CFKane 16th Aug 2007
I'm criticizing your habit of twisting facts and leaving out important information. Since you refuse to admit mistakes, I must assume your posts are intentionally deceptive.

To limit it to the single occasion discussed here, I again quote your statement:

"Intel’s XEON X5365 3.0 GHz quad-core CPU which shipped back in April was deliberately omitted from these results"

That undoubtedly suggests that AMD would have had benchmark numbers they could have used in the slides created in April for products that were already shipping at that time. But the facts are: There are still no SPEC*_rate results published for the product that was shipping in April. It was in June that Intel published results for a product which is only available now. With that additional important information, which you chose to ignore, the central claim of that statement - "deliberately omitted" - is revealed as what it is: an outrageous lie.
They sure omitted all through July when they kept pushing those benchmarks. The accusation is that they kept pushing the same benchmarks from April through July. It might have been excusable (though that doesn't excuse simulated scores) back in April, it wasn't excusable in May, June, and July. It certainly wasn't excusable in mid July AFTER they removed the bogus benchmarks weeks earlier.
.
[Update: In my message that was deleted above, I made a humorous analogy comparing George Ou's attacks on AMD to a dog barking up a tree without a cat. Apparently that is considered over the line. Let's try it again without that joke...]



George Ou says...

"...AMD got caught red handed only last month of passing same old bogus benchmarks even AFTER they yanked it from their own website. They were yelling at me for accusing them of leaking it to the press WHILE at the same time they were actually handing it out to the press. You, AMD, and everyone else accusing me of falsely accusing AMD of leaking it to the press have all been proven wrong. So you take your accusations and can it..."

Say it with me slowly: There is nothing wrong with AMD or Intel offering estimated benchmark results for an unreleased processor so long as they are clearly labelled as such. AMD clearly labelled them and did nothing wrong.

For George's penance, he must go to the top of the highest mountain nearby where he lives, sit in the lotus position, and repeat the above phrase until he stops lying about AMD!

I already thoroughly debunked George's rabid blog-a-thon attack on AMD (1)(2). But, I never accused him of leaking the AMD estimated benchmarks to the press just so he could rip them to shreds over it in the blogosphere. But, since he is so willing to tell provenly-false AMD stories over and over, I cannot discount the possibility he is behind it all.



George Ou cites "Tech ARP" story THAT HE FED TO THEM

"...George Ou of ZDNet has just alerted us to the fact that..." -- "Tech ARP" (3)

Shameless George Ou FED the story to "Tech ARP" and then cites it here as some kind of support for his own rabid attacks on AMD. George Ou says it's true because Tech ARP tells you it's true because George Ou told them so??? One question: What??!!! shocked laugh

Here's another quote from "Tech ARP" that George does not cite since they show AMD acting responsibly. AMD's estimated benchmark results originally released under NDA LAST FEBRUARY that George Ou is going ballistic over will soon be replaced with more current results (and probably actual SPEC benchmark results).

From "Tech ARP":
Note : AMD has requested that we remove the benchmark results as they are outdated and inaccurate. They have promised us updated results soon, so keep an eye out for these results. With the launch looming so close, we think they will be doing everything they can to have the results out ASAP. (3)



------------------------------------------

(1) Debunking George Ou's rabid attack on AMD
http://talkback.zdnet.com/5208-10533-0.html?forumID=1&threadID=35931&messageID=661568&start=-1

(2) Baloney! This post FURTHER PROVES George Ou is unfair to AMD!
http://talkback.zdnet.com/5208-10533-0.html?forumID=1&threadID=36193&messageID=665742&start=-1

(3) AMD Quad Core Opteron (Barcelona) Technology Report (Tech ARP)
http://www.techarp.com/showarticle.aspx?artno=424&pgno=6
0 Votes
+ -
Unless AMD has a rabbit in their hat
No_Ax_to_Grind 15th Aug 2007
They are in deep, deep trouble.
Apparently it somewhere because if they still are trying to market "Phenom" as a Core 2 Killer.
0 Votes
+ -
Memory and CPU saturation
DevGuy_z 15th Aug 2007
George, there was a review that I read the other day on these two chips (sorry I don't have the link handy) and the general performance for these two chips was outstanding when compared to previous Intel editions. Generally the review favored the two new chips because of the leap in performance.

But the article went on to say that the chips did not appear to scale very well and that indications were that memory access was being a significant bottleneck.

Basically as the number of processors climb or if the application is heavily memory intensive, then AMD will likely win out. So for big iron or HPC AMD still really looks good.

But if I was making the ultimate gaming machine or setting up a small 4 way server, Intel it would be.

Both Intel and AMD are working on additional virtualization extensions.
0 Votes
+ -
Its looking bad for AMD
Stuka 15th Aug 2007
I was a really big fan of AMD for many years as thy beat the ever living snot out of the P4 clock per clock. And they used quite a bit less power. But it seems AD has woke up the giant that is Intel and it may be to AMD's doom. I love my C2D processors, fast, good price, and run cool.

I can only hope that AMD pulls a rabbit out of its hat. What I fear also is if AMD dies, so will ATI.
0 Votes
+ -
ATi wouldn't be missed...
Linux User 147560 15th Aug 2007
nVidia is a much better card in my opinion... as for AMD "dying", I don't think so. It's like any other time in the CPU battle, one will falter then eventually find it's course again and surge ahead. Look at what AMD has done so far and where they came from. devil
0 Votes
+ -
The story starts off with the usual skewed perspective. Official AMD announcements of low power parts going to be launched together with standard parts are cast aside in favor of leaked channel information. The fact that the sources said nothing about low power parts is also reason enough to dismiss them entirely and compare AMD's upcoming standard parts to Intel's premium low voltage Xeon. It's also quite obvious that the 2 GHz Barcelona will compete directly with the 2 GHz quad core Xeon, despite the absence of any benchmarks and the very unusual new pricing policy on AMD's side this would suggest. In the end, wherever AMD's product may have an advantage, it must be a niche market...

But that didn't quite satisfy you and you probably realized how thin your story was so far. So you ventured into new territory: the financial aspects. This is where it really gets funny. While HPC - and this probably covers everything including applications where AMD's advantages in scaling are already accountable today - is a niche market, the higher clocked parts are where the money is and AMD's only chance is to ramp up clock frequencies fast. Let me tell you something: AMD's financial situation is much more complicated than that and the product portfolio is just one aspect of it. Margins are generally not modest in the server space - they're great, and that's why AMD launches the server parts first. The most important aspect of Barcelona is that it is a quad core part, which AMD does not have at all at the moment. No market share, zero revenue. It also happens to be a product which flourishes in exactly that "niche" which you are trying to dismiss as insignificant. Really negligible is the 10-20% performance increase you may get by using a higher clocked part from a competitor that works within the same thermal envelope, when the main reason for the purchase is the 70-80% performance increase you get from doubling the number of cores in combination with suitable applications. The impact Barcelona will have on AMD's future financial results will be nothing but positive. It will not miraculously pull them out of debt, but neither would a 3 GHz version at launch. The conclusion of your post is just a testament of your ignorance regarding the complexity of such questions. Stick to your benchmarks. Those are simple raw numbers you cannot go wrong with.
0 Votes
+ -
The most important aspect of Barcelona
thetruthhurts 15th Aug 2007
"The most important aspect of Barcelona is that it is a quad core part, which AMD does not have at all at the moment. No market share, zero revenue."

Barcelona is the most un-anticipated event of the summer...fall...whatever. Already destined to go into Pat Gelsinger's 'dustbin of obsolete technology.'

QED
.
This is another George Ou blog, straight from an Intel press release (or was it the other way? laugh ).

Since George Ou puts so much faith and trust in press releases, I thought I would cite some AMD news posts balance things out.

New Intel Server Processors Provide Ultimate Choice in Speed and Energy Efficiency (George's source for this blog post)
http://www.intel.com/pressroom/archive/releases/20070813comp.htm

AMD Technology Analyst Day
http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_15161,00.html

AMD Technology Analyst Day -- News in Brief
http://www.amd.com/us-en/assets/content_type/DownloadableAssets/AMD_Technology_Analyst_Day_News_Summary_FINAL.pdf

AMD to Ship Industry?s First Native x86 Quad-Core Processors In August
http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~118193,00.html

AMD Announces Broad Software Ecosystem Support for Industry?s First x86 Native Quad-Core Solution
http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~118786,00.html

European Commission Charges Intel with Antitrust Violations
http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543_15008~118586,00.html

New Economic Study Finds Intel Extracted Monopoly Profits of $60 Billion Since 1996
http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~118720,00.html
0 Votes
+ -
You are funny, really...
tek_heretik 15th Aug 2007
George is a little full of himself, lets hope your post deflated his ego a little. Lmao. XD
0 Votes
+ -
Thanks! Glad you enjoyed it! (NT)
TechExec2 16th Aug 2007
.
0 Votes
+ -
...you come here in an attempt to blog desperate AMD marketing spew. Guess what? None of it has worked.

AMD bleeds as a consequence of getting fat and happy with K8, just like Intel did with P4. Now, AMD shall suffer the same consequence as Intel, except to a much greater degree because this time Intel owns AMD on performance, power/performance, and price/performance.

MC Hammer says: "Can't touch this"
0 Votes
+ -
hmm
Badgered 16th Aug 2007
except to a much greater degree because this time Intel owns AMD on performance, power/performance, and price/performance.

I agree on overall performance, and power consumtion. But not so much on price/performance.
0 Votes
+ -
ROFLMAO
thetruthhurts 17th Aug 2007
Amd OWNED by Intel
0 Votes
+ -
Q6600 quad is best bang for buck
georgeou 18th Aug 2007
E6600 is not worth it because the Stepping G based E6550 is a better buy. Even the E6750 2.66 with FSB1333 is cheaper than E6600.

If you got more budget, go for a quad core at $280 for the Q6600.

Join the conversation!

Formatting +
BB Codes - Note: HTML is not supported in forums
  • [b] Bold [/b]
  • [i] Italic [/i]
  • [u] Underline [/u]
  • [s] Strikethrough [/s]
  • [q] "Quote" [/q]
  • [ol][*] 1. Ordered List [/ol]
  • [ul][*] · Unordered List [/ul]
  • [pre] Preformat [/pre]
  • [quote] "Blockquote" [/quote]
ie8 fix
Click Here
ie8 fix

The best of ZDNet, delivered

ZDNet Newsletters

Get the best of ZDNet delivered straight to your inbox

Facebook Activity

White Papers, Webcasts, & Resources
ie8 fix
ie8 fix