AMD the saviour
posted by Sharikou, Ph. D. @ 11/21/2006 06:09:00 PM
Journal of Pervasive 64 bit Computing
Main Blog Page
Site Feed:
sharikou.blogspot.com/atom.xml
Blog Roll
Hot pages
sharikou.blogspot.com/atom.xml
Blog Roll
Hot pages
- Price War Strategic Analysis
- AMD and Intel Performance
- Intel's false performance claims
- AMD's Architecture Lead Persists
- AMD and Intel Performance
- AMD and Intel Architecture
- Bensley/Woodcrest and Athlon MP
- Bensley/Woodcrest is 2P only
- 64bit CORE Performance!
- Conroe Busted
- Capacity: AMD vs Intel
- Conroe's Impact
- AMD and Intel in China
- A Beautiful Picture!
- Temperature and System Reliability
- Google and AMD
- AMD's Grand Masters
- Intel's Predicament
- AMD and Dell
- NGMA Core and Colwell 6th Gen Arch
- AMD and Intel History
- The Tale of Intel and Anand
Analysis on IT trends and competitive strategies, with emphasis on micro processors, computer systems and networks. Based on latest news, backed up with real data, this site intends to provide a true and realtime picture of the fast changing IT landscape. This journal strives to be accurate on facts and sharp on criticisms. You may email your opinion to sharikou@yahoo.com or post comments here, be cool and intelligent.
About Me
- Name: Sharikou, Ph. D.
Freelance journalist on IT matters. Some of my writings have been published on online IT journals. Any original content on this journal is Copyrighted, but it's free for non-commercial use. Any Trademarks used on this site belong to their respective owners. Some of the pictures are links. If there is any issue with the content of this site, please email sharikou@yahoo.com .
Previous Posts
- How to analyse the Intel vs AMD competitive situation
- Intel Clovertown quad fragged by Opteron dual
- Hector and the one
- Turion 64 for high class people
- Constitutionality of $750 price tag for a song
- Intel graphics not good for kids games
- AMD shifting production to mobile
- vPro rejected at Round Rock
- DELL to be 50% AMD soon
- Charlie at INQ reckons Intel in deep dodo
44 Comments:
Good for Dell. Do you know what this means? Intel’s technical iron grip on the world seems to be waning as she is robbed of her underworld sales tactics and uninspired engineering. AMD’s technical advantages continue to throw salt on the old guards wounds with regularity. Due to constant blood letting of market share the whores grip is loosening. Amazing Intel really wanted this war, now the slut is in serious pain.
OH, Yeh…. I like this part “AMD Now Shipping Across Full Product Line”
That’s gotta hurt!
I’ve heard rumors APPLE is next!
Do you have the special edition of the numbers that give a break down of AMD vs Intel related income?
I like the sentence
"In the Americas, revenue was $9.2 billion on unit growth of negative 4 percent"
negative growth :)
Looks like they now offer more choice with AMD and better customer service, so they should be doing better. They should have done this a long time ago though.
http://theinquirer.net/default.aspx?article=35902
Hmmm...... Interesting!
In the third fiscal quarter of 2006 Dell claimed only 17.18 percent of the PC market with approximately 9,803,000 units sold.
It would be nice to see how many of those were AMD before starting starting to claim that AMD saved Dell.
How could AMD "save" Dell when Dell has several times higher revenue than AMD? Does Dell earn insane profits from every AMD CPU it sells? If not then how much did AMD CPU's increase it's revenue and profits?
I’ve heard rumors APPLE is next!
makes lot of sense, one of excuses for not going apple is lack of hardware choice,
Apple can have its own line of AMD based products, offering more choice to potenial customers
Do you have any info abt revenue form AMD based hardware ?
Yes APPLE is next. The deal is done and includes all the CPU line of AMD which is very strange because I think Intel helped APPLE in migrating from IBM PowerPC to x86. Maybe APPLE is not happy with something. Or Intel is charging them too much for the CPUs.
I think AMD will no have capacity to deal with all this, they market share will ramp up especially when Intel deliver the complete WEAK line of Intel Core 2 CPUs.
You are cute.
Except of course that Dell shipped hardly any AMD product - if any at all - during this period.
It’s good to see that Dell has found a new partner who it can suck dry having been denied Intel’s pricing discount.
I still don’t see how this is good for AMD who now has become Dell’s parasitic host.
Dell’s negative growth and “focus on higher margins” shows that it can no longer compete on pricing alone. Having no history of good technological advantage, poor customer service and poor quality, it is perfect timing that AMD is now on board in this sinking ship. Second rate Athlons and Opterons behind 40-100% in performance, price/performance and performance/watt. Yes, a match made in heaven.
I’m sure Intel is now laughing its way out of the margin sucking deal it had with Dell and try and sell its unchallenged products to much higher paying customers. When you have the best product that everyone wants, you control who the customer you want to have the best margins.
Obvious Prediction
Due to the worlds shift to 64-bit and AMD’s advantage clock for clock, graphics and power to watts Intel has hit a wall. I predict Paul will attempt suicide by 4Q-07 and many Intel fanboys will follow in his lead. I’m writing this post in advance to say, we’ll miss you guys.
Merry Christmas Dell.
What Intel-ies need to know?
Intel has put most of its stock in MOORE’s LAW to seek advantage, where’ as AMD has chosen design over just transistor size. AMD 90nm silicon is completive to Intel’s 45nm and in most cases out performs its smaller brethren. How is this possible? Innovative designed architecture and open platform will lead future advancements and Moore’s law is reaching the end of its cycle.
Ever CPU Intel sells this year just prolongs the inevitable conclusion. Without a major break-thru Intel’s near future looks bleak.
AMD has announced it will give Intel fanboys amnesty. You must first throw your Intel based machine over the nearest bridge and come to the light. AMD has been advised not to take prisoners’ after the release of Vista.
Of yeah...That's why DELL's income was boosted by Woodcrest sales.
AMD is so money.
So money they don't even know it!
LOL another turd from the pretender!
Remember when Dell grew faster then anyone in the industry, remember when they were no 1.
Now they got AMD and what has happened. Slowere growth, lower profits and no longer #1. HP rules..
Dell is broken and getting worst. Crappy and ugly machines with 2nd class slow CPUs.. yup what a recipe for success.
You're right on there Sharikou. You get your PhD at an interent university. Your logic is so ridiculous...
Ha ha ..it is funny to read the comments of some Intel fan boys ..who are still in a state of denial :-))
A dumb idiot wrote:
"Intel has put most of its stock in MOORE’s LAW to seek advantage, where’ as AMD has chosen design over just transistor size. AMD 90nm silicon is completive to Intel’s 45nm and in most cases out performs its smaller brethren. How is this possible? Innovative designed architecture and open platform will lead future advancements and Moore’s law is reaching the end of its cycle."
Hahaha, nice joke! Intel doesn't even have 45nm out yet, and won't for 9 - 12 months.
How can AMD 90nm 'outperform' Intel 45nm if it's not even released?
If you're talking about Intel 65nm, then I don't know what you've been smoking, but AMD's 90nm K8s sure as hell can't keep up with Intel 65nm C2Ds.
Nice try though, fanboy.
Just don't commit suicide when Intel doesn't BK, it's not worth killing yourself over.
Yes APPLE is next. The deal is done and includes all the CPU line of AMD which is very strange because I think Intel helped APPLE in migrating from IBM PowerPC to x86. Maybe APPLE is not happy with something. Or Intel is charging them too much for the CPUs.
//////////////////////
just a question of good biz practice..having two vendors just stops an exclusive from getting too fat and too happy
paarl
For the Turkey's here is a fork for you to noodel.
1) In the past AMD beat INTEL with one generation inferior silicon. The cost was reduced output as die was much larger but ran faster. They days of Opertron crushing Netbust is over... Yes Craig and Paul fucked the company with their stupid failures in communication, DSP, frequency, and Itanium.
2) No look at the landscape.. INTEL core2 kicks AMD's ass by a good 20-30%. I expect Barcelona to level the playing field. So for 2007 INTEL and AMD will be close. The difference will be INTEL has 4 65nm factories ramped and partially deprecicated while AMD sucks cash and depreciation to ramp 65nm.
Finally thru NO management of technology Paul Otellewennie got luck. Merom has been in the making for a couple years. Its not like Paul woke up and saw the light . Nor is AMD going to be able to look at COre2 and fix something on Barcelona.
The bottom line is both companies will be in a price war where the consumer wins. AMD will suck wind as its inefirior manufacturing and delayed 65nm comes to haunt him. Finally the ATI purchase will distract their very best people and cause paralysis across the company.
In the end AMD falls behind and INTEL continues to make billions. Paul will claim credit but it'll be because Hector screwed up. Not different from the past 4 years where AMD's apparent success had more do do with AMD not fucking up and Paul and Craig being absoluate idiots.. making every wrong technical call for the past 5 years.
Hope Paul and Hector are enjoying their Turkeys because both are real Turkeys
Dell switches to AMD.
Intel boy says, Dell sucked anyway.
If Apple switches to AMD
Intel boy says, Apple sucks anyway.
Google switches to AMD
Intel boy says, Google sucks anyway.
IBM, Sun switch to AMD
Intel boy says, who needs them
AMD buys ATI
Intel boys says, ATI sucks
AMD becomes the super computer CPU of choice
Intel boy says, who needs it anyway
AMD continues to eat at Intel’s market share
Intel boy says, were still the biggest anyway
Intel boy eats crow instead of turkey on Thanksgiving and he says, “I love crow anyway”.
NOTE: Intel boy doesn’t see what is happening, he wouldn’t understand anyway.
"If you're talking about Intel 65nm, then I don't know what you've been smoking"
I stand corrected, 65nm.. Thank You
netrama said...
“Ha ha ..it is funny to read the comments of some Intel fan boys ..who are still in a state of denial :-)) “
I agree and they’re getting worse as time and technology roles on.
Happy Thanksgiving to all, even those in denial.
"Remember when Dell grew faster then anyone in the industry, remember when they were no 1.
Now they got AMD and what has happened. Slowere growth, lower profits and no longer #1. HP rules.."
Intel fanboys, at lease those commenting on this blog, have no logic. They can't get their argument's causality right. Maybe they think Intel ought to be able to travel faster than light?
Dell had growth slowdown before it chosed AMD. HP beated Dell partly because it started to use AMD CPUs long before its (previously) bigger foe.
"Except of course that Dell shipped hardly any AMD product - if any at all - during this period."
Yeah right.. all those AMD's capacity people claim to have been taken by Dell were just vaporized at Dell's factories...
"If you're talking about Intel 65nm, then I don't know what you've been smoking, but AMD's 90nm K8s sure as hell can't keep up with Intel 65nm C2Ds."
He is not smoking anything, because AMD's 90nm outperforms or "keeps up with" Intel's 65nm on pretty much everything other than SSE and SuperPi.
And if you're talking about 65nm Netburst...
What drives CPU performance is not (just) Moore's Law, but more importantly the architecture. Even Intel engineers will tell you that.
Edward - you really should not talk about Si; you have demonstrated time and again you lack insight in this area.... You are blending DESIGN and Si process technology.
As an example how does AMD's 65nm 3800 compare to AMD's 90nm 3800? Does it kick the ass of the 90nm ? Should it?
It really depends on whether the product is a shrink, or is designed to take advantage of the newer node.
And for the anonymous idiot who said AMD's 90nm process is competitive with Intel's 45nm (65nm); this is clearly not the case - it is a case of product differences not Si technology.
Oh, and Moore's law enables more transistors which DOES enable design (and power consumption reductions).... Design of course is equally important but best design in the world on 1um technology would not keep up with todays 65nm technology and would cost the end consumer a lot more!
"The company suspended its share repurchase program in mid-September and, therefore, only spent $335 million to repurchase 15 million shares."
Wonder how much this impacted EPS? Unfortunately we don't know because Dell couldn't even meet it's commit to post official Q3 numbers!
Also interesting to note EPS last year for this quarter was $0.25; that's a good -25% boost YoY!
As always, Shari-kook, your financial acumen is truly astonishing...
Who has the best performance?
Who has the best performance/watt
Who has the most factories and MS
Who leads in mobile the fastest growing market
WHo has the biggest bankaccount to for R&D.
Who helped Apple launch their big resurgence in PCs.
Who the fuck cares if Dell is using AMD.. Dell used to be the hot PC company. Now the hottest technology company is Apple and they are all INTEL inside.
Too bad loser
"Who has the best performance?"
Performance of what? Supercomputers? Servers? Or SuperPi on your desktop?
"Who has the best performance/watt"
65W typical TDP certainly looks like a great achievement over 65W max TDP.
"Who has the most factories and MS"
monopoly...
"Who leads in mobile the fastest growing market"
monopoly...
"WHo has the biggest bankaccount to for R&D."
R&D is great, but unfortunately the bankaccount spends even more for marketing and monopolistic strategies.
"Who helped Apple launch their big resurgence in PCs."
"Helped"?
"Edward - you really should not talk about Si; you have demonstrated time and again you lack insight in this area...."
And what insignt do you have better in this area? All I've said were correct and more, only that some people have problem to grasp the big picture.
"You are blending DESIGN and Si process technology."
Nop, it's you who can't read. I'm blending nothing. I said clearly, "What drives CPU performance is not (just) Moore's Law, but more importantly the architecture." That clearly distinguished design (architecture) from process technology (Moore's law).
Or let me spell it out for your undergraduate or less brain: There are two major factors at play for CPU performance: #1, design; #2, process technology. A better technology can be beaten by a less one because the former's design is inferior.
"As an example how does AMD's 65nm 3800 compare to AMD's 90nm 3800? Does it kick the ass of the 90nm ? Should it?"
Did I say it should? Doesn't it prove my point that design is more important for CPU performance than process technology?
"It really depends on whether the product is a shrink, or is designed to take advantage of the newer node."
It really depends on a lot more other things. A shrink may increase speed, may reduce power, and may economically allow more transistors. However, if your basic design sucks, no shrink can save you. That is why people said AMD's K8 @90nm beated Intel's P4 @65nm with a better design.
"And for the anonymous idiot who said AMD's 90nm process is competitive with Intel's 45nm (65nm); this is clearly not the case - it is a case of product differences not Si technology."
And for the anonymous idiot who is you, that AMD's 90nm and Intel's 65nm really mean the products at those feature size. You really can't read, can you?
"Design of course is equally important but best design in the world on 1um technology would not keep up with todays 65nm technology and would cost the end consumer a lot more!"
I don't think it's even feasible to make 64K L1 cache at .8um. At that point the design is obviously impossible.
But such an argument is moot. It's like saying "it really doesn't matter how smart you invest in stocks if you only have $100" - and what do you invest with that? Buy your kids a meal? Most stock brokers will tell you that having 50% more money doesn't give you as much advantage as investing smarter; but if you could have 16x money... of course you can do a LOT more.
edward said...
"Performance of what? Supercomputers? Servers? Or SuperPi on your desktop?"
Supercomputers with co-processors - AMD
Supercomputers without co-processors - IBM
Servers 4P and up (FP) - AMD
Servers 4P and up (INT) - Intel
Servers 2P and down (FP) - AMD
Servers 2P and down (INT) - Intel
Desktop 1P (FP & INT) - Intel
Mobile - Intel
It's all relative except Desktop and Mobile.
"65W typical TDP certainly looks like a great achievement over 65W max TDP.
When comparing it to AMD's future 65nm 65W TDP no.
At the same time he made a good point about cache being a large portion of the power for Intel.
What do you think AMD's TDP for K8L (Rev H) will be with 4MB of cache?
"Who has the most factories and MS"
monopoly...
"Who leads in mobile the fastest growing market"
monopoly...
Intel.
Cant just answer the questions huh Ed?
Until the trial is over that is speculation, and only showing how pro AMD you are.
Innocent until proven guilty.
Now before you go on a rampage...
Here is my opinion, I would not be supprised if Intel did throw money around, but people like you and Sharikou are ready to hang them before the trial is over.
"R&D is great, but unfortunately the bankaccount spends even more for marketing and monopolistic strategies."
Intel has a larger R&D than AMD, but is it larger than IBM?
"Helped"?
What is so hard about this question Edward?
It would seem very simple, Intel has better performing processors for Mobile and Desktop (the majority of Apple sales), and has a huge marketing department.
There has been an array of Apple commercials lately, and they almost all have the Intel addon at the end.
Who do you think pays the majority of the cost of those commercials?
Hence Intel has helped Apple's resurgance.
"Intel.
Cant just answer the questions huh Ed?
"
enumae, I'd really like to ask you not to respond dumb comments after my statements, but I guess it's your freedom...
"Having the most factories" and "leading in a market" say little about how good the products are. The straight answer as the one you're only capable of is meaningless; that's why I did not answer it that way. But I guess it's beyond your comprehension...
Do you really think Microsoft's operating systems were best in the world since Windows 95, or Toyota and/or GM make the best cars, or Intel makes the best CPU?
OMG, you do, don't you?
Then please know that I'm tired of responding to you (or anyone else the like) and your lack of thinking and logic.
Edward, I am sorry you can not counter a point I made, but you really need to grow up, and quite crying about my lack of logic due to your lack of comprehension because it will only lead you to further disappointment later in life, though you might as well get used to it and quit making these ridiculous post.
Some idiot thought his points were correct.
- Single-core INT and FP performance numbers are of little meaning for server 2P (4 cores) and up. Core 2 performs well only on simple benchmarks at those areas.
- AMD already has 65W CPUs out. The 65W AX2 4600+ is 20% cheaper than the C2D E6400, with better performance except SSE2 and SuperPi.
- Rev.H has 2MB cache; plus cache is the relatively cool part of the CPU.
How many errors can one make in one post? Enumae, you just did the impossible; you just lowered your own standards.
"- Rev.H has 2MB cache; plus cache is the relatively cool part of the CPU."
Oops... I mean 2MB (shared) L3 cache.
Edward said...
"- Single-core INT and FP performance numbers are of little meaning for server 2P (4 cores) and up."
Did you even read my post?
I stated very clearly that in 2P Core 2 has a INT advantage, and K8 has an FP advantage.
Wasn't the original post in which you replied in the general context of Intel and AMD, and if so that is why I had done a break down.
"AMD already has 65W CPUs out."
Did I deny this?
"The 65W AX2 4600+ is 20% cheaper than the C2D E6400, with better performance except SSE2 and SuperPi."
Newegg doesn't have them, but the price would be the same as an AMD Athlon 64 X2 4600+, and the price was $243.00 retail.
The E6400 was $218.50 retail, that would make the E6400 10% cheaper.
Here is a review you may or may not like (source), but the lower clocked E6400 is right there competing with the Athlon 64 X2 4600+EE sometimes better, sometimes not, but it is not limited to SSE2 or Super Pi (unless gaming is SSE2, I am not a programer)...
-link-
"Rev.H has 2MB cache"
It has 512KB L2 per core plus a 2MB L3 shared, 4 cores (2MB) plus a 2MB shared L3 = 4MB cache.
"plus cache is the relatively cool part of the CPU."
I had always thought it was the hotter part, if it is not then that is my error.
"How many errors can one make in one post? Enumae, you just did the impossible; you just lowered your own standards."
Edward at least I research what I am talking about maybe you should take some notes?
Lets review...
Was the 4600+EE cheaper... No.
Does the 4600+EE perform better... No.
Does the Rev H Quad core have 4MB cache... Yes.
You are incapable of debate, and your immaturity is showing, grow up.
PS: Wasn't it you who had stated that when one resorts to name calling he has lost the argument?
"I stated very clearly that in 2P Core 2 has a INT advantage, and K8 has an FP advantage."
Right, talking about using a single value of (single-threaded?) benchmark to compare two machines with 4 cores, where one is SMP, while the other is NUMA.
You have no clue, do you?
"Did I deny this?"
Yes you did. You claimed to compare Intel's Core 2 with AMD's future 65nm 65W CPU. If you did not deny AMD's current 65W CPUs, what's that "future" for?
"The E6400 was $218.50 retail, that would make the E6400 10% cheaper."
Oops... my bad. I didn't know it dropped so fast in just 3 months time. Guess most people rather opt for E6600?
"Here is a review you may or may not like (source), but the lower clocked E6400 is right there competing with the Athlon 64 X2 4600+EE sometimes better, sometimes not"
I actually like that review, even though it's tailored to market E6400, but it at least proves that gaming performance is not CPU bound.
"It has 512KB L2 per core plus a 2MB L3 shared, 4 cores (2MB) plus a 2MB shared L3 = 4MB cache."
It has 2MB extra cache than 4 K8 cores. That'd be the only source of additional power usage you talked about. Unless you lost track of what you were talking about?
"I had always thought it was the hotter part, if it is not then that is my error."
Thank you.
"Edward at least I research what I am talking about maybe you should take some notes?"
Yes, I noted that I lost track of C2D pricing 2 months ago.
I also noted that, 1. you do not understand CPU performance (please stop using marketing websites for performance purposes), 2. you didn't know what you were talking about when relating the "4MB" cache with power consumption.
"You are incapable of debate, and your immaturity is showing, grow up."
No, I'm incapable of debate with you, and I'm sorry to say that you don't even think/know better than someone "immature" like me.
I don't know why I'm wasting time to argue with you who can only grab marketing info off the web. You don't listen to truth, do you? It was exaggerated to say X2 4600 performs (absolutely) better than E6400, because, as I've always said, it depends on the workload. But the point is if one is not running SSE2 nor apps like SuperPi (with complex pointer operations), X2 4600 will serve him better than E6400. Of course those Intel-marketing websites will tell you differently, but please, please, get two boxes and do some real benches yourself, okay?
"PS: Wasn't it you who had stated that when one resorts to name calling he has lost the argument?"
Nop, I never said that. But I do tell you that I call people idiots for good reasons. You have no reason to believe that I was calling you, though. ;-)
(unless gaming is SSE2, I am not a programer)
That is what most of it is. That is why Conroes look good on gaming benchmarks. But I still would not get a hot chip for my gaming box. AMD 3800+ with a good graphics card will more than suffice.
edward said...
"Yes you did. You claimed to compare Intel's Core 2 with AMD's future 65nm 65W CPU. If you did not deny AMD's current 65W CPUs, what's that "future" for?"
Here is the context in which I had answered...
Anonymous said..."Who has the best performance/watt"
You had said... "65W typical TDP certainly looks like a great achievement over 65W max TDP."
So, I said "When comparing it to AMD's future 65nm 65W TDP no.", I had interpreted your comment as sarcastic becuase AMD has 65W max TDP, I actually agreed with you that Intel's 65W typical was not as good as AMD's future 65W average, as opposed to there current selective 65W parts.
If it did not come across that way, sorry for the confusion.
"I actually like that review, even though it's tailored to market E6400, but it at least proves that gaming performance is not CPU bound."
I only look at the graphs and test setup, I am not interested in the testers opinion.
I did not link it because it showed that the E6400 was faster than the 4600+EE, but because I searched Google and it was the first one there, and like I said, looking at the graphs they looked about equal.
"It has 2MB extra cache than 4 K8 cores. That'd be the only source of additional power usage you talked about. Unless you lost track of what you were talking about?"
Edward I am trying to be patient with you, but your little immature comments are getting old. You had said that Rev H had 2MB cache, now suddenly I don't know what I am talking about... WTF!!! I had proven you wrong, so what are you talking about now?
Or are you now talking about my question to you about the TDP of Rev H, which when made was under the assumption that the cache was one of the warmer spots on the chip?
"Thank you."
Could you link to an article about the cache being the cooler part of the chip, thanks?
"I also noted that, 1. you do not understand CPU performance (please stop using marketing websites for performance purposes),"
Its just a review, show me a better one instead of bitching about it, like I said it was not selective.
"2. you didn't know what you were talking about when relating the "4MB" cache with power consumption."
Wow, you can say what I had already admitted to, and again it was originally in the form of a question.
"You don't listen to truth, do you?"
Show me the truth in published documents, or some form other than you telling me, I have said before I am not narrow minded, and I enjoy debate, just show me where the information is.
"It was exaggerated to say X2 4600 performs (absolutely) better than E6400, because, as I've always said, it depends on the workload."
Understood.
"But the point is if one is not running SSE2 nor apps like SuperPi (with complex pointer operations), X2 4600 will serve him better than E6400."
Understood.
"Of course those Intel-marketing websites will tell you differently, but please, please, get two boxes and do some real benches yourself, okay?"
If I had the money I would like to, I know that you use both, is there a way you could document what you have seen and done when comparing the two.
"Nop, I never said that. But I do tell you that I call people idiots for good reasons. You have no reason to believe that I was calling you, though. ;-)"
I had thought it was you, maybe it was on MMM site, oh well.
Edward I do not want to argue with you, but would rather debate, now if you want to resort to name calling, then so be it.
Post a Comment
<< Home