Charlie lost a few IQ points reading Anand
posted by Sharikou, Ph. D. @ 12/21/2006 08:54:00 AM
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
- Solaris 10 fails to generate interest
- Intel lost another 5% of market share
- My predictions for DELL all come true
- AMD capacity estimate again
- Intel will have to copy AMD again
- Anand made another negative scaling discovery
- Intel stuck with the hot and slow image
- Patty sounds desperate
- AMD's channel supply problem finally solved
- AMD X2 4600+ will replace X2 3800+
33 Comments:
It's IMO one of Charlie's reasonable article(s) - that is if there are others. ;-p
I applaud Charlie for pointing out the obvious ignorance and bias of the Intel fanboys.
I applaud Charlie for pointing out the obvious ignorance and bias of the Intel fanboys.
Most people today know Pentium is a dead brand and crap chip.In a way, AMD has achieved brand parity with Intel.
No wonder, charlie had to disguise himself as a bunny, in a bunny suit to get into the last IDF
Wow, indeed on of Charlie's most sane posts ever,(if not the most sane)he hit the situation with the hammer right on the head. Far too many intel fanboys expected too much and assumed far too many things that AMD never said about with the debut of the AM2 socket/DDR2 support and thier 65nm, 300mm production process.
Sure thier OC ability isn't great right now compared to what C2D is capable of right now, but that doesn't matter to those who run the show. The Athlon64 X2 has a 10% to 15% performance deficit average clock vs. clock against C2D, no news there. What is important to AMD right now is yields, production output and technology support for the future. AM2, Torrenza, Fusion, HT 3.0, APM and FAB 36 have addressed all those issues. In return, they have OEM wins with some of the biggest in the business, namely Dell, HP and so on, this is what truly matters to a company. Not some minority group of overclockers and enthusiasts who consitite about %1-%5 of thier total market.
In time,(approx. 6 months) AMD's bump in performance will come with K8L and IMHO it will overtake C2D due to having a far more efficent IO,(HT 3.0) among other things in store. In the meantime, they aren't hurting as much as some fanboys would hope. The bottom line does not rest with the bleeding edge of technology. It's a feather in one's cap for sure for marketing, but Joe Six-pack could care less, much less know anything about it. AMD will do just fine come next year, especially as the momentum shifts in thier direction.
So you agree with Anand and Charlie that AMD's 65nm sucks.
Same situation as core 2 release.
AMD made no performance commits on 65nm - AMD fans than said it will OC much better and outperform 90nm...not AMD's fault...
Intel said Core 2 will perform 40% better than previous Intel product - Intel fans and press re-interperted this to mean 40% better than AMD...not Intel's fault. yet when Core 2 was not 40% better than AMD, you were all over Intel...
"In a way AMD has achieved brand parity"
WTF? Intel is selling chips (P4) that underperfomrm lower priced chips (Athlon) - you can't do that with brand parity!
Intel is selling chips (P4) that underperfomrm lower priced chips (Athlon)
95% of people don't know anything about Core 2 Duo. Even my friends in electrical engineering don't know. But they are starting to know AMD64.
"In a way, AMD has achieved brand parity with Intel."
Now that we know Dr. Shari-kook is a wafer start, yield expert; he is now also a marketing and brand expert?
The gold standard for brand awareness and value is the Interbrand/BusinessWeek study each year. Intel is the #5 most valuable brand in the ENTIRE WORLD. AMD didn't make the top 100.
http://www.ourfishbowl.com/images/surveys/BGB06Report_072706.pdf
A quick summary of what this means - if you threw away every fab, office building, and copier, just the name "Intel" is worth over $32 billion.
I guess it matters what the meaning of the word "is" is...
Brand value link didn't come through - click here:
http://www.interbrand.com/best_brands_2006.asp
and then click on the "Best Global Brands 2006 Report" link
I think AMD's engineers lost a few IQ points by raising the L2 latency for 65nm, making them slower than 90nm.
http://techreport.com/ja.zz?comments=11486
AMD frags itself.
Brand value link didn't come through - click here:
http://www.interbrand.com/best_brands_2006.asp
I heard Pentium is a 30 billion brand, where is it now?
I think AMD's engineers lost a few IQ points by raising the L2 latency for 65nm, making them slower than 90nm.
These minor things are inconsequential. Right now, Intel enjoys a 10% lead with Core 2 Duo, that only emphasizes the fact AMD frags 75% of Intel's chips. By the time Intel ramps up Core 2 Duo to about 50%, AMD will make a quantum leap in performance.
Sharikou said...
"I heard Pentium is a 30 billion brand, where is it now?"
Are you really trying to be right?
Coca Cola... they know Coke, Diet Coke etc...
Microsoft... Windows 98, 2000, XP and soon to be Vista
IBM... GE... Intel
Yes AMD is growing, but it is no where near Intel on brand recognition (brand parity), accept defeat on this one.
Sharikou said...
"These minor things are inconsequential."
If it was Intel you would be all over them for it, you should be the same way towards AMD.
"Right now, Intel enjoys a 10% lead with Core 2 Duo"
Just out of curiosity have you seen the newer reviews with an 8800GTX?
The margin has grown with the GPU bottleneck out of the way (I am talking about gaming).
"that only emphasizes the fact AMD frags 75% of Intel's chips."
I have shown this before with no response from you, please comment...
Total market capacity is 100 units.
Intel = 75
AMD = 25
AMD can only frag 33% of the Intel market, thats all, your 75% claim is incorrect.
Charlie needs to get some perspective. Yes - Brisbane is what it is...and it is what AMD said it would be. However, anyone doing a technical review of the product would conclude that from an end user POV, it's nothing to write home about at this point. And those same folks would only be doing their dilligence in pointing out that it's not the best choice in the marketplace right now. Just as they did w/ Netburst, Smithfield, etc.
Nowhere did Charlie mention Anandtech, in fact Anandtech did the most thorough testing to determine why 65nm is slower than 90nm.
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2893
hey, look at it this way:
P3 evolved to beat K7, K8, and K8 65 nano. Imagine what would happen if intel reinvents the P4!
My favorite quote in that article from Charlie: "AMD is obviously going down the tubes."
Thanks for poing that out. I agree. That has to be the sanest post because finally something was said correctly. In case you missed it, here it is again: "AMD is obviously going down the tubes."
And the reason I hold Charlie in such high regards is that he apparently also thinks the plural of 'die' (in reference to more than one microprocessor being produced on a wafer is) is 'dice'.
Nay, nay, Moosebreath... the plural of die is die... Just like moose is the plural of moose. When a journalist covering a specific sector cannot even correctly grasp a simple concept like that it clearly makes me doubt their knowledge of anything more advanced.
The exception to that rule that helps him save face in this post is/was: "AMD is obviously going down the tubes."
Attaboy, Charlie!
I have shown this before with no response from you, please comment...
Total market capacity is 100 units.
Intel = 75
AMD = 25
AMD can only frag 33% of the Intel market, thats all, your 75% claim is incorrect.
I believe Sharikou was talking about Intel's market share while your talking about global market share... apples and oranges!
But regardless, I saw your blog about your 33% and it makes no sense to me!
Somewhere along the line you must have gotten confused because I didn't quite follow your calculation!
Jeach said...
"I believe Sharikou was talking about Intel's market share while your talking about global market share
Somewhere along the line you must have gotten confused because I didn't quite follow your calculation!"
Whats the difference between market share and global market share?
I will try and make it really simple...
AMD produces 1/4 (25 units) of the market, and Intel 3/4 (75 units).
Hence AMD's 1/3 (25 units) can only compete against 1/3 (25 units) of Intels capacity, which leaves 50% or 50 units of the market open to Intel.
The remaining 50% is where AMD can grow, but they need capacity to do it.
Maybe that explains it better.
All the websites do is reflect the hopes of their readers, and these readers are people interested in new technologies and PC parts, not share traders and bankers.
Therefore it is not unreasonable for them to reflect the hopes and wishes of the readership rather than tittle tattle on about marketshare and gross margin for AMD .. that job is better suited here to be honest as it is more relevant to the readers of this blog.
Nobody who likes the stuff I mentioned about would want to buy a 65nm chip if it runs some of their applications more slowly, they will select 90nm from the same company and get the benefit of that more mature and speedier process. The do not knows can have the 65nm parts, so everybody is happy ....
"Look where 90nm got them after that.... FX74 is the pinnacle. 1+ghz scale and two cores.... Winnie 3000+ being 1.8ghz 512kb single core 90nm if I remember correctly."
And how long was it between Winnie and FX-74?
--> if folks are going to wait this long for improvements might as well buy a C2D in the interim! (faster, cheaper and uses 1/2 power of the FX74 solution).
I think the reason folks are excited is because AMD said 65nm process was 30 or 40% better.
No they didn't say the first 65nm processor would be better, but when they announce 65nm release, make a big deal they are only 1 year behind Intel's 65nm process and announce 65nm will be 30-40% better... I think folks coupled these together and assumed we would not have to wait 1-2 years for the process to iterate/mature to see the improvements (especially with APM3.0!)
It turns out while AMD may only be 1 year behind when it comes to printing 65nm, if you factor in actual process performance the technology looks to still be several years behind Intel.
Also I'm a bit concerned that when they talked about 45nm they said they were expecting only a 15% improvement - if this is all they are projecting this is not a good sign for the next technology node and means they have hit a wall (and are not ready to implement high K gate oxides yet).
"Look where 90nm got them after that...."
Well, intel has scaled from 1.86GHz C2 to 3GHz C2 and 2.66GHz quadcore. That is almost the same as AMD has had during its entire 90nm product cycle :)
WTF? Intel is selling chips (P4) that underperfomrm lower priced chips (Athlon) - you can't do that with brand parity!
/////////////////////
after all AMDs output is sold...after all C2D output is sold..what else is left but P4s and Celeries...of course they are sold in large numbers...nothing else is left
*Edit*
"Hence AMD's 1/3 (25 units) can only compete against 1/3 (25 units) of Intels capacity"
Hence AMD's 1/4 (25 units) can only compete against 1/3 (25 units) of Intels capacity...
Fat fingered that one.
To be fair to AMD, it's a good, natural evolutionary step moving to 65nm. No BIG deal, a bit of power savings, but other than that it should just be treated as a milestone rather than something revolutionary.
As such - it doesn't really look all that fantastic when your 65nm parts don't perform as well as your 90nm parts do.
And about brand - when you ask the common person what they think of when they think of AMD - it's nothing, a processor. That's it! Some people think good things, some people think bad things. There is very little imagery though.
When the common person thinks of Intel - they associate that with the sticker on the front of their PC. "Intel Inside". They envision little guys in silver suits jumping around. They have heard the name over and over. That is worth a TON. That is what marketing consultants drool over, because there is a recognition beyond 'processors'.
Before you argue, remember, it doesn't matter what YOU think, it matters what everybody ELSE thinks.
My favorite quote in that article from Charlie: "AMD is obviously going down the tubes."
And the reason I hold Charlie in such high regards is that he apparently also thinks the plural of 'die' (in reference to more than one microprocessor being produced on a wafer is) is 'dice'.
1. I find it comical that you identified with this quote. It was a sarcastic stab at ppl like yourself. Brain dead fanboys and clueless reviewers. You couldn't have proved his point better.
2. The "dice" misuse as well as the MANY misspellings and obvious grammatical errors are INTENTIONAL. They have been part of The Inq. since the start and are present in every single writers articles. Get ready, here's your clue. The Inq. is "modeled" after the tabloid it's name resembles and in tabloid "spirit" they add these "errors".
Sheesh. I never realized there were so many mouth breathers who don't get the obvious.
Hey Necro --
Apparently you missed the sarcasm in my comment which was due to the sarcasm in Sharikou’s post that was begat by the sarcasm in The INQ link he provided. So let me spell it out for you... (heck... I'm even going to type this at 20 WPM instead of 80 WPM in hopes it helps you understand -- more sarcasm):
I was picking and choosing a sound byte/blurb to summarize the article in the same manner our host the good Doctor picks out one minor thing and a whole different post evolves.
As for your second point... I can only THANK YOU for proving my other point. The INQ is on par with a grocery store tabloid... which hardly equates to responsible journalism. I agree their niche is, as you put it -- tabloid "spirit". They mix some truth, with many rumors, and are the laughing stock of The Valley. Inserting misspellings or inside jokes intentional only makes them look like more of joke. It is sad because there are still people in this world that believe everything they read. After-all.. .if it was found on the internet, it has to be true, right?! Unfortunately the good Doctor uses The INQ far too often as a credible reference instead and this also takes away from his credibility.
I'm sorry you opted to take the "dice" and roll over my post, but you did give me another chance to pull the wool o'er your eyes my lil' lamb, Necro.
Ok Sheepshagger...
I missed your sarcasm and point. kudos. I am man enough to admit my mistakes. kudos as well for your sarcastic reply to my "mouth breather" comment. 20 WPM vs 80 WPM. I like it and will have to remember it for future use. (Giving you full credit of course.)
As for the Inq., I lose my temper a bit where they are concerned because so many ppl miss the point. I will disagree with your opinion that they are a tabloid. They used this approach to separate themselves from the 10,000 other tech news sites and I believe it has worked very well. The AMD/ATI merger was a good example of their many industry contacts giving good info. When you report from the inside using that many "moles" you are bound to end up wrong or of the mark sometimes. They do NOT EVER sign NDA's and so far have not been swayed in their opinions by ad money. Personally I find this refreshing in contrast to the other 9,999 tech news sites that just act as extended PR departments for the companies they report on. Just my opinion, and as my Grandma always said, opinions are like a$$es, everyone has one and everyone else's stinks.
"95% of people don't know anything about Core 2 Duo. Even my friends in electrical engineering don't know. But they are starting to know AMD64."
Do your EE friends live in a cave in Austin?
I have no idea what Sharikou is complaining about. I had no expectation that there would be any dramatic change with 65nm. AMD does not deliver true shrinks with the first process change. This wouldn't actually be 65nm; it would probably be closer to 78nm. By the time AMD releases K8L they should have a transistor that is more optimized for this process.
As to the cache latency increase; I have to admit that that one has me puzzled too. I really wouldn't have thought that there would be any big reason to increase cache beyond 1MB before K8L arrives with additional L3.
As to the cache latency increase; I have to admit that that one has me puzzled too.
My guess is the latency is designed for Z-RAM.
Post a Comment
<< Home