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.
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 .
We re-reported that Intel's forced channel stuffiing has BKed a few Intel distributors. Now, there is a growing concern that more will be stuffed to death.
posted by Sharikou, Ph. D. @ 8/01/2006 10:50:00 AM
58 Comments:
you know the dummest thing about your blog.. You offer your stupid other blogs as evidence to support your current or next blog..
Do you know how credible that makes you doc?? No PhD has does that execpt maybe you!
I hope those channels that went BKed would gang up to sue Intel.
-----------------------
One a different note, any thought about the power leakage at 65nm? I would speculate that Intel's Memron at 65nm would cause laptop to drain battery too quickly.
Maybe it's better to stay at 90nm for low power leakage (static leakage) and work on the power management (dynamic power consumption)???
-Longan-
Hey anonymous above, you know a far "dummer" thing. Wasting time reading a blog you don't understand or don't care for.
Its not Sharikou's job to supply evidence, this isn't a trial, its a blog. He's simply quick linkink his related articles.
you know the dummest thing about your blog.. You offer your stupid other blogs as evidence to support your current or next blog..
Do you know how credible that makes you doc?? No PhD has does that execpt maybe you!
As randy said, there's no demonstration needed in sharikou's comment. He is just linking to a related article in his blog, and then to the source of the new story. I don't see what's wrong with it. I believe it's quite sad that people come here and make posts of the kind "Dude, you're dumb!!!!!". Probably my 14 year old cousin is able to write smarter comments than yours.
Wow, I told people just last week on Tom's HW that Intel would run a lot of small PC builders out of business. They didn't listen. I expect that MANY companies won't survive this price war.
I expect that MANY companies won't survive this price war.
What has CPU prices to do with retail sellers profit? Could you explain?
At least in Estonia it's relatively fixed around 2-10% of CPU price. Sure, if CPU prices drop by 50% then profit per cpu will decrease but on the other hand there will be more buyers and that will balance thigns out quite nicely.
Anyway I as a client can see only good about the price war because I can only win
"you know the dummest thing about your blog.. You offer your stupid other blogs as evidence to support your current or next blog..
Do you know how credible that makes you doc?? No PhD has does that execpt maybe you! "
Proof of another dumb Intel fanboy... I don't think they ever say anything different, and always end with a personal attack. Not trying to lump all of them in one group, but mainly directed at those childesh ones with no intelligent comments. Geesh!! Enough already!
Off topic and not relevent opinions related to the topic here...!! Create your own blog!
Back to the topic:
Nice read of the article on proof of how Intel are channel stuffing to make their bottom line.
I feel terribly bad for those who operate in the channel and have to go BK because of this. I highly doubt that those going thru this will ever deal with Intel again, and there should be some law protecting those businesses.
if CPU prices drop by 50% then profit per cpu will decrease but on the other hand there will be more buyers and that will balance thigns out quite nicely.
That's very naive thinking. A CPU price drop has almost zero effect on PC demand. It just shift the parts people will buy. People were buying Athlon 64 3500+, now they pay the same money for X2 3800+. As I pointed out earlier, Intel execs are morons. They sell Celerons at $25 while they have pile of Pentium 4. A PC with a $25 celeron will still cost $250 once you add memory, HDD and other stuff. Even with a free CPU, a Windows PC would cost $200 because of other parts.
dummer or dumber? The spelling on this discussion is terrible. The grammer is much better.
Merom seems to be a bit, and I mean a bit, more efficient than the core duo laptop proc. Around 7%?
I am most curious about the number of desktops Dell is cranking out everyday with various p4 and celeron flavors? Maybe Dell is solely responsible for cleaning out the channel. The advantage to having a 75% or so market share is that many people that own intel systems maybe upgrading like all the AMD owners are right now via newegg and TD at unbelievable prices.
I am however waiting for merom. Nothing against the T64, but the wider availability and configuration counts for something. Now if we could just get OEM's to go back to clean Windows disks instead of bundled image disks like HP.
Who will win in a price war?
INTEL, lowest cost producer, highers volume. They made 8 billion pushing a infeior product this past year.
Now that product is pushed to the bottom of the stack with Conroe/Merom/Woody bringing up the premium line.
Chipset shortage is gone, pushing the cheap 90nm dual-cores out at firesale prices. Forcing AMD to push high cost 200mm 90nm CPUs at 50% price drop
INTEL wins the war, may lose the 4x4 top end...
But its about the bottom line baby.
Doctorate candidate you just failed your defense.
How you passed your prelims is beyond me. You always thought so illogically?
Anychance AMD/ATI can make a quasi integrated minipci -like minipci wireless cards - to put with the new turion laptops? Even if you put a lower end ATI chip that runs on passive cooling with 64 or 128 of lower clocked ram on it, it would differentiate the Intel IGP and keep a very low cost.
This might even work for super small form factor pcs.
A CPU price drop has almost zero effect on PC demand.
If so then perhaps you can tell me why I've seen so many people who want to upgrade their CPU's now? Most of them said they were waiting for the price drop and would most likely not have upgraded otherwise, at least not so soon.
As we speak about price war, has anyone calculated how much the prices actually fell? I don't know the nubmers about Intel but I do remember that AMD had an average price drop of around 50%. Intel low-mid range dualcores were considerably cheaper than X2's already before the price drop and their prices didn't fell that much. If so then would it be fair to say that AMD's huge price drop is even worse than Intels?
Even with a free CPU, a Windows PC would cost $200 because of other parts.
Actually it would cost quite a bit more if it had Windows on it but I do agree that nowadays you can put together insanely cheap PC if you like to and dirt-cheap CPU plays quite important role in the overall price. Such a PC would suite perfectly for most people who only need it as communication device or a typewriter.
Sharikou wrote:
"That's very naive thinking. A CPU price drop has almost zero effect on PC demand. It just shift the parts people will buy. People were buying Athlon 64 3500+, now they pay the same money for X2 3800+. As I pointed out earlier, Intel execs are morons. They sell Celerons at $25 while they have pile of Pentium 4. A PC with a $25 celeron will still cost $250 once you add memory, HDD and other stuff. Even with a free CPU, a Windows PC would cost $200 because of other parts. "
Ooh! The doc got that right. By pumping up Conroe, Intel openly admitted their netburst was trash. This is the Osborne syndrone X2 (shall I say Osborne DUO??). People will wait for the better stuff.
If they are in the urge to upgrade and cannot wait, with the same $$$ budget for CPU, people will try to get better CPU grade. Who are going to buy the low-end Pentium in Intel's inventory!!!
-Longan-
"A CPU price drop has almost zero effect on PC demand."
If so then perhaps you can tell me why I've seen so many people who want to upgrade their CPU's now?
You have to understand that the percentage of people who upgrade their CPU is tiny. Furthermore, those bought an Intel PC in the last few years won't see any benefit from upgrading: Intel's P4 clockspeed stayed almost the same. If you upgrade from a Northwood to Prescott, you get worse performance.
"INTEL, lowest cost producer, highers volume. They made 8 billion pushing a infeior product this past year."
That would only be true if Intel's Core 2 Duo could ramp up, AND there are enough motherboards out there that support it stably.
So far I don't see neither.
I don't know, maybe it'll start to change starting next month?
"One a different note, any thought about the power leakage at 65nm? I would speculate that Intel's Memron at 65nm would cause laptop to drain battery too quickly."
There are some early benchmarks ouyt already Merom has slightly better battery life at improved performance over Core. I think Tom's Hardware has a benchmark and there were a couple of others as well.
That would only be true if Intel's Core 2 Duo could ramp up, AND there are enough motherboards out there that support it stably.
Its ramped baby... You guys are in denial. Like when you all refused to believe the benchmarks.
Why wouldn't INTEL be ramping this baby.
The Doctor
PCs are very elastic... guys.
Drop the price and people will upgrade. Even I am tempted to upgrade my 900M PIII to PentiumD.
Lot of bang for the buck there
You either go for the value/performance INTEL product or gaming highend Conroe EE. There is really no reason to buy the middle AMD product.
Longan said...
"Who are going to buy the low-end Pentium in Intel's inventory!!!"
My answer for you is based on most of the chips being lowend dualcore.
The general public who does not understand what they are buying, they see a price, dualcore and speed.
Thats about all most people understand, had they understood K8 ability with out the clockspeed it would be a different story (myself included), but AMD is not marketed any where near Intel, as I am sure you have seen on TV or in a magazine.
Intel talking so much about speed and dualcore has definetly been accepted by the general public.
Intel's marketing team had better get a raise if they clear there inventory... lol
"You either go for the value/performance INTEL product or gaming highend Conroe EE. There is really no reason to buy the middle AMD product."
Your Intel Pentium-D 940 CPU, compared to an AMD X2 3800+, will be 40% more expensive, 50% more power-hungry (don't forget the memory controller), 5% slower (see AnandTech's Core 2 Duo benchmarks), and can't upgrade to quad-core.
The only Pentium-D 9xx that's cheaper than X2 3800+ is P-D 900, which saves about $50. Yet the motherboard for P-D is like $15 more expensive. So to save $35, probably 5% of the total PC price, you would prefer a hotter, noiser, and MUCH slower machine? I can't imagine the logic, frankly, because that won't even afford a good heatsink and fan.
So while you feel no reason to buy AMD, there's no reason that everyone would think like you.
"Its ramped baby... You guys are in denial. Like when you all refused to believe the benchmarks."
It turns out that we guys were correct to refused the benchmarks, because the initial round of benchmarks were totally skewed and exaggerated.
On the other hand, you guys the Intel fanboys are in denial that Intel cheated you about Conroe's super good performance.
No, it's not okay to pump the 15-20% in-average lead to 30-40%, as Intel did. Of course you can hand-pick a few apps (SuperPi?) that C2D wins 40% over K8. So can I pick some that C2D wins none or even loses (for the same price).
Oh BTW, Core 2 Duo took 6 months to ramp. Great indeed. Now on Dell the earliest shipment date is what, 8/14! What's Intel (and Dell) been doing during the past few months?
Its ramped baby... You guys are in denial. Like when you all refused to believe the benchmarks.
Why wouldn't INTEL be ramping this baby.
The Doctor
Go buy any conroe chip, and any conroe compatible motherboard from any major online retailer, in stock, today. You cant, sorry, and if you can - tell us where.
Thats not denial thats realism.
Why wouldn't Intel be ramping it? I don't know. My guess is that they are either morons, or it faced heavy technical problems getting on the production line. You tell us - why aren't Intel ramping this baby?
"you know the dummest thing about your blog.. You offer your stupid other blogs as evidence to support your current or next blog.. "
I guess the AMD fannies did not get the main point.. L E T M E S P E A K S L O W L Y. Y o u c a n n o t r e f e r e n c e y o u r s e l f i n a n a r g u m e n t .. G E T I T !
OR do you want me to speak slower
You scum make me laugh with your laughable denials.
"I couldn't buy a Conroe E6600 at Wal-Mart the day it launched, Intel is not ramping Conroe!!"
God it's going to be a GREAT next quarter. Intel will take a beating, but the funny thing is that AMD is going to absolutely get reamed in the back side. When AMD stock is at 15 I will be back here to laugh at you ridiclous fools.
"Merom seems to be a bit, and I mean a bit, more efficient than the core duo laptop proc. Around 7%?"
"There are some early benchmarks ouyt already Merom has slightly better battery life at improved performance over Core. I think Tom's Hardware has a benchmark and there were a couple of others as well."
Core 2 Duo T7400 2.16Ghz Battery Life:
http://www.pcper.com/images/reviews/276/Intel%20Core%202%20Duo%20T7400%20-%20Battery.png
Wait for more benchs...
Hmm...the link:
http://www.pcper.com/images/reviews/276/Intel%20Core%202%20Duo%20T7400%20-%20Battery.png
shows that Merom has shorter battery life!!!
I am an engineer and came up with this "layman" series of formula for power consumtion:
1. Same design rule (i.e. 90nm) bigger die burns more power.
2. Same die size, smaller design rule burns more power (65nm die runs at higher freq. for more dynamic power consumption and has more static leakage)
Merom is both bigger die size than the previous Pentium-III core and at a smaller design rule, should it consume more power?
----------------------------
On a different note: Someone got to do more stuffing:
http://us.ft.com/ftgateway/superpage.ft?news_id=fto080120061834292789&referrer_id=yahoofinance
"
Intel's chip stockpiles create semiconductor glut
By Chris Nuttall in San Francisco
Tuesday Aug 1 2006 18:15
Intel, the world's largest chipmaker, has almost single-handedly created the biggest semiconductor glut in the industry since it was hit by massive stockpiles two years ago, according to a leading research firm."
-Longan-
Intel’s Q3 numbers may reflect the widely reported channel stuffing, but most likely the loss of goodwill could come home to roost sooner. Intel partners keep vaporizing while AMD’s partner list keeps growing. More important is the constant turmoil and controversy of spurious marketing claims, dark business practices and mounting legal problems. Keep an eye open for creative accounting as performance pressure builds. Dell switching to AMD has had enough and needs steam in that big engine. Dubious missing product at launch can’t help their image. Have I missed anything?
Dickwad said:
It turns out that we guys were correct to refused the benchmarks, because the initial round of benchmarks were totally skewed and exaggerated.
Back that up. Show where any benchmarks were cooked by Intel. Everyone was pissin' and moanin' about how Anandtech fixed their initial Conroe benchmarks and didn't compare it to an adequately configured FX-62 (overclocked BTW). So Anandtech listened to all the AMD fanboy pissin' and moanin' and re-did the benchmarks. And guess what, Conros still kicked crap out of FX-62 (AMD's top of the line on desktop). So Dickward, please supply a link that proves that Intel made up benchmark results or SHUT YOUR PIEHOLE.
Some anonymouse coward said:
Go buy any conroe chip, and any conroe compatible motherboard from any major online retailer, in stock, today. You cant, sorry, and if you can - tell us where.
I'll do you one better. You can buy complete Core2 Duo systems today at:
Dell.com
Gateway.com
HP.com
Falcon Northwest
Voodoo PC
and many more.
You just have to be willing to pay for the superior performance since this is the new king of the hill you're getting. As far as ramp up... 1 million processors as an initial shipment isn't bad for a new architecture. It will certainly ramp faster as Intels 3 (count them) 65nm fabs start producing Core 2 Duo. How many 65nm fabs does AMD have? Sorry I didn't hear you, was that Zero you said?
Whey aren't there any AMD designs like this?
I love the Shuttle X100 and plan to get one immediately.
Intel brand gets stuffed in Brazil?
A small company in Brazil sues Intel and is winning.
This is funny.
Link
Wirmish thanks for posting the battery benches. It looks like Merom's battery life is going to be a lot better than the AMD fanboys think. Sure to burst their little bubbles.
In case you missed it.
>Some battery benches
Love to see more of that. I think my next PC to replace my very aging Northwood core desktop will be a Merom based Notebook. No more desktop PCs for me....
Quoting from the link supplied on the battery life:
"The platform we are using to test the T7400 doesn't support all the new power features of the CPU, so the following battery benchmarks should be taken with a grain of salt. With an updated BIOS from Asus we are going to be re-running these tests to get more accurate results." further quote: "Something ain't right with the above results since we know that the new T7400 Merom CPUs should have better (or nearly equal) power consumption than the older Core Duo Yonahs. We are awaiting more details about power consumption since it's something that is, perhaps, more important than the performance numbers. We will report more as we hear it."
see link : http://www.pcper.com/article.php?aid=276&type=expert&pid=9
Looks like Tom's had a bit better luck but did not see significant improvements. As I said earlier 7% or so.
In my field it is very important not to take statements and results out of context. It is best to give the full picture as the statements that pc perspective supplied above and below the graph.
Funny, isn't it?
2003:
AMD Fans: Haha, our chip outperforms yours!
Intel Fans: That hardly matters when you can't buy it anywhere!
AMD Fans: You're just in denial because you got beat!
2006:
Intel Fans: Haha, our chip outperforms yours!
AMD Fans: That hardly matters when you can't buy it anywhere!
Intel Fans: Hah, you're just in denial because you got beat!
Swings and roundabouts people. I bet you we'll be having this discussion again when K8L launches in 2007/08.
@ Graham,
Just shut up, you are a joke with your stupid comments all the time.
Who gives a shit that you don't want a desktop anymore, I don't and I think many more.
You just go and buy your merom notebook and leave us alone.
The only reason your on this blog is to slag AMD and Sharikou, but no sensible comments so far.
You are nearly as bad as "The Doctor"
The Nurse
I guess the AMD fannies did not get the main point.. L E T M E S P E A K S L O W L Y. Y o u c a n n o t r e f e r e n c e y o u r s e l f i n a n a r g u m e n t .. G E T I T !
OR do you want me to speak slower
1st: It's not an argument, it's just a story, and he is linking a related article, just the way every site in the Internet does.
2nd: While referencing ones self articles is not the best practice, it is not prohibited. Please see these two articles written be L. Adleman. The first was written in 1994 and published in "Science". The second is a draft written in 1995, and in it there's a reference to Adleman's first article. Maybe you would like to send a mail to Adleman calling him dumb and demanding him to give back his ACM Turing Award.
1994 article on DNA computing
1995 draft with reference to 1994 article
Who is Leonard Adleman?
"When AMD stock is at 15 I will be back here to laugh at you ridiclous fools. "
BUY!
Laptop Logic
Clash of the Titans ~ Dothan vs Turion:
http://www.laptoplogic.com/resources/detail.php?id=17
Batterymark - Life test:
http://www.laptoplogic.com/data/resources/images/17/batterymark-life.gif
" Please examine the results remembering that the true competitor to Dothan in battery life is the AMD's MT lineup, which uses 30% less power than the ML models. "
~~~
ZD-Net Australia
AMD vs Intel ~ 10 notebooks tested
(Nine Intel vs only one AMD)
http://www.zdnet.com.au/reviews/hardware/notebooks/0,39023407,39203488,00.htm
"If your mobile need is raw grunt then you just cannot go past the AMD Turion 64 equipped Asus A6000 -- in business and multimedia applications the A6000 was well ahead of the other notebooks and for gaming the Asus had the highest performance in DirectX9."
"2003:
AMD Fans: Haha, our chip outperforms yours!
Intel Fans: That hardly matters when you can't buy it anywhere!
AMD Fans: You're just in denial because you got beat!
"
Well... what about 2004 and 2005 that were conveniently skipped over?
Remember that K8L is something within a year, not till 2009.
"Funny, isn't it?
2003:
AMD Fans: Haha, our chip outperforms yours!
Intel Fans: That hardly matters when you can't buy it anywhere!
AMD Fans: You're just in denial because you got beat!
2006:
Intel Fans: Haha, our chip outperforms yours!
AMD Fans: That hardly matters when you can't buy it anywhere!
Intel Fans: Hah, you're just in denial because you got beat!
Swings and roundabouts people. I bet you we'll be having this discussion again when K8L launches in 2007/08."
I don't remember finding it hard to get over here. Expansive, yes. Hard, no.
On the other hand all the stores here have yet to receive your Conroes.
Anonymous wrote:
"Ooh! The doc got that right. By pumping up Conroe, Intel openly admitted their netburst was trash. This is the Osborne syndrone X2 (shall I say Osborne DUO??). People will wait for the better stuff.
If they are in the urge to upgrade and cannot wait, with the same $$$ budget for CPU, people will try to get better CPU grade. Who are going to buy the low-end Pentium in Intel's inventory!!!"
You gravely overestimate the technical prowess of the public. How many mom's buying PC's for their kids for back-to-school are going to look over system specs? Probably not a lot. Most will look for the cheapest solution they can find on Dell or Gateway for their needs.
In all honesty, the peripherals will probably be much more scrutinized by the average consumer than floating point performance or energy efficiency.
The Core 2 Duo line is aimed at the high-end/enthusiest, and guess what 90% of the high-end enthusiest users have expressed great excitement over? That's right: Core 2 Duo.
Expect Intel to clear inventory and be supply limited on Core architecture products through the end of the year. Maybe knock AMD back under 20% MSS. Who knows?
With the greatest respect for Hector and the AMD team, even they are not above singing the tune that suits their purpose. In this case about benchmarks pre and post the launch of Conroe/Woodcrest:
Before Conroe/Woodcrest:
“We are very confident that we'll continue to have superior benchmark capability based on DDR1 well into 2006.”
Hector Ruiz, AMD Q1’05 Earnings Call, Apr 13, 2005
“With Opteron, AMD set the performance standard that the competition is now chasing. Performance per watt leadership, a more relevant or customer-centric measure of system performance and our unique direct connect architecture will help us continue to set the bar for system level performance for years to come.”
Hector Ruiz, AMD Q1’06 Earnings Call, Apr 12, 2006
“Price performance, AMD64 last time I checked, had broken every benchmarking record and continually raises the bar for performance.”
Marty Seyer, AMD Analyst Day, Nov 15, 2005
“…we have a sustainable advantage here around performance per watt that the competition cannot touch.”
Marty Seyer, AMD Analyst Day, Nov 15, 2005
Post Conroe/Woodcrest:
“In this past quarter, we have entered into a new stage of industry transition. The world has spoken and is really a new world that is about differentiation…. It is not about benchmarks.
”Hector Ruiz, AMD Q2’06 Earnings’ Call July 20th, ‘06
“It is interesting from perspective, the performance crown and the microprocessor business is something that’s been passed back and forth several times over the past 10 years and I expect that to continue. Interestingly who has the performance crown at a given instant as relevant – is relevant really to pretty small fraction of the marketplace. That fraction is one refer to as the enthusiast market.”
Marty Seyer, AMD Q2’06 Earnings’ Call July 20th, ‘06
All corporate hacks are the same. They'll say whatever they need to make the next sale.
Expect Intel to clear inventory and be supply limited on Core architecture products through the end of the year.
Yes, they will be supply limited, but what makes you think that they will clear inventory? In the past year they have been building inventory consistently. Right now about 95% of Intels production is still Netburst. Unless they slow production of Netburst while they transition to Core (which they said they wouldn't do), they will continue to build inventory.
Well... what about 2004 and 2005 that were conveniently skipped over?
I was comparing major product launches - Athlon 64 in 2003, and Core 2 this year. I'm not going to defend the Prescott and Pentium D launches because, well, what IS there to defend about those?
I don't remember finding it hard to get over here. Expansive, yes. Hard, no.
I actually worked in a major computer retailer at the time of the Athlon 64's launch, and I remember that for most of the first month, availability was pretty awful. It was only when I bought my own Athlon 64 in late November that year that availability really started going, probably so that AMD could shift the Athlon XPs still in the channel.
On the other hand all the stores here have yet to receive your Conroes.
"My" Conroes? I still use the Athlon 64 I bought that year, and upgraded from a P4 1.6A. Believe it or not, the world isn't divided purely into AMD and Intel fanboys. 95% of the population, myself included, couldn't really give a damn what brand processor they use, so long as it works well.
Channel partners will survive. Some always go out of business as a response to market changes. If they were stupid enough to stock far more inventory than they could sell, too bad, they're out of business.
They may lose on their investments, but a successful partner will expect that to happen from time to time and will live on.
Why the hooplah today? Because INTEL caused it. I'm sure the same was caused by AMD in the past. And this blog is ran by the most biased individual I've ever heard, so of course this is the big news.
But it's all good, survival of the fittest is a good thing in the big picture.
Smart people that have held out on upgrades or system purchases will get top-notch CPUs at an affordable price. Tech freaks will get their Conroes soon, they just have to get on the waiting lists.
Companies that choose to align themselves with only AMD or only Intel will die off. Look at Dell.
Again, the successful people are people that evaluate everything as unbiasedly as they can (Sharikou, read that over again). The rest can hang out with Mel Gibson, whose bias is also destroying him.
While I'm thinking of it...
Sharikou, why don't you get rid of your bias? You're a smart man/woman...you could benefit the world a lot better if you didn't tailor your evaluations and comments to devalue Intel. In fact, people would probably learn a lot from visiting this blog instead of hearing the same one-sided opinions over and over.
I know I've learned a lot here, but usually from the people who don't choose sides, not from Sharikou or Intel/AMD fanatics.
The rest can hang out with Mel Gibson, whose bias is also destroying him.
You mean the Jews will ruin him?
"All corporate hacks are the same. They'll say whatever they need to make the next sale."
Actually, which of the statements is false, except Hector's first comment about performance crown "well into 2006"? (So he was surprised that Intel would pre-release the benchmarks 7 months ahead... IMO, Intel probably did so to beat his words.)
And you are just plain silly to call 2nd quarter "post" Woodcrest/Conroe. Systems based on those chips were no where to buy for another 2 months at the end of May!
As far as I see AMD's executives only make claims that they themselves hold true. Why don't you make a list of Intel's statements and see how much more rediculous they are? In terms of corporate morality, AMD & Intel are of two completely different breeds, and IMO Intel fans just pretend blind to that difference.
Anonymous said "1st: It's not an argument, it's just a story, and he is linking a related article, just the way every site in the Internet does."
You are correct if a scientific article is referencing another scientific article written by the same person who does not want to re-list a bunch of conclusions he made in the past..
However, when you are makeing an argument, referencing a blog you wrote to enforce your argument seems lame and unconvincing.. In effect, you may be saying: I know it is true because I said before it is true!
I am glad you are making an argument, though, very politely, something I rarely notice on this blog
"Its not Sharikou's job to supply evidence, this isn't a trial, its a blog. He's simply quick linkink his related articles."
You know, Sharikou brought that on himself.. If I come and say, I am better than you, I have to prove it. If I say I have a faster car than you, you would like to see what car I have. If you keep telling the world that you are a PhD and keep reminding them of that every article and blog you reply to, then, it is only fair to surface proof on that.
Most people who reply to sharikou's blog, intel fans or amd fans, know he is not really a PhD. He just uses it to enforce an argument and he thinks it is working. But he is forgetting that having a PhD does not necessarily enforce an argument except if the argument is stuffed with facts and hard evidence.
I admire Rahul's blogs, although I disagree with him sometimes, but he never claimed he was a PhD or an MS or even a BS. When he states his opinion (which he reminds people as it is only his opinion) he supports it with facts!
Hope the Self Proclaimed Doc understand that.. But again, may be he would not! GROW UP!
Sharikou, Ph. D said...
The rest can hang out with Mel Gibson, whose bias is also destroying him.
You mean the Jews will ruin him?
Yeah, they ARE responsible for all the wars you know. This Intel/AMD price war is all because of the Jews too.
Here is some supporting evidence:
Adolph Hitler - Jewish
Gordon Moore - Jewish
Hector Ruiz - Jewish
Hezbollah - Jewish
Osama Bin Laden - Jewish
Saddam Hussein - Jewish
Aerosmith - Jewish
Britain and Iceland - Jewish
AMD - Jewish
Intel Conroe - Jewish
Mel Gibson - NOT Jewish, however, I think it's important to understand that if he is going to 'fuck' you and you're Jewish, it's A-okay.
Hey! My word verification says 'Yummy'! My lucky day!
I wonder if Intel is holding back releasing more conroes until August 7 to clear the channel of heatburst and slowleron and the lower prices after price cuts?
Just a thought. I just dont see Intel only having 1500 or so conroes for the release on July 27.
Longan's joke:
I am not Jewish and you guys are a bunch of Youwish.
Intel fanboy "Conroe beats FX-62", "yeah right, you wish" AMD fanboy.
AMD nuts "Athon-3800 frags Pentium 965", "Pfftt, you wish" Intel craze.
AMD fanatic "Socket-F crushes Woodrest", Intel die-hard "Baloney, you wish".
Intel junkie "Conroe overclocked 4Ghz burns FX-64", "You wish..." AMD buff.
-Longan-
P.S. A friend of mine worked for Intel and went to Israel for a on-loan project in 2005.
He said the Israel engineers not only worked hard and long hours, they worked intensely as if Armageddon is coming... He was so reliefed when he got sent back to United State.
Gotta respect them with such work-ethic or they beat you hand-down.
"I am an engineer and came up with this "layman" series of formula for power consumtion:
1. Same design rule (i.e. 90nm) bigger die burns more power.
2. Same die size, smaller design rule burns more power (65nm die runs at higher freq. for more dynamic power consumption and has more static leakage)
Merom is both bigger die size than the previous Pentium-III core and at a smaller design rule, should it consume more power?"
To the idiot, I mean, "engineer" who provided this astute analysis: Any idea on the supply voltages and amount of voltage overdrive between 90nm products and 65nm products?
If I told you the supply voltages on 65nm products were lower, do you have any guesses on what that would do to power consumption?
Any thoughts on whether improved chip design specifically in sleep state management might help?
And finally if more prowerful chip allows an application to run under a smaller CPU load than the respective 90nm part, do you think that might help battery life.
Also 65nm does not necessarily have to mean higher (I think you mean clock?) frequency - that is dependent on transistor sizing and chip design. Does 65nm Core 2 product run at higher frequency than 90nm P4 Netburst chips?
Other than those things, analysis is spot on!
"Any idea on the supply voltages and amount of voltage overdrive between 90nm products and 65nm products?"
Would you mind explaining how shrinking from 90nm to 65nm automatically or intrinsically implies lower transistor voltage or and the overshoot? or are you just dumping things that you don't understand to argue with him?
"If I told you the supply voltages on 65nm products were lower, do you have any guesses on what that would do to power consumption?"
We all know that dynamic power consumption is proportional in the first order to the square of working voltage; it is advanced by fabrication materials and techniques, and not specific to any feature width.
"Any thoughts on whether improved chip design specifically in sleep state management might help?"
That's again orthogonal to the transistor size shrink. I don't know why do you even bother to bring that up, when he was talking only about the latter.
"And finally if more prowerful chip allows an application to run under a smaller CPU load than the respective 90nm part, do you think that might help battery life."
In short: NO. Under the same conditions otherwise, a 4 IPC @2GHz will consume more power than a 2 IPC@2Ghz, even if the former works at 50% load. That's because in order to achieve double computing capacity, more than double logics are needed, and even if they are not switching, they still consume static power, which happens to be more severe under 65nm than anything larger.
"Also 65nm does not necessarily have to mean higher (I think you mean clock?) frequency - that is dependent on transistor sizing and chip design. Does 65nm Core 2 product run at higher frequency than 90nm P4 Netburst chips?"
I have no idea why you mix size and design together in your discussion. P-D in 65nm consumes more power than Conroe in 65nm because the former has higher frequency. P-D in 65nm however consumes less power, and overclocks better (higher frequency) than P-D in 90nm.
In summary, what that 'idiot' engineer said were far more credible than what you mumbled above. I mean you'd have been a failed engineer - one can never argue logically with two or more factors varying at the same time!
Oh man, the "layman" is kinda rude. I am the clever engineer who try to communicate with the rude and dumb layman like you....
I happened to work on ASIC so let me help you a bit more. At Core 2 has less number of pipe line so it run at lower frequency than the 90nm P4.
P4 with same number of pipe line (super-pipe line) at 65nm will run at a slightly lower voltage but at a much higher frequency than P4 at 90nm.
Go take some ASIC and digital design course and come back here and talk. Tweaking voltage on the control panel does not count. Don't forget to go back to your parents and ask them to teach you some manner. Ask them "how to be nice course 101".
-Longan- ASIC guy
Longan wrote:
"Merom is both bigger die size than the previous Pentium-III core and at a smaller design rule, should it consume more power"
Please Intel fanboy, read what I wrote "Pentium-III core" ok? I am talking about the laptop chip. There is no reason to compare it with Pentium 4.
If you don't work on ASIC, please please stay off this subject ok.
I strongly suspect the Merom will consume a lot more power than the Pentium-III laptop core. Previous link I checked show shorter Merom's battery life with some excuses of non-support function. I have yet to find a bench mark showing Merom has longer battery life.
-Longan-
More food for thought
Do you have any clue as to the operating voltage of INTEL or AMD's 90nm and 65nm technology?
Go get that information before you speculate on power consumption and any reason why it might or might not be better.
The Doctor.
Longan- ASIC guy
No wonder you work on ASICs. You program in RTL and let the tools do everything else. Any monkey with a BS in CS can do your job.
Lets look at this silly line
P4 with same number of pipe line (super-pipe line) at 65nm will run at a slightly lower voltage but at a much higher frequency than P4 at 90nm
You are truely and idiot. Thank god AMD, INTEL and the high end CPUs aren't designed by a monkey like you.
Do you understand yet the fallacy of your statement? If not the Doctor can educate you.
Post a Comment
<< Home