Friday, January 26, 2007

K8L frag starts in April

40% will open a major gap that will take Intel three years to catch up.

It's a cruel world for Intelers out there. Intel's BK in 2Q08 is unavoidable due to AMD's massive capacity ramp.

Once K8L is out, Clovertown will be sold at $150 and Woodcrest will be $95. QX6700 will be $150 and Conroe will be $87. Quad FX will be $999 per pair (two K8L quads). AMD will flood the hell of the market with dual core K8L chips, fragging Conroe by 40%. People think AMD will be slow to transition AM2 chips to K8L core, think again.

For anyone who plans to buy a server there, he needs to consider how to wisely spend his money. Buy a expensive Clovertown machine, and find it's about half the speed of the K8L, or buy a cheap Opteron box, and then pluggin the new K8L chip and get massive performance boost. It should really be a no brainer.

Yet, Hector Ruiz promised a killer architecture in 2008. With this architecture, AMD will frag Intel by 3x at least and permanently neutralize the Intel threat.

72 Comments:

Blogger Roborat, Ph. D. said...

Sharikou: "K8L frag starts in April"

* You also said AMD frags 75% of Intel's CPUs. Outcome: massive ASP/margins drop for AMD
* Intel frags itself with Core2Duo. Outcome: 11% Q/Q revenue gain for Intel vs AMDs 3%
* QuadFX frags Kentsfield. Outcome: 4x4 5-20% behind while sucking twice the energy.

Since you're still using the word "frag", AMD should be worried!

2:37 AM, January 26, 2007  
Anonymous Anonymous said...


I don't necessarily think that this should be taken with a grain of salt. I mean, to make claims like, "This is going to be the biggest enhancement to the x86 architecture since we released Opteron. It will take the power-per-watt equation to levels people have never seen before." and then not perform to those standards is worse than not having a competitive product.
It seems like Barcelona was created for server, and if we get a cpu initially meant to make a horse of a server, then I cannot complain. I look forward to a Conroe killer, that would just be ironic, don't ya think? It would be like a cruel joke. I'm not a fanboy, I just think if amd stole back the crown after having it for 3 years and losing it for less than one, that would be funny in a nerdy sort of way.
And if it kills a Conroe and uses a max of 65w there's no stopping it. I look forward to it. Anyone else? Admittedly I am a bit biased because I have AMd products, but I also have Intel products, and I don't think one company is better than the other, but I kind of want Amd's solution to be a winner.

from here
... couldn't say it better myself ...

4:44 AM, January 26, 2007  
Anonymous Anonymous said...

Just keeps on getting better and better. xD

5:17 AM, January 26, 2007  
Anonymous Anonymous said...

Yeah, it'd kick everyone's butt if it were released TODAY, but it will be released 6 months from now. You expect nothing to change on Intel's end in these 6 months?

We'll all have a good chuckle about this article in a few months - just like we did with all your other commentaries in the past.

"Intel's CPU's cause Sony's batteries to burst into flames!" I'm still cracking up over that one.

6:37 AM, January 26, 2007  
Anonymous Anonymous said...

Read the article carefully junior. It states that sampling "shop around" in April-June, not shipping in April.

It also notes that there won't be an impact on AMD's financials until Q4. That's 3 more quarters of hemorraghing if you couldn't do the math. 3 more quarters of Intel bringing in >$1B in cash a quarter- and not getting any closer to the locak Burger King.

7:07 AM, January 26, 2007  
Anonymous Caldwell40 said...

Engineering samples of Barcelona MAY be out in April buy wide availability to OEMs and online retailers? Nope! Now you're just being silly.

9:13 AM, January 26, 2007  
Blogger They said...

You'll get the full force of Christ, don't think you won't. He was sheer weakness and humiliation when he was killed on the Cross, but oh, he's alive now--in the mighty power of God! We weren't much to look at, either, when we were humiliated among you, but when we deal with you this next time, we'll be alive in Christ, strengthened by God. Test yourselves to make sure you are solid in the faith. Don't drift along taking everything for granted. Give yourselves regular checkups.

9:32 AM, January 26, 2007  
Anonymous enumae said...

Off topic...

DEC buys AMD's low-cost PC assets

Do you think this will efect there OLPC project?

11:18 AM, January 26, 2007  
Anonymous Anonymous said...

Can you explain why AMD only promises (= vaporware) while Intel proves (-> IDF etc.)?

I dont get it cause following this pattern either you or AMD is wrong.

12:12 PM, January 26, 2007  
Blogger PENIX said...

AMD's game plan is flawless. I don't see how Intel can possibly compete with this.

AMD's hardware is a scalable powerhouse. Intel's hardware is slow, hot and unstable. Their exploding laptops have not helped their reputation. It is no wonder that they continue to lose more and more market share.

I wouldn't be too surprised if Dell and HP are in talks with AMD about going exclusive. Why would Dell and HP continue to base their company around faulty Intel hardware? It is in their best interest to dump Intel completely and go exclusive with AMD.

1:53 PM, January 26, 2007  
Blogger Jeach! said...

There should be (unless I'm mistaking) added excitement to AMD's 2007 competitive edge: CO-PROCESSORS. There should be a few different types this year.

You expect nothing to change on Intel's end in these 6 months?

Sure! Manage FAB transition from 65nm to 45nm, try to sell obsolete Pentium inventory, find buyers for non-profitable units, spin-off the flash business, cook-up benchmarks, fire 10% of workforce annually, buy back stocks to make the book look a bit better, bribe Sun and Google. And well, there's no much more time innovating the C2D architecture. But a reassuring factor is that you don't have to innovate when you can just follow in AMD's foot-steps.

"Intel's CPU's cause Sony's batteries to burst into flames!" I'm still cracking up over that one.

Don't crack up too much, YOU might burst into flames :)

It also notes that there won't be an impact on AMD's financials until Q4. That's 3 more quarters of hemorraghing if you couldn't do the math. 3 more quarters of Intel bringing in >$1B in cash [...]

You 'may' be right about AMD's financials, but if AMD can show early demos and benchmarks that will cripple Intel's server sales. Intel will be forced to significantly lower its prices which will affect its overall ASP's. And do remember this: If you were to graph Intel's profitability curve on a graph it is NOT a straight diagonal line (like a backslash). It is a curve that the more you go right, the faster it drops. So for every ASP hit Intel takes, it's profitably moves to the right. What I mean is that if you cut Intel's revenues by half, it won't cut Intel's earnings by half, but more like by 90%. Most of you Intel'ers don't understand this. Sharikou could probably post an image of what I mean on some blog. Scientia is good at explaining the financials, he can probably do a better job of explaining it then me.

2:09 PM, January 26, 2007  
Blogger PENIX said...

Anonymous idiot said...

"Intel's CPU's cause Sony's batteries to burst into flames!" I'm still cracking up over that one.

You Intel fanboys cannot grasp the most basic logic. It DOES NOT MATTER if it is Sony's fault or Intel's fault. Intel laptops are exploding, and that's all you need to recognize. You don't see AMD laptops exploding left and right like Intel. Intel latops are a hazard and should be banned from all public areas.

2:19 PM, January 26, 2007  
Anonymous Anonymous said...

Do you think this will efect there OLPC project?

look here found it on amdzone.com:
----------------------------------
http://www.amdzone.com/modules.php?op=modload&name=News&file=article&sid=7115

2:20 PM, January 26, 2007  
Anonymous Anonymous said...

K8L will dominate Intel the way 4x4 dominates Core 2.

ROFLMAO!

2:23 PM, January 26, 2007  
Blogger TheKhalif said...

Though I applaud your "devotion" your "enthusiasm" tends to be a little bit much.

I just don't understand why people don't understand why Barcelona will achieve a large increase over Opteron X2.

1. Barcelona was designed for the dual stressor technique (said to switch 40% faster than 90nm)
2. Intel successfully used L3 to get 10-20% increases.
3. Barcelona has dual FP, dual SSE4, dual loads, enhanced prefetch, OoO loads(maybe stores), more branch history, larger fetch buffer.

4.65nm is naturally cooler.

2:49 PM, January 26, 2007  
Anonymous Anonymous said...

"AMD will begin shopping the Barcelona chip around to customers in the April-June time frame (so, in about three months)."

Does this mean paper launch in April or actual meaningful chip availability? I find "shopping it around" to be a peculiar choice of words why not say released? available?

"Allen said Q4 2007 will be when the first real impact of Barcelona comes through in AMD's financial statements."

Hmmm...chips in Q2 yet Q4 will be the first real impact, 2 quarters to meaningfully ramp JUST a server chip? Find this funny after all the AMD fanboy whining about Core 2 ramp rate from Jul - Dec (especially when server was 75% converted to Core2 over that time period!)

"When I pressed him on what exactly that 40 percent includes, it was slightly less clear; he mentioned a number of metrics, including performance per watt."

Is anyone else thinking Sharikou = Allen? Oh nevermind he didn't use the word "frag"

3:04 PM, January 26, 2007  
Anonymous Anonymous said...

Hey boys, don't fear and don't cry if a bad day the good Doc Sharikou disappears, there's the qualified successor: Penix for president! I suspect Penix may be a Sharikou's body part, hidden in the darkness...

3:27 PM, January 26, 2007  
Anonymous enumae said...

Penix said...

"I wouldn't be too surprised if Dell and HP are in talks with AMD about going exclusive..."

AMD can not supply them at there current size or production capabilities.

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

Sharikou said...

"AMD will flood the hell of the market with dual core K8L chips, fragging Conroe by 40%."

Looking at the words chosen in the statements about the 40% it would seem he was being vague.

(When I pressed him on what exactly that 40 percent includes, it was slightly less clear; he mentioned a number of metrics, including performance per watt.)

It could all be FP related task in which they already have a 20%+\- advantage.

3:50 PM, January 26, 2007  
Anonymous Anonymous said...

Hmmm....performance claims for chips that will be available in Q3/Q4...wasn't it not that long ago when people were yelling "Osbourne effect" when Intel provided ACTUAL BENCHMARK DATA, as opposed to verbal claims, in March for a chip that was set to ship in July.

...but...but...but... it's drop in upgradeable for current Opteron dual cores. So basically I should buy an Opteron chip now, use it for 6-9 months and then what, throw it away?

Yeah, these claims may help those who currently have Opterons and may be debating about waiting for K8l or jumping ship to Intel. But for those looking to buy a new server right now it says either wait 6-9month, buy an Intel chip now, or double spend on chips by buying a AMD dual core now and then throw it away after buying a quad in 6-9months.

So I understand - when Intel did this it was very bad and stupid...when AMD does this it is OK? At least Intel provided actual benchmark data when they did it!

4:15 PM, January 26, 2007  
Anonymous Anonymous said...

"AMD's game plan is flawless. I don't see how Intel can possibly compete with this."

Well,since penix gives it his/her stamp of approval, all of us AMD fans can sleep easy now as penix has time and again shown himself to be the superior intellect on this board!

4:17 PM, January 26, 2007  
Anonymous Anonymous said...

"Intel's CPU's cause Sony's batteries to burst into flames!" I'm still cracking up over that one.

----->You Intel fanboys cannot grasp the most basic logic. It DOES NOT MATTER if it is Sony's fault or Intel's fault. Intel laptops are exploding, and that's all you need to recognize. You don't see AMD laptops exploding left and right like Intel. Intel latops are a hazard and should be banned from all public areas. <-----

Penix how can you be soooooo narrow minded. If the sony battery CAUSED the laptop to blow up then how is it the cpu's fault. In reality it was the actual battery catching fire something completely independant of the cpu.

Using your same logic, example, you fall down and break your arm becasue you slipped on ice, but in "reality" its your legs fault becasue they didn't keep you upright, therfore your legs are inferior and you should have them cut off. See where i'm going with this get into the real world.

Fact is RIGHT NOW Intel has a rather good product, I like AMD and have AMD systems but face it right now the C2D is very good competition for the current A64 cpu's. None of this is even set in stone so stop getting your fangirl panties in a knot...

6:12 PM, January 26, 2007  
Anonymous Anonymous said...

"AMD's hardware is a scalable powerhouse. Intel's hardware is slow, hot and unstable."

Not according to Google, who today dropped AMD as a supplier and switched to Intel. I'd bet they know more about what is happening behind the scenes at both companies than you do.

7:24 PM, January 26, 2007  
Anonymous edward said...

"Can you explain why AMD only promises (= vaporware)"

Can you give us an example where AMD fail to deliver a product the way it promised?

8:00 PM, January 26, 2007  
Anonymous Anonymous said...

Yea, though I walk through the valley of the shadow of death, I will fear no evil: for thou art with me; thy rod and thy staff they comfort me. Thou preparest a table before me in the presence of mine enemies: thou anointest my head with oil; my cup runneth over. Surely goodness and mercy shall follow me all the days of my life: and I will dwell in the house of the Lord for ever.

8:09 PM, January 26, 2007  
Anonymous Anonymous said...

Hey Pat, when you get to BK could you bring me back a large onion rings? I love onion rings!

8:40 PM, January 26, 2007  
Anonymous Anonymous said...

Looks like Intel revealed one of its tricks up their sleeves - high K/metal gate will be rolled out at 45nm node - this will put them 1 if not 2 generations of IBM (and therefore AMD as they don't do major process technology development) who doesn't intend to implement this technology until at least the 32nm node.

There is an article on the INQ and it is actually somewhat accurate (surprisingly enough). This should lead to significantly better leakage performance (thus transistor off state power) and/or allow them to scale speed much higher and remain within the desired power envelop - Intel will likely do a bit of each.

It'll be interesting as this should significantly increase transistor/power performance over tradition technology scaling of SiO2.

http://www.intel.com/technology/silicon/45nm_technology.htm

The article is dumbed down enough even Sharikou should be able to understand it - other than the phonon scattering and Fermi level pinning part! :)

9:59 PM, January 26, 2007  
Anonymous anti-Intel said...

Well, well Intel "funnyboys": I don't think that Intel could survive more than 7-8 months after K8L launch! Why, because Intel's tech. hasn't a match for K8L architecture. And, yes, I like AMD more than Intel the monopolistic super sh*t.

PS: 3 days ago I bought a Intel E6300 (now oc. to 2.8 GHz) and I must tell you Intel guys: is not so super as you all say. I've run some benchmarks and compared to a AMD X2 5200+ and (e6300/2.8GHz)is just 5-10% faster. And even more: you're beloved Intel cpu is about to explode. It runs at 65-68 deg. C at full load with stock cooling (which is so big and so inefficient)... you call this 65nm technology ???

11:11 PM, January 26, 2007  
Blogger Ho Ho said...

jeach!
"Manage FAB transition from 65nm to 45nm"

Intel doesn't transition, it builds new fabs that are finished bu 2008. Older 65nm ones will continue to produce 65nm stuff for a while before they are converted.


jeach!
"You 'may' be right about AMD's financials, but if AMD can show early demos and benchmarks that will cripple Intel's server sales. Intel will be forced to significantly lower its prices which will affect its overall ASP's."

Well, Intel demoed its 45nm quadcores and let people play with them. Where are real-world Barcelona demos? After all, Barcelona should be availiable much sooner than Intel 45nm CPU's

penix
"Intel laptops are exploding, and that's all you need to recognize"

As much as these are Intel laptops they are laptops with Sony batteries. It's not Intel's fault Sony hasn't used those batteries with AMD cpu's.

thekhalif
"3. Barcelona has dual FP, dual SSE4, dual loads, enhanced prefetch, OoO loads(maybe stores), more branch history, larger fetch buffer."

What is different compared to Intel 45nm CPUs?

"Yeah, these claims may help those who currently have Opterons and may be debating about waiting for K8l or jumping ship to Intel."

I don't think there are too many companies with DDR2 Opterons who want to upgrade their CPU power that soon. Uusally servers aren't upgraded once per year but once after several years.

edward
"Can you give us an example where AMD fail to deliver a product the way it promised?"

4x4? When you consider bang per buck then it is so bad that it's scary.

12:40 AM, January 27, 2007  
Anonymous Anonymous said...

"40% will open a major gap that will take Intel three years to catch up."

"Buy a expensive Clovertown machine, and find it's about half the speed of the K8L"

"With this architecture, AMD will frag Intel by 3x at least and permanently neutralize the Intel threat."


So, AMD (who as we all know, never lie or exaggerate anything) claim K8L will be faster than Clovertown. Then you claim that it's going to be twice as fast, then go on again to claim that it's THREE TIMES faster? Giving absurdly inflated performance estimates is one thing, but you could at least try and do it with some consistency!

1:20 AM, January 27, 2007  
Anonymous Anonymous said...

http://news.com.com/Chip+companies+entering+their+metal+period/2100-1006_3-6153962.html?tag=st.prev
AMD and IBM have "metals", too. With immersion lithography, they may even be one step in advance regarding Intel's process (in lab conditions) ?

5:23 AM, January 27, 2007  
Anonymous Anonymous said...

Start the BK clock for Shari-jocksniffer.

7:05 AM, January 27, 2007  
Anonymous Anonymous said...

Just for your information IBM/AMD announced in december they would use ultra-low-K interconnect dielectrics for their 45nm proces, estimated to be used in 2008, when they will produce guess what 45nm chips.
At least know what you are talking about!
This just shows intel is behind! If they just now present a new technology, how long is it going to take to get everything perfect and sorted out??

source: http://home.nestor.minsk.by/computers/news/2006/12/1301.html

7:23 AM, January 27, 2007  
Anonymous Anonymous said...

"edward said...
"Can you explain why AMD only promises (= vaporware)"

Can you give us an example where AMD fail to deliver a product the way it promised?"

Hmmmm - better than 2 months after "launch" and the ONLY 65nm part you can buy from AMD is the x2 3600.

Dell shipping 4000+ or greater based on 65nm tech??? Go try and configure a system....

HP??? Nope.

Alienware. Miss again....

VoodooPC (Rauls a huge AMD Fanboy).... Nothing there either.

Given that it took about 6 weeks AFTER "launch" for the 3600+ at 65nm and still no parts greater than that I'd say that's failure...

Much like the douche who runs this blog...

7:57 AM, January 27, 2007  
Anonymous Anonymous said...

1. Barcelona was designed for the dual stressor technique (said to switch 40% faster than 90nm)

Let's skip the buzzwords. Is core frequency going down or not?

2. Intel successfully used L3 to get 10-20% increases.

Apples and oranges. Intel uses L3 to average down a larger memory latency. Besides, Woodcrest has 4MB of L3 for 2 cores and Barcelona will have what, 2MB for 4 cores?

3. Barcelona has dual FP, dual SSE4, dual loads, enhanced prefetch, OoO loads(maybe stores), more branch history, larger fetch buffer.

Now you're talking architecture.

4.65nm is naturally cooler.
That's not related to Barcelona. All chips will do that and 45nm is even cooler :-)

9:31 AM, January 27, 2007  
Anonymous Anonymous said...

Sharikou - I was wondering if you could do a blog about K8l fragging?

There are so many of your recent blogs on the AMD Q4 earnings and the obvious correctness of your 40% market share runrate prediction that it would be refreshing to see something on the K8l speculative performance.

11:59 AM, January 27, 2007  
Anonymous Anonymous said...

Looks like Intel revealed one of its tricks up their sleeves - high K/metal gate will be rolled out at 45nm node - this will put them 1 if not 2 generations of IBM (and therefore AMD as they don't do major process technology development) who doesn't intend to implement this technology until at least the 32nm node.

I read somewhere that AMD/IBM also have this technology, and immersion litography process scheduled for 45nm nodes, but Intel already talked about 22nm nodes ... so - race continues.

In my humble op., it will all get down to timing, marketing - and the smoothest upgrade path for the existing customers.

One is for sure imho: when K8L hits the shelves, C2D will go down in prices. Makes one wonder how will motherboard prices scale.

12:13 PM, January 27, 2007  
Anonymous Anonymous said...

AMD will have a paper introduction in late June with K8L and press samples. Do not expect production units to start appearing until August, general retail availability in late September , wide spread by December but still only 15% of total production output. AMD will not cross the 50% of production totals until the second half of 2008. AMD has already stated this to the motherboard suppliers. End of Subject.....

12:50 PM, January 27, 2007  
Anonymous Anonymous said...

" edward said...
Can you give us an example where AMD fail to deliver a product the way it promised? "

Brisbane and 4x4.....

12:52 PM, January 27, 2007  
Blogger IT Kitty Cat said...

I read the article and it seems to me it is more about marketing than anything else. I am looking forward for real benchmarks. I have a feeling that Barcelona will be a great product if it comes on time, yet, I still have my doubts about the 40% performance advantage about "current products".

Allen's tone reminds me of Intel's tone when they promised the 40% performance advantage that was coming with C2D.

I guess we will have to wait and see. But till then, everything from both sides is a vaporware.

4:30 PM, January 27, 2007  
Anonymous Anonymous said...

"Can you give us an example where AMD fail to deliver a product the way it promised?"

K8Late!

4:31 PM, January 27, 2007  
Anonymous Graham said...

The only reason I read this blog these days is for the humor factor. I used to try refuting the daily lies, spin and illogic but there is just too much bullshit spewed out by Sharikou that I can't keep up. Now it is just entertainment. I look forward to your backpedaling a lying once Q2'08 comes around and Intel is NOT bankrupt. Till then keep up the good humor "doctor".

4:48 PM, January 27, 2007  
Anonymous Anonymous said...

This blog is now just silly....hey I know, why don't we discuss the new Saw 3 movie??? Who in here thinks that Saw III is the best of the series? I would have to say though that Jason from Friday the 13th would probably frag them all though. "FRAG" by the way is a military term that means to destroy it also means to fragment i.e. I need to defrag my hardrive once a week because I run Vista.

9:03 PM, January 27, 2007  
Blogger hyc said...

Well, despite all the drivel, I think this is a good point:

...but...but...but... it's drop in upgradeable for current Opteron dual cores. So basically I should buy an Opteron chip now, use it for 6-9 months and then what, throw it away?


Though for different reasons. I think the switch to DDR2 was a year too early. There's still no real benefit, DDR2 pricing is still high and production is still low. If the K8L was a drop in to S939/S940 it would be a far better story. They would still have needed a transition phase, which would have been executed about the same way as the current 939->AM2 phase, just that they jumped the gun.

Another thing - the Intel fanboys are right - bandwidth just hasn't been that big a deal yet. There still aren't any CPUs that are using enough bandwidth to max out the DDR400 memory controller. As I said a while back, AMD should have sat out DDR2 and switched at DDR3.
Working with DDR2 in-house for research purposes, fine-tuning and preparing for DDR3, yes. Releasing a product - no.

As it is, there's a large user base on S939/S940 that may not wait to make the transition to K8L, simply because it may cost them as much as switching to Intel.

I might have upgraded my X2 3800+ to a newer X2 on S939, but not now. So it'll be the better part of a year before AMD has an opportunity to get more money from me.

12:21 AM, January 28, 2007  
Blogger Ho Ho said...

"I read somewhere that AMD/IBM also have this technology, and immersion litography process scheduled for 45nm nodes, but Intel already talked about 22nm nodes ... so - race continues."

http://www.realworldtech.com/includes/templates/articles.cfm?ArticleID=RWT012707024759&mode=print
So far, Intel’s key competitors in the world of device fabrication, the IBM/Toshiba/Sony/AMD partnership appear to have pushed back high-k dielectrics and metal gates to the 32nm process. Unfortunately for AMD, the 32nm process is not expected to enter production until mid to late 2010

On Thursday, Intel held a briefing to demonstrate their advances at 45nm. While most people have read that Intel recently booted four operating systems on the first silicon from the 45nm shrink of the Core microarchitecture (codenamed Penryn), what is more interesting is the underlying silicon that Penryn was fabricated on. Intel’s presentation described advances in their 45nm process; the first commercial manufacturing process to use high-k gate dielectrics and metal gate electrodes. The combination of the two will give Intel a significant advantage for the lifespan of their 45nm process, by substantially reducing leakage and improving performance.

So basically, with high-k Intel is about three years ahead of AMD.

12:57 AM, January 28, 2007  
Anonymous The Sheepshagger said...

Anonymous said...

Just for your information IBM/AMD announced in december they would use ultra-low-K interconnect dielectrics for their 45nm proces, estimated to be used in 2008, when they will produce guess what 45nm chips.
At least know what you are talking about!
This just shows intel is behind! If they just now present a new technology, how long is it going to take to get everything perfect and sorted out??


Gosh... Pretty impressive. IBM & AMD announced something in December (1 month ago calendar-wise) that they are going to release in 2008 on 45nm while Intel's announcement last week said Penyrn (due to be released in June/July on 45nm) will incorporate the new high-k and metal gate transistor. So by my count, I'd say it won't take them very long at all. I mean, they just demonstrated it on 5 different systems, are in the middle of shipping vendor samples, and are ramping up the process in their development fab and in the midst of constructing the 2 high volume fibs to run that node and new process a full year ahead of IBM/AMD.

What I read in the IBM/AMD release is that they are using immersion lithography (read $$$$$$$ re-tooling) for 45nm that Intel won't need until 22nm. What that says to me is that Intel has found a way to print pattern on existing and proven lithography tools for another 2-3 years before making that expensive leap. But don't kid yourself... you have to know Intel probably has been testing and perfecting immersion technology from various suppliers and will be ready for the transition.

Clearly Intel is not / was not behind. Their holding aces up their sleeves.

3:40 AM, January 28, 2007  
Anonymous The Sheepshagger said...

I've been doing some calculating:

Where X=S

Penix = Penis

Penis = Sharikou

Sharikou = Penix

3:43 AM, January 28, 2007  
Anonymous Anonymous said...

http://www.techworld.com/opsys/news/index.cfm?newsID=7065&pagtype=all

I have re-googled this article.

Q: If AMD/IBM share the process research, what really can we expect of new 65nm AMD cores (and 45nm when ready) ?

And why clock drop on Barcelona ?

9:59 AM, January 28, 2007  
Anonymous Anonymous said...

According to the cnet article,http://news.com.com/Chip+companies+entering+their+metal+period/2100-1006_3-6153962.html?tag=st.prev
it looks like intel if pretty far behind in production expertise compared to IBM and AMD.
Intels dry litho process is outdated and antique and will destroy most of the hafnium benefits.
Super charged pentium 3s, no random numbers, notebook fires, antique platforms,quarterly market share loses, another 11000 plus layoffs, stock in toilet, increasing inventories, and now dropping prices to compete with AMD, wont be long.
ho ho, just because you get numbers does not mean they are random, check out what makes a set of random numbers statistics 101.

12:27 PM, January 28, 2007  
Anonymous Anonymous said...

AMD has now 25% of marketshare (Q4 2006). This is Mercury Research.
Intel is going doooown, slowly but 30% is another goal to reach in 2007!

2:30 PM, January 28, 2007  
Blogger PENIX said...

Anonymous idiot said...

Penix how can you be soooooo narrow minded. If the sony battery CAUSED the laptop to blow up then how is it the cpu's fault. In reality it was the actual battery catching fire something completely independant of the cpu.

ho ho said...

As much as these are Intel laptops they are laptops with Sony batteries. It's not Intel's fault Sony hasn't used those batteries with AMD cpu's.

You fanboys have completely missed the point that I just stated. As I just told you before, it DOES NOT MATTER if it is Sony's fault or Intel's fault. It DOES NOT MATTER if it's the CPU, the battery, the motherboard, the LCD, the speakers, the keyboard or any other component. All you need to do it analyze what has already happened:

Intel laptops are exploding. AMD laptops are not exploding.

Listen up fanboys. No one cares who or what is at fault. Stop pointing the finger and wake up.

3:32 PM, January 28, 2007  
Anonymous Anonymous said...

For folks who think moving to immersion litho is an advantage....

Cost of 193nm dry litho tool ~20mil. Cost of immersion tool >$30Mil.

Why is using a more expensive tool a good thing as opposed to extending a tool? Unless you are trying to MAXIMIZE cost per wafer?!?!?

Given the choice between printing a 45nm feature (actually smallest feature on this technology will be in neighborhood of 30nm) with a new tool that has new potential defect modes, and is >50% more capital cost vs a tool that you have used for the 2 previous technologies... which would you choose?

3:56 PM, January 28, 2007  
Anonymous Anonymous said...

For folks who think moving to immersion litho is an advantage....

Cost of 193nm dry litho tool ~20mil. Cost of immersion tool >$30Mil.

Why is using a more expensive tool a good thing as opposed to extending a tool? Unless you are trying to MAXIMIZE cost per wafer?!?!?

Given the choice between printing a 45nm feature (actually smallest feature on this technology will be in neighborhood of 30nm) with a new tool that has new potential defect modes, and is >50% more capital cost vs a tool that you have used for the 2 previous technologies... which would you choose?

3:57 PM, January 28, 2007  
Blogger Dr. Yield, PhD, MBA said...

some anonymous twit uttered while drooling on himself:
it looks like intel if pretty far behind in production expertise compared to IBM and AMD.
Intels dry litho process is outdated and antique and will destroy most of the hafnium benefits.


What the @#%$! are you talking about?
Can you explain to me how the electrical benefits of hafnia dielectrics is offset by the lower k1 and resultant improved CpK of immersion litho techniques?
Can you explain to me the negatives for implementing immersion lithography (not including the new defect modalities)?
How about the COO (cost of ownership if you didn't know) drawbacks resulting from higher equipment prices, more expensive resist stacks, lower TPT on immersion tracks, and more expensive service compared to dry tracks?

Let me give you a hint- next time you want to play buzzword bingo when slamming a company's technological prowess, you might want to know a little something about the subject.

As to dry litho being antique, immersion microscopy has been around for well over a 100 years. The only reason people are breaking down and applying the technique is because they don't have any other choice. Intel clearly has figured out a way to achieve production worthy yields and CD control for 45nm logic that they find acceptable. Will it give them the same theoretical capability as immersion? No. Might they equal or better immersion in reality? Yes. Will you actually learns something and maybe contribute something of substance to this discourse? I can only hope.

10:16 PM, January 28, 2007  
Anonymous Anonymous said...

Well the good news is that it doesn't seem that long to find out how good K8L, April to June. This is a lot better than my expected July-September I was thinking for intial benchmarks.

Maybe it will be close to the Q1 earnings release if they are pretty bad so as to put some sugar on things ..that would be great!

10:55 PM, January 28, 2007  
Blogger Ho Ho said...

some anonymous guy obsessed by random numbers
"ho ho, just because you get numbers does not mean they are random, check out what makes a set of random numbers statistics 101."

I'll repeat for N'th time: please share the benchmarks you used so I could verify them. I can test them on P4, C2D, P3, P1, Athlon XP and s754 K8, all running 32bit Linux. You have my contacts but you are unwilling to use them. Too bad for you.

Btw, if Barcelona is going to frag Intel then why aren't there any real-world CPU's shown? Intel has shown its 45 nm dual and quadcore laptop, server and desktop CPU's running real-life applications, months ahead of predicted release date. All AMD has shown is task manager and simulated performance numbers.

http://enthusiast.hardocp.com/article.html?art=MTI2OCwxLCxoZW50aHVzaWFzdA==

4:27 AM, January 29, 2007  
Anonymous Anonymous said...

Does this mean paper launch in April or actual meaningful chip availability?

Yes it will, like Intel paper lunched Core 2 Duo.
AMD with that will cut down the profits of Intel!

Intel will have to lower their prices to levels that will put them in Bankrup curse!

4:44 AM, January 29, 2007  
Anonymous The Sheepshagger said...

Holy freaking moron, Batman!!!

Some Penix (where x=s) said:

You fanboys have completely missed the point that I just stated. As I just told you before, it DOES NOT MATTER if it is Sony's fault or Intel's fault. It DOES NOT MATTER if it's the CPU, the battery, the motherboard, the LCD, the speakers, the keyboard or any other component. All you need to do it analyze what has already happened:

Intel laptops are exploding. AMD laptops are not exploding.

Listen up fanboys. No one cares who or what is at fault. Stop pointing the finger and wake up.


If no one cares who is at fault, why in the name of Zeus' Butthole are you saying "Intel laptops" are exploding?!

You can't purchase an "Intel notebook" nor can you purchase an "AMD notebook" as they are only a vendors of components in the same manner that Sony is, who has multiple times released statements accepting blame for fires in laptops containing Sony made batteries.

Your statement is only correct with the following statmenent:

"Laptops from (insert the following OEMS: Dell, HP, Toshiba, Acer, Sony, Lenova) are/were catching fire due to exploding Sony batteries."

If it doesn't matter who is to blame (in your own words) why trip over your dick to make Intel look bad and make yourself sound like a moron?! This story was over and done with long ago... Quit spreading FUD.

4:54 AM, January 29, 2007  
Anonymous Anonymous said...

"Intel laptops are exploding. AMD laptops are not exploding.

Listen up fanboys. No one cares who or what is at fault. Stop pointing the finger and wake up."


I see what you're saying, but you're still completely wrong. Ordinary Joes who see this on the news aren't gonna say "Oh, those laptops with Intel chips are blowing up, so I'll get one with an AMD chip," they're gonna say "Oh, those Dell laptops are blowing up, so I'll get an HP one." What processor the laptops use isn't gonna cross their mind - the news reports hardly even mention them! It's like saying that people are gonna blame ATI for the high failure rate of the early Xbox360s, therefore the PS3 is bound to beat the 360!

6:26 AM, January 29, 2007  
Anonymous Not penix said...

----->Intel laptops are exploding. AMD laptops are not exploding.

Listen up fanboys. No one cares who or what is at fault. Stop pointing the finger and wake up. <-------


Ok so if i'm finger pointing what are you doing. You seem to be caught up in your fanboi tactics yourself.

Just becasue a laptop blows up and happens to have an intel chip doesn't make intel bad (plz clarify why i'm wrong) If you were to put one of those sony batteries (pretending it was the right model #) into an AMD laptop the same thing would happen, hence amd laptop blowing up, would you still be spewing the same garbage. Once again i'm not a fanboy i'm trying to understand how its intel's fault for a sony problem.

1:06 PM, January 29, 2007  
Anonymous TheKingRich said...

Either way... the winner will be

ME!

I'm going to be able to buy processors that can perform the pants off anything 3 years ago, for a fraction of the cost!
WOOHOO!

3:26 PM, January 29, 2007  
Anonymous Anonymous said...

Yes it will, like Intel paper lunched Core 2 Duo.
AMD with that will cut down the profits of Intel!


That's strange. On launch day I got my Core 2 Duo E6600. Stock was limited, but it was available. Of course the supplies are going to be limited and high in demand when you have an awesome new CPU that everyone wants. AMD is the king of paper launches and vaporware. 65nm? Announced in December.. took two months to be able to buy one in retail. R600? Nowhere to be seen. K8L? We've got some marketing crap from Randy Allen, and a demonstration from AMD running Task Manger and some "simulated" performance gains.

5:37 PM, January 29, 2007  
Anonymous Anonymous said...

Quote ho ho: "So basically, with high-k Intel is about three years ahead of AMD."

Hey, ho ho, I heard that you tried to post this same crap at RubyWorks and got OWNED!
IBM are ALREADY using the process at their East Fishkill Fab and it will be used at the 45nm node in 2008, but you already know this, so stop spreading FUD.
See here

6:03 PM, January 29, 2007  
Anonymous Anonymous said...

"Allen's tone reminds me of Intel's tone when they promised the 40% performance advantage that was coming with C2D."

Intel promised 40% better than their previous products and it delivered. It was only the stupid press and fans (on both sides) who re-interperted what Intel said to mean 40% greater than their competitor. Every time I have asked someone for a story/link where INTEL claimed 40% better AMD, no one has been forthcoming.

On the other hand VP Allen clearly stated 40% better than Intel products in the recent interview, so I guess we'll see...

6:24 PM, January 29, 2007  
Anonymous Anonymous said...

Lookie here..

INTEL annouced the biggest thing in semiconductors in 40 years. 45nm technology, HighK, Metal gate, 45nm CPU on track for 2007 production in 3 factories.

http://www.intel.com/technology/silicon/45nm_technology.htm


AMD can't even get a wet dream going on 45nm let along deliver the good at 65nm.

BK, I'd like a double cheese burger with extra onioins. Hey is that you Sharikou? I guess your PhD was a internet special heah?

7:02 PM, January 29, 2007  
Blogger PENIX said...

In response to "The Sheepshagger", "not penix" and an Anonymous coward:

Since you people still cannot get a clue, I'm going to walk you through the logic, step by step.

We know the following:
1.) Sony provides the batteries for dozens of OEM companies
2.) These OEM companies provide both AMD and Intel laptops
3.) Both the AMD and Intel laptops use the Sony batteries
4.) Sony has claimed responsibility for the exploding batteries
5.) Sony batteries have only exploded in laptops which are Intel based


Batteries should not explode under ANY normal usage conditions. This is an obvious flaw in the battery, but why only Intel laptops? Why hasn't a single AMD laptop exploded? Something is triggering the flaw in the battery, and it is only present in Intel laptops. It could be the CPU, the motherboard, the charging circuit, etc. They have not released enough info to make this judgment. This is why we can only make conclusions based on what we know for certain: Intel laptops explode. AMD laptops do not explode.

If you had a stick of dynamite (Sony) duct taped to your penis, which would you rather have thrown at you? Unlit matches (AMD) or lit matches (Intel)?

9:55 PM, January 29, 2007  
Anonymous Anonymous said...

Penryn and Yorkfield will frag Barcelona and Nehalem will finally put AMD to its grave. I predict AMD BK in Q1'08.

9:19 AM, January 30, 2007  
Anonymous ed said...

"Hmmmm - better than 2 months after "launch" and the ONLY 65nm part you can buy from AMD is the x2 3600."

You Intel fanboys just can't act better than FUDers, like the company itself.

Less than a month of Brisbane release, it's available to the channel all the way to 4800+ (search "Brisbane" on that page).

11:23 AM, January 30, 2007  
Anonymous enumae said...

Penix
Since you people still cannot get a clue, I'm going to walk you through the logic, step by step.

It happened in Macs as well while using IBM chips.

What does that tell you?

Sony accepted the blame, paid for the replacements along with the other PC manufacturers, not Intel.

Can you accept the fact that Intel didn't pay for the replacements?

If you can, you should then realise that it is in fact do to the Sony batteries, and not related to Intel.

Dell laptops blew up with Sony batteries, or whom ever, but to try and link it to Intel would be untrue.

4:27 PM, January 30, 2007  
Anonymous Not Penix said...

Ok fair enough, but since there hasn't been enough information released why bother saying that it could be Intel's fault, just leave it at "laptops are exploding with sony batteries"

7:59 PM, January 30, 2007  
Anonymous Anonymous said...

"Less than a month"? Since when is January 31st "Less than a month" after December 5? That's nearly TWO months.

Nvidia and Intel do it right:- Geforce 8800, hard launch. Core 2 Duo, hard launch. Both with limited availbility, but the stock was there nonetheless.

8:14 PM, January 30, 2007  
Anonymous Anonymous said...

"ed said...
"Hmmmm - better than 2 months after "launch" and the ONLY 65nm part you can buy from AMD is the x2 3600."

You Intel fanboys just can't act better than FUDers, like the company itself.

Less than a month of Brisbane release, it's available to the channel all the way to 4800+ (search "Brisbane" on that page).

11:23 AM, January 30, 2007"

And you're an idiot. The day I posted that those processors were *not* on newegg - I spend half my life on that site. Those were posted in the last week.

Also, these procesors were announced more than month ago r-tard - closer to two months now. So please - talk shit to someone else...

9:33 AM, January 31, 2007  
Anonymous ed said...

"And you're an idiot. The day I posted that those processors were *not* on newegg - I spend half my life on that site."

Wow, sad you. :-D

I knew people got hold of a Brisbane (better than X2 3600, BTW) right after the new year. Guess they are more capable than spending half their lives on newegg. :-)

I linked newegg only because it happens to have those chips up to X2 4800 the first time I cared to look. What were you claiming then? No better than 3600? Truly a retard...

4:41 PM, January 31, 2007  

Post a Comment

Links to this post:

Create a Link

<< Home