Will Intel's 45nm send AMD to the grave?

[Viditor]

Originally posted by: dexvx
Half the posts in this are fanboism fiction and the sad thing is they don't even realize it.

Originally posted by: Diasper
You do realise cost is not just about what nanometer process level but what yields of maturity you have.

I remember reading an article that AMD always have very high yeilds - and that they always aim to launch into a new process with mature yields.

Intel I suspect will just launch into a new process flinging as much $ at it as possible - of course that doesn't necessarily mean it's cost-efficieient or mean that you are producing low cost chips if yields aren't as great. Perhaps taking a more cynical attitude moving to a lower process and trumpetting it is at least good for your stock price and all its manipulations.

Such as this statement. No company will put an immature process into mass production. No company.Some people just cannot accept the fact that Intel has better fab technology.

No company has ever launched a chip at fully mature yields to date...the closest that I am aware of was AMD's 90nm launch. A mature yield means that you are producing chips at the same percentage yield on the new node as you were on the previous node. I don't really know what you mean by an "immature process"...

 

[carlosd]

Originally posted by: fatty4ksu
The end of AMD is here.

Intel has stopped joking around, it over AMD.

And mods should stop joking around with you. You should be banned for saying such ridiculous stuff. I am sorry for you, how can somebody be so dumb to say that kind crap?

 

[Hacp]

Originally posted by: Viditor

Originally posted by: dexvx
Half the posts in this are fanboism fiction and the sad thing is they don't even realize it.

Originally posted by: Diasper
You do realise cost is not just about what nanometer process level but what yields of maturity you have.

I remember reading an article that AMD always have very high yeilds - and that they always aim to launch into a new process with mature yields.

Intel I suspect will just launch into a new process flinging as much $ at it as possible - of course that doesn't necessarily mean it's cost-efficieient or mean that you are producing low cost chips if yields aren't as great. Perhaps taking a more cynical attitude moving to a lower process and trumpetting it is at least good for your stock price and all its manipulations.

Such as this statement. No company will put an immature process into mass production. No company.Some people just cannot accept the fact that Intel has better fab technology.

No company has ever launched a chip at fully mature yields to date...the closest that I am aware of was AMD's 90nm launch. A mature yield means that you are producing chips at the same percentage yield on the new node as you were on the previous node. I don't really know what you mean by an "immature process"...

Intel's 65nm. Read it in a THG article.

 

[Aenslead]

Originally posted by: biostud
Doesn't IBM have a partnership with AMD in chip production.

Yes, I beleive they do.

 

[Viditor]

Originally posted by: Aenslead

Originally posted by: biostud
Doesn't IBM have a partnership with AMD in chip production.

Yes, I beleive they do.

IBM and AMD have had a joint development team at East Fishkill, NY since 2003. They are developing 65nm, 45nm, and 32nm processes for SOI...

 

[Viditor]

Originally posted by: Hacp

Originally posted by: Viditor
No company has ever launched a chip at fully mature yields to date...the closest that I am aware of was AMD's 90nm launch. A mature yield means that you are producing chips at the same percentage yield on the new node as you were on the previous node. I don't really know what you mean by an "immature process"...

Intel's 65nm. Read it in a THG article.

Not that I ever believe THG, but that is possible...could you find a link for the article when you get a chance?

Cheers!

Edit: BTW, did they say fully mature yields based on percentage or volume? At 65nm, volume would require only a 40-45% yield to equate to an 80% yield on 90nm...

 

[dmens]

BTW, did they say fully mature yields based on percentage or volume? At 65nm, volume would require only a 40-45% yield to equate to an 80% yield on 90nm...

Consider the increase of dual core parts in total 65nm shipments, plus the fact that afaik there is less 65nm capacity compared to 90nm at launch, due to chipsets and other "lesser" parts. Also, the 90nm supply dearth was a design issue, not production.

 

[Wahsapa]

Originally posted by: Dadofamunky

Originally posted by: dmens
Didn't an overclocked yonah do well against FX-60? yonah @ 2.7ghz comes close on power consumption to a stock FX-60.

http://www.oc.com.tw/article/0601/readocarticle.asp?id=4895

Yeah, that's a great link, doesn't display in Explorer without crashing and probably installs a malware program on your PC. Sorry, I'll wait until the pros wade in with their assessments.

visit taiwanese websites much?

 

[hans007]

Originally posted by: Viditor

Originally posted by: dexvx
Half the posts in this are fanboism fiction and the sad thing is they don't even realize it.

Originally posted by: Diasper
You do realise cost is not just about what nanometer process level but what yields of maturity you have.

I remember reading an article that AMD always have very high yeilds - and that they always aim to launch into a new process with mature yields.

Intel I suspect will just launch into a new process flinging as much $ at it as possible - of course that doesn't necessarily mean it's cost-efficieient or mean that you are producing low cost chips if yields aren't as great. Perhaps taking a more cynical attitude moving to a lower process and trumpetting it is at least good for your stock price and all its manipulations.

Such as this statement. No company will put an immature process into mass production. No company.Some people just cannot accept the fact that Intel has better fab technology.

No company has ever launched a chip at fully mature yields to date...the closest that I am aware of was AMD's 90nm launch. A mature yield means that you are producing chips at the same percentage yield on the new node as you were on the previous node. I don't really know what you mean by an "immature process"...

even amd's 90nm process neeeded another revision to get venice out.

it was just a shrink before that and did them almos tno good. so it wasnt a fully finished process ever.


i have no idea what type of crack you guys are smoking like AMD has been this shining light of all cpus. i know the enthusiast community loves them, and most of that was borne out of havnig cheaper CPUs. i have bought a ton of amd chips in the past for that same reason, but at this point their CPUs at many price points actually cost more than an equivalent intel platform.

when the k6-266 .25 came out , they resorted to shipping CPUs directly out of their prototype fab line in san jose because they did not have anything coming out of fab25.

this is a pretty cyclical business, but intel's track record is waaaay better than amd as far as fabs go. amd's .35 node was awful too, the k6-233 was an overvolted by 15% space heater of a cpu.

intels main advantage is they have so many fabs that they can prototype a process at one fab and replicate it at the 6-7 other fabs they have since all the plants are identical. amd has much much less trial and error room, not to mention every time they build a new plant they never transition the older ones over, since they do not do identical plants anymore and do not want to buy the equipment.

intel is not going to lay down and die, they slip up here and there and amd takes a lead (like the p60 fpu bug, or the coppermine problems which gave the k7 a window). but they always come back.. when the williamette sucked, the northwood came out.

i mean it is very much a nvidia vs ati type battle, in the same way that ati is sometimes in the lead, but more often than not nvidia hsa the lead longer.

 

[Viditor]

Originally posted by: hans007

even amd's 90nm process neeeded another revision to get venice out.

it was just a shrink before that and did them almos tno good. so it wasnt a fully finished process ever.


i have no idea what type of crack you guys are smoking like AMD has been this shining light of all cpus. i know the enthusiast community loves them, and most of that was borne out of havnig cheaper CPUs. i have bought a ton of amd chips in the past for that same reason, but at this point their CPUs at many price points actually cost more than an equivalent intel platform.

when the k6-266 .25 came out , they resorted to shipping CPUs directly out of their prototype fab line in san jose because they did not have anything coming out of fab25.

this is a pretty cyclical business, but intel's track record is waaaay better than amd as far as fabs go. amd's .35 node was awful too, the k6-233 was an overvolted by 15% space heater of a cpu.

intels main advantage is they have so many fabs that they can prototype a process at one fab and replicate it at the 6-7 other fabs they have since all the plants are identical. amd has much much less trial and error room, not to mention every time they build a new plant they never transition the older ones over, since they do not do identical plants anymore and do not want to buy the equipment.

intel is not going to lay down and die, they slip up here and there and amd takes a lead (like the p60 fpu bug, or the coppermine problems which gave the k7 a window). but they always come back.. when the williamette sucked, the northwood came out.

i mean it is very much a nvidia vs ati type battle, in the same way that ati is sometimes in the lead, but more often than not nvidia hsa the lead longer.

Huh? Venice was a new stepping...nothing to do with the manufacturing process. The original 90nm also included the strained silicon.

As to the rest, AMD's biggest market is nowhere NEAR the enthusiast market...it's the server segment (by a LONG shot!). For the desktop, I can't think of a single Intel chip (currently) that performs anywhere near as well as an AMD in the same price range, let alone for less. (Please post any examples you can think of, as I'm willing to learn).

For Fabs, yes Intel uses the "copy everything" method...this is both good and bad. For large scale production, it reduces costs. For quick changes, it's far more cumbersome and can actually cost much more.

 

[Viditor]

Originally posted by: dmens

BTW, did they say fully mature yields based on percentage or volume? At 65nm, volume would require only a 40-45% yield to equate to an 80% yield on 90nm...

Consider the increase of dual core parts in total 65nm shipments, plus the fact that afaik there is less 65nm capacity compared to 90nm at launch, due to chipsets and other "lesser" parts. Also, the 90nm supply dearth was a design issue, not production.

I haven't seen an increase in dual core 65nm part shipments (with the exception of Yonah), though I am not privy to Intel's inside numbers. If you mean that with the addition of Yonah, there are more dual cores going out the door then I suppose so. But the footprint on Yonah is pretty small and I can't see that as taxing 2-3 300mm 65nm Fabs...is there something you know that we don't?

 

[Griswold]

What is this? Instead of clock speed race we now have a die size race to predict doom for one or another company? Thanks for giving me the first smile of the day.

 

[Zebo]

Originally posted by: AnandThenMan

Originally posted by: ZeboAMD will pay for thier spend thrift come late this year early next and definity when 45nm rolls from intel. Those chips cost 1/4 to make than 90nm. $$profit$$ Add in the fact intels newer chips will be fast and cool creating more demand I see AMD selling $40 chips again like t-bred days in order to survive.

That's one of the most stupid things I have ever read in this forum.

Seems you didnt read the the tread then - the OP predicted death for AMD - I'm only talking about AMD's going to have to scale back their "ambitions" when intel gets back on track.

 

[A554SS1N]

Who reckons their's a load of Intel Paid Fanboy's in this thread purposely provoking others, and coming up with crap like 'the end of AMD is close'...? Guys, seriously, get a life.

I hope the mod's close this thread because some of the comments are just plain silly and useless.

 

[Griswold]

Originally posted by: Zebo

Originally posted by: AnandThenMan

Originally posted by: ZeboAMD will pay for thier spend thrift come late this year early next and definity when 45nm rolls from intel. Those chips cost 1/4 to make than 90nm. $$profit$$ Add in the fact intels newer chips will be fast and cool creating more demand I see AMD selling $40 chips again like t-bred days in order to survive.

That's one of the most stupid things I have ever read in this forum.

Seems you didnt read the the tread then - the OP predicted death for AMD - I'm only talking about AMD's going to have to scale back their "ambitions" when intel gets back on track.

And that is why? Because you think you know when intel has their 45nm ready for prime time? Because you think you know when AMD has their 65nm ready for primetime? I agree with what he said, your line could as well be coming from the intel FUD department.

 

[AnandThenMan]

Originally posted by: Zebo

Seems you didnt read the the tread then - the OP predicted death for AMD - I'm only talking about AMD's going to have to scale back their "ambitions" when intel gets back on track.[/quote]

I read every post in this thread. You are falling into the same, tired old trap that every Intel fanboy seems to (which is surprising considering you are one of the last members here I expected this from) ....When Intel does this, that, and the other thing to "get back on track" AMD will be devastated and back to selling CPU's for next to nothing.

AMD is being very smart IMO, and is preparing to dominate the x86 server space even more. They are going to continue to eat into Intel's precious server profits. If anyone is going to be reducing prices to move product it is Intel not AMD.

 

[thereaderrabbit]

as if 45 nm were everything...

AMD will be just fine.

 

[StrangerGuy]

Back in the Celeron-A days, Intel didn't killed off AMD despite having superior CPUs and chipsets, so I don't think the same thing is going to happen in the future.

 

[hondAS2ooo170]

Originally posted by: Mik3y
wrong forum. of course amd will be able to match up with intel.

wat he said

 

[Zebo]

Originally posted by: AnandThenMan

Originally posted by: Zebo

Seems you didnt read the the tread then - the OP predicted death for AMD - I'm only talking about AMD's going to have to scale back their "ambitions" when intel gets back on track.

I read every post in this thread. You are falling into the same, tired old trap that every Intel fanboy seems to (which is surprising considering you are one of the last members here I expected this from) ....When Intel does this, that, and the other thing to "get back on track" AMD will be devastated and back to selling CPU's for next to nothing.

AMD is being very smart IMO, and is preparing to dominate the x86 server space even more. They are going to continue to eat into Intel's precious server profits. If anyone is going to be reducing prices to move product it is Intel not AMD.

[/quote]
Expected?

I'm not bias in any way - indifferent to AMD or intels fortunes. I am saying with a cheaper processes - evenly matched chip - lower power chip - better name brand - intel will smack down AMD next year. If you think that's the stupidest thing you ever heard then you need to get out more and look at why people buy stuff.

 

[Viditor]

Originally posted by: Zebo
Expected?

I'm not bias in any way - indifferent to AMD or intels fortunes. I am saying with a cheaper processes - evenly matched chip - lower power chip - better name brand - intel will smack down AMD next year. If you think that's the stupidest thing you ever heard then you need to get out more and look at why people buy stuff.

I guess the point is that you're assuming an awful lot from Intel and very little from AMD...

 

[dmens]

I guess the point is that you're assuming an awful lot from Intel and very little from AMD...

Considering how AMD's 90nm is being stretched to the limit and their 65nm isn't coming for another year, you tell me what new breakthroughs to expect from them this year.

 

[openwheelformula1]

nothing wrong w/ "stretching the limit of 90nm SOI" when you have better performance w/ lower power consumption at a cheaper price. Keep in mind I am not even Intel bias. Once Intel has better performace (both speed and efficiency) like they did w/ my 2.8ghz Northwood, then I'd jump all over it.

 

[sxr7171]

AMD is far too entrenched in the market now. Intel had their chance but they screwed up. Don't get me wrong, I think AMD will hurt, probably badly if they and IBM don't get their process technology up to speed, but they won't die.

If anything I'd like Intel to make a comeback in the enthusiast market so that we can have some competition and lower prices.

 

[Viditor]

Originally posted by: dmens

I guess the point is that you're assuming an awful lot from Intel and very little from AMD...

Considering how AMD's 90nm is being stretched to the limit and their 65nm isn't coming for another year, you tell me what new breakthroughs to expect from them this year.

Huh? I thought you knew that their 65nm was being released this year...and that we have no idea how far the 90nm is being stretched. Those are the kind assumptions I am talking about.