intel is.......

[cqcc]

part1
http://www.theinquirer.net/?article=19105
part2
http://www.theinquirer.net/?article=19110

where r all the intel fans website? i will post these links there, they should learn

 

[DAPUNISHER]

Interesting read, thanks I liked the close

Even if everything goes perfectly for AMD during this time, Intel is far from dead. AMD can provide about 30% of the market demand for X86 CPUs, so even if Intel came out with nothing from now to Merom, AMD would only double its market share. At the far ends of the relevant period, Fab 36 is set to come on line with a 65nm process. Then, and only then, will AMD have a chance to take more than half of the market.

But by then, everything will have changed. Up until late 2006, there is little Intel can do. 2006/2007 brings us Merom/Conroe, 65nm, CSI and the K10. We'll be starting from square one then. An Intel with something to prove versus a resurgent AMD should be great fun to watch. µ

That seems a good incapsulation of the situation facing both companies. AMD should be able to parlay the temporary advantage into some market share gains and increase their financial health substantially. Of course Intel isn't sitting on their hands either, and as the threat increases, so will their efforts to meet it.

For the present however, Intc hasn't suffered due to their mistakes, but has instead prospered. They have intelligently redirected their focus to other higher growth markets e.g. mobile and flash. Personally I'm stoked to see things going as they are, because now the fight is really ON :beer:

 

[iwantanewcomputer]

wow...good articles. really tore intel a new one, oh wait, intel tore themselves the new one and the enquirer is laughng at them.

i have a question, when is amd slated for any real archetectural changes, not just revisions of earlier cpus? it looks like they pretty much have been rolling with the same design since the early athlon days and just tweaking it for each new process shrink, and adding a memory controller. nothing like the different pentium lines or pentium m and itanium. i guess this could be cause they don't have the buget to screw around with other lines.

another thing. i've asked this before but got no conclusive answer. why doesn't intel just slap a memory controller onto a prescott? it seems like with the deep piped archetecture it would benefit a lot more than the k8 by the better bandwidth/latencies. intel's second biggest weapon with the netburst arch, besides speed, has been high fsb. they've got to realize the benefit an onboard controller would give them since the k8 came out.

 

[BW86]

good read

 

[DAPUNISHER]

Originally posted by: iwantanewcomputer


i have a question, when is amd slated for any real archetectural changes, not just revisions of earlier cpus? it looks like they pretty much have been rolling with the same design since the early athlon days and just tweaking it for each new process shrink, and adding a memory controller. nothing like the different pentium lines or pentium m and itanium. i guess this could be cause they don't have the buget to screw around with other lines.

They have certainly done more to the 64bit line compared to the early Athlon days than just add a memory controller and tweak for die shrinks unless you consider the 64bit features and support tweaks In response to your musings about why they haven't done anything as remarkably diverse in approach to their microprocessor line-up like Intel has, I'll refer you to the addage "If it ain't broke, don't fix it"

another thing. i've asked this before but got no conclusive answer. why doesn't intel just slap a memory controller onto a prescott? it seems like with the deep piped archetecture it would benefit a lot more than the k8 by the better bandwidth/latencies. intel's second biggest weapon with the netburst arch, besides speed, has been high fsb. they've got to realize the benefit an onboard controller would give them since the k8 came out.

Intc makes a great deal of revenue from their chipsets, why canabalize sales there by moving part of it to the CPU? If they add on the cost then the price/performance ratio becomes skewed further against them, so they would be facing a lose-lose situation by doing so from the CFO's perspective IMHO.

 

[iwantanewcomputer]

DAPUNISHER, i guess taht makes sense, but now it looks like they need the performance more than anything. i doubt they would loose much in chipsets, they'd probly just charge the same for something that does less or charge more for the processors

 

[DAPUNISHER]

Originally posted by: iwantanewcomputer
DAPUNISHER, i guess taht makes sense, but now it looks like they need the performance more than anything. i doubt they would loose much in chipsets, they'd probly just charge the same for something that does less or charge more for the processors

For the average desktop user a 3ghz HT DEll, Gateway, HP, ect. system is more than fast enough already, so performance isn't that big an issue for the time being. OEMs aren't going to rush to AMD anytime soon because Intel desktop systems aren't meeting their customers needs, Not that AMD could meet the demand if they did

If you look at the different chipsets for A64 you see very little variance in performance because of the memory controller being taken out of the equation. So features and price are the primary focus. AMD doesn't like being in the chipset game so this works out nicely.


Intel on the other hand, thrives on selling their CPUs and chipsets together. Put the memory controller on-die and performance+stability, the 2 things that have traditionally given them a clear advantage would be largely nullified. Looking at both your speculatory ideas for how they might approach going to the on-die memory controller I offer up the following for you to consider. Increasing the CPU price to help make up for lost revenue from the chipset divison would make AMD a more attractive solution than it already is. Trying to pass off a less feature rich chipset for similar money means OEMs *except perhaps Dell* will find competing solutions much more attractive because the performance+stability would likely finally be on par as with A64 chipsets, but the competing chipsets cost significantly less given the proposed Intel strategy.

I suppose they could refuse to license the rights to make supporting chipsets to anyone else but then they lose that revenue, and risk seriously undermining their relationship with these manufacturers. That could result in SiS, VIA, and ATI throwing their full weight behind AMD and put Intel "on an island surronded by enemies" if you will.

Of course this is just my speculation but the alternatives you mentioned don't make putting the controller on the CPU a good alternative for whatever performance gains it might yield. I'm certain since they fully cross-license everything that Intel is R&Ding in that area but taking away advantages of their chipsets, making it easier for others to design their chipsets for Intel microprocessors, and attempting to recoup the revenue in the ways thus far discussed doesn't strike me as being viable alternatives.

Intel has very deep pockets and can afford to be 2nd best while redirecting their efforts to a better roadmap, so I place their need for performance lower on the priorities list than you is all Cliffnotes: I believe it is far more necessary to AMD's financial health and future to lead in performance for the immediate future than it is for Intel.

 

[Zebo]

Good posts DP. I have some q's Will Conroe/Merlin have 64 bit support? Why does'nt Intel just buy AMD.. I mean 4 billion market cap, Intel could buy them and not even notice like us spending $20.

 

[DGath]

I admit it... I used to be a pretty big intel faboy, but over the past 18 months, the two companies are going in opposite directions. My 18 month old 2.8C will last me a little while, but come upgrade time, I'm probably leaning towards AMD at this point. Intel at this point is just living off their name, but then again... their name is worth billions.

 

[Matthias99]

Originally posted by: Zebo
Good posts DP. I have some q's Will Conroe/Merlin have 64 bit support? Why does'nt Intel just buy AMD.. I mean 4 billion market cap, Intel could buy them and not even notice like us spending $20.

AMD's more like $6B nowadays (at least according to a couple quick stock quotes), and I doubt Intel could get them for much under $10-15B in a forced buyout. While Intel *could* afford this (they're worth about $150B), that's still a lot of money. I mean, they only made $900M last year; $10B is nearly 10 years worth of profits for Intel, plus they'd undoubtedly spend a lot more trying to transition the fabs, etc. even if they just scrapped AMD's processor lines.

Plus, I *really* doubt the SEC would let them buy their only serious competitor, even if Intel thought it made financial sense.

 

[piasabird]

According to an article that I beleive was written by Anand Intel is really into strong arming OEM's into doing what they want. Intel probably is more of an Evil Monopoly than Microsoft ever was. They just do it quietly and carefully and quite often outside of the USA.

http://www.anandtech.com/mb/showdoc.aspx?i=2264&p=2

 

[Accord99]

Originally posted by: Matthias99
While Intel *could* afford this (they're worth about $150B), that's still a lot of money. I mean, they only made $900M last year; $10B is nearly 10 years worth of profits for Intel,

Intel made over $7Billion in net profits last year. But regardless, they have almost no chance of buying buy AMD without anti-trust issues in the US and Europe.

 

[Zebo]

Ya and intel has billions in bank and lots of stock it can sell, AMD is a joke in size about the same as nV or ATi..intel could swing all three and not notice. Assuming it made financial sence to buy AMD, and this argument could go both ways, anti-trust is the issue but why? You have MS with total control of desktop why would they stop Intel? Especially when IBM apple Sun and cyrix and others are out there?

 

[Cerb]

Originally posted by: iwantanewcomputer(snip)

another thing. i've asked this before but got no conclusive answer. why doesn't intel just slap a memory controller onto a prescott? it seems like with the deep piped archetecture it would benefit a lot more than the k8 by the better bandwidth/latencies. intel's second biggest weapon with the netburst arch, besides speed, has been high fsb. they've got to realize the benefit an onboard controller would give them since the k8 came out.

It would cost more money than in the chipset, gives reason for buying their chipsets instead of SiS or VIA, and oh, did I mention it owuld cost more? Part of their hold on big companies is that they can cheaply produce large numbers of chips. AMD, packing them like they are, cannot. They've got more punch, but it is still more expensive to slap 20 of them in a blade server than some LV Xeons. It would only be of benefit to Intel if they were to stop pushing their chipsets and motherboards. But that just cuts their protentials for profit.

AMD is doing well, and if they keep going, will only thrive, but they are doing it the way Intel just refuses to, by working with others in the industry. The way they are going, they should be around and competitive for years to come, but Intel still has the massive manufacturing capability that cannot be touched by anybody else.

 

[Matthias99]

Originally posted by: Accord99

Originally posted by: Matthias99
While Intel *could* afford this (they're worth about $150B), that's still a lot of money. I mean, they only made $900M last year; $10B is nearly 10 years worth of profits for Intel,

Intel made over $7Billion in net profits last year. But regardless, they have almost no chance of buying buy AMD without anti-trust issues in the US and Europe.

Ah, misread that source; that was $900M in profit for a *quarter* of 2003, not the whole year. The numbers you quoted are accurate.

Still, I think the antitrust argument is why they haven't tried to do anything about AMD. They're certainly a thorn in Intel's side (especially in the server segment right now, as Opteron is definitely a better MP solution than Xeon ATM), but trying to get rid of them would be more trouble than it's worth. Intel knows there's little risk of their being dethroned anytime soon.

posted by Zebo:

anti-trust is the issue but why? You have MS with total control of desktop why would they stop Intel? Especially when IBM apple Sun and cyrix and others are out there?

Intel and AMD are the only companies that produce CPUs for Windows-based computers (which is, if you haven't noticed, a pretty big market ). Getting rid of AMD would give Intel a monopoly in that market segment. They probably wouldn't let MS buy out Apple (their biggest competitor in desktop operating systems), either, for much the same reason.

 

[Zero Plasma]

Better then butter.

 

[RadiclDreamer]

Originally posted by: Matthias99

Originally posted by: Accord99

Originally posted by: Matthias99
While Intel *could* afford this (they're worth about $150B), that's still a lot of money. I mean, they only made $900M last year; $10B is nearly 10 years worth of profits for Intel,

Intel made over $7Billion in net profits last year. But regardless, they have almost no chance of buying buy AMD without anti-trust issues in the US and Europe.

Ah, misread that source; that was $900M in profit for a *quarter* of 2003, not the whole year. The numbers you quoted are accurate.

Still, I think the antitrust argument is why they haven't tried to do anything about AMD. They're certainly a thorn in Intel's side (especially in the server segment right now, as Opteron is definitely a better MP solution than Xeon ATM), but trying to get rid of them would be more trouble than it's worth. Intel knows there's little risk of their being dethroned anytime soon.

posted by Zebo:

anti-trust is the issue but why? You have MS with total control of desktop why would they stop Intel? Especially when IBM apple Sun and cyrix and others are out there?

Intel and AMD are the only companies that produce CPUs for Windows-based computers (which is, if you haven't noticed, a pretty big market ). Getting rid of AMD would give Intel a monopoly in that market segment. They probably wouldn't let MS buy out Apple (their biggest competitor in desktop operating systems), either, for much the same reason.

I know its easy to do, but you forgot VIA's line of cpu's

 

[formulav8]

I remember reading that a Intel Rep said that they ARE working on adding the memory controller to the cpu. So I think they will do it even if it means having to copy the little guy. I don't remember the article that he said it in. But he did say they are working on it. Just because it isn't on their roadmap doesn't mean they are not working on it



Jason

 

[Matthias99]

Originally posted by: RadiclDreamer
I know its easy to do, but you forgot VIA's line of cpu's

I didn't forget them -- let's be honest here; VIA is not a serious competitor to Intel in any way right now, at least in the desktop market. If they're hardly worried about AMD, VIA isn't even on the radar. I don't know of any Tier 1 OEMs (companies like Dell, HP, IBM, etc.) using VIA chips in their desktop machines.