Keeps getting worse for intel..

[P4Man]

If this article is right:

<A HREF="http://www.theinquirer.net/?article=22498" target="_new">http://www.theinquirer.net/?article=22498</A>

Intel will be virtually standing still for yet *another* 12 months. Even on 65nm they will be launching 2.8-3.4 GHz dual cores and 3.0-3.8 GHz single cores, all still on 800 MHz FSB. The only significant difference between now/soon and a year from here, is 2 Mb caches.. wow..

Also, we've had 3 GHz P4s since 2002, WTF is intel thinking releasing 3 GHz P4s chips <b>four</b> years later ?? while not quite the same, still, between 2000 and 2004, we have seen a ~4-5 fold clock increase ! If this keeps up much longer, we might actually see AMD catch intel on clockspeed.

Now to add injury to insult, thermals don't look all that .. hmm. hot.. either: <A HREF="http://translate.google.com/translate?u=http://www.x86-secret.com/index.php?option=newsd&nid=869&langpair=fr|en&hl=en&ie=UTF8" target="_new"> this article </A> claims the single core 65nm Cedar Mill will still have a TDP of 86W for a single core.. ouch ! No wonder Pressler won't clock any higher.

As it is, even todays AMD FX55, a <b>130nm</b> product, seems on par, if not superior over intels 12 month out <b>65nm </b> products. AMDs 90nm athlons will spank them, and just imagine what could happen if indeed AMD realises its goal to ship 65nm in Q1 2006..

Time to sell that intel stock, I tell you.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[o1die]

No disrespect intended, but when's the last time Intel lost money? Amd just reported a 46 million loss for the quarter. Got to give Intel credit for selling product, even if some of it is inferior.

 

[Rugger]

One of the reasons Intel is still making money is the diversity in their product line and the fact that they are the largest supplier for OEM builders.

__________________________________________________
<font color=red>You're a boil on the arse of progress - don't make me squeeze you!</font color=red>

 

[Flinx]

Until AMD cracks the OEM market (DELL), that is to say corporate customers yell AMD, Intel will stay PHAT.

Course it can't last forever but you have to maintain your lead for quite awhile. Corporations refresh their computers only every 3-4yrs. Now some of that may be on a yearly cycle if their smart.

Its like the US $$$ maintaining it's status as the prime currency although they have huge trade deficits.

Intel maintains its' status as the prime currency of computers even though they're sucky (no insult implied to the USA)

If AMD does manage to up clock speed with the technolgy they got from IBM, strained silicon, they might do it. This was supposed to up clock speeds by 25%. Where the heck is that 25% ?

The loving are the daring!

 

[P4Man]

I didnt claim intel would start losing money shorty, did I ? But if their MPU products get this far behind AMDs, it will most likely impact their ASPs, and therefore their bottom line, which should result in a (further) drop of their stock value.

As for AMD, their CPU business actually had an extremely good quarter, best ever and they reporter YoY sales increases over 30% and a net profit. However their flash business did terribly, resulting in an overall (smallish) net loss. But their flash business (spansion) is about to IPO'd, leaving AMD just with their MPU business.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[P4Man]

>One of the reasons Intel is still making money is the
>diversity in their product line and the fact that they are
>the largest supplier for OEM builders.

None of their diversified products bring them a penny, they are loss generating, the only business that makes them (tons of) money, is cpu's. You are quite right about the second argument though. But as opteron progresses (and it does quickly), I can imagine resistance against non Intel based desktops and laptops will also further reduce in the corporate market.

But AMDs biggest handicap is not "corporate stupidity", but simply the fact they lack capacity to guarantee a steady supply of enough parts to supply a Dell or HP with enough cpu's for their high volume products. That will probably change once FAB36 goes online next year (new fab, as well as 300mm), and even more when the deal with Chartered to outsource some of AMDs 90nm production becomes reality. Don't be surprised to see a whole lot more OEMs start using AMD in their popular products then, I'm not even sure I'd completely rule out Dell. AFAIK, AMD is about to triple their capacity in the next 18 months.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[Guest]

Im probably late but I just saw that intel's Pentium D will all be multiplier unlock...Thats probably the first good news in a while. Altough it mihgt not help all that much with these realllly high thermals its still a little plus...

Anyway I can only agree with you it keeps getting worst...The last good product I saw from intel was the P4C! And its been quite a while... The good news is that my P4C at 2.6 is probably the processor that I bought that will last me the longer, it's impressive to see just how well this 2+ years old CPU is holding its ground against intel's latest(affordable) cpu...

Asus P4P800DX, P4C 2.6ghz@3.25ghz, 2X512 OCZ PC4000 3-4-4-8, Leadtek FX5900 w/ FX5950U bios@500/1000, 2X30gig Raid0

 

[pat]

If AMD does manage to up clock speed with the technolgy they got from IBM, strained silicon, they might do it. This was supposed to up clock speeds by 25%. Where the heck is that 25% ?

Clockspeed is nothing compared to processing speed. If AMD increase clock speed, without valuable improvement on the processing speed, then that will not be a good option.

What AMD did good with the A64, is that they concentrate on a new architecture that could process more, while keeping it at lower clockspeed. If they could keep their core between 2-3 GHz, and still improve on processing, then it will be another winner.

High clock speed is not just hard to manage on the CPU, but on the board too. That increase cost of research and developpement, increasing price for motherboard. So it has to worth it

Having a 2ghz CPU that outperform a 3GHz one is good.. having a 2GHz cpu that could outperform a 4 GHz cpu will be much better. Havinf a 3.5GHz that could outperform a 8 GHz cpu would be heaven

<font color=red>Sig space for rent. make your offer.</font color=red>

 

[Mephistopheles]

This sucks. Without competition, there will be no need for AMD to rush their own R&D and product launches. We could be getting 2.6Ghz desktop parts by now (not FX-priced) and possibly 2.8Ghz by now if it weren't for Intel's lackluster performance.

That said, there still could be some improvements in Presler and Intel's 65nm tech which yield increased IPC - clock is not everything, mind you; IPC could increase. It terribly needed just that.

Not saying that it <i>will</i> happen, I'm just saying it <i>could</i>.

 

[P4Man]

when i wrote that post, I was going to add at then end "and now lets hear Mephistopheles spin all this positively, in his endless optimism". You dissapointed me only slightly

As for lack of competition; it only exists in the ultra high end, a market segment that is pretty much irrelevant for both companies for anything but PR value. There is plenty of competition below that, driving prices down. just look at smithfield, does anyone seriously believe intel would have launched these as affordable mid range cpu's if it wheren't for AMD killing them otherwise ? So, the competition still works for me, since I never buy topend anyhow

As for your second comment; I don't know. From what I read, its seems like Cedar Mill is just a die shrunk Prescott with increased L2. The only thing I could expect, is that they not only enlarged, but also improved L2 cache latency. This was increased fairly dramatically going from NW to prescott (probably to allow ultra high clockspeeds), and the main reason for prescotts disappointing performance, but since these ultra high speeds will apparently not be attained anyhow, it would make sense to focus more again on a lower latency cache a la Northwood, at the expense of clockability. Who knows, maybe they overdid it *again* and that explains the low clocks for Cedar Mill

A faster cache aside (which effect should not be underestimated, just look at Pentium M), I don't expect anything significant. maybe another handfull of new instructions called SSE-4 to make the thing marketable..

No, it just doesn't look good for intel. Especially not when <A HREF="http://www.c627627.com/AMD/Athlon64/" target="_new">this roadmap </A> is to believed.. 3 GHz K8s this year, 3.2 GHz dual cores next year.. ouch!

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[wolverinero79]

"None of their diversified products bring them a penny, they are loss generating, the only business that makes them (tons of) money, is cpu's"

I don't even know how to respond to this.

Without going into all the examples, chipsets make Intel quite a bit of money. Also, their communication parts, while sometimes positive, sometimes negative, are being packaged with their CPUs and chipsets and as a package interest more buyers. So you can't claim the CPU sales here are purely from desire for the CPUs by themselves - your recent comments about how AMD chips are superior would refute that.

Obviously people are buying from Intel because they are getting a benefit they don't get from AMD.

And since Intel is consistently making record revenues and profits Q on Q, Y on Y they have an incredibly strong business model that consists of a broad mix of products, good marketing, and definitely smart investment.

You know, for as much bad press Intel has received regarding Prescott and at one time HT (there were tons of skeptics regarding HT upon its release, and there still are) and with as much disdain as they get from techno-geeks for game performance, Intel has somehow managed to get both Prescott and HT enabled CPUs purchased by many people.

Also, quite a few of Intel's "missteps" or "failures" have been products that they see will not be money makers, so they drop them. Unless they are getting some benefit from a product, there's no reason for them to keep it, unless to support products that do make them tons of money. And it that case, it's difficult to differentiate the part from the package.

I'm just your average habitual smiler =D

 

[over_c]

None of their diversified products bring them a penny, they are loss generating, the only business that makes them (tons of) money, is cpu's

I believe that quote was referring to AMD, not Intel. AMD just posted a huge total loss due to their flash memory business. For AMD, only their CPUs are making them money.

 

[wolverinero79]

See that would make more sense. But after reading what he quoted, it seems like he's making the statement about Intel...

I'm just your average habitual smiler =D

 

[endyen]

A year ago, Intel was forcing the price of flash memory higher, saying it was needed to cover costs.
Now they have halved the price, making all flash memory producers loose money. Wonder why?
They are a much larger player in the flash game. That leads me to also wonder what the real state of thier flash division is. I would expect that they have covered the books by selling some very "high end" stock to another division, at an inflated price.

 

[apesoccer]

Or since AMD <i>was</i> making a lot of money off of that buisness...They're [intel] seeing AMD as real competition now, and are fighting tooth and nail. I read an article a while back on AMDs flash memory; they used to make very good money there. They're [intel] the indians in a 'circles the wagons' scenerio, and they're hoping AMD runs out of ammo first. And they were right, according to rumors circulating that AMD is getting out of the flash buisness.

Of course i could be wrong, since they aren't even the top 2 competitors in this buisness** [please see quote below](ie perhaps intel doesn't have enough <i>pull</i> to cause this). And AMD is the smallest of the top 4 companies overall. Samsung is huge...Intel is pretty big, and Toshiba invented Flash and is like a mini-Samsung (ie seems to have their hand in about every cookie jar...).

**edit: Correction...I don't know much here, so i'll just quote P4's statement.

...AMD/Spansion and intel are number one and two (or two and one as they switch positions regulary) in the NOR flash markets. Both are indeed relatively small players in the NAND, and overall flash markets though

Flash Mem:
<A HREF="http://www.theinquirer.net/?article=18526" target="_new">http://www.theinquirer.net/?article=18526</A>
<A HREF="http://www.theinquirer.net/?article=18177" target="_new">http://www.theinquirer.net/?article=18177</A>
<A HREF="http://www.theinquirer.net/?article=19416" target="_new">http://www.theinquirer.net/?article=19416</A>

<A HREF="http://www.theinquirer.net/?article=22516" target="_new">http://www.theinquirer.net/?article=22516</A>

...just call me...ANALOGY MAN...

Current machines running F@H:
AMD: [64 3500+][64 3000+][2500+][2000+][1.3x2][366]
Intel: [X 3.0x3][P4 3.0x2][P4 2.4x5][P4 1.4]

"...and i'm not gay" RX8 -Greatest Quote of ALL Time<P ID="edit"><FONT SIZE=-1><EM>Edited by apesoccer on 04/14/05 05:49 PM.</EM></FONT></P>

 

[P4Man]

Warning: Long rant below.. it has an I-rating, meaning intel fanboys are strongly cautioned not to read this without parental supervision.

>Without going into all the examples, chipsets make Intel
>quite a bit of money. Also, their communication parts,

Nope. Communications and flash is easy to debunk, read intels financial statements; ICG (Intel communication Group) has yet to earn a penny. These last three years, this division has posted a ~800M loss on~$5B sales. Imagine it would be IPO'd LOL ! Before that, I don't know, but there was never a profit in the last 10 years as far as I remember.

Chipsets and motherboards is harder to prove, since they mix those numbers with the CPU's. But from what I heard/read on investor CCs, these products on their own are not profitable for intel either. They are what intel calls "enablers".

Lastly, their consumer electronics group has been moved into the architecture group as well, hiding their losses on almost non existing sales (things like HDTV chipsets) behind their x86 cash cow. No one seriously believes those products are anywhere close to profitable as yet.

>So you can't claim the CPU sales here are purely from desire
> for the CPUs by themselves - your recent comments about how
> AMD chips are superior would refute that.

oh what does desire have to do with anything ? Sure, intels cpu sales are still very healthy, its margins more than fat enough to support their (logical) enormous overhead as well as the losses from all the other divisions/products and still post terrific profits.

>Obviously people are buying from Intel because they are
>getting a benefit they don't get from AMD.

People don't buy from intel, they buy from an oem or retailer. As to why, lots of reasons. i'll name just a few:

* violating antitrust laws in Japan and Europe, and most likely in the US as well, but the FTC doesn't seem to care
* $4B marketing budget (also includes administrative cost)
* enough fab capacity to supply the entire world and then some
* ignorant consumers
* some pretty decent products

>And since Intel is consistently making record revenues and
>profits Q on Q, Y on Y they have an incredibly strong
>business model that consists of a broad mix of products,
>good marketing, and definitely smart investment.

We shall see.

Lets see what happens in Europe and Japan when it no longer strongarms OEMs to not use AMD cpu's (or not more than 10% of their product mix). FWIW, in certain markets where this doesn't play a role, like the DIY market, AMD has a marketshare of >50%.

Lets see what happens with Lenovo, who, after the purchase of IBMs PC division has expressed ambitions to "outdell" dell, using "alternative" suppliers. Lenovo -including IBM- is the worldwide number three not far behind HP and Dell, and the clear number one in the booming Chinese market.

Lets see what happens when AMD is no longer capacity constrained as it is now, when the migrate eveything to 90nm, bring FAB36 online and outsource to Chartered. Think about it.. why would they sign a deal with Chartered when they are about to <i>triple</i> their own capacity with FAB36 ? What could possibly require such massive increase in production capacity.. hmm ? A hint: its not HDTV decoders

Lets see what happens when dual core opterons hit the market this year, and bring the cost of 4 and 8 core x86 systems down to only a fraction of their current costs/prices. Lets see how intels multibillion dollar Xeon business holds up with eroding ASPs and losing marketshare.

Lets see what happens when Sun and HP have been selling a broad opteron portfolio for more than a few months.

Lets see what happens when AMD spins of Spansion (flash business), so intel can no longer harass AMD by putting tremendous price pressure on their flash business.

Lets check back in 12 or 18 months shall we ? Somehow, I think Intels business model will no longer look so incredibly strong, and a lot of its skyhigh investments won't look any smarter than its guestimated $10B (!) dollar investment in Itanium to date.

intel is a huge company, with a lot clever people, with near endless resources and with more money on the bank than many african countries, so its not going anywhere anytime soon.

But its also a company whose shareholders have been used for decades to high growth rates, fat margins, huge profits and lucrative dividents. Thats the kind of expectation its current stock valuation is based upon.

Intel is also not a company that has ever sucessfully (profitably) competed on anything except CPU's. Worse, anything except <i>x86</i> cpu's with IPF going nowhere, relegated to a small and shrinking niche, and other previous fruitless attempts to move away from x86 (remember the i432 "micromainframe " ?). Thing is, in exactly in this x86 market, its all time cash cow, its seriously losing its edge. In fact, they lost it a couple of years ago, but now it seems, with no hopes of regaining it anytime soon. Its CEO is now saying it is going to transform intel from an cpu vendor, into a "digital consumer manufacturer". They have tried so often, and failed with each attempt. The only thing different, is that this time they may *HAVE* to transform the company if it wants to remain a highly profitable, cutting edge company.

Call me an AMD fanboy if you like, a Fudder, I don't care. But I will not be surprised to see intel going the way of DEC, Wang, SGI, Cray and other formerly proclaimed "untouchable" IT tech companies. In fact, I would be far more surprised if intel wouldn't. My only real question is: will it still happen this decade or not, and will they precede HP or will HP precede intel ? I think its going to be a close race..

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[P4Man]

AFAIK, AMD/Spansion and intel are number one and two (or two and one as they switch positions regulary) in the NOR flash markets. Both are indeed relatively small players in the NAND, and overall flash markets though.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[P4Man]

>And they were right, according to rumors circulating that AMD
> is getting out of the flash buisness.

Its not a rumour, its official.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[apesoccer]

Coo, i'll make the changes!

Current machines running F@H:
AMD: [64 3500+][64 3000+][2500+][2000+][1.3x2][366]
Intel: [X 3.0x3][P4 3.0x2][P4 2.4x5][P4 1.4]

"...and i'm not gay" RX8 -Greatest Quote of ALL Time

 

[apesoccer]

=P

Have they changed yet? I figure it's rumor until they actually let go of it.

...AMD sold $447 million in flash memory, but that wasn't nearly enough, so they racked up a $110 million operating loss which took them from an overall profit to a loss, again.

The biggest news was on the flash front with AMD filing for an IPO on the Spansion business. It looks like this is the first baby steps to spinning off the flash business so AMD can get back to doing what AMD does best...

=D I'm waiting on the kick in the rear for nit-picking. Your point is well taken.

Current machines running F@H:
AMD: [64 3500+][64 3000+][2500+][2000+][1.3x2][366]
Intel: [X 3.0x3][P4 3.0x2][P4 2.4x5][P4 1.4]

"...and i'm not gay" RX8 -Greatest Quote of ALL Time

 

[darko21]

I take it you are now re-invested in amd maybe some calls or leaps? (just a guess BTW) .. I made some money on my amd long position but not nearly the money I would have if I had sold in December. I dumped amd after the intel orchestrated flash disaster then rebought and caught a dead cat bounce for a 1.50 and exited because the intel dumping flash at a massive loss is just to risky for my liking. There really should be something illegal about intel selling flash at a billion dollar loss just to make amd look less profitable. Flash ='s 7% sales for intel but 50% for amd. I truly believe once amd spins of spansion intel will stop dumping nor flash at such a huge loss as it won't effect amd's bottom line anymore. Who knows but that's my guess.

When I exited amd after the dead cat bounce I stumbled across lexar and their court case with Toshiba. I made a lot of money on that.. lexar for a day or 2 posted the opening statements on the lexar web site I think the judge made them remove it. After reading both opening statements I could not see how lexar could loose. So I bet a lot of money on lexar and sold after the verdict.

I pulled almost all my money out of the market and bought a new truck and I am in the process of selling current home and buying much better home.

The small amount I still have in the market is now in lexar calls in june 05 (not advised very risky) If that pans out the way I hope I will buy amd leaps and insure it with intel puts just incase the market crashes 3rd world war whatever.

Not trying to bash intel here but I would not think buying intel right now a good idea although I have been wrong before and will be wrong again. Intel has name brand recognition and fully knows marketing. However Japan and Europe are not Intel’s only
<A HREF="http://www.reuters.com/newsArticle.jhtml?type=topNews&storyID=8184892" target="_new"> headache </A>

My personal opinion is with the spin off of spansion intel wont be able to hurt amd without hurting themselves a whole lot more do to lower asp’s in cpu’s. With fab 36 and charter windows x86-64 bit coming soon the clear advantage in higher asp DC server, work stations makes me think the future looks bright indeed for amd but if that dose pan out will take WS a while to figure it out. And Intel often has an ace up it’s sleeve but IMHO all Intel’s card have been played.

If I glanced at a spilt box of tooth picks on the floor, could I tell you how many are in the pile. Not a chance, But then again I don't have to buy my underware at Kmart.

 

[Jito463]

I'd have to agree with P4man's comments

FWIW, in certain markets where this doesn't play a role, like the DIY market, AMD has a marketshare of >50%.

Where I work (a small OEM/DYI store), we sell about 90-95% AMD, and only those few, misguided souls who refuse to run AMD still buy the Intels. Whatever floats their boat, though. If they want to spend more for less, it's their money. I'm not a fanboy, but having been using and/or repairing computers for about 14 years now, I've never once owned an Intel and never been sorry for it. And now with XP x64 about to go live and AMD essentially prepared to dominate the market (despite Intels feeble attempts to get in the game at the last minute), I foresee a very bright future for AMD. Heck, if Intel could produce a comparable CPU for less money than AMD, more power to them. I'd probably switch. Whichever gives me more for the money. But Intel is such a giant and spends so much on marketing that they can't afford to be losing money on their cash cow. Only time will tell what the future holds, though. But for my money, I see AMD in my computer for a very long time.

 

[P4Man]

I invested in neither for the moment, in fact pretty much in no stocks. What I ranted about might take a couple of years until the results would really show, and I currently can't invest more than some pocket change on such a long time horizon (need cash at hand for other plans). The irony of the stock market is that intel is seen as the industry barometer, if they go down, AMD usually goes along. Not sure how long it will take for the market to figure out AMDs situation coulc be quite different from intels.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[P4Man]

>'d have to agree with P4man's comments

You really don't <i>have</i> to
I'm fairly surprised no one seems to disagree any more on a rant where I basically predict the downfall of intel.

>I'm not a fanboy, but having been using and/or repairing
>computers for about 14 years now, I've never once owned an
>Intel and never been sorry for it

Well.. there have been times when AMD didn't make a whole lot of sense unless you where buying really lowend and even then. The K5 days come to mind, the end of the K6 period and early Athlon days (with a choice between 2 motherboards and 1.5 chipsets, one of which was terrible) and the late athlon XP days versus northwood. In the K5 days I owned intel (pentium 133), in the K6 days I even owned my first and last Dell (P3-500) and I'm still in the late AXP days

> But Intel is such a giant and spends so much on marketing
> that they can't afford to be losing money on their cash cow

Bingo. But to be fair, they could drop prices by 50% and still make money on cpu's, but they as a company, they couldn't support such drastically lower ASPs to pay for their overhead and to compensate their other losses. They can do this perhaps for one niche (xeon seems like a safe bet for dramatically lower ASPs to counter opteron), but not across their product line, and they are under serious pressure almost across their entire product line -"low end" dual core desktop chips and mobile chips aside, but low end only makes you that much, especially with a 2x the diesize, leaving them with mobile where they are pretty secure for now.. until Turion and windows x64 show up at least.

As for marketing here something to consider; Intels billion + (?) co-marketing campaign really is a clever trick, but its very much like a accounting trick comparable to mail in rebates as well: say intel fabs a given cpu for $30 sells it for $60, but in return the oem gets $15 comarketing dollars; what really happens is that it will publish a fat 50% margin and high $60 ASPs, and hide a good part of the cost of sale (the comarketing) in the $4B marketing budget. In reality its margins are only 25%, and its ASPs $45 making them much more vulnerable than you'd think at first glance, since they can't just cut their marketing budget without directly affecting sales, margins and ASPs. Now this all legal and all that (I assume), and it works, but it does give a wrong impression of intels business. It doesn't just spend such huge sums on marketing because it *can*, but because it *has to* as part of their sales cost.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

 

[Mephistopheles]

I'm fairly surprised no one seems to disagree any more on a rant where I basically predict the downfall of intel.

Oh, you actually <i>expected </i>us to disagree?...

In any case, I think Intel is a very strong company, which means its downfall would probably take some time, if it were to happen. But indeed, Intel's products have been sad little processors lately...

They'd have to keep doing sad little processors for quite a while to lose all of their resources. Hopefully, we'll get more respectable products from them at some point, when the company has felt the impact of change... maybe in 2006 or 2007, who knows. Hey, competition is good...