Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
That's the first time I've heard of erasing by tunneling. It sounds like mirrored bit is more of a NAND solution than NOR? How many write cycles will mirrorbit support, I thought it was similar to floating gate.
I don't think many AMD investors are very happy, but I'll reserve judgement until after earnings and the CC. Somethings wrong with flash, what that is remains to be seen.
your right of course. That has been INTC's plan all along. The trouble is that the number of INTC centers able to support all the losing ones is shrinking all the time. I sure hope AMD gets Turion out there soon as anything that slows down centrino is good for AMD.
I would feel a lot better about AMD if it had lined up another source to produce all the 32 bit chips. I'm not sure AMD can force INTC pricing with just fab30. The best I think we can hope for is that AMD is able to produce enough chips that INTC has to react to AMD pricing. Hopefully the move to 90nm will help in that regard.
As far as flash goes I'm not that concerned about INTC as I view their tech as being inferior, but who knows what AMD's yeilds etc. are. From the way mirrored bit was described you would think it would have taken over the market by now. Or maybe it has but INTC is able to keep prices low? Hopefully the CC will clear things up?
Was it coincidence that INTC announced that X86 and IPF would merge in 07 about the time AMD sprang AMD64 on INTC. INTC was caught flat footed and as usual did a song and dance. There were rumors about a Yamhill that would have 64 bits, but it looks like this was nothing but a Prescott with extensions that didn't work. Personally, I think that's the reason Prescotts have so many transistors, they're filled with Yamhill junk that never went anywhere.
INTC blew it on X86-64. As a matter of fact I remember Otellini denying vociferiously that INTC even had an X86-64 project when he was asked about Yamhill. IPF was to be the next phase. X86 would be INTC's 32 bit solution and Itanium would be INTC's 64 bit solution. However, without backwards compatibility Itanium was not a viable 64 bit solution, hence the eventual pigeon holing of Itanium as a high-end solution only.
You've crafted a very curious world you live in. How do you keep fact separated from fiction?
yeah, CPG should do just fine. It sure would be nice to have a quarter where both flash and CPG performed. I guess that's too much to ask for. Very frustrating.
Anyway, there seem to be serious problems in the flash side of things, something I hope gets talked about a lot in the CC. In a way I hope it's INTC pressure and not Samsung NAND causing the problems. I don't know what's going on with AMD, but I'm pretty sure INTC isn't keeping up in the flash design area. Samsung looks like much more of a problem than INTC. Especially if the flash markets stagnate in 05.
INTC forces MoBo makers to buy 915/925 chips
http://www.theinquirer.net/?article=20646
Me..
Typical INTC arrogance/hubris. How much longer will these guys put up with such bs. This must be doing wonders for AMD CPU sales. No wonder AMD mobos are now 50% of the market. INTC continues to drive customers to AMD. You have to love it.
Maybe flash is weak, but CPG seems to be doing just great with all the help INTC is giving.
Yeah, well not from where I'm seated. It sure looked like INTC decided to go with Itanium when MSFT gave it the choice of either X86-64 or IPF. INTC has been scrambling every since the very terrible decision to catch up with AMD64.
At the time it looked like INTC had given up completely on X86-64, and ensuing events only reinforced that view. IPF was going to replace X86 and that was that. However the success of AMD64 soon showed the utter idiocy in that decision and INTC was forced into a hasty retreat and the acceptance of AMD64. As they say the rest is history and you view is definitely warped.
I suppose dumping X86 for Itanium was a stroke of genius? Current management inherited a golden goose and chose to eat the bird. There was no reason for AMD to have ever gotten to where it has. INTC management has screwed up one thing after another only saved because those that came before did such a great job at building defenses.
AMD may have gotten hurt this quarter in flash, but that's not where INTC lives. This year could be very interesting particularly if there is any slow down in global processor sales. But next year will really be interesting as fab36 comes online. INTC had the chance to get rid of AMD once and for all, but fortunately for AMD B&O ruled INTC and all they could ever do is waste money on projects that never go anywhere.
Hard to say just what happened, but for sure flash took a big hit? Flash is going to be a lot less of AMD after this quarter.
INTC has a lot of money to throw at flash. Given the small % of INTC's total sales (some where around 7% I seem to remember) INTC can afford, from a strategic point of view, to put pressure on AMD there. It should be interesting seeing just what percentage of flash sales are now mirrored flash. If the percentage is high then either there's something wrong with the cost advantages we have been told are associated with mirrored flash or INTC took a big hit. It's probably a little of each, but given the way INTC reports we may never know.
However, I suspect the real damage is coming from Samsung with its' stacked(MCP) NAND, Dram, and SRAM
http://english.chosun.com/w21data/html/news/200501/200501100038.html
http://www.eet.com/semi/news/showArticle.jhtml?articleId=57700151
Who knows AMD may just be a incidental fire victim as INTC and
Samsung fight it out.
yeah, this is 2 quarters in a row the rug has been pulled out from under us at the last minute. I'm not too happy. This is the sort of thing Jerry used to do.
Knowing INTC I certainly wouldn't put such a thing past them, but it's unlikely as everyone in the world would be pointing at INTC and the glare would be more of a problem than AMD is.
Besides, it would have to be a pretty big missile/bomb to really put AMD out of business. From what I understand fabs are built like bunkers.
Every once in awhile, if one is very lucky, one can read something and learn a lot about oneself without having to go through the empirical pain normally associated with epiphanies. Such is the article posted by TGPTNDR on SI.
http://www.gladwell.com/2002/2002_04_29_a_blowingup.htm
Y2k was another perfect example of the black swan. Personally, my style is to bird dog my investments figuring that in any case less the global nuclear warfare companies like AMD will continue keeping on and any hits the stocks take because of general market depressions will be recovered as long as the company is sound. Even as Y2k took effect AMD resisted the fall and only began to drop when the production problems led to loss of market share and actual earnings losses. INTC never really took a Y2K hit, but has continued to suffer because of weakening fundamentals.
In the end companies unlike options don't generally loose their value all at once because of some event. Of course a terrorist attack on AMD's Dresden fab might put AMD out of business, but what are the odds? Hmmm...
HP and AMD nestle up
http://www.hardwaregeeks.com/comments.php?shownews=2689
Me...
Is it me or does it seem like AMD and HP are getting a lot friendlier. I know I've seen HP flyers lately with just AMD products, something I had never seen before. Then there's the DL385 and blades coming along with the media center. While they may not be lovers yet it does look like HP is a lot more receptive to AMD's advances than in the past. One can only wonder how close the relation would be if AMD were better able to supply more of HP needs?
SFF needs wind tunnel to cool INTC processor.
In order to be able to cool down the modern processors and graphics cards Shuttle had to use its new P-series chassis equipped with a total of 5 fans, fourth generation of Shuttle’s patented ICE cooling system for central processing units and improved airflow design. The chassis changes resulted in slightly higher size and weight compared to typical barebones from Shuttle: 32.5*22*21cm (length, width, height) versus usual size of 30*20*18.5cm. The XPC95PV2 is equipped with 350W PSU.
http://www.xbitlabs.com/news/mobile/display/20050110052952.html
AMD chipset penetration
http://anandtech.com/tradeshows/showdoc.aspx?i=2317&p=14
We talked to one motherboard manufacturer whose motherboard shipments were 60% AMD and 40% Intel. The rest of the motherboard manufacturers in Taiwan are still shipping more Intel than AMD but we're seeing a continued shift towards a 50/50 AMD/Intel split when it comes to motherboard shipments. By the end of Q2 all Socket-478 CPUs will have dried up, meaning that anyone looking for an Intel platform will be forced to move to 915 or 925X(E), which should obviously increase shipments of those two platforms but may also increase the number of AMD platforms sold.
VIA's best ally among the Taiwanese motherboard manufacturers is currently shipping around a 60/40 split when it comes to VIA vs. NVIDIA chipsets. The rest of Taiwan however appears to be more along the lines of 50/50 between VIA and NVIDIA. The manufacturers we talked too indicated that as soon as nForce4 was available on a more widespread basis that their AMD motherboard shipments would soon be much stronger in favor of NVIDIA.
Me....
More good stuff in the article.
It looks like platforms are going to take on a lot more significance than before, both for AMD and INTC. Not too surprising given the number of current players in the consumer convergence and the winnowing that is going to take place. Fortunately it doesn't sound like INTC has quite got it together yet, but it does seem like platforms will be taking most of the sales in another year or so. Not that there won't be room for boutique producers, particularly in the high end, but platforms are going to be what the biggest part of the market is all about.
Things could get a lot tougher for AMD, assuming INTC gets performance competitive platforms to market. On the other hand there will be tremendous pressure placed on companies like NVDA to come up with alternate solutions as they find themselves frozen out of markets by the INTC juggernaut.
There's going to be a lot more pressure on smaller companies to merge, or at least get very close to each other. This will be a real test of AMD's partnering model in comparison to the vertical integration model INTC is pushing. We'll have to keep both eyes on this.
Fortunately for AMD, I think the whole convergence thing is going to be so big that companies with the best ideas will probably do very well. Just having the ability to produce billions of widgets won’t necessarily result in sales. Ease of use is probably going to be the biggest factor with performance close behind.
More on vassili from SI
http://www.amdboard.com/fic_010505.html
It seems to me that the only thing lacking in this product to make it a truly killer app is the lack of 3g/802.11g-n wireless connects. With wireless connects this would seem to come pretty close to my view of what the personal communicator will become. Not bad for a first stab at a universal platform.
To me this sort of low-cost, low-risk, possible big payoff thinking is what really differentiates the new AMD from the competition. While the PC market is going to remain the mainstay of computing you can bet that that beige box in the den is going away, far away.
I saw the posting by MAS_ on SI. I had read about it a few days ago but didn’t pay attention to the date. Anyway the DL385 and the blades should be here soon, hopefully. Now why wouldn't INTC want Itanium compared to Opteron, seems rather flattering to Itanium?
Actually I’m sort of relieved, having a media announced good news story at earnings has been a surefire recipe for a price drop.
Maybe they will announce both? Time frame seems right. Where do you see Integrity? All is see is something about enterprise solutions in the HP announcement.
Here's something from the register.
http://www.theregister.co.uk/2004/09/08/hp_opteron_servers/
What makes you think the DL385 is not Opterons. I seem to remember having read more than once about it and the blades coming about now. Doesn't Integrity have numbers almost like DL385? I'll have to look it up?
DL385
We've been waiting for quite a while for this one and the blades.
http://messages.yahoo.com/bbs?action=m&board=4687810&tid=amd&mid=1126554&sid=4687810
This may tie in with
http://h10054.www1.hp.com/ent_event/landing.html?mtxs=home-ent&mtxb=B2&mtxl=L1&jumpid=in...
The 18th may be a major media event day with earnings.
Trouble is I get this funny feeling whenever AMD anounces good news with earnings.
Yeah, but I'm not sure how long the $ recovery will last. The administration has been trying to do the LBJ "guns and butters" thing, which has eliminated the surplus the administration inherited and created gaping deficits. Clearly the administration never expected the Iraq conflict to last this long.
Anyway, the higher deficits should require higher long term interest rates which means foreigners will have to be more supportive since Americans don't save. As the $ falls one would expect foreign investors to demand higher rates because of the likely real negative return, but so many factors are involved that foreign reserve banks will probably continue to invest no matter what. However, long-term rates will probably have to go up, at least some, to maintain deposits from central banks of countries where trade with the U.S. isn’t a big factor. This in turn should slow the housing market which has been the preferred investment medium for Americans since y2k.
One might think that the lowered $ would result in more U.S. jobs, and undoubtedly it will, but most of the jobs sent overseas are in labor intensive areas where U.S. labor costs are still no match for say Chinese/Indian labor costs. This also includes some technical jobs such as programmers. The jobs outflow problem isn’t going to get fixed any time soon and the best the U.S. can do is try to improve productivity in areas where it is still competitive. Which of course leads to fewer jobs at home, but is good for technology companies such as AMD.
So the U.S. can fix the deficit by either reducing government spending or by importing fewer goods. Since the war will continue to siphon off huge amounts, for way into the future, defense spending reductions are not an option. That means cuts will have to come in other areas. Entitlement programs such as SS and Medicare will probably have to take the biggest hits since their the biggest outflows. Of course that would probably prove to be political suicide if the republicans were to push the changes through without the democrats helping, which seems unlikely. So I see little headway being made in reducing the government deficits under the current administration. They will probably choose to blame the democrats for the deficits and leave the solution for another administration to deal with.
So the only real way I see to fix the deficit is to stop importing so many goods, something that would hurt everyone in the world, which probably means that the effects will be ameliorated some way, out of self interest. Most likely by continued foreign investments in U.S. long bonds. Long term though the $ is going to lose its' status as the worlds preferred reserve currency and that means a less valuable $. This will happen as the U.S. becomes less of a factor in global trade and tying currencies to the $ becomes less desirable as raw products(petroleum) will become more costly in $’s .
The thing that will be interesting is inflation. Personally, I think once the housing market cools off there will be decreased demand for goods, both foreign and domestic, such that inflation won't be much of a problem. A few years ago I read that something like half of consumer spending was being financed by home refinancing. While that has undoubtedly slowed down simply because rates have ceased to fall, housing inflation has still allowed lots of people to tap housing lines of credit. These lines of credit are mostly at a couple of points or so above prime which means that as the Fed continues to raise rates these loans will have higher monthly payments attached to them. We could be getting ready for another period of rising bankruptcies, which would also lower the value of real estate and slow down economic growth. Of course as inflation slows the Fed will have fewer reasons to raise rates.
Interesting times.
I'm not going to argue that AMD had a good week in comparison to INTC, but again it's like looking at a snapshot. Take a look at the 5-year chart it's more like the movie I was alluding to.
I know you're still invested in AMD and you certainly wouldn't be if you could find something else that you thought was going to do better. Further, I seem to remember you being located in Germany which means you have a double whammy to overcome in the sinking $. I assume what comes across, as negativism is really healthy skepticism. A lot of the criticism you point at AMD is very justified, but a snapshot can't tell the story a movie can.
How prudish of your wife. Just tell her the bee in you wants to dust a few lilies, she’ll understand. Just keep her away from the knives.
Yes, the way I see it AMD now has the management, the products and the money to compete head to head with INTC. What it lacks is the capacity and probably the marketing. Hopefully those things will start to get resolved in the 06-07 time frame.
It is hard to see out 6 months let alone a couple of years. For us long timers that went through the times when it was a toss up as to whether AMD would survive this seems like nirvana.
AMD does have a couple of wild cards to play in PIC and vassili, or whatever it is called, that could change things a lot. Whatever, the semi business is entering a period of big changes as convergence takes hold. Computing power is going to get dirt cheep and start showing up in everything.
AMD has had some design problems over the year but the production problems have in general dwarfed those problems. This commingling of resources with IBM, Chartered, SNE etc. seems to be paying off big time, and probably not just for AMD. Bio diversity/cross fertilization, that's the ticket.
It seems like people either take a simple picture or a movie. Taking a picture of AMD at any time during the last 5 years produces an image of a frog, warts and all, when compared to INTC. If however you take a movie of those five years and compare it to INTC then AMD comes out looking much more like prince charming.
Let's face it AMD is still basically a boutique producer and as a result has to depend on partners that don't always respond to AMD's desires or schedules or with products that we would like. INTC benefits from platforms in ways AMD can't as yet, however this also tends to lead to a NIH mentally that causes a lot of problems. Bio diversity is the best insurance a species has to longevity, inbreeding leads to idiots.
When I started selling my INTC stock and moving into AMD back in 98 I expected that it would take about 5 years for AMD to reach critical mass (1/3 of the market). Boy was I off, but thankfully INTC management was kind enough to let AMD off the hook and now it looks like that date could occur as early as late 06 or early 07.
In any case the point is that the movie shows constant progress. Not that there haven't been more than a few set backs, but it looks more and more like the worst is behind AMD and a bright future lies ahead. This dithering because a product is late or doesn't fill a space completely seems a waste of time; a longer perspective is needed. However, I do appreciate the intimate knowledge and play by play color commentary that is provided so keep up the good work. Just don't get too rattled by the minutiae in making investment decisions, this is a trek not a sprint.
one view of today
http://www.schaeffersresearch.com/commentary/observations.aspx?ID=12195
It's all about earnings.
Faster is definitely better than slower. AMD would really have to drop the ball to screw up 05, but then there were those pony chips.
I'll agree about the server market, but I'm not so sure about 50 m chips. This whole capacity issue is fraught with imponderables. For one thing I'm pretty sure that the 12.5m/q estimate is a theoretical one based on a complete fab30 conversion to 90nm. That's something that won't happen until H2 at the earliest and probably later. Then there's the problem of just what chips are they talking about in that 50m estimate. For another, in most production environments getting much above 90% of theoretical capacity in sellable parts is very difficult, for a variety of reasons.
I'm figuring about 45m chips max in a full 90nm year, something that won't happen until 06. Personally, I'll be very happy if AMD produces around 40m chips this year. So basically I see AMD capped at something less than 20% of the unit processor market in 05. Given rapidly rising demand as the year progresses, particularly after W64 is introed, I figure by q3 AMD should be snuggling up against the capacity wall. Anyway that's where I'm coming from.
Despite my rather dim capacity view I expect AMD to make lots of money on processors alone in 05. And then there's flash or PIC or vassili and who knows what's next. All in all things do look much better than they ever have. Despite the latest market gyrations I have never felt better about my AMD investment, even when the stock was in the $90s.
Well, all I know is what I read. When Itanic was introed it was going to be the thing that killed X86, then it was going to merge with Xeon and that would take over the server market, then it wasn't going to be in workstations, then it was going to be an exclusive high end enterprise solution. What next?
I've said it many times before, but the way things are going it sure looks like multiple dies, each with multiple processors, using PTP topology and clustered together with high speed switches will replace the "big Iron" that Itanium so desires to emulate. From my point of view, what remains of Itanium is chasing a dying market. Oh and did I mention servers, of all types, are going to be inexpensive also.
One things for sure though, Itanium isn't going away any time soon, INTC has too much ego invested, and killing it outright would just invite law suits. So I guess INTC investors are going to be treated to the humongous profits only Itanium can provide for quite some time.
DivX
http://home.businesswire.com/portal/site/google/index.jsp?ndmViewId=news_view&newsId=20050105005...
from saxman
Them...
By working with the world's best technology and entertainment companies, we intend to create a seamless consumer experience that will enable anyone to use widely available consumer electronic devices to manage and share a complete library of digital media," said Kevin Hell, Chief Operating Officer at DivXNetworks, Inc. "When DivX Connected devices and services hit the market later this year, consumers will be able to simply pick up the remote control for a DivX Certified Connected Media Player and gain instant access to all the video, music and photos stored on their PCs via any connected TV or home theater system in the house. Anyone will also be able to access DivX Connected online services to check local weather, listen to digital music or get new movies from premium content providers."
I thought I read this before going on vacation? Anyway It's great news. One has to wonder how much longer INTC will throw money down the rat hole. Hopefully, a lot longer. Talk about a shrinking market. Pretty soon Itanium will only exist in the imagination of chipguy.
Good post.
I've never seen even the most rabid AMD supporters, myself included, ever express the opinion that INTC is going away anytime soon. What we usually comment on is how far AMD has come despite having to fight the many advantages INTC has.
In fact one of my favorite themes is that if INTC management had an iota of the ability the press gives them AMD would never have gotten this far.
Critical mass for AMD is a very valid concern, but for all practical purposes that argument disappears when fab36 comes online. To my way of thinking the only thing keeping AMD from gobbling up a lot of market has been its' inability to supply the market. While the OEMs may want to use more AMD products I'm very sure they’re leery of making too much of a commitment lest AMD not be able to supply their needs. This probably includes DELL.
One might say, "but this doesn't apply to the server market", but in truth it does. It's ironic, but no matter how much one can prove compatibility between INTC and AMD products there are still shops that go all INTC because they are afraid of compatibility problems. So robbing Peter to pay Paul just isn't an option and AMD cannot concentrate on say servers to the detriment of other products. AMD needs to be able to supply the entire business needs across all boundaries and that's where fab36 comes in. FAB36 introduces a whole new processor game.
In the meantime it's nice to see AMD creating new products such as PIC and the FIC Vassilli (wasn’t he the bad guy in "Princes Bride"?) that can be manufactured using non AMD fab space.
AMD won't take much market share from INTC this year, but they will continue to hurt INTC's margins and AMD's margins will continue to grow very nicely. 05 should be a very good year for AMD. For INTC 05 will depend on general market demand. A slow demand year for INTC is going to be a real problem whereas for AMD it won't make much difference.
This is just the big money playing games. They drive AMD up and then down making money both ways from those willing to play their games. AMD's high beta and the legions of momentum players following AMD make it a ripe candidate to bounce more during market gyrations. One needs to keep an eye on the fundamentals which have never, never looked better.
Yes INTC still has tons of money, INTC inside, all those fabs and an army of mindless automatons buying its' products, but in the end it's the ability to innovate that determines the success of a company, and INTC is floundering about like a cork in a hurricane, at best only able to mimic what AMD is doing.
AMD is like a 100 to 1 long shot that's leading going into the stretch. If it wasn't for inept INTC management AMD would have never have gotten this far. There is absolutely no reason for AMD having ever been allowed the huge success it has enjoyed/ is enjoying except for the hubris and disastrous decisions that INTC management have made.
Looking at INTC currently is like trying to drive forward while looking in the rear view mirror, there's going to be a hairy crack up. I'll rephrase that for you in case it's meaning was lost. Looking at INTC's past is no predictor of it's future and I'm not the only one that thinks that way.
http://finance.yahoo.com/q/bc?t=5y&s=AMD&l=on&z=m&q=l&c=intc
Yada, yada.
INTC is blessed with constipated thinking. That, sooner or later is going to result in a big backup that will only be relived by a massive dose of cascara sagrada. The only thing saving INTC's butt right now is the good growth in the overall markets. Once thing slow down INTC will be in big trouble.
INTC still does have lots of advantages, but the mindset largely makes those things irrelevant.
I'm back, be gone you Baal acolytes.
Well I can hardly contain my glee over the Alchemy announcement. I was excited about PIC, but this portable audio/video solution is pure ecstasy. This is clearly the way of the future, talk about your killer app. This is exactly the sort of merging of consumer electronics and the PC I have been wanting AMD to get into. Clearly, the people running AMD have a very good view of the future and where things are going.
Meanwhile INTC continues to be run by a bunch of baboons locked in a mid 80's mindset. How can such a great company have gone so wrong so fast? I'm sure most at INTC still think faster is better and all INTC has to do is come out with a faster chip and AMD will be once more relegated to the dust bin. It really is sort of pathetic.
The trouble is that INTC isn't even able to keep up with AMD in the traditional markets as AMD is first with dual processors, owns the gaming market, and companies like Pathscale continue the march towards making mainframe processors irreverent. Even the laptop market is likely to be removed as a sinecure for INTC as the melding of entertainment and communication continues. That's where the market will be, who wants to run spreadsheets on their T&L, that's just the sort of thinking that permeates throughout INTC?
One can only imagine what things would be like if AMD had more capacity. Boy things really look great.
Looks like I should pay more attention to the dates, still a good read. I fully expected something like this from INTC given the test Tom has been running. Should have read the whole article first. It still doesn't change my view of INTC. They may be more subtle about things now but it's still the same old game.
INTC arm-twisting of Tom.
http://www4.tomshardware.com/blurb/19970310/index.html
Looks like INTC is up to its' usual tricks. While I'm not that fond of Tom this is just BS that INTC is doing. Glad to see it backfiring. INTC really is a whore/horror, your choice.
From the Saxman on Yahoo.
Nice move going on. I'm not sure how meaningful it is given the low volume, but nice just the same. There does seem to be a core of believers out there that view every drop as a buying opportunity. God bless them, one and all.
It still seems more risky to be out of AMD than in, especially with the general markets moveing up the way they have. Nice to see AMD doing better than the SOXX and INTC for a change.
Still, I wouldn't read too much into what's going on. A lot of this is just Y/E bonusus being put to work. Last week was probably some big boys taking gains. Markets are generally volatile at Y/E.