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.
No.
2 CPUs are on the motherboard.
2 more CPUs are available on a CPU board add on.
Only 2 or 4 CPUs.
upc
EDIT: There is plentiful information on this and other questions you've been asking on sun's website. Whitepapers, datasheets, benchmarks, etc. It's linked on the front page: www.sun.com
Aside from the rather impressive announcement, what software does the thing run?
Now: 32-bit Linux, 64-bit Linux, 32-bit Solaris 9, 32-bit Windows Server 2000/2003.
Q4/Q1 05: 64-bit Solaris 10, 64-bit Windows Server 2003
It's right there in the datasheet:
Solaris™ 9 4/04 OS, x86 Platform Edition (32-bit)
• Software Express for Solaris OS, x86 Platform Edition
• Red Hat® Enterprise Linux 3 for AMD Opteron
(32-bit/64-bit)
• SUSE Linux Enterprise Server 8 for AMD Opteron
(32-bit/64-bit)
• SUSE Linux 9 Professional* 64-bit
• Microsoft Windows 2000* (WHQL certified),
(customer provided)
• Microsoft Windows Server 2003* (WHQL certified)
(customer provided)
upc
The 40z uses either 2 or 4 8xx series Opterons. You must have meant something else. The W2100z workstation?
upc
If you are having trouble with an answer of your own then you are involved in a game of pure chance.
I will not engage in a debate with anyone, let alone a board moderator, who resorts to unnecessary putdowns like that one.
upc
My last attempt, I promise:
A = discussions properly undertaken on this board
B = discussions about information and events that could conceivably affect AMD stock price
My claim is that A is a very small subset of B.
For B contains:
- All discussion about the economy in general, for that can affct the market, and thus AMD
- By extenstion, all geopolitical discussion, for that can affect the economy, hence the market, hence AMD
etc., etc.
I am NOT saying that AMD investors do not need to keep the B information in mind, just that it should not be discussed on THIS board.
I hope that makes my position clear. This time, I mean it. No more on this from me.
upc
Tiger, no. To use your metaphor, I DISCUSS the wheel-spinnings here, and DISCUSS the price of gas and potential for bad weather and road conditions elsewhere. But I will not continue this argument here, as I just posted to sgolds. If I had private messaging, I would be happy to continue talking with you about this.
upc
Why is the subject being changed? My question to you is: at what point is a topic so general that it should be discussed elsewhere, despite its possible ramifications on AMD stock price? I would argue the latest economic thread was clearly so general as to be well over the line.
For example, shall we also discuss the possibility of terrorism on any given day? That could affect the market, and AMD.
How about the likely outcome of the presidential election? That could affect the market, and AMD.
upc
p.s. Your "boss" seems to agree: http://www.investorshub.com/boards/read_msg.asp?message_id=3645410
p.p.s. This will be my last word on this subject. For the sake of the board, I hope this politico-economic debate is continued elsewhere.
It's not production capacity, but product itself that has increased, specifically, K8 unit volumes, in response to greatly increased 130nm newcastle starts last quarter.
upc
The point is not that macro-economic conditions have no bearing on AMD as an investment. The point is that discussion of those conditions is probably better handled on a different board, as there is nothing AMD-specific about it. Think about it. Do you want the same non-stock-specific information on every specific stock board?
upc
I´d like to see AMD do something to counter INTEL´s faster bus speed.
Isn't that what boosting HyperTransport to 1GHz was about? Also, DDR-500 with excellent timings is even available from Crucial now, so it's pretty mainstream. Aren't those two together the equivalent of boosting the FSB?
upc
Why do think a moderate price cut has anything to do with 90nm? These are the cuts announced in the memo posted by BUGGI. As AMD explained in the CC, as they continue the unit # shift from K7 to K8, the K8 must enter more mainstream ( cheaper ) price bands. So that's what this indicates: AMD is making more K8 desktop parts now, and desktop parts will probably be the last thing ramped on 90nm, after mobile and server.
upc
It's right here. http://www.investorshub.com/boards/board.asp?board_id=1606
Welcome!
This is now the Off-Topic [OT] sub-board of the official Advanced Micro Devices investment board
Enjoy!
upc
no i meant over here on iHub.
There are lots of things that could affect the value of AMD stock which would probably not be appropriate to discuss here. It just seemed to me that the discussion was getting rather general and political. Isn't there a special other AMD board meant to continue these sorts of topics?
upc
Interesting roadmap. In the Athlon64 939 section there is this:
A64 4000+ Q404 (that is earlier than previously suggested)
but then what is this?
A64 -4000+ Q205
A64 -4200+ Q305
How can the same part (4000) be released again quarters later? Except it is not exactly the same, it is "-4000+" instead of "4000+" Does this refer to a necessary rating change due to Intel speeding up their FSB? Something else?
Also Q305 is getting to the point of dual cores, even for desktop.
upc
What's with all this off topic stuff? Shouldn't this go on another board?
upc
Buggi, Red Storm is the large supercomputer, and yes, full functionality there has been delayed. It was also announced that it would be upgraded to 90nm Opterons next year if I recall correctly. But the XD1 is not Red Storm related. The XD1 is the (acquired) Octiga Bay product, and that timeline was early-adopter beta sites in this H2, with general availability in Q1 05 or something. I will look for the press release.
Here it is.
Early shipments of the OctigaBay product are expected in the second half of 2004, with general availability in early 2005. Pricing, to be announced later this year, is expected to range from under $100,000 to about $2 million.
http://phx.corporate-ir.net/phoenix.zhtml?c=98390&p=irol-newsArticle&t=Regular&id=499043...
upc
I accept that the NYT leans towards the Democrats in the editorial opinion, but I find their reporting of facts still to be second to none
Is this meant to have a smiley? Their political opinion is manifest throughout their news stories, particularly anything billed as 'news analysis', their story selection, and story placement. I suggest you track down a number of blogs that follow the NYT (and others) and point out the bias. It's become *very* bad the past few years, and ruined what used to be the newspaper of record.
upc
I'm sorry, where was the review site showing the average performance of an A64 3400+ to be 5-10% over that of the 3.4 GHz Prescott? (Let's not speculate about how a 3800+ should perform relative to a 3.2 GHz P4, when we can do a direct comparison.)
As for Celeron versus P4, actually that did cause some issues for Intel in the space of overlapping frequencies, which contributed to their development of the 300-series / 500-series model numbering scheme, (along with Banias being a high-IPC part, which also pressured for the change).
-------------------------
I'll say it again: You have to tell consumers what you are offering to them (for a given price) in the simplest possible way.
That means rating the parts as accurately as you can with respect to the competition, not forcing the consumer to know about some review somewhere that tells them that they should be willing to pay a bit more because they're getting 19.37% more performance for your product than the rating implies.
This is about as basic as marketing gets:
Make it *easy* for the consumer to decide on the price / performance / branding tradeoffs.
Building unlabeled, undescribed, unknown amounts of extra performance into your products, and then expecting them to know about it is confusing as hell to the consumer.
upc
well, there is a chance the bottom will go lower tomorrow. MSFT just missed by a penny.
upc
A smart sale. With MSFT missing, tech will be down tomorrow.
upc
Intel is not way above the levels of a year ago. AMD was artificially low last spring on a concern they would not survive.
I suspect your "buyer beware" ironically coincides with a mid to long-term bottom in AMD stock, barring an extreme terrorism event.
upc
I have answered this question multiple times. Athlon XPs, as they were released, were benched and rated and marketed against the P4's "GHz" rating. But what happens over time? New Athlon XPs are released, at higher speed grades. The older Athlon XPs find themselves in a segment where Intel is marketing Celerons. Celerons are also marketed by GHz, even though this is a different rating system, and low end consumers cannot tell the difference. AMD, being without 2 brands, hestitates to remark the parts, because their one rating system would then be internally wildly inconsistent, and they would cannabalize mid-high XP sales, as consumers confuse the ratings. How much more clear can I make this? Low end consumers read 2 things off the tag: the price, and the rating. Simply with the passage of time, Athlon XP processors migrated into the Celeron segment, and once there, their unadjusted rating killed their prices. Do I need to sketch an example?
This all happened because Intel is using two rating systems that sound the same to the mass consumer. GHz and GHz. One is P4. The other P4 Celeron.
Please cite a benchmark roundup comparing a 3400+ to a 3.4 GHz Prescott (let's forget about EE and FX, okay?), which exhibits, on average, a 5-10% advantage for the 3400+. I don't think you can find such a thing, for it would be remarkable, and imply AMD could have rated the 3400+ as the 3600+ or 3700+.
upc
Most of AMD's sales are international. And I'm sorry, but did you say AMD and Intel have "seen a lot stock appreciation" ??? You may be holding your chart upside down. :)
Shorter beware.
upc
No, everyone does not know that. 99% of low-end buyers have no clue about XP performance ratings versus Celerons. Why would you think otherwise? Low-end Athlon XP prices are low precisely because they are effectively mislabeled with respect to their competition, and that's all the consumer looks at.
And I'm sorry, but the A64s do not have an average 5-10% performance cushion vs. Prescott at a given rating level. They have held their ASPs because AMD has ramped slowly and carefully, and because Intel has virtually no supply of high-end parts, and the parts they do have do not yet possess the features ( amd64 and NX protection ) that AMD's parts do.
upc
Is that enough to overcome the Intel brandname ?
The performance rating is not the place to overcome the Intel brandname. The performance rating should be accurate. Until AMD has a stronger brand, they will not get price parity, but so be it. They get even worse pricing by rating their parts too low.
By your logic, AMD should rate the 3800+ A64 as the 2800+ A64. Surely the reviewers would love it, and the public would recognize the value, and AMD would get a lot more for the parts than Intel gets for a 2.8 GHz Prescott... whoops.
Now's not the time to be lowering the defences and softening the PR ratings
It's raising the marketing and pricing defenses and correcting the rating giveaway now that the part is clearly positioned against Celeron. You could argue that AMD should produce higher-performance value parts and therefore counteract some of the brand/pricing effect. But to argue that AMD should mislabel the performance of their parts is folly because you are dealing with mass consumers here and they believe the ratings, and have no knowledge of any un-rated extra performance. There is much discussion of this on the silicon investor thread and in fact I must credit those posters with much of my thinking on this issue.
Another way to look at it is that it is not "softening the PR ratings". It is changing from having one rating system (a disaster when your competitor is using two) to having two systems, and moving some parts from into the new brand and its rating system.
upc
Whats up with that?
HP profits.
upc
yes, high end computing is a "niche", but that is what we were discussing.
upc
Realize that is a NYT opinion piece and take it with a very large grain of salt, especially in an election year, as even their news coverage has been playing up the bad, downplaying the good.
upc
I would say you may be confusing the goals of the software and hardware divisions of Sun. There is still a benefit to offering Solaris on hardware platforms that Sun does not sell, like Power.
Whether or not Sun is directly assisting in the design of the K9, it seems clear they are heavily invested in developing Opteron enterprise infrastructure, and have said as much. Here are two examples, one from the partnership launch, one from last month.
http://www.sun.com/smi/Press/sunflash/2003-11/sunflash.20031117.1.html
Future AMD Opteron Processor-based designs: Sun and AMD will collaborate on a portfolio of future AMD Opteron processor-based systems and scalability beyond 4-way AMD Opteron processor systems. The parties will also collaborate on coherent HyperTransport technology implementations.
http://biz.yahoo.com/prnews/040601/cltu051_1.html
"The Broadcom-AMD partnership is a critical aspect of the AMD Opteron processor ecosystem that Sun is helping to develop," said John Fowler, Executive Vice President, Network Systems at Sun Microsystems, Inc. "We are pleased to see Broadcom join AMD as one of Sun's long-term partners in the server space."
upc
On that plane of existence
IPF and PowerPC are the *only* real choices in looking
beyond the increasingly irrelevent SPARC.
If I read the archive correctly, you used to say that IPF was the only choice. Now it appears you've added PowerPC. How can you be certain you won't need to add Opteron until you see what Sun and Serverworks are putting together?
upc
This news shows there is no special arrangement there. :(
How does it show this? There seems to me to be some possible confusion between software (porting Solaris to multiple platforms) and hardware (what type of systems Sun chooses to produce) going on. News of an alpha port of Solaris booting on an Itanium (and an attempt to port to IBM's Power cpus) allows one to conclude that Sun is not going to work with AMD on future K9 designs? Or is there some other information that I am missing that is sufficient to establish your claim? Thank you in advance.
upc
What makes you say that Opteron ever had Sun's "undivided attention"? I would say that is too strong of a statement. But how exactly are you defining "high end computing"? Does that exclude HPC clustering? And also 8 way Opteron systems? Isn't it known that Sun is in a partnership with Serverworks to develop 8 and 16 way chipsets for the Opteron? With dual core parts, the 16-way might even become 32-way.
I would add that beyond that I believe that Sun should consider acquiring Cray to get ahold of their Opteron technology, but of course, this is not something that is happening, merely my opinion.
upc
It seems evident now that the whole Opteron promotion was a short-term fling.
Forgive me for expressing my disagreement but I cannot fathom how one draws such a conclusion from the facts at hand. Is there some other point of information beyond an alpha port of Solaris to Itanium that you are using to come to this rather remarkable conclusion? It seems to me that the Opteron workstations and 4 way server launching this month, together with the forthcoming work of the company they acquired earlier with the old Sun employee Bechtelshom would strongly suggest otherwise.
upc
As far as I know, there are no K8-based "Athlon XP" parts. There is an unofficial K8-based Athlon XP-M being used at HP in their socket 754 notebook.
upc
p.s. here is an earlier post about this.
http://www.investorshub.com/boards/read_msg.asp?message_id=2934232
sgolds, that sounds right, Athlon XP-M and Mobile Athlon-64 are marketing names. So an Athlon XP-M can have a K8 core, if it is 32-bit "only", even due to a fuse or bios setting.
Yes, it is at least an early "version" of Dublin, made for HP. HP may have lost interest since Intel launched iAMD64, and HP got on board with Opteron, but I think that having "ordered" the part, they are using it as an option on one machine (and it is also the processor in the Compaq nx9105 business notebook in the UK.)
I don't think this actually means Paris is coming sooner.
There should be a record of all of this discussion on this board back in early March. Search "Dublin".
Doug
Also, Intel could be funding the port.
upc
demonstrates how *no* OEM can ignore IPF and still hope
to stay in the game
Could you elaborate on this point? I could see that it might demonstrate that someone attempting to grow marketshare for an enterprise class operating system should port the operating system to a large number of enterprise platforms, but how can one conclude anything about hardware OEM attention to IPF from this information? As an example, would it be fair to say that Dell very nearly ignores IPF while remaining arguably the most successful hardware OEM?
upc
Crossover (in terms of waferstarts) K8/K7 Summer 05.
No, that was 90nm / 130nm output unit # crossover. Check the transcript.
upc
AMD appears to have locked up the desktop performance title across the board. An interesting review especially the apparent synergies exhibited by the nvidia chipset motherboards with the nvidia graphics card. The performance advantage over the via chipset boards is surprisingly high in some benchmarks.
12% faster than via in Veritest 2004 Business Winstone
10% faster than via in Veritest 2004 Content Creation
15% faster than via in Halo
upc
AMD says 90nm revenue shipments "likely within the next few weeks". See this post on the silicon investor thread.
http://www.siliconinvestor.com/stocktalk/msg.gsp?msgid=20324610
upc