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.
wbmw, Yes, but with Prescott and with Itanium you'd have the additional advantage of being able to fry your eggs for breakfast on the case, as well as keeping your coffee HOT! ;)
Yeah, I know, cheap shot - I'm just playing. :)
Paul
Keith - re: VoodooPC - silence is golden. ;) Sales OK?
Paul
EP, but "full" and "good" are two entirely different things.
Paul
Anyone know how far-reaching the cooperation between IBM and AMD is? This could have some interesting outcomes if there's any cooperation here.
Paul
http://www.reed-electronics.com/electronicnews/article/CA340355?stt=000&industryid=22113&ind...
NaS, My point is that more and more people are waiting to buy until after Christmas, especially with more and more gift cards, etc., being given. People know that there are frequently good deals to be had afterward. The Christmas deadline is still a factor, it's just becoming less of one.
Paul
wbmw, he is more balanced than his compadre from RWT, Paul DeMone, who has become the most outspoken Intel/P4/Itanium advocate I've seen(that excludes shrill nuts like Waitress, whom I lump into the same irrelevant category as facsnotfiction, subzero, etc). Paul also spends hugely more time there than Dean. I must admit, though, I'm very offended by the way he is often portrayed and demeaned there. I don't think he's dishonest or deceitful, I just disagree with his conclusions.
Paul
Dan3, correct me if I'm wrong, but I thought that was the whole concept of registered RAM, to take the direct load off the bus to allow more modules to be accessed without (necessarily) reducing speed, at the cost of a cycle of latency for the registers. If I'm understanding this correctly, I see no reason why Itanium couldn't use DDR400 if they had a chipset that could do it. Assuming, of course that the FSB side didn't have issues with it when more processors are used; if it can't handle adequate speed to make use of the increased memory bandwidth, then it'd be a waste of time to develop such an expensive MP chipset.
Paul
Reseller Mike, well I hope the word doesn't get out too much, or that Osborne effect will certainly happen. Right after Christmas is not a long time to wait.
Paul
Jerry, of course there is that relationship, but the factor with even more impact is obviously voltage(or current, depending on how you want to look at it), so what they are able to do there is even more important than frequency.
Paul
BUGGI, don't worry about your "imperfect" expressions. Most of us here are capable of digging out your meaning in those instances when it is less than clear. Your input is valued.
I imagine that if I had to join you on a German board, I'd sound no better, since I learned my German as a child, which means my technical vocabulary is almost non-existent.
Your boldness in participating in fairly sophisticated conversations here is commendable. Keep it up!
Paul
Doug, I suspect the last may be closer to the truth. But even if that 3.8 comes along, I'm fairly confident at this point that AMD will be right where they need to be to at least compete effectively, if not maintain their lead. Add the power advantage on top of that and things will be looking good.
Paul
dougSF30, I think I'd disagree. The 125W would not only require a new cooler but re-designed MB and different PSU from cost-conscious OEMs. I think the speed is possible for Intel to get around, but the power issue is becoming more and more important to everyone involved, consumers and OEMs alike. But in any case, both sides can be argued.
Paul
Sorry, I was out of the loop for a while and forgot to check the time stamp. <:)
sgolds, dougSF30 - the clock speed of Prescott is not what is the main concern, I think the power consumption is the real issue.
Paul
dougSF30, sgolds - c'mon, guys, chill. Doug, you've made your opinion known. You do not have the responsibility to continually warn others about the dangers of TA, though. If someone is interested and wants to pursue it, hey, we're supposedly all adults here, able to make decisions for ourselves. Some folks have an interest in it, so let them deal with sgolds in peace. I don't think anyone would mind a "Be careful with TA!", but let them decide if they want to ask you why and pursue that side further. sgolds, you're right that it's a waste of time to try to convince Doug, so give it up(Doug, I'm not saying you're ignorant or anything, just that you clearly have a set opinion on the matter, based in your own convictions). You guys both have too much of value to offer this thread to be wasting time on this argument.
re: Ace's server review. I'd love to hear some analysis by folks here who know a heck of a lot more about this field than I do. What do you think of the Quad? Any glaring weaknesses in the review? Remarkable things to note?
Paul
Keith, re: confident about flawless execution on the 90nm process. Agreed, I'll believe it when I see it, but it beats the heck out of not hearing anything or hearing stumbling statements about it!
Paul
Keith, what makes it worse is that Pravin is usually on the ball, with rather uncanny instincts for timing his transactions, so when he speaks, a lot of people will consider what he says very seriously.
Paul
Looks like it may be another buying opportunity!
Paul
Ixse, these links are all in the bottom of the info portion of the main page for the message board. But, for the sake of easing your access, here they are:
Please read the Terms Of Usage before posting!
http://www.investorshub.com/boards/complex_terms.asp
Also keep this in mind:
http://www.investorshub.com/boards/read_msg.asp?message_id=1109442
http://www.investorshub.com/boards/read_msg.asp?message_id=1218679
http://www.investorshub.com/boards/read_msg.asp?message_id=1273556
Posts that clearly violate these terms will be deleted without prior notice. There will be no discussion of deleted posts on this board. If you have any questions regarding this issue, please report them here: #board-504
Update to Anandtech article:
http://www.anandtech.com/systems/showdoc.html?i=1925
UPDATE 12/01/2003: Elite PC has provided an update on Dual Processor usage. "The only AMD processors that can be run in DUAL mode are the 200 and 800 series Opteron processors. The motherboard in the Titan FX machine will run all 940 pin processors, however only the 200 and 800 series Opteron processors will run as Dual processors." AnandTech just tried to benchmark the Titan FX system with 2 FX51, and while the system worked, the boot screen only identified ONE CPU even though two were installed. Oddly, the few benchmarks we ran were higher than single CPU, but not at the levels we would expect with Dual Opteron chips. Elite PC recommends the 200 or 800 Opterons for Dual CPU operation.
(underlining above is mine, boldface is Anand's)
Interesting - wonder what the reason for the increase was?
Paul
All, re: Feb Intel launches. Before we assume too much and get too wrapped around the axle, let's recall that xbitlabs has been dead wrong on a number of occasions. I'm not quite satisfied that things are going that poorly for Intel, a few more data points that are more reliable are needed. But it's certainly info to be considered. If true, then AMD has a great opportunity if they can execute smmoothly on 90nm.
Paul
Doug, don't forget that there are those on the other extreme, like facsnotfiction(counterpart to subzero) and Dan3(counterpart to the other group?).
Paul
wbmw, prepare to be assimilated - resistance is futile. ;)
Keith, "Nervous Nellies" who are used to the old AMD taking profits?
Paul
P4EE finally available at newegg. $1025! They also have an A64/K8V/cooler package for $529 after $25 rebate. Hmmm... I think I'd have to take choice B!
Paul
Doug, I don't know, that's a pretty radical thought there! ;)
Paul
sgolds, wbmw, EP - maybe the Itanium cache situation is what's driving the building spree. Taking Itanium to the desktop? :P
Paul
Ouch! Hot, hot, hot!! Tejas is supposed to be even toastier than Prescott.
Paul
http://www.digitimes.com/NewsShow/NewsSearch.asp?DocID=CEC31C4F878E15BA48256DF000424927&query=AA...
Aavid and Enlight to battle computer heat
Original date: 2003/12/3
Translator: Chinmei Sung
Taiwan-based case maker Enlight yesterday announced a strategic alliance with US-based CPU cooler supplier Aavid Thermal Technologies. The new cooperation will address the growing heat problem in computers as processors run faster and devices get smaller.
Aavid estimated that Intel’s Prescott processor, manufactured with a 90nm process, will generate 20% more heat than a current Pentium 4 (P4) chip, made using a 0.13-micron process. Tejas processors, which will be introduced in 2005 and made with a 65nm technology, will generate 50% more heat than a P4.
Taiwan’s Market Intelligence Center (MIC) estimates that small form factor mini-barebone systems will account for 1.5%, or 1.659 million units, of the forecast 110.6-million-unit desktop shipments this year. The number is likely to exceed two million units next year, following the launches of new models based on Intel’s introduction of the Balanced Technology Extended (BTX) form factor specification.
Enlight and Aavid will work together on computer cases designed to more effectively dissipate heat. Enlight chairman and president C.M. Liao estimated that the company can save up to 10% of costs through the alliance.
EP - Where did this come from? Sounds like they've done well.
Paul
wbmw, I don't think you need to worry about that; Johan is a man of integrity.
Paul
Nice, if true. Opteron leading Itanium in server sales.
From DRBES on SI.
http://www.bayarea.com/mld/mercurynews/business/7397043.htm
Bloomberg News is reporting that Advanced Micro Devices Inc.'s newest microprocessor chip for server computers runs twice as many of the servers sold in the third quarter as a competing Intel Corp. semiconductor, researcher IDC said.
Computer makers shipped 10,746 servers with Opteron, the chip from Sunnyvale, California-based Advanced Micro, Intel's biggest rival in processors, IDC analyst Mark Melenovsky said. Intel's Itanium powered 4,957 servers, which run networks and Web sites.
Opteron has won clients such as Sun Microsystems Inc. and International Business Machines Corp., giving Intel's processors for servers their biggest competition ever just as businesses start to increase spending, analysts said. Revenue at Advanced Micro is forecast to grow more than twice as fast as Intel's this year and next, helped by Opteron sales, according to analysts.
I would think it reasonable to do the benchmarks both in the standard configuration and the tweaked configuration, which is a very common convention among hardware reviewers.
Paul
PS. As long as it's properly labelled.
Ya know, it'd sure be nice if the drive-by insult raids, condescending attitudes and such were ratcheted way down. After all, this isn't a sandlot fistfight center, it's supposed to be a place where relatively intelligent thoughtful people of differing opinions can come to discuss AMD and the environment that affects the stock so that all can come away with a better understanding of how to invest. No, I'm not trying to be a nanny, I just fail to see any benefit coming of some of the rather immature antics some are displaying here(on both "sides" of the fence), and I certainly see it working to the detriment of this board.
Just my 2 cents.
Paul
Keith, why do you say that, when that's about the timeframe when AMD is supposed to introduce 90nm K8? Assuming, of course they do and it goes relatively smoothly...
Paul
alan81, I don't imagine tuning will be so much of an issue with Xeon systems. They've been around longer, the tricks have already been learned tested and instituted. Opteron systems are still new enough for lots of room to check out tweaks.
Paul
Hmmm... 8-way Tsunami soon to come?
http://www.tempestcomputers.com/amd/amd_index.htm
from aceshardware forum
http://www.aceshardware.com/forum?read=105055794
Paul
Keith - the main reason I've visited his site is more for Joe's heatsink reviews, but they've gotten rare lately. Ed does have some good editorials, but he has slanted more toward Intel lately. He does periodically make a good point, though. I don't ever rant about his rants, though; just take them with a grain of salt, use what I can and move on.
Paul
EP- true, but it does indicate room for improvement. IT will only be proven reliable with time.
Paul
EP - not likely. Johan doesn't work that way. More likely it has to do with something new the BIOS allows, or just plain ol' discovering tricks about how to better configure the BIOS, which will rapidly be adopted by the community, and at a slightly slower rate, by the OEM. Just because the OEM didn't think of something doesn't mean it's not valid. And if it provides a legitimate 10-30% boost without inducing instability or other problems, then you can bet your bippy they'll institute the changes after thoroughly validating them.
Paul