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.
chipguy - He didn't say they'd(ORNL or Cray) bought them yet, just that AMD had very likely already produced at least a significant portion of what is required, which makes perfect sense.
Besides, who said they'll stay on the shelf for 2 years? The system won't just pop up overnight, I'm sure they'll be starting on it very soon, if they haven't already.
Intel doesn't seem to be able to crack 20. Maybe after Conroe's out. Don't think their Q report is likely to help.
alan81 - Not any real guesses yet that I recall hearing.
chipguy - Watch out, those sour grapes are bound to produce some nasty wine!
Will start to take delivery when? Don't you love how journalists today seem to forget the 5 W's(and H)?
Part 2 of Z-RAM story at digitimes.
http://www.digitimes.com/bits_chips/a20060329PR201.html
Thanks for the word, we'll be watching for it. What kind of press are you involved in?
CJ - Nah, even gov't has its limits. I imagine it's a joint CIA/NSA venture. ;)
CJ - Kinda like harnessing nuclear fusion, eh?
cruzbay - Fascinating! Thanks.
Thanks. I was curious if they'd gotten to point where they'd developed computer-based tools that could directly convert the file into light patterns to do the job. Guess that's still future..
Ignorant question here:
Are mask sets actual physical entities(sort of like a stencil)? Are they now (or are they likely to become) data files that are directly used by what I would assume would have to be new types of tools?
Congrats, Keith!
pfosse - The problem being that now is when the price war might help Intel, and AMD doesn't have any imminent products containing Z-RAM, even if it does pan out as nicely as they say.
Maybe more resources than you realize.
http://www.realworldtech.com/page.cfm?NewsID=359&date=03-28-2006#359
Oops, I see Joe already noted it.
chipguy - re:"Here is yet another embedded control customer, one with a
natural leaning towards an AMD based solution, that AMD
is not able to satisfy because it backed the wrong ISAs for
low power applications."
Oh, the IRONY!! ;)
DoU - re: "AMD becomes nothing more than roadkill."
Gosh, doesn't that sound familiar!
cruzbay - I especially like this line:
"(Oh yes indeed my fellow Mac users; Apple is poised to have a repeat of the G4's infamous FSB bottleneck shortly after the switch to Intel. Look on the bright side, though: the situation won't be quite as dire... at first.)"
Ouch! They won't like the sound of that!
DoU - Conjuring up unfounded FUD again, eh? mmoy said nothing about buying the trial, much less for $250; he did mention buying the OS for $89 after his trial expired.
DoU - Can you say "trial"?
mmoy - No kidding - just the CPU, HD and memory are darned close to being worth that much!
CJ - He's made it quite clear he's clueless about the situation. Looks an awful lot like it's a hefty case of sour grapes. Even with Otellini on the board, Intel couldn't swing the sale, which is likely quite substantial. For him to claim the software problem is somehow an AMD-related issue just causes him to further discredit himself.
drjohn - re: "Thanks but no thanks you guy's can be the genuine pigs I will wait untill the real deal is ready for prime time which may not be untill 6 months after launch."
You need to either learn how to spell or stop insulting people. Which is it?
drjohn - Perhaps one might try reading the story rather than just the headline?
CJ - Sincerely wrong is still wrong. ;)
cg - Bookmarked for posterity.
That's also only single channel memory, though I don't know how memory-bound Cinebench is.
mas - It'll be awhile, though, before it's here in enough volume and has the widespread support/acceptance to make any real difference for AMD.
mmoy - It certainly can, depending on what the BIOS recognizes. I use mobile XPs in several machines, and depending on how I clock them, they may or may not be recognized as legitimate Athlon XP models. This is particularly an issue when you delve into the territory that corresponds to unreleased models, which BIOSes may not yet even be updated to recognize; the 2.8GHz FX would beone of those, right?
chipguy - c'mon you know there's more to technology leadership than process feature size. Though I don't know if I'd assume AMD has taken the lead. It certainly is closer to it than it has been in years past.
No doubt, as it stands now, Conroe looks tough. What we see when independents get their hands on them will likely look a little different, but I have no doubt it will remain quite impressive. To think otherwise would be foolhardy. The $64 question now is what AMD will produce in response. :)
Well, Conroe looks pretty impressive.
SP - IIRC, transition to 65nm production to begin 2H06, substantially complete by summer 07.
Joe - Or it could mean that increasing die sizes will eat up the capacity. Just to play devil's advocate; which I think we have to do to do our due diligence. I'm sure Intel will do their best to push die sizes up if it can pressure AMD.
wbmw - Better look closer. It said full-speed at 68W. So it's not so much an increased power envelope as a marketing decision to use their new process advantages to give the market what it wants - a reduced power envelope with full performance.
Opteron 185 is "available" at Monarch (Special order, 4-6 days) $1189 Interesting - that's $50 more than the 285. They probably have more demand for 1xxs than 2xxs, though.(Both retail box versions, BTW)
Link
V
V
http://www.monarchcomputer.com/Merchant2/merchant.mv?Screen=PROD&Store_Code=M&Product_Code=1....
chipguy - Fortunately, AMD didn't count solely on 64-bitness in its plan for success with K8.
They seem to have done quite well, and frankly, when the time comes that 64-bit software is more pervasive, who's going to have had the most experience dealing with it, as well as having the most mature and established x86 64-bit processor?
That's likely to mean something to a lot of customers, don't you think?
Interesting little tidbits of data from Charlie re: Conroe vs. Hammer:
-----------------------------------------------
http://www.aceshardware.com/forums/read_post.jsp?id=115156611&forumid=1
Mikem wrote:
A Conroe is 20% faster then a Hammer clock for clock.
A Conroe E6600 316$ is as fast as a AMD 5000+ 2,6Ghz,
AMD is usually 10% cheaper then Intel, so a AMD AM2 5000+ would cost 285$.
Nope, not clock for clock. Bzzzt! Next.
-----------------------------------------------
http://www.aceshardware.com/forums/read_post.jsp?id=115156650&forumid=1
Eric Bron wrote:
Mikem wrote:
Intel Conroe 65Watt
E8xxx: 3.33 GHz / FSB 1333/ 4 MB shared L2 cache 1000$
E6300: 1.86 GHz / FSB 1066/ 2 MB shared L2 cache 209$
(Eric Bron)- Same TDP for these two ? you're a really optimistic guy, aren't you ?
XE has a much higher TDP. Twitboy is factually wrong. Again.
-Charlie
---------------------------------------
cantonese - If you're out there, could you possibly hunt down and look at the article in the Economic Daily News referenced here:
http://www.digitimes.com/systems/a20060303PB208.html
(Sorry, no direct link given)
See if you agree that Quanta and Gigabyte are the likely builders of the AMD Google servers? The Register thinks Sun is the most likely candidate.
http://www.theregister.co.uk/2006/03/04/goog_opteron_sun/
rlweitz - re: "...as you end up with alot of Semprons attempting to make a few FXs.
How so?