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.
Petz - That is correct, at least with the nF3 based MSI K8 Neo Platinum board. One of my grid crunching compadres has one and can OC about 20-45MHz higher using IDE drives rather than SATA. And that's using the nVidia SATA in the chipset! The external chip-based SATA does even worse.
Paul
Joe, Indeed it's very handy - I've been using it for a while. Works great for using mobile CPUs in a desktop and manipulating the multiplier and/or Vcore.
http://crystalmark.info/?lang=en
mmoy, if it didn't before, it supports "x64" now.
OT - my "noggin" status update :)
http://www.nforcershq.com/forum/viewtopic.php?p=437003#437003
Good news. :)
tecate - I'd be astonished if it were. As many times as it's been rumored, why would it already be priced in this time?
Paul
mmoy - how much L2 does your 754 3200+ have, 1MB? That might fit. Gets a little confusing at times with all the options and ways of getting to the model #. Newegg has three 3400s: 754/1MB/2.2GHz(ADA3400BOX - retail), 939/512KB/2.2GHz(ADA3400DEP4AZ), and 754/512KB/2.4GHz(ADA3400AEP4AX)
Paul
Anything look odd about the DIMM modules on this Broadcom chipset AMD board?
Paul
http://www.theinquirer.net/?article=19684
chipguy - In a high-bandwidth, low-latency system, wouldn't they want to take advantage of every CPU's free HT link? Is it feasible for a SeaStar chip to be built with 3 more of those HT interfaces as well as all the system interface goodies? I don't know, just wondering if there's a legitimate functional/performance reason for them to do it the way they did.
Paul
UnD,rlweitz - the 754 3400+ is a 2.4GHz chip according to newegg. They do carry a 512K 2.2GHz S939 3400+ as well.
Paul
chipguy - What are the L1 & L2 cache sizes on those chips? Don't you think that the fact that Itanium is dealing with an L3 rather than main system memory is a large factor?
Paul
mas - I'm sure Opteron will eventually have full 64-bit addressing(or whatever is considered adequate/necessary), they just decided it wasn't necessary until they'd established a secure foothold in the server market.
Paul
wbmw - Not for long. It's going to be far more widely used, so that's a huge motivation for developers to work hard on it. Not to mention hardware makers and their drivers.
Paul
SmallPops - Sounds like "whistling past the graveyard" to me.
gollem - Fortunately, it's been handled by jail cell bars! :)
chipguy - re: "What I won't do is ignore the hard lessons about the relative importance of technology, business relationships, marketing, and economics in this industry learned from observing it for more than a quarter century. Lessons that as a chip design
engineer are sometimes hard to accept."
All very valid points, perhaps you could point out what you see AMD doing right in those areas a little more often.
Paul
jjayxxxx - Thanks! eom
nVidia already shipping nF4 A02 & A03 steppings
Paul
Nvidia already shipping A02 and A03 versions of nForce4
Charles Chou, Taipei; Steve Shen, DigiTimes.com [Friday 12 November 2004]
"Nvidia has already starting shipping the A02 and A03 versions of its nForce4 PCIe-enabled K8 core logic chipset, stated Drew Henry, general manager of platform business at Nvidia, yesterday during a telephone interview with DigiTimes.
Henry made the remarks in response to market speculation that Nvidia would probably delay shipping the A02 version from November to December, because the chip cannot support HyperTransport (HT) speeds of 1GHz.
Henry said that Nvidia designed the nForce4 A02 version to support AMD (Advanced Micro Devices) 754-pin Athlon 64 CPUs and 800MHz HT. Nvidia is also offering an A03 version of the nForce4 chipset, which supports AMD 939-pin Athlon 64 CPUs and 1GHz HT, Henry added.
Henry affirmed that Nvidia has started delivering both the A02 and A03 versions to its downstream clients."
http://www.digitimes.com/NewsShow/MailHome.asp?datePublish=2004/11/12&pages=A7&seq=40
Keith - re: free Opteron publicity - :D
Paul
Folks, not being a chart-following person, I've not even looked at Mr. Nazdaq's stuff. I assumed someone would speak up if it were spam. If y'all feel it is just spam, I'll start eliminating it. Let me know what y'all think.
Paul
bobs10 - So, maybe hacking out a chunk of my brain didn't affect my memory too badly after all!! :)
wbmw - Forgive me for not having a link, but unless I'm mistaken, I seem to recall one of AMD's "high mucky-mucks"(Sanders/Ruiz/Rivet?) saying back after they worked things out with Intel (a couple of years ago?), that those restrictions were no longer in effect, or at least no longer an issue.
Paul
Keith - Looks like someone knew; kind of helps explain the recent runup, doesn't it?
Paul
Keith - I had my doubts about ATi pulling it off first time out - I'm tickled to be proven wrong! :)
Paul
Keith - WOW! Looks like Anand thinks ATi really got it right!:)
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2269&p=22
This could be good news for AMD, too. A lot of these folks will be getting, or have gotten, K8 chips.
Paul
http://www.xbitlabs.com/news/video/display/20041105095529.html
rlweitz - You're assuming it was the CPU that was the unavailable item. Do you have any basis for that assumption? They do have some of their own specialized components that are just as "key".
Paul
Duke- Yes, I see a pattern, but all the Intel products being discussed are future as well, so what's your point?
Paul
Keith - re: limiting advantage - I don't know that AMD will be "disadvantaging" itself with that tighter envelope. I don't think there's nearly as much patience for loud, power-hungry server systems as there used to be. As long as they can make sure they've drawn that envelope line in the right place, they'll be fine.
Paul
From Anand: Maybe this is why nVidia has delayed on IGP K8 chipsets?
Paul
http://www.anandtech.com/mb/showdoc.aspx?i=2264&p=3
"The performance enhancements that go along with Revision E chips include some optimizations to the Athlon 64's memory controller. The more optimized memory controller improves bandwidth efficiency with regards to unified graphics memory accesses; given that the only type of graphics that uses system memory (and thus the on-die memory controller) is integrated graphics, it's safe to say that the Rev E chips will offer better integrated graphics performance."
fpg - That is hefty density!eom
Keith - Yes, there are always those folks, but most gamers who are really serious build their own rigs. And word of mouth is very swift and very potent in this community, and it trickles down to those who are not as serious, even among those who don't know all that well how to build their own systems. For now, those who are serious all-around gamers know the system to have is A64. No ifs ands or buts. There may be specific games where Intel can hold its own, but they're far fewer than those which A64 "owns". So why would they choose to limit their gaming? They won't. This can certainly change, but for now AMD is in the catbird's seat re: gaming.
Paul
Keith - re:"I only hope that the target group for this CPU, as small as it may be, does their dd and avoids this chip."
Rest assured that the huge majority of them are well informed and will do so.
Joe, Keith - Is ATI's IGP a dual-channel setup? If not, that may be a problem for any I/O intensive uses.
Paul
Keith - re:HP - :D eom
OOPS - My bad - I meant first silicon, not tapeout!! :(
Paul
Sun tapes out.
I suspect it's supposed to be Niagara rather than Niagra. Perhaps the author is excessively familiar with Viagra?
Paul
V
V
V
V
http://www.siliconstrategies.com/article/showArticle.jhtml?articleId=51200997&_requestid=325838
What are the chances of Samsung becoming a foundry partner for AMD, now that they've announced their intentions to get into that field? They do at least have experience with high-performance CPUs. If I recall correctly, didn't they fab Alpha for Compaq? And arguably make a better chip? They shouldn't have much problem with Sempron K7, wouldn't you think? Or is that all ancient history, now irrelevant?
Paul
Keith - Certainly some truth to that. However, I still expect that they will wind up with the premier "chipset" for K8. It just would've been nice if they'd gotten nF4 right on the first go-round. But then, Via screwed up their first versions of most of their K7 chipsets(remember all the "A" versions?), and they're doing just fine now. nVidia is in the unenviable position of having to deal with 2 tough competitors on different fronts, but they'll get it together, I just hope it's sooner rather than later.
Paul
wbmw - Connectivity Solutions, for one.
Paul
V
V
V
V
V
V
V
http://www.amd.com/us-en/ConnectivitySolutions/ProductInformation/0,,50_2330_6631,00.html
wbmw - You're better than that! It only took me a couple of minutes to find AMD's solutions on their website. Try again.
Paul
fpg - I believe it's a mix of Intel's hypocrisy - "Intel's "leadership role" in getting 64-bit computing to the desktop""
and Keith's irony - e.g. Intel is at it's best in its hypocrisy.
Paul