Slashdot Mirror
Intel Encounters Another Problem with RAMBUS
Palin Majere writes, "News.com is reporting that Intel is once again having problems with its RAMBUS memory chipsets. This time, it's affecting the i820 and i840 chipsets, and is located in the chipsets (MRH and MTH) that allow customers to use regular SDRAM memory instead of RAMBUS memory.
It causes memory corruption and has already caused Intel to cancel three motherboard designs as a result. " With the continuing shortage of high-end Pentium processors, and stuff like this, it's no wonder that AMD has been doing better and better.
"Welcome to the Twentieth Century where Trusts crush personal liberties and individual rights and the government calls it capitalism"
Well guess what kiddies, that's what capitalism is all about, and you dopes thought it meant freedom! BWA-HA-HA-HA!
The true mantra of Capitalism is: "Free Markets, Not People!"
Heh you must be joking seing that every test shows that cpu for cpu the athlon beats equal Mhz Cpus from intel. Wake up and smell the coffee
got one...Dell is about the only place u can get rdr and a coppermine in under a month
You are comparing apples and oranges. Beside the fact that you are not comparing systems running at the same clock, the SPECint scores for the Athlon system are based on older versions of Microsoft's compiler and libraries (with different optimisations, as well), which completely invalidates this "comparison".
All the tests comparing the Athlon and CuMine when running the same code show the Athlon massively outperforming CuMine in floating point, and with similar performance in integer provided the Coppermine is using Rambus DRAM (it really sucks otherwise).
Also, one can afford a much faster Athlon system for the same money, since there is no need to spend hundreds/thousands extra for RIMMs, and this also should be taken into account when comparing systems (unless money is no object, which isn't true for most users). A price/performance comparison of Athlon vs. Coppermine would completely blow the CuMine out of the water.
You wouldn't happen to own any Intel stock, would you?
It seems that trollers are posting less frequently now and that they are posting more interesting stuff. This is pretty kewlio y0!!!1
I just dumped my RMBS stock anyway. What a fiasco this has been.
Intel doesn't own Rambus, so they can't make this decision. Rambus is a separate company (NASDAQ:RMBS) that licenses their designs
I had an AMD DX4/133 processor, so I'm more 3133T3 than you there, bub.
YOU LAFF CUZ YER AN ASSRAMMER
blah, you fuckin suck
I agree fully with you, especially those people who complain about ONE little story. However, i've tried to submit 4 very relavent and interesting stories in the past few days and they've all been "surpassed" by a much more poorly written submission. More terse, less descriptive and stimulating...
I just think Rob & Hemos are out to get me. :-/
30% is probably too low but is not that far out. It's wrong to assume that adding a CPU will double performance.
Kernel compilations are alternately CPU and IO bound so benefit rather a lot from 2 CPUs.
I recently timed seti@home on my new BP6 with 2 x Celeron 500s vs my old Celeron 300A overclocked to 450.
Running a single SETI client gives pretty much the same timings as the old machine (faster CPU, but slower bus)
Running two of them at once gives a total of about 1.5 times the performance though each task takes 40% longer to complete. The seti client isn't multithreaded but is CPU bound and is likely not optimised in any way for SMP.
This is on kernel 2.2.14. Hopefully the 2.4 kernels will see an improvement.
My system is now not so much faster for everything as more responsive.
It's not a problem with SDRAM either. It's a problem with the versions of the i820 and i840 chipsets which ALLOW you to use the faster, cheaper SDRAM. When you use RAMBUS with it, you get corrupt memory problems.
Dell also claims to be running W2K for their corporate website... which means you probably shouldn't beleive anything Dell sez.
And by the way, I'm still running my Celeron 300As overclocked to 450MHz while awaiting the next generation of AMD processors...
DROP IT NOW!
A.C. --pretty sure his next big PC purchases will be AMD Caspian dual K7 the way Intel is behaving
You've got it all wrong!
They're posting it as a cover so you don't think their journalistic "integrity" has been compromised.
hey yo, rambo says to pour a hot bowl of grits down your pants, or i'll shot you in the eye. thank you.
im a tR0lL
fuck da m0dZ
I encountered a problem trying to RAM my cock up Mariah Carey's ass -- no KY Jelly! Do you think this is related?
Why was this post moderated down? It is ontopic, it talks about the connection between Intel and slashdot. The story is about Intel and it's on slashdot therefore the message couldn't be more ontopic!!!
What does it matter? The story got posted and now there's a discussion... That's all that would have happened if it'd been you that posted it.
> using these chips because they're DAMNED GOOD,
> not because intel's chips are crappy
I bought an AMD because PIII have CPUIDs.
The fact that the AMD kick ass is an added benefit
Intel will have a tough job getting me back as a customer.
Side note: Beware AMD + Asus K7M motherboard sucks
until you have reset BIOS to default. The two levels
of CPU caches are disabled [but marked enabled in the BIOS]
You can check this by running TestCPU.exe in (erk) windows.
I got about the performance of a PPro200 from an AMDK7/550.
If you overclock the CPU too much, the BIOS reset itself,
and the same problem reappear.
Cheers,
--fred
a motherboard with "BP6" features or better for those Athlon cpu's...
with a little group of hired gcc developers by AMD,
with a little non-NDA agreement to let every purchaser download every information on motherboard asic's, ic's, SMP documentation, and official online e-mail/usenet/www/irc support,
with a little support from Linux, NetBSD, OpenBSD and FreeBSD.
Caspian dual K7---Q3 2000 Right in your time frame :-)
Caspian is the dual cpu chipset used in ALpha 21264 machines, I'd expect this one to be targetted at the high end K7, Mustang.
Intel has stopped making SMP-able celeries as of 11/1/99 (source: evergreens slocket info page)
ever notice how the intel logo has less of a drop-shadow than the very pronounced ones given to the other icons, like the linux penguin (just a hunch that it'll be on every page), and the slightly lesser one on the slashdot.org logo for stories about slashdot?
the dropshadows just look amateurish. drop them and nothing will be lost.
Nah, Intel reported that every 486 PCI chipset was defective in someway or another. PC AT Forever!
A modern 'gaming' video card has 16 or 32MB of memory -- can't imagine that AGP would being doing much for you then.
AGP is really only a 'feature' for older video cards or crap PCs that use main memory for video ram.
"ASSRAM" would be what exactly, "Another Silly Specification-RAM"?
-- Guges --
Man, you know what gets me? All of these ignorant people who blindly follow AMD simply because they are the "little guy" picking on a big corrupt giant like Intel.. They don't seem to realize that AMD is no different than Intel in the long run. They're a company like any other, trying to fill their pockets and win market share. Picture this: Ten years down the road - Intel is but a distant memory, and AMD is standard.. I'm willing to bet that if that scenario ever comes to be, AMD will be the exact same Intel we see today. Competition won't be as fierce, so they won't have to work as hard. Which means they'll be able to slack off and charge more for an inferior product. People want AMD to be on top... They aren't thinking about the consequences of that. I think things are great the way they are right now. AMD and Intel are pretty much tied. If it stays that way, then you know both companies will be on the ball. Pentiums have never been so cheap in cost since AMD started gaining market share. - Just my 2 cents.
Intel has made a sizeable investment in VA Linux (even before VA went public). We don't know the details, but it's interesting that VA Linux doesn't offer any Athlon boxes. If Intel suffered, it's not too much of a stretch to think that VA Linux would be in a pickle to the extent that they were wed to (possibly inferior) Intel hardware. If VA were having troubles, then Andover and Slashdot might suffer as well. In other words, follow the money, and the web of dependencies it fosters.
Since VA now owns Slashdot (via their purchase of Andover), It's interesting to see how "independently" Slashdot will report on stories that cast a poor light on one of the prinicpal investors (Intel) in Slashdot's parent company (VA Linux).
Another thing screwed up with athlon boards/chips is limit of 768MB max memory.
Moderate this up! +3 Insightful
I don't know about your computing needs, but I do know Rambus wasn't designed with my needs as a consumer in mind. Rambus, theoretically, is good at streaming large amounts of data--or it will be good for that oneday--but as others have mentioned that ability comes with a price: a higher latency than conventional DRAM. The dope from CS people is that for large servers it may be very well and good to serve databases and streaming media from Rambus based systems, but for the short hop, two carryon-bags in the overhead compartment kind of computing most civilians do its pointless to have that kind of bulk throughput--at the expense of higher latency. OK, me not computer science student, just consumer; but accepting the technical charaterizations put on the Rambus v. Dram divide by the people-who-know [1], what I think we have here is an economic and moral question. Is Intel trying something new and noble? Or are they just trying to subsidize their assault on the enterprise computing space by making Joe Sixpack pay out the wazoo for a technology that is actually illsuited to his needs--especially considering that a more appropriate, more proven technology is at hand and vastly cheaper.
Now I do feel a little warmth in my heart for INtel knowing they have given some encouragement to the development of Linux; but I also know that they have done that for the most businesslike and self-interested of reasons, not out of idealism. Not that I fault them for it in any way: no one is being cheated or deceived. But Rambus is a different matter. In this it appears that Intel has sought to use their overwhelming market presence to make all of us on the lowend pay for their higher ambitions. What is of geatest utility to the greatest number should reflect the economies of scale effect in its price, relative to commodities that benefit fewer people and are in less demand; Intel's bid to push all the RAM makers to Rambus would invert the natural effect of that rule & arbitrarily deny the beneficial operation of marketforces to consumers. IANAL but that looks like a textbook distortion of the free market process, which if sucessful, would warrant an FTC/DoJ inquiry.
Is AMD doing something cutting edge? I couldn't say. I'm sure some people would say they must be doing something that looks like hard work if they can build a better Intel cpu than Intel can. Other than that, you could say they are synthesizing some technologies of the Intel standard architecture with some that are fundamental to the Alpha processor world. That's probably "nontrivial"; still even if they were sitting around drinking beer and tossing darts at photos of Grove and Barret, I don't think that would make me feel Intel was justified in ripping me and a couple hundred million other people off with Rambus.
oooh, now I'm really peeved! Fie on them!
[1] Tests have shown first generation Rambus equipped systems are actually slower than the current generation DRAM systems they are supposed to replace.
WTF? VA/Andover would have to own part of Intel for Intel's stock price to affect Andover's.
Athlons support SMP fine, it's just there aren't any SMP motherboards available yet. :-(
Though they should be relatively quick in coming, since the Athlon uses the same bus as the Alpha.
Thing is, Athlons are so fast that SMP (as opposed to clustering) is unrealistic until memory bandwidth picks up. So we need 266MHz DRAM ASAP, and the Rambus fiasco is slowing down the change to faster SDRAM.
AGP 4x Versus AGP2x, feh --don't worry: whether by plan or accident you made the right decision and you're not missing anything even if you did want to 3Dgame intensively.
Ah but they haven't pulled their heads out--not fully. Willamette boards will be RAMBUST only. If you're Intel fans, better get ready to prove your love for Andy!
No amount of volume will get the cost down if they stay a lot worse than SDRAM.
They closed up the consortium or whatever, it was in EE Times a few months ago but some of their work will be folded into the DDR-II spec according to the story.
The AMD DX4s were cool - so long as you had a PCI motherboard to go with 'em. (As opposed to VESA, etc.)
So, when the hell are there going to be SMP Athlon motherboards to take advantage of this? I remember there being much fanfare about this capability when the Athlon come out.
The IBM PC project was CHEAP, relatively speaking for IBM... they only thought they'd sell 250,000 of them and wound up selling MILLIONS, and had good market share until the PS/2 fiasco (basically IBM's RAMBUS, but at least Micro Channel worked... in fact PCI is quite similar to it, only Intel knew how to sell it :)
If IBM had kept refreshing the original design and/or made MCA open they would have been fine... they could move tons of machines on name alone, and _did_.
- Chad
I can't beleive anybody would be stupid enough to moderate that down. That's like feeding the conspiracy theory (which seems to be pretty correct from what I've read...).
It's like ten people in your city saying they just saw a UFO, and then they "disappear" the next day.
I don't even have to explain why moderating that down what stupid. It had better be obvious to anyone out there. If it isn't, I'm damn surprised you know how to work a computer.
Fortunately, I never joined the scourge of moderators (the reason for that is obvious too), so I can't change the moderation on the parent post.
I see the offtopic tag being used correctly (parent post has 1, offtopic score).
:-)
-1, offtopic is a useless waste of moderator points. -1 is for garbage, but it seems to be applied to things that are offtopic. Why? If it is offtopic it is not garbage. If it is offtopic garbage, then it is a troll/redundant. It shouldn't be marked offtopic!
Just a minor rant, sorry to bother you.
(Note I'm talking about PC 100, 64 MB SDRAM here...)
;-). If you pass all the 1 per person companies, and look for the first non-unknown company (ie, one that plans to stay in business long enough to keep their name), you'll find the cheapest memory is $42 US.
:-(
Yeah, it looks cheap on pricewatch, but it isn't (as usual
That is $60 CAN (where I live). I bought memory last summer for $55 CAN from a retailer. That should be about $50 CAN internet purchase price. Now, memory should have gone down in price over the year, so I expect to pay $45 CAN retail.
And until I can pay $45 CAN again, I won't buy, because it is overpriced (To me, I don't care if it is lower than the manufacture cost. Once you sell something low, some people will expect that to continue forever. I'm one of those people.).
Once SDRAM is my price again, I plan to make a HUGE purchase (about 512 MB) of it. I won't be stuck with a pokey 64 Mb computer for a year again (actually, it was 128 MB, but I decided I needed yet another computer, and had to split the memory between the two)
I won't buy memory at inflated prices ever again. I bought 4 MB extra for my 486 in (I think) 1996. It cost me $60 a stick (4 x 1 mb). Within 2 years, it was worth $5 a stick. Never, ever, will I again put money into something that loses 90% of it's value over two years. Sure my computer ran faster, but at what cost? I have never seen something crash like that, including all other computer parts.
Oh, but I do agree, RAMBUS sucks for all the reasons you mention (it is WAAAAY overpriced, ridiculously more than SDRAM is overpriced right now).
Now, since RAMBUS is "special" RAM, you might be interested in this story:
In 1998 I received a bondwell 486 laptop (made in about 1995/1996). After repairing it (bad power supply), I decided it needed more memory (it only had 2 mb or so). I noticed the memory expansion was proprietary, so I phoned bondwell. bonwell told me I was looking at paying $600 for a 4 MB upgrade. They never changed their price from what it was back in 1995, and they didn't care one bit that it was ridiculous today.
That is what could happen to your RAMBUS systems if you are unlucky enough...
Why -1 for the parent post? Maybe the person posting it doesn't know AMD Althlons support SMP, but AMD doesn't offer motherboards/chipsets for that purpose.
And no, offtopic is NOT an excuse to give something -1. Offtopic means offtopic, not garbage. Simply mark it as what it is (offtopic, and up/down if you feel it necessary). If it is offtopic crap (ie, troll), then the post deserves -1. The parent post isn't crap by a mile. Maybe a little redundant, but otherwise, it seems fine, but offtopic, to me.
Why not use your moderator points to mark down a troll, or mark up something useful?
Nobody noticed that the title of this article is grossly inaccurate?
The problem is not with RAMBUS, it's with SDRAM (and ECC SDRAM at that) in RAMBUS native chipsets (i820, i840). The problem is strictly related to the translator chips that translate between SDRAM and RAMBUS signals. This is like saying Japan sucks because all the japanese translators screw up the translations.
Perhaps people should pay more attention to what they write.
I'll trade you a 386@40 (also AMD).
yeah but the stock market is so fucked up you never know what will make you gain or lose money.
I have a new Dell 420 workstation w/ RAMBUS memory. Does this have any affect on me? Or is it just isolated to Intel boards w/ support for SDRAM too?
if it's not the fucking rambos, it's fucking taco and hemos begging me to sodomize them....
I agree with you in general, but there are some exceptions to this rule-
Like MS.
MS secrecy and control have allowed it to sabotage software competitors. And it has not really caught up with them yet.
shutup whore
>What IS this?? /. like 4 times in the last few days,
>I have been SHAFTED royally by
>putting up a good story with a great writeup & link and then NOT
>getting it posted. Then a shittier writeup gets put up like a day
>after i post my submission.
>Is it possible they hate me??
You know I hope there's a blacklist that's been created just for people like you who claim they submitted stories and then bitch about it if they don't get "credit" about it. Maybe then asses like yourself will learn to shut up.
What IS this??
/. like 4 times in the last few days, putting up a good story with a great writeup & link and then NOT getting it posted. Then a shittier writeup gets put up like a day after i post my submission.
I have been SHAFTED royally by
Is it possible they hate me??
From a motherboard manual, error beep codes: S-L-L-L-SS: Speaker Error
Just received my Athlon yesterday... man, not only are these things badass performers, but they actually FEEL beefy!
Esperandi
Whenever I hear RAMBUS, I try not to laugh too hard because I once almost choked to death on my own spit just pondering RAMBUS...
If I ran Intel, I'd certainly shaft the public for every penny they're worth and be a complete capitalist corporate cunt of a company.
I think that's why their CPUs are so expensive...
Except for you unfortunately.... oh wait... DOH!
...don't let the door smack you in the ass on your way out.
;) to need a heatsink/EMI shield in normal (FCC class A/B, near-room temp) environments.
You linux-heads should hate rambus...half the reason a 128mb stick of rambust costs $700 is to make up the 'IP royalties' for rambust and its shitty R&D costs. Try www.pricewatch.com, see how GODDAMN cheap sdram is now. Read on to find out why this matters.
DDR is going to breath some hot fire up rambust's ass. After intial speculative price costs (aka hoarding and exploting shortages of a new tech, i.e. coppermines) ddr will be aroung 10% more than single rate. The actual costs to the mem makers will fall to 2-3% more per chip a year after the lines start running. (some already started lines, like micron)
Why does rambust suck?
Oh the REASONS, the REASONS!
Its the first type of standard memory (uh, Cray computers cooled with special carcinogenic coolants don't count as standard
POWER drain, BIGTIME. Contrary to DDR, where it gets FASTER, then uses LESS power. One of the first markets expected to convert to nearly 100% penetration of DDR is the laptops.
Because of the above, you are damn lucky if more than a thrid of any module is getting used at one type.
LATENCY. As many cable users can tell you, ping is king, bandwidth is not. (DSL tends to have better pings, and most important to hardcore users, more stable over time...not 120 one minute and 40 the next, which screws up your instinctive responses, like leading your prey with the crosshair) Online gaming can do fine with 5-10k/sec, but you will be staring at your lifeless corpse more often than not if your latency is high. Bandwidth is important, but in the age of 8x and higher CPU mutlipliers, latency reduction can easily double or triple actual performance.
COST COST COST. The equivalent for computers that LOCATION LOCATION LOCATION is for retail sales and real estate. Who gives a shit is rambust does 15% better in intel/rambust crafted benchmarks...I can outfit a server with a freaking gig+ of sdram for the cost of 128 rambust. Anyone wanna wager which one would get left in the dust under heavy use? "7:1 odds, come and get em!"
/end-o-my-rant...almost
If any linux user supports rambust, a word comes to mind - Hypocrite! IBM and others are GIVING away tech info on DDR. Want a free predone module design? Here ya go! Want test results on trace timings? Here ya go! And so on and on and on and on...
AMD has said since long before Athlons were released that it's chipset was only a prototype until third parties made some. Also that's why they haven't produced an SMP chipset. I've had mine for three or four months (550 Mhz) and haven't had a problem with it. I'm just not into all the 3D games, so I don't care about AGP 2x, 4x, etc. But as a real workhorse it's fantastic, keeping pace with a friend's BP6 dual Celeron/466.
Hopefully soon VIA will take the chipset to the next level, SMP.
Though I too have no idea where you get the 30% schtick, these systems were using Linux 2.2.14. I compared compiling a kernel with the same config, as well as some kde2 cvs source (kdelibs I think it was). With the kernel, I could do it in 2-2.5 minutes on my Athlon/550. Using just one process on the BP6, it took about 4.5. Going with -j2 it cut it down to a tad over 2 minutes, well better than your 30% rule. With similar performance in the other things I've compiled since then, sometimes I win by a few seconds, sometimes the BP6 does, I've come to the conclusion that they are pretty much the same at compiling. To me, this says a lot about how well the Athlon runs. Once SMP Athlon boards come along, there will be no comparison to the BP6 anymore. ;) Maybe by then the PPGA P3/500 will be cheap enough to load in.
Thanks alot! Very informative!
Pax -- Ob
... and your source for these numbers is what?
Who did the benchmarking?
Linux is only free if your time has no value. Windows is only free if you threaten to use Linux.
FYI, AMD750 is history. VIA has finally produced their KX133 chipset, which among other things allows AGP4X.
People have complained about the inferiour chipset for the Athlons. It's just that picking on the little guy is not what people like to do.
AMD's U.S. fabs have completely switched over to 0.18 micron aluminum interconnect for Athlons. Dresden is sampling copper interconnect chips (Thunderbird core, 256K on-die L2 cache running full CPU speed), with volume shipments in June and buyable in July. (They demoed a 1.1GHz Thunderbird a few days ago, no special cooling.)
Yeah, I want a multiprocessor Athlon box too.
As I read over the specifications for the Intel CC820 Motherboard (the 820 chipset with the SDRAM converting MTH chip on board), it specifically states that the board does not support ECC SDRAM. This has been included in the docs since the first CC820 board we put in a PC. Sounds like it's a problem they've known about. *Shrug*
Well, things are changed very, very rapidly for past 2~3 years. Now you can't differentiate personal workstations from PCs, due to rapid speed-up with PC processors.
PowerPC technology, and Pentium technology have had some problems in them. Although the RAMBUS technology is not specifict to the Pentium processors, it shows that things should be changed less rapidly.
I personally think that PC industry makes lots of wastes. Unlike TVs we have, it's very hard to use PCs more than 10 years. ( Well, you can, but you will be tempted by new S/Ws, faster graphics, larger S/Ws.. ) Sometimes, people may need not-so-powerful machines, but people buy them. PC companies always makes faster machines, instead of lowering the prices of existing machines.
For example, Pentium 133 Mhz machines are still usuable for many games, wordprocessing, etc.
But you can't buy new Pentium 133Mhz machines with prices under $300.
Hmm..
Am I too pessimistic?
I don't know about it... In general, I buy the best product I can with the money I have to spend... If AMD's chips are for some reason inferior to Intel's when your Caspian motherboard arrives, are you still going to buy the AMD chips anyways?
Are you nutty? Memory speed doesn't need to pick up to make SMP Athlons attractive purchases... So many apps times are spent rerunning the same chunks of code, and manipulating the same chunks of data for more than a few clock cycles at a time, that so long as a CPU's cache is efficient, they'll stand to benefit greatly from SMP, so long as the apps and OSes in question support the extra processors.
Just because one particular chipset design has problems doesnt mean that Coppermine is going to lose out. At the computer shop I work at we are using good old pc133 RAM with the Advance 5 133 Chipset from VIA. So far we have had absolutely no problems with these boards. I would highly recommend them to someone looking for a coppermine solution.
ps If your looking for fast reliable RAM you can pick up pc100 RAM that will run 100% at 133 (look for the 222 refresh rating) pretty cheaply.
AMD 386/40... nyah nyah.
Karma: Good! Napster: Baad!
BTW, does anyone know if Athlons are being made with .18 micron processes (last I heard they were still at .25) and/or copper interconnects yet, and if not what the current planned dates are? </i>
The answers are yes and probably. Probably, because I am not 100% sure. Try checking out overclockers.com
Karma: Good! Napster: Baad!
I can't tell where you pulled that 1.3x-figure (pulled from sleeve or asked wife I s'pose), but if you have good software, 2 or more different threads and ideal application (compiling stuff, rendering animations or encoding mp3 for example) it's more like 1.8-1.95x (depending on io-rate). My friend bought an Athlon 550 and I went for BP6+2x466 oc'd to 2x545. Neither of us have regretted. I can rip and encode 3 cd's to mp3 in about half an hour (using Grip, see Freshmeat) and compile a 2.2.14 kernel (lots of stuff enabled) under 2 minutes. With athlon and a bit less hefty configuration, 2.2.14 takes about 2.50. So, all in all, it depends what you plan to do with your box.
-- v --
Can you say, 'loser'? No? Reread that first paragraph.
Can't you see that you're bragging? If you want to provide stats, give a link to one of the many publicized reviews.
or how certain OEMs save a couple million here & there by removing true on/off switches from the front of their computers.
Come on people, why is this comment being moderated up? This person is announcing the imminent death of Intel merely because of reports of a few problems in their new chipsets. And open source/specs is not a magic potion that solves every problem, especially not from a large corporation's point of view!
On the contrary, history has shown that opening specs can be a very _bad_ thing for revenue, all ethical concerns aside. IBM suffered greatly because they let others produce clones of their hardware. Microsoft and Sony, with very closed strategies, are thriving.
Don't moderate up a comment just because you like, in principle, what it's saying. Moderate comments that really are insightful and well thought out, regardless of your personal point of view.
Broccolist
The new VIA chipset changes that
Your statement, "Intel has never had a processor of [sic] equal processor speed that outperformed Intel", is patently false. AMD's Athlon outperforms a Coppermine of equal clock speed by greater than 17%, in both integer and floating-point operations. Why do people still believe that Intel remains unchallenged?
*cough*microsoft*cough*
I honestly dont think AMD will do that well. See intel has a whole line of technology lined up and they have always done the best with their technology. Using the CISC way of doing things, amd has never had a processor, or equal clock speed, that outperformed intel. Also the sse componant in intel chips. However I also think that AMD's risk with their own motherboard was not a good dicision and that could also lead to less chips being bought, but, amd can turn themselves around a little.
"If a man watches 3 football games in a row he should be declared leagaly dead" - A
I'm concerned. AMD's got a nice processor. I have no first hand knowlege of it, because I'm still a step or two behind (CELERY 300a! Hell yeah), anyway.
Intel's major foibles seem to occur when it tries something new, like this RAMBUS thing. Has AMD tried anything new, besides the change to slot 1 and a new chipset. Intel seems to be trying to break new ground in a major way, not just steps.
I'm a usual supporter of the underdog, but only if they are better for what I need.
So, what do you guys think? Is AMD just sitting still, not really doing much besides evolving, where Intel is running into trouble because they're trying something new.
Of course, I could be wrong and missed something that AMD has done.
later
Dan
aparantly you never heard about the airline company who saved several hundred thousand dollars every [x] months by removing 2 olives from every salad.
However, I also know they are wasting other people's money, that they made with their ultrahigh margins, the highest in this industry.
:o)
Funny. Dell says "PIII is more stable..." what a nonsense
Sigged!
I have a question: you say that your Athlon 550 MHz system is faster than a friends dual Celeron 466 MHz (guess he uses the Abit BP 6). What OS are you and your friend using? I am asking this because NT and Linux have mediocre SMP support. It's pretty hard to take advantage of SMP in these OS-es, so I guess the typical gain of adding a second CPU, in NT, is about 30% (if used as a typical workstation, not server), and Linux w/kernel 2.2 would do about the same (again, working typically as workstation).
Now, if I do the math, 466*1.3=606. Taking into consideration Athlon's advanced risc-like core, it becomes clear why your system outperformed your friend's.
However, if you were using something like BeOS, which has gains of 70 to 99% with the second CPU, the dual celery would smoke your system bigtime.
So what kind of OS do you guys use? And how did you perform the performance measurements?
Why I am asking: I am planning to buy an Abit BP-6 w/2 Celeron 466 MHz myself, but I need to know whether it makes sense, or should I rather buy an Athlon 550, with the AMD chipset (I don't give a shit about AGP 4X, and the mobo with the AMD chipset is 35 US$ cheaper here in Helsinki!), which is the system you have.
I plan on using BeOS and Linux.
Sigged!
I think things are great the way they are right now. AMD and Intel are pretty much tied. If it stays that way, then you know both companies will be on the ball.
tied? AMD's market cap is about $6 billion. Intel's is over $350 billion. While their chips may be similar in performance, thse companies are not equal. Intel is quite profitable while AMD is usually losing money. AMD's profitability is slowly changing as a result of Athlon, but they are still not on equal footing with Intel.
I think that the x86 cpu market would be served well by a somewhat larger AMD. I'm not saying I want them to dominate the market, but I think that if they were closer in size to Intel they could build more fabs and spend more on R & D (though they seem to be doing quite well on the R & D front these days). I would also like to see competition from other companies. Via and Transmeta are starting to provide that competition. It will be interesting to see where things go in the next couple of years.
joe
To use the FireWire protocol doesn't require royalties, you need to pay Apple to use the FireWire brandname and logo. This doesn't count as part of the technology in my book.
Where is my mind?
mfspr r3, pc / lvxl v0, 0, r3 / li r0, 16 / stvxl v0, r3, r0
Check out Project Upper/Mute, an all-around awesome compiler fra
Nice try, but you're just confused on the whole subject. You see, Intel can't get RAMBUS to work and if they did it would be even more expensive than their overinflated underperforming chips.
I think the whole thing is a scam by Intel to get people convinced that computers are SUPPOSED to be tremendously expensive (they were supposed to be 5 years ago, but not now) again so that people won't laugh so hard when they see that a P3 is $100 more than an Athlon at the same clock rating and the Athlon grinds its ass into the dust on performance.
Esperandi
LONG LIVE AMD! It's nice to see that justice is eventually meted out on those who engage in bad practices. Now if only it would happen soon to some other companies...... *cough* mpaa *cough
With Intel trying to play catch up with AMD and dropping RAMBUS in favor of traditional SDRAM after having committed to RAMBUS, it shouldn't surprise us that they are having a hell of a time. Frankly, I am surprised that they have not had more problems. True, they have been plagued with bugs, but they redesigned the whole board in a matter of weeks. Imagine if they didn't have a talented team...
Hmmmm I will say that AMD is truly in good shape when we can get MB's to make dual athlon systems. Without those AMD can't seriously challenge Intel in the x86 server/workstation markets. Sometime this fall or winter I will be looking to build a new system and it will probably be a SMP system. I hope AMD has SMP enabled chipsets read for that time so we can get SMP Athlon MB's cuz I really want a dual processor Athlon system with RAMBUS memory since it runs at 200mhz.
So the problem here is really with SDRAM compatibility.
Doesn't anyone see this as BENEFICIAL to RDRAM support? I'll bet Intel takes these chipsets, pulls the SDRAM controller back out of them, and re-releases.
Whammo. They then have an excuse to not support SDRAM. Plus, this makes SDRAM look unreliable (which it's not). Maybe this is part of a plot to push people to RDRAM. I, for one, wouldn't put something like that past the Big I. But then again, I'm notoriously paranoid...
--Superunknown[GP]
The above comment is CopyWrong (K) Erisian Entertainment. All Rights Reversed. Ewige Blumenkraft!
I was thinken about getten the asus p3c-d motherboard (2 intel 800mhz cpus overclocked to 1ghz each can anyone say supercomputer)anyway if i was to use sdram with it insted of rdram (with the asus dr2 dimm riser)and not use the ecc would there be any problems? P.S. dont flame me for goin with intel you dont see any dual amd k7 motherboards out there do ya.
We have recently been in contact with a swedish company that sets up computing cluster to run a program called Fluent to do Chemical analysis I believe. According to them the Athlon systems that they have outperform equivalently clocked PIII's by 25-30%. Talk about blowing away...:o)
They just can't get it up with the 820, 840 chipsets and Rambus
Intel has been rushing their products, but AMD is still struggling to make industry connections. Macs are still running on 100 MHz busses. I don't know about the more exotic architectures, except that they cost an arm and a leg.
WARNING: there is a trojan on your
Just where are you getting your facts from? My athlon smokes my p3 of the same clock speed.. okay.. well maybe not SMOKES but it is noticeably faster..
----------------------------
----------------------------
Esobofh - Currently drinking fresh mango juice.
I'd also like to add that RAMBUS isn't an open technology -- you're free to make all the RIMMs you want, as long as Rambus Inc. gets a royalty. I was under the impression that everyone had learned from the ongoing FireWire fiasco that royalty-based consumer technologies don't get adopted.
I understand their motivation for earning some kind of money for all of their R&D, but RAMBUS has failed to provide either a compelling reason for most of the world to drop SDRAM and switch to RDRAM -- the price doesn't match the performance.
There may be a place for RDRAM at the high end, but with the memory requirements for current machines, I'll be damned if I go back to paying more than US $1/MB for RAM, much less $8...
--
--
fat lenny's gonna lick your brain today.
I think it's good that the two companies are competing (it gives buyers lower prices and better selection), but it appears Intel tries to rush products out the door too fast... I for one am not clamoring for Rambus the same way I would clamor for the next beta of Action Half-Life... From the reviews I've read of currently-available i820 setups, you don't get much of a performance increase for all that extra money you pay... right now, I'll stick w/ AMD chips, or maybe the good ol' Intel Celeron... can't afford a Coppermine right now... but I'd rather have an Athlon anyway... I don't wish Intel would crumble... but I like 'em to feel the heat of close competition! This Rambus story brings to mind the Pentiums when they first came out... they had 'problems'. So I waited until the Pentium was nearly obsolete to buy one : ) just my 0.02
Some days I feel like I've been tarred and gzipped...
...also, there is something vaguely uncool about the acronym RIMM... I like "DDRs" better... kind of like Deutsche Demokratische Republik... but much cooler...
Some days I feel like I've been tarred and gzipped...
1) To Intel: "Ha ha ha ha ha ha. Nyah nyah!"
R andom-Access-Memory, for those who aren't in the know.)
2) Let's hear it for DDR-SDRAM!
3) We don't have enough letters in front of "RAM" yet, no! I want SCFLEADDRAM!!!
(that's Super-Cali-Fragi-Listic-Expi-Ali-Docious-Dynamic-
4) AMD rulez! Oh man, I want a Crusoe. My K6/300 is just sucking lately.
---
pb Reply or e-mail; don't vaguely moderate.
pb Reply or e-mail; don't vaguely moderate.
This is just a minor nitpick, but the marginal cost will not move to zero, but to the per unit license fee that Rambus Inc. charges.
What I really don't understand is why Intel hasn't cut its losses on this. From an uninformed observer's perspective, it seems that there is a strong incentive for _all_ players to move to DDR in the near term and offload Rambus tech towards "special projects" which need tomorrows bus speeds today (which is to say not servers, or PCs, but research devices).
I'm guessing from an economic perspective that they are weighing the advantages of control (they do get sole distrobution rights from Rambus, right?) of the market greater than the traditional advantages of low fixed capital transition costs. This seems a bad strategy to me, since intel really isn't in the memory market, and that they're primary sales are in semiconductors.
Ah well... I'm sure their decision makers have more info than I do. :)
Pax -- Ob
For all their billions, I don't see how they can survive this one. A break-up, whether horizontal (the DOJ's preference) or vertical (which is what I'd like to see), could be fatal to Microsoft. It has no experience in dealing with a free market. And it has a large number of blood-thirsty competitors sharpening their knives, just around the corner.
Even if they did escape, this time, what would that do? The tower is wobbling, and each new battle will shake it a little bit more. The longer it survives, the worse the final crash will be. Sooner or later, investors, supporters and backers will realise this and jump ship. Not because they're cowards, but because nobody hangs around under those conditions.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
Intel *had to* release these chipsets & boards to get people to use Coppermine. That's because the price of RDRAM is over 6 *times* the price of standard SDRAM. The RDRAM-to-SDRAM translation was added as an afterthought when intel finally pulled its head out of its ass and realized people are not gonna pay 6 times the price for a product that has no real advantages over the existing one.
___
___
If you think big enough, you'll never have to do it.
I've read a quote from the CEO of Rambus that it would have succeeded if Intel hadn't gotten in a hurry and pushed it.
Wouldda, Couldda, Shouldda...
Supporting the AMD Athlon is great, but alas, I haven't heard any news about any Athlon-compatible motherboard chipsets that will support the upcoming DDR SDRAM running at 266 MHz.
Once that happens, THEN we might seriously consider the AMD Athlon as a serious alternative for high-end server and workstation applications.
Raymond in Mountain View, CA
Opening specs in a stupid way (like IBM lisencing an OS and chip without any control over them) might be stupid, but building entirely closed systems is even stupider. Look at Apple. In the late 80's it was pretty clear that the Mac OS was superior to anything on the PC side, and Macs were competitive with anything on the PC side. In large part what killed them is that they got fat and arrogant. They overcharged their customers, ignored their needs, and generally pissed away their lead. Now that's not strictly a result of having a closed system, but without the competition from Mac clones, they got fat and lazy. As a result, PC's ate their lunch with inferior products, simply on the basis of better prices, and the increased varieaty and responsiveness of a competitive marketplace.
If IBM had successfully prevented cloning and Apple had pursued a substantial cloning program, we might very well all be using Macs right now. And in that case, IBM's PC revenues might very well be even less than it is now.
I believe what you meant to say was that royalty-based consumer technologies don't get adopted unless they're supported by Intel.
_________________
rooooar
... the harder they fall. It would be nice to see Intel taken down several notches, IMNSHO. While Intel's stuff generally works (unlike the computer industry's other monopoly), it is over-priced, and Intel has a record of engaging in anti-competitive practices. I don't want to see them fail, but I think their grip on the PC industry needs to be loosened up a bit.
Intel tried to design a system that would be expensive to clone, and would corner them the market. It's failed.
Sure looks that way, although I would be warry of counting my chickens before they are hatched.
Back in the late 1980s, IBM tried a similar tactic, with a closed, proprietary, and expensive system bus called MCA. It completely flopped. People never learn.
Rule #1 of the Hardware Industry: Don't Try to Make Money Licensing Your Design. It is too easy for someone else to make their own design without paying you.
Intel's best hope of survival, never mind market domination, is to open the RAMBUS specs completely.
I wouldn't go that far. They haven't bet the farm on RAMBUS. Intel has other products outside of the world of memory. Perhaps you've heard of their Pentium line? :-) The failure of RAMBUS won't exactly feel good, but it won't kill Chipzilla.
dragonhawk@iname.microsoft.com
I do not like Microsoft. Remove them from my email address.
The problem with AGP 2x isn't with the chipset, it's with manufactures who decide to save .005 cents on a motherboard by not putting in an extra capacitor.
Good post!
Moreover, Rambus costs $1,000 for 128MB. Check out:
http://www.mushkin.com
..one of the few online dealers where you can even find it. Intel it betting a LOT on RAMBUS, and pissing off everyone with their flaky i820 and i840 chipsets. Check out how many big OEMs now offer Athlon system. Intel has really left the door open for AMD. If AMD can get to market with a chipset that supports DDR-SDRAM, 4X-AGP, and SMP, they will put a serious hurt on Intel.
The one thing that will keep Rambus Memory, Inc. afloat is the fact that Intel supports them, and it will be the memory in Play Station-2s. But if the price does not fall down to somewhere within the range of DDR-SDRAM (which already exists for video cards) it is going to fail in the desktop market and take Intel's i820, i840, Pentium-III, and Williamette with it.
I'm really looking forward to buying an Athlon box as soon as SMP and DDR-SDRAM support are a reality. The VIA KX133 chipset is already a very nice stable platform for uniprocessor/SDRAM setups.
---
True - but as I understand it the issues have been more to do with volume up to this point
What I really don't understand is why Intel hasn't cut its losses on this. From an uninformed observer's perspective
I think that long-term Intel probably wants to move the memory controller onboard for all but high end MP systems - one advantage that RamBus drams have today is that they allow more concurrency in the dram system - in particular more 'open banks' (ie sense amps with cached data) and more concurrent RAS cycles in the array. Making use of this sort of stuff is very difficult for a memory controller which necessarily sees transactions serially over a (relatively slow [compared with rambus speeds or cpu clocks]) slot1 bus. With CPU clock speeds getting faster and memory not the CPU architects are in a bind - they are spending big on things to make up for the slow memory latency (not memory transfer rate - read latency is the first order effect here that's the killer) like big caches. But their wonderfull superscalar and/or VLIW CPUs are stuck talking to a potentially fast memory system through a slow serial pipe.
The concurrency in the multi-bank architectures in the memory system can really only be used effectively directly from the CPU where the concurrency from the CPU architecture is directly expressed. My guess is that long term the Intel designers would like to pull the RamBus controller onto the CPU die so that they can attack their latency problems.
For the record there's another way this can work too - RamBus is narrow - you can toss 2-4 of them onto a die (if you can afford the area and power costs) where you can only afford the pins for a 64 or 128-bit bus. You don't have to run the RamBuses in lock step - instead you interleave the shit out of them (4-way for 4 interfaces - every 4th cache line from a different bus) this aagain allows you to increase your concurrency - at the expense of the customer having to stuff all 2/44 buses identically (2/4 simms at a time).
Having said all this I think that competing technologies are trying to push at the multi concurrently active bank thing too. I think that Rambus just started evangelizing that first.
MCA was a great idea. It worked like a charm on their 'larger' systems, and filled the need for a faster-than-ISA bus. VLB wasn't yet commercially produced, PCI was still a twinkle in someone's eye. IBM had peripherials available for MCA, it was better than ISA by an order of magnitude, and they could control it. What would you do?
.sig: Now legally binding!
I think the fact that there's just NOW coming to market a decent Chipset for the Athlon has hurt AMD quite a bit. I also think AMD should come out of the closet a bit and share what they know of why their Irontgate chipset isn't always compatible with AGP2x as it's spec'ed to be.
.18 micron processes (last I heard they were still at .25) and/or copper interconnects yet, and if not what the current planned dates are?
OTOH, Intel is having a MUCH harder time with the new boards (i820 and i840) - the number and seriousness of the errors on these things in crazy.
And combined with the disaster-in-the-making known as IA-64 (personally, I think it seems like a good idea on paper, but there are so many problems I don't think that anything good will come of it), and their production problems on high end Pentium III chips, Intel is not doing at all well.
By comparison, AMD is doing good. The Athlon is doing great and it seems that the architecture will hold up for quite some time (unlike Pentium IIIs, which IMHO are pretty much on their last legs as a viable design for new chips - hence Willamette and IA-64, neither of which will be here for at least 6-9 months). The chipset problems are a disadvantage, as is the lack of availability of SMP Athlon boards.
BTW, does anyone know if Athlons are being made with
Really??? I got an AMD 486 Dx-2 50 I'll let go really cheap.
AMD doing Better and Better??
I think the fact that there's just NOW coming to market a decent Chipset for the Athlon has hurt AMD quite a bit. I also think AMD should come out of the closet a bit and share what they know of why their Irontgate chipset isn't always compatible with AGP2x as it's spec'ed to be.
And while Intel may have they're bugs, they're very public these days, and therefore they're fixed very quickly, and you can typically get the fix without too much trouble or cash. Granted, that's not necessarily the case with the RAMBUS issue, but who has the money to buy the stuff at this point??? (:
Yes, I may get flamed for saying that bit about the Irongate, (Some think it's the Athlon MB Manufacturers not meeting spec, and nVidia hasn't entirely sidestepped blame..) but it's AMD's Processor AND Chipset. If something's not meeting spec, they should do something about it, or at least make the consumer aware of a problem, and what to watch out for.
So while AMD may truly look to be doing "Better and Better" both chip makers still have their own problems to deal with.
AMD is making doing better and better because their chips are getting better and better. I used a K6 back when no average user had ever heard of AMD. Now the athalon is all the rage. people are using these chips because they're DAMNED GOOD, not because intel's chips are crappy - that's just a perk :)
-FluX
-FluX
-------------------------
Your Ad Here!
-------------------------
"It is seldom that liberty of any kind is lost all at once." -David Hume
Historically, not a very good argument. Remember the original IBM PC? They published all the specs openly, and then were immediately surpassed in the market by a startup known as Compaq, who could build on (& tweak) the design without having to recoup the millions (or more?) that went into the original research. IBM all but died in the PC market, shortly after initiating it.
Of course, then there's also the Apple story. Keep it all locked up & sue anybody whose product shows the slightest resemblance to yours. I remember reading a quote several years ago that went "You're nobody in the computer industry until you've been sued by Apple."
This seems to indicate that neither strategy is a good way to go. The companies that end up the winners are the ones that wait for somebody else to make a breakthrough, then make a small improvement (be it speed, on-board cache, or whatever) and sell at discounted prices. The only expenses to be recouped are the costs of reverse engineering the original, not development from scratch.
- - - -
Correction: DDR-DRAM is much faster then Rambus. 100MHz DDR-DRAM has bandwidth of 1.6GB/s, same as Rambus. 133MHz DDR-DRAM has a bandwidth of 2.1 GB/s (that's giga *bytes*, btw, not bits). DDR-DRAM, as well as regular SDRAM, also have a significantly lower latency.
And as it happens now, memory bandwidth is not the bottleneck. Even 800MB/s of regular PC100 SDRAM is plenty for 99% of applications, including the latest 3d games. Just about the only thing that would make use of the higher bandwidth is large databases. Too bad you can't put more than 512MB of RDRAM in a machine... ;-)
However, lower latency is guaranteed to boost performance a bit, no matter what kind of application you are running. This is where standard SDRAM and the upcoming DDR-DRAM have an advantage over Rambus. Not to mention the cost...
So, the whole situation can be summed up in one sentence: Rambus is just an inferior product with a ridiculously high price.
___
___
If you think big enough, you'll never have to do it.
DDR-SDRAM is great, but while you can get them in quantity, FINDING a motherboard that supports it natively is quite something else. :-/
I want to know when will the VIA Apollo KX133 chipset be upgraded so it will support DDR-SDRAM.
Raymond in Mountain View, CA
While I don't think that this is the death of Intel, for they have too many fingers in too many lucrative pots, it suggests that they have misstepped badly with RAMBUS are are going to lose their market dominance in CPUs if AMD keeps it's act together... which other than Irongate and a scarcity of Athlon MBs, they have been doing fairly well.
As far as Irongate's AGP issues, there isn't a whole lot of difference performance-wise at this point between AGP 1x and AGP 2x. Maybe in the next iteration of video cards we'll see a more significant difference, but I'd rather have a CPU that is 15%-20% faster than sweating about a 4% hit on the AGP bus. Some people feel that Athlon is not being entirely honest & ethical with the issue, which may taint their reputation in the long run.
My next machine will be an Athlon based system. I've suffered extreme technolust since they were released, and they just get better and better.
It will be quite some time before Intel has anything in market to compete with Athlon, and by that time it might be too little too late. Their most recent efforts have yielded uncertain results in comparison to the Athlon.
-- benton.
Rambus (Inc.) is a company!
They(Rambus Inc.) have designed a memory type called DRDRAM that only uses a 16bit wide external databus, and 8*16bit wide bus internaly.
As always when it comes to Intel it's only MHz that counts, not what they do with those. DRDRAM can handle 800MHz but as the bus is only 16bit wide it wont be very much faster than the 64bits(At most twise.).
I'd put my money on SLDRAM, it will be atleast twise as fast as DRDRAM and is, unlike DRDRAM, an open standard. SLRAM shouldn't have a problem doing 3GB/s+, at much lower clock speeds.
"Last words are for fools who haven't said enough." - Karl Marx
It's failed.
I hope Intel, and other chip manufacturers learn from this. Secrecy and control aren't cool. They can, and will, turn around and bite you.
IMHO, Intel's best hope of survival, never mind market domination, is to open the RAMBUS specs completely. Do a hardware variation of the GPL. If they don't, it's going to bleed them dry. If they do, sure, there'll be clones, but Intel will still exist.
Given the choice of pride or survival, Intel needs to think about that survival option a bit more.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
The MRH's are Memory Repeater Hubs, and come in two flavors - MRH-S and MRH-R. The -S does translation from rambus to normal SDRAM. The -R is to let you get aronud the 2GB limit on rambus. A rambus channel can have up to 32 256Mb devices, or 1GB. An MRH-R has two channels dangling off it, doubling capacity. Put two MRH-R's on one channel from the chipset, and you can have 4GB.
One interesting thing about this solution is that it takes time to go through the chips, increasing the already high latency of rambus.
I wonder if Intel would alter their decision on Rambus, were they able to go back in time and do so. They might pull it off yet, but it won't be easy. If it does work, it will only be because they are Intel.
The enemies of Democracy are
Reading the article my take on it is that the problem is in the device that does the SDRAM to RamBus conversion (ie it's a channel adaptor that lets them mix and match rams types) - and the problem only occurs when you use ECC.
I can think of 2 reasons this might happen - either they got the ECC logic wrong (probably likely), or there's a noise problem on the sdram side when they drive 72 data pins [for ecc] rather than the usual 64 (less likely). Either way it isn't a RamBus problem.
There's a lot of noise made about the various merits of memory types - my personal take on it is that it's mostly a wash, RamBus drams do have some advantages - but for main memory systems they are more in the future (and revolve around how many chips it takes to make a minimum memory sized system as memory continues to move down the memory density curve - M$ may of course make this moot). Their main disadvantage is cost - and it's rather a chicken and egg sort of thing - if people use them a lot the marginal cost of RDRAMS will probably go close to 0 - but if people don;t use them in volume because they cost more that won't happen. Remember in the core of a RDRAM is the same core that's in an SDRAM it's just the interface circuitry to the pads that's different.
Rambus is a design for a memory system from Rambus Inc. It is extraordinarily fast on paper. Intel chose their design and decided to support it on a lot of their new products.
The implementation took a long time to get around to getting around. It is now here. Intel bet a LOT on Rambus, because it would give them significant control over a lot of markets. (IE: They own rambus designs)
Rambus is significantly different from the DRAM used commonly today. It requires changes to how stuff is laid out on the motherboard. And it is manufactured differently, to very demanding tolerances.
It is now in production and is competing with DDR-DRAM, which uses existing manufacturing processes, generally works with existing chipsets, and is easy to support. And it doesn't require a fan setup for the memory alone. And runs far cooler. And gives almost as good performance when set up correctly as a RAMBUS setup. And is also capable of being manufactured in quantity, whereas RDRAM is extremely difficult to manufacture. DDRDRAM is also about a fifth of the cost of a RDRAM setup.
You do the math, and read up on it a bit.. I think you will agree that for all intents and purposes (read: mainstream pcs, servers, et al), Rambus is DOA.
Toodles.
-troll taker