Slashdot Mirror
AMD Thunderbird And Duron Set For June Launch
Chris Tom writes: "Mark Hachman has an article up on EBNS discussing the AMD Thunderbird and Duron. Topics include release date, die size, performance, and positioning against Intel's CPU offerings. The Thunderbird and Duron will both include on-die L2 cache. "
Can anyone with inside information explain to me the reasoning behind the Duron's cache setup? It just seems that having an L2 cache smaller than the L1 cache wouldn't increase performance much over L1 with no L2 cache, while it does increase die size. Is there a reason why 64/64/64 was chosen over 64/64/0 or 32/32/128?
Dell's argument about the supply chain is laughable at this point. Over the past year we've had i820 delays, i820 bugs, i820 bugs again causing a recall, processor delays (although generally speaking not critical ones), and processor shortages from Intel. From AMD we've had... a lot of CPU's! ;-) Admittedly their Irongate Chipset has been far from ideal, but once you throw VIA into the mix, things look a lot better.
Seriously, AMD's record over the past decade hasn't been great, but over the last year it appears to be much better than Intel's. If supply was really the issue I'd think at this point Dell would have taken the initiative and dabbled a bit with AMD (nobody says you have to commit your entire product line to AMD chips).
sigs are a waste of space
AMD is keeping the wraps on this new processor for a reason, and I bet this is it.
Thanks for the insightful information! (^_^)
At least now I know there will be Socket A motherboards available when the "Thunderbird" Athlon ships in June. I didn't want a repeat of last August when CPU's were easy to come by but motherboards were hard to come by.
Raymond in Mountain View, CA
Actually, it's quite the opposite. I pick what processor I want, and what price I want to pay for it, and then I wait for it to hit that price.
I usually buy processors at around $150.00. It's a little high compared to the low end, but it's still high enough that I can feel close to the bleeding edge.
--- "So THAT's what an invisible barrier looks like!" - Time Bandits
The article commented that AMD has supieror production yields.... which puzzles me. I know Dell (gee wonder why I know) is iffy about using AMD procs because of supply problems. It'd be nice to see AMD push stuff out the door so larger OEMs can begin to really consider them for servers and higher end markets. Even if their procs are faster,better, etc than Intel.... overall Intel can provide more quanitity to a company that demands high supply. If AMD is sitting around going "well, we're a little low" that'll drive the OEM's prices up (unexpectatly)...
---
Well, with Athlon, you have a superior chip (than the P3) at a lesser price. What more can you ask?
The Duron will be a major improvement over K6-2 though.
So what would a two-proccessor machine be? A Duron Duron?
I don't like fish. Reverse the fish to e-mail.
AMD Moron processor as well...
BlackNova Traders
What happened to VIA's Celeron competitor? I wanted to buy two of those and MP those things.
Imagine a .....
I can see the future news releases now...
Today Intel, in response to the strategic alliance of AMD and Ford, is planning to team up with Chevrolet.
Intel's plans for the future include a chip named the Intel Camaro, which is going to be a direct competitor to the AMD Mustang.
-- Dr. Eldarion --
It's not what it is, it's something else.
Duron is kinda interesting... sounds vaguely like a Klingon name... "Duron will destroy its enemies on the field of silicon battle and drink their blood! Qapla'!"
Then again, that's not the worst naming gaffe that could be made. I will *never* forgive Intel for not naming the Pentium II the Sexium. Even worse is them not then calling the Pentium III the Septium. That way if something went wrong in the manufacturing process it could have been the Deviated Septium. *rimshot*
In all seriousness, the Duron should have enough of a price/performance ration to seriously give Intel some competition. The next computer I plan to be will be a Duron box, and I suspect I'm not alone on that one.
Of the top 10 PC manufacturs, 9 use AMD parts (incl. Compaq, Gateway, IBM, HP...), with Dell being the only hold out.
.13 micron process technology - Mustang may well be a ".15" .18/.13 micron process hybrid).
Guaranteed volume may be a factor of why Dell hasn't yet adopted AMD parts, but another major reason is that it would jeapordize Dell's special relationship with Intel.
Still, performace and price are likely to force Dell to use AMD processors this year in order to remain competetive. Thunderbird is likely to be around 10% faster than PIII at the same clock speed, but will in fact also be available at much higher clock speeds than PIII.
Ignoring Dell's limited edition 1GHz PIII made possible by their Intel relationship (1GHz PIII won't be in volume production until Q4), the fastest PIII currently available is the 866MHz.
1GHz Thunderbird vs 866MHz PIII is a big performance gain for Dell to expect it's customers to give up. The latest Computer Shopper is even advertizing the as yet unannounced 1.1GHz Thunderbird, which may be a surprise announcent at the launch on June 5th.
Intel are not going to be able to match Thunderbird clock speed or performance until Willamette is launched towards the end of this year. At that point AMD Mustang will be out, and may well again be the faster processor (no benchmarks for either Willamette or Mustang are available, but AMD technology roadmap indicate they are about a year ahead of Intel in the move to
Then Intel went to coppermine: 256K of L2 cache running full speed. This yielded about a 10% speedup: for instance see Tom's comparison here.
Meanwhile, though faster Athlons were released (at 700, 750 MHz), they had no faster L2 cache: they had to run at 40% or 33% of the core speed. The result? Coppermine beats Athlon, narrowly. You can see it happening in this graph, if you imagine the blue line extending about straight (which is more-or-less what happened.) The two chips are quite comparable at the lower end, or maybe Athlon wins. But in the 1G processors, PIII is the winner (see this, for instance).
Now with Thunderbird, Athlon will again be more or less the same in L2 cache as PIII: 256K full speed on-die cache. So, we should expect the Thunderbird to kick PIII's butt, by about 10% or so plus the fact that it will actually be available.
Duron should also beat celeron, though perhaps by a smaller margin.
AMD's Beofulf startup script: 5.... 4.... 3.... 2.... 1.... *sound of rocket taking off* Thunderbirds are go!
AMD's next chips (the Stingray series) will cause office buildings to retract into the ground, whenever you run a difficult program.
Reports that using the AMD Thunderbird causes you to walk in a strange, stiff-legged manner are denied, as are rumours that Brain has bought out Red Hat.
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)
Nothing I've seen indicates that newer celerons and newer P3s perform the same for a given clock speed. This used to be the case. For example, I have a celeron 300a overclocked to 450. The performance is similar to P2 450 and a P3 450 (the older flavor of P3 that had 512k off-die cache) for most applications.
The newer P3s have 256k on-die cache that makes the chip much faster than the old P3s. The newer celerons are the same, but with half of the L2 cache disabled. The benchmarks I've seen have shown the new celeron lagging considerably behind a new P3 with the same clock. I think the days of celerons offering the same performance as a P3 are over.
Two notes: I'm basing this on my memory of reviews. If someone knows better please correct me (I know, I know, it's tough to find people willing to at Slashdot). Also I wonder if the celerons aren't slower than the P3s because the L2 cache's bandwidth is halved. I'm no expert, but memory systems can use interleaving similar to RAID striping on hard drives. Maybe when Intel chopped the L2 cache in half (really just a crippled coppermine P3) they halved bandwidth too.
"You can bypass the click-though licence agreement on Microsoft's PAC specification by opening the file with WinZip instead of running the file". There. Have you violated the DMCA today?
"The good die first." "Most of us are morally ambiguous, which explains our random dying patterns." --- MST3K
Some people are asking about SMP, and since I recently attended an AMD partners conference, I thought that I would share.
.25 micron process, you can't put the L2 on die, it'll produce too much heat. However, you can't access the L2 cache very quickly through a socket, there's too much resistance.) The Duron will use the new Socket A, and the Thunderbird will be available in both Socket A and Slot A configurations, so that all of the old motherboards aren't immediately obsoleted. Later processors will only be available in Socket A.
First, the Duron and Thunderbird processors are going to integrate the L2 into the die. The AMD techs seemed excited about that since the Athlon is already outperforming the Coppermine. The move from the L2 on the module to L2 on the die should prove a massive boon to their already good performance. This also means that there is no longer a reason to use a slot. (With
Second, SMP is coming, but not in the Duron or Thunderbird. Mustang will bring us SMP later this year, in Socket A configuration. The point that AMD stressed is that Intel processors use a shared bus to the North Bridge, whereas their Athlons will each have their own bus to the North Bridge. This will provide the processors a whole lot more bandwidth to their memory and peripherals.
Theoretically, a dual Athlon MB could be made using the Tsunami chipset (Compaq's blindingly fast dual-processor chipset for the Alpha). Unfortunately, the chipset would then cost more than the two Athlons (about $1000), so I don't see this happening soon.
Yes, it will be competitive with the Celeron. Even though it does have less Level 2 cache, the effective 200mhz bus will make all the difference in the world -- Not to mention the Athlon core which is still far superior to the PPro core of the Celeron. The Celeron 2's had relatively dissapointing performance, and it's likely that Duron will perform much better and cost less.
As for the memory, not much will change probably. Duron's will be using basically the same chipset for Socket A (KZ133 and AMD 750). There are several places to buy "Athlon-tested" memory though, so I don't really see why this is a problem.
Thunderbird? Duron? Athlon?
They sound like secret weapons from a bad sci-fi movie...
Hosting for Creators: http://rpg-works.net
From theregister.co.uk:
AMD practises safe computing
Relax, you'll just feel a tiny prick
Chimpzilla's new - ahem - entry level chip, Duron, would appear to share advanced technology with condom giant Durex, if the contraceptive company's website here is to be believed.
The site states: "New technology has considerably improved the condom and enabled the production of far more sophisticated versions than our ancestors were used to. The latest development is DUREX Avanti made from a unique polyurethane material, DURON, which is twice as strong as latex enabling a thinner, more sensitive film."
AMD would not confirm rumours that Duron would ship in packets of three rather than trays of ten, nor that the company was planning a 'Duron Inside' ad campaign. ®
Looks like you'll need to lap that Duron processor as it may be ribbed for her pleasure.
The chip may grow flaccid if not in use.
Feel free to make your own juvenile jokes.
--Shoeboy
(former microserf)
I currently buy SuperMicro P6DBEs (dual cpu 440BX boards) 10-20 at a time @~$160 a piece - I'd jump at the chance to be able to buy something similar that's Athlon based - but at a similar, or lower, all up (stuffed including CPUs) price - sadly I doubt this is going to happen any time soon
Dell's PCs are so goddamned expensive I don't see how their "special relationship" benefits anyone.
Dell has a huge advertising co-op with Intel, so the special relationship mostly benefits Dell, by getting them millions of dollars of free advertising. Also, they got price protection on RDRAM, which allows them to peddle some of Intel's otherwise worthless 820 motherboards. Also, many of Dell's computers are Intel barebones systems.
I'm sure if anyone implied that their special relationship benefits the customer. It definitely benefits Dell, though, which is why they will have a strong resistance to upsetting Intel, even if it hurts their profits.
If Dell goes AMD, that will be a warning sign to Intel that they no longer dominate the market.
--- "So THAT's what an invisible barrier looks like!" - Time Bandits
I can tell you from personal experiance that Dodge Neons really suck.
Molog
So Linus, what are we doing tonight?
So Linus, what are we going to do tonight?
The same thing we do every night Tux. Try to take over the world!
We already know that the AMD Duron and the "Thunderbird" variant of the Athlon CPU will be great, if only because they are based on the modern Athlon CPU core with its 128 KB L1 cache and the all-new FPU unit.
The big considerations in regards to these new CPU's are chipset support and motherboard availability. In regards to chipset support, will AMD finally make the 760 chipset avaiable? And when will VIA Technologies ship the Apollo KZ133 chipset, the one that will support the Socket A design natively? And how about availability of motherboards that use Socket A? Once the 760 and Apollo KZ133 chipset motherboards become widely available, then AMD will avoid the issue of plenty of CPUs but no motherboard situation that plagued initial Athlon shipments last summer.
Raymond in Mountain View, CA
For slot-A/socket-A mobo information, you can check http://www.slota.com (socketa.com points to the same page).
The KZ133 has been sampling for a long time, and will be available in volume along with socket-A motherboards to coincide with the Duron/Thunderbord launch in early June. There are at least 3 socket-A motherboards that should be available at launch (the FIC mobo is already on pricewatch.com - search for "Duron" or "Thunderbird").
There will also initially be some slot-A Thunderbirds to ease the transition, and these will work with existing AMD 750 based slot-A motherboards, although don't expect to be able to buy a slot-A Thunderbird in retail - they are likely to only show up in OEM computers.
The AMD 760 chipset is for DDR and will be available in Q3. There is also the 760-MP (aka 770) which will support 2-way SMP as well as DDR.
There are also many other AMD chipsets coming, such as Micron's DDR/SMP ones, Via's KZ266 for DDR, ALI (Acer Labs) support, etc.