AMD's Duron Slated For June

Devil Ducky writes: "AMD announced that they will release the Duron sometime in mid June, instead of last April. The Duron is intended to compete on price with the fabled Intel Celeron. Duron will include 128KB of primary and 64KB of integrated cache, meanwhile Athlons contain 128KB primary and no integrated cache. When released it will be available in 600, 650, and 700MHz with plans for 750MHz soon. The story even makes some quick comments on the names Celeron and Duron."

Comments

Why AMD moved to socket A

[SpinyNorman]

Slot A packaging is very expensive and is only necessary because the Athlon (like the Intel Katmai PIII) has off-chip L2 cache and therefore needs a CPU module mini PCB (slot A package) to integrate the processor and L2 cache RAM chips. By moving the cache on-chip, not only does performance increase but you no longer need the PCB - the CPU can be a simple ceramic socket part, which is what the socket-A Thunderbird and Duron are (same as Intel's move to the Coppermine PIII socket packaging).

Sci-fi acid trip marketing names

[Anonymous Coward]

-I am Duron, ruler of Athlon, the fifth Pentium of Celeron, holder of the sacred Katmai, oracle of the Klamath. Step forward and speak, mortal!

-Where d'ya come up with dem names, dude?

-Silence! Your insolence must be punished! I sentence you to the depths of Xeon! You shall be devoured by the PowerPC... what? "PowerPC"? Who has changed the sacred scripts?!?

The Klingon Processor

[kawlyn]

Duron son of Athlon

Re:The Klingon Processor

[luckykaa]

This code has no honor. It deserves to be executed.

AMD are really on a roll...

[toofast]

I used to use Pentium chips in all my machines, but since the K6 I'm pure AMD.

True, the K6 was no top-performer. But it was compatible and was damn cheap. Today, a few of my servers run Athlon... Linux, NT, even NetWare. Damn fast. And they're still cheaper than PIII's. The only thing missing is SMP boards.

I think the Duron will be a strong contender in the home-user market, and will be a good replacement for Socket 7-based compuets.

Is AMD making a serious run at Intel?

[miguelitof]

All of a sudden, AMD doesn't seem like the also-ran in the processor wars any longer. The Athlon has garnered favorable reviews, new marketshare, and even more loyal customers for AMD. Now the Duron will re-secure AMD's hold on low-end PCs.

No matter what happens, this should be great for consumers. At the very least, Intel will once again have to drop prices on the Celeron.

Re:Is AMD making a serious run at Intel?

[SpinyNorman]

Clock for clock Athlon holds it's own against PIII, win some/lose some benchmarks.

Thunderbird adds 256K on-chip exclusive L2 cache, and will see similar speed gains as when Intel moved from Katmai to Coppermine PIII (i.e. when they did the same thing). Thunderbird will clock for clock beat PIII.

But here's the real kicker: Athlon is available in volume now at 1 GHz, while PIII is at 866MHz (except for a few 1GHz samples at Dell). Intel claim 1GHz volume in Q3/4, when Thunderbird will be at 1.3-1.4GHz.

Duron will also beat Celeron II due to having more total cache (L1 + L2) as well as a full speed bus, unlike Celeron's deliberately restricted 66MHz one.

PIII Coppermine core is at the end of its life, and according to Intel may reach 1.1GHz max. Intel's future lies in the PIII Coppermine replacement core - Willamette which may sample in Q3, but is not expected in any quantity until late Q4 or 2001. In the meantime AMD have Mustang coming out... AMD also have the advantage of a modern copper interconnect fab process in their Dresden fab (licenced from Motorola), while Intel are still using slower aluminum interconnect and won't catch up for around a year.

Conclusion: For at least the remainder of this year, AMD's high end processors will be faster than anything available from Intel. In 2001 things may be a little more even as Intel catch up to AMD in updating their fabs, and getting an up-to-date processor core out, but some analysis are suggesting that AMD Mustang (in .13 or .15u process) will still outclass Willamette - time will tell.

industrial cement floors

[Odds]

Wow, I'm glad to see someone working on increasing the connectivity of the web. That industrial cement floors link was so valuable.

Re:industrial cement floors

[Rupert]

I thought it was rather interesting, actually.

My window commands a great view of "Target Plaza South", currently under construction in downtown Minneapolis. I have to tell you that the concrete floors being poured there recently have not anywhere near approached FF/FL "Superflat".

See, news for nerds does not have to be about computers. Many of us, particularly those with houses of our own, find concrete and the pouring of it, rather interesting.

Re:Typo??

[barleyguy]

Nope. That's correct. However, the Athlons do have 512K of external cache, which they don't bother to mention.

Re:Typo??

[JabXVI]

No, what he meant was that the Athlon doesn't have integrated L2 cache (which is quite a bottleneck), it has 512KB of external L2. I think AMD will also be releasing a new Athlon (code-name Thunderbird) about the same time as Duron (I've heard rumors that the Duron could actually be faster than the old Athlon, so it'd make sense to release Thunderbird and Duron at the same time). Thunderbird has 256KB of integrated L2.

Re:Got a silly question

[stripes]

What I don't understand is why make the secondary cache smaller than the primary cache?

The L2 duron cache is exclusave, that is does not contain cache lines from the L1 cache. Many other processors don't work this way, in part because it is simpler to make a multi-processer system if only the L2 caches need to do bus-snooping. My guess is the Duron won't be too happy in a multi-processor system for this reason.

It'll still make a decent single CPU system, especally at the dirt cheap prices listed in the last article.

Re:Slot or Socket? (VIA)

[TuRRIcaNEd]

I don't think it was the chipset manufacturers that got slammed by Intel as much as the motherboard makers. The whole reason that Intel went for a Slot rather than a Socket design was not primarily out of ease of fitting (as claimed by the marketroids), as much as it was that other CPU makers would have a much harder time, as they could then patent Slot 1 and price the license effectively out of range. AMD responded with Slot A which did not only for them what Slot 1 did for Intel, but also solved a few of the bus limitation problems along the way (Getting those disaffected former DEC guys was obviously a canny move ;-).

To cut a long story short, Intel figured this out, and put the smack down on anyone who supported the Athlon. Not just VIA, (whom Intel despise for matching their chipsets in many respects, and then selling them cheaper), but motherboard makers too. ASUS didn't even put the K7M on the front page of their website for months, and that was widely regarded as the best all-round K7 board! Personally I think Intel need to be taught a lesson, which is why I'm a confirmed AMD advocate now. Hopefully with Duron, and maybe Joshua (Heh! VIA get their own back!) assaulting the basic Celeron, the Athlon taking on the Celeron II and PIII, and the Thunderbird and Itanium slugging it out, those of us who are more tech-savvy can return the smack to Intel.

Anyway, to avoid going too OT here, it will be interesting to see exactly how AMD will make up the shortfall in mobos when the Duron is released. If the vacuum that faced the Athlon upon release can be avoided, we can hope for an interesting fight, which I hope makes Intel question a couple of fundamental issues (Such as, don't fire your older workers, who've probably been in the game for a while, because they will go elsewhere). On another note, It would be cool to see AMD and VIA team up and use the same socket standards on their lower-range chips, or even better, to implement a standard everyone will use, and stop these tit-for-tat architecture changes, that renders building/upgrading machines a logical and financial grind! (Unlikely, but I suppose I can dream.....)

Re:When to upgrade? (slightly OT)

[Psiren]

I'm also looking to upgrade fairly soon. I've read a lot about the Abit KA7 m/board which supports PC133. Unfortunately it apparently has problems with some DIMM's (due to the KX133 chipset so maybe other boards suffer the same). Specifically 128M DIMM's don't work until a 64M DIMM has been in. Has anyone had any experience with this? In all other aspects the board is really good, but I don't want to buy one and find it doesn't like the memory.

Now weary traveller, rest your head. For just like me, you're utterly dead.

outstanding troll thread

[Blue Lang]

i was just lamenting last night how poor the troll threads have gotten lately.. this one is great. keep up the good work.

--
blue

Re:Slot or Socket?

[Shostykovich]

As previously stated, goto anandtech and look at AMd's roadmap, and you will see that this is a "socket a" chip, and you'll probably need a motherboard with the KZ133 chipset for it.
No, in fact, this isn't justifiable, and it is all Intel's fault. Perhaps you've not seen this before, but remmeber how Cyrix was sued 7 times or whatever, and VIA is in a lot of hot water right now with intel? Its because they basically used technology intel decided it owns in their products. I am not debating whether this is "right" or not, but it is the reason AMD will not make a chip that will go on mobo's designed for Intel chips (eg, slot 1): Intel would sue them.
It is most certainly NOT AMD's fault that they can't just go make chips using the same interface as Intel's chips. See, Intel is able to controll most motherboard and chipset manafacturers, because those companies get most of their money making stuff tailored around Intel's line of products. Because Intel is a competitive company, it will obviously try to limit AMD's hold in the market. Thus they force AMD to design their own interface, and they also make chipset manafacturers (right now, VIA) suffer if they design chipsets around AMD's proucts. It costs money to design these chipsets, and even more money to get the market to accept them as viable and adopt them. Thus, AMD suffers because they have to waste resources creating and marketing their technologies, while Intel is accepted as the defacto standard.

When to upgrade?

[Black Parrot]

I wonder if someone could pull together several variables for me.

I am thinking about building a new system this summer. I never buy the fastest CPU, but rather pick one that is several notches down from the top, right at the sharpest bend in the price/performance curve. (That would put me at about 750-800MHz for an Athlon right now, but of course that will creep up over the course of the summer.)

The problem is, I will need a new motherboard as well. And that's where the unbound variables start expanding. So here are some questions. Alternatively, ignore the questions and just summarize what you would do if you were going to build a high end (but not top) x86 system in the near future.

• Is PC-133 "there" yet? Is it a cost-effective improvement over PC-100?
• Is PC-133++ anywhere near? (I.e., PC-150, PX-whatever.)
• Is DDR memory "there" yet? Cost-effective?
• How do you expect the various AMD *ons to shake out in price/performace for a given clock speed? (Assume general purpose developer's desktop plus scientific number crunching in the background.)
• Should I wait a few months? Six months? A year? Or is the price/performance improvement going to be more or less continuous during that time?
• Any other variables I should be asking about?

Thanks.
--

Re:When to upgrade?

[ToLu the Happy Furby]

If I were you, I'd have a mid-range Thunderbird at the top of my list. It will debut at 700-1000 Mhz, and probably hit 1.1 GHz soon thereafter, so we're probably talking somewhere in the 800-900 MHz range. The process will have a lot of headroom in it, so you'll probably be best off getting a 700 MHz T-bird or Duron and overclocking it, assuming you're comfortable with that.

The big question with this system is whether DDR is worth it. The actual cost of DDR should be barely more than the cost of PC133--it's just as easy to make, but it might cost a bit more because of limited supply. Instead the "cost" of going DDR is measured in time; DDR mobos don't look to be available until late this summer, a month or three after T-bird is released. If you can wait, don't buy until you see the first benchmarks of T-bird on DDR vs. T-bird with PC133. Don't make your decision on benchmarks of the PIII with DDR, which will probably be available sooner: the PIII's chip-to-Northbridge bandwidth is only 1.07 GHz/sec, so it is usually saturated by PC133. DDR ought to show a bigger performance increase on the T-bird's EV6 bus, which will run at 266 to match the DDR. On the other hand, don't plan on going DDR before you check out the benchmarks; no one really knows exactly how well it will perform.

As far as chipsets, you want to be looking at a VIA KZ133 if you're sticking with PC133 (NOT the KX133, which is apparently incompatible with the T-bird), and probably a KZ266 if you're going DDR, although there may be other DDR chipsets available for T-bird as well, most notably Micron's Samurai chipset.

The big unknown in the future x86 market is Intel's new Willamette core. It probably won't be out in volume until January or so, and thus might not be an option for you. On the other hand, they'll be looking to do a paper release in late summer or early fall to compete for press time with AMD, so we should have a good idea of how well it'll perform months before we can actually buy one. (And maybe if you're lucky and willing to pay a lot, you might even be able to buy one in 2000!)

The reason I mention it is that while the general consensus seems to be that Willy won't be such a huge deal, Paul De Mone, one of the most respected semiconductor analysts on the net, has had some very positive things to say about it. If you're interested after reading the two mondo articles there (and if you're any kind of hardware geek, I can't see how you won't be), you might want to check out what he's had to say about Willamette in Ace's Hardware's technical forum. Frankly, Paul really knows what he's talking about, and he seems to think Willamette will solidly give Intel back the performance crown. Whether Intel will use that as an excuse to price Willy out of the upper-mainstream market where you're looking to buy is another question entirely.

Any other variables I should be asking about?

Well, depending on what you're looking to do with this computer (and with what OS), the video card is probably the most important component these days--certainly more important than chipset and arguably more important than CPU. If you want good 3D performance, then at the moment that means running Windows. Period. This is changing relatively quickly, though, so it may be less of an issue when you get your computer. (It will NOT be a non-issue.) 3dfx has always had some of the better Linux drivers, so if you're going Linux one of the new V5 cards is probably your best bet. nVidia is well known for having terrible, and closed, Linux drivers, although they claim that that's changing. If you're going to be running Windows, a DDR nVidia GeForce 1 is probably going to be the best bet to match your computer (i.e. just-below-really-high-end).

If you're going to be running Windows games a lot, then this will be the most important part of your purchase, hands down. The two obvious choices will be the V5 and the GeForce 2; the GF2 is a bit faster with full-screen anti-aliasing off, while the V5 is faster with it on. The GF2 (and GF1) has T&L to speed up future high-poly games; the V5 has Glide which provides the fastest play in games like Ultima IX and anything using the Unreal engine (Unreal Tournament, plus many upcoming games like Duke 4). To really decide, however, you should go to some indepth benchmarks from a gaming site and look at the resolutions/settings you'll be playing at in the games you'll be playing and see which card performs better. Also ATI has a card aimed at the high-end 3D market due out this summer.

If you don't need good 3D, take a look at an ATI card for great DVD playback (depending on your ethical opinion thereof), TV-tuner, and general features, or take a look at a Matrox G400 for top-notch quality and the best dual-monitor support around.

Other than that, you should be fine performance-wise. (I'm sure I don't need to tell you that the monitor is the most important piece of equipment for overall computer satisfaction, and that a nice keyboard and mouse are close behind.) Of course, for "general purpose developer's desktop plus scientific number crunching in the background"--i.e. compiling stuff and running distributed.net--IMO any computer sold today is more than adequate, although of course extra compiling speed is always nice, as is a higher ranking. In this case, I'd say the most important factor is how much cache is typically consumed by compiling. (Anyone care to inform me?) If it's a low-cache amount low-memory access activity, go with a Celeron or Duron; if it's low-cache amount high-memory access, go with a Duron; if it's medium-cache amount, go with a T-bird or Coppermine; if it uses up a whole lot of cache you may be better off with an Athlon "Classic" or a Katmai PIII for the money.

Hope this helped!

Re:When to upgrade? - Typo alert!

[Black Parrot]

> Is PC-133++ anywhere near? (I.e., PC-150, PX-whatever.)

That should have been "PC-whatever".
--

Re:Got a silly question

[Ed Avis]

What I don't understand is why make the secondary cache smaller than the primary cache?

If a memory location is used frequently enough to be stored in the secondary cache, it will surely also be in the bigger primary cache, which is faster and looked at first. Thus the secondary cache is effectively useless, except maybe as a write-back buffer. Maybe this is a mistake in the article, or I've not understood?

(BTW wouldn't it be cool to write software that fits entirely in the primary cache - 128Kbyte used to be considered a lot! If there were some way to initialize the cache at boot time, you wouldn't need RAM at all :-)

Sargon

[ch-chuck]

Sargon - it was also the name of a cool Z80 chess program by Dan and Kathy Spracklen

huh?

[ejbst25]

Intel makes Celeron to compete with AMD K6. AMD makes Duron to compete with Celeron. Intel makes Emeron to compete with Duron. AMD makes Furon to compete with Emeron.

for (int i=1; i>0; i++){
compete(INTEL, AMD_LATEST_CHIP);
compete(AMD, INTEL_LATEST_CHIP);
compete(Microsoft, COW_DUNG); //Can they ever win this one?
}

God I love endless loops.

Good news for the consumer

[cadfael]

Frankly, I don't care if AMD or Intel makes the faster/better/sexier chip, just as long as there is competition. If some other Lage Unnamed Corporation had more competition, it might turn out better software. As it is, the two chip makers are forcing each other to work harder to be better and cheaper, which can do nothing but good things for you and me when we go buy our next machine.
-- The Hollow Man

Re:Good news for the consumer

[Devil Ducky]

You didn't say that it was a monopoly, only that it needed MORE competition. So Oracle fits the description of a Large Unnamed Coroporation with little/no competition that could make better software.

Don't everyone get upset, I've never had any real experience with Oracle software, but everyone who makes software can make it better with time.

Devil Ducky

Re:Where d'ya come up with dem names, dude?

[Manaz]

Turns out Klamath and Willamette are rivers in Oregon... since Intel is largely in Oregon as well, that makes some sense. Besides, those are just codenames anyway (along the same line, IIRC, Katmai is a mountain).

Intel name ALL their processors (at least the code names) on rivers in the Oregon area... Coppermine and Katmai are rivers as well AFAIK....

OT, You don't know any typical Americans

[Lord Kano]

You make some good points, the American public education system is a joke. Our students are the least capable in math and science out of all of the students from the industrialized world.

If one relies solely on what one is taught in our schools, that person is lost. However, our universities produce some of the most competant professionals on the planet.

Our public education systems churns out semi-literate, semi-educated consumeroids. However not all of us are beer bellied football junkies.

Where does the majority of the world's software development occur? Why? Because we've got the economic resources to draw the best and brightest from around the world, foreign and domestic.

Where did the power of the atom firse get harnessed? Why? Because our free society didn't allow any antisemitic dictator to drive the best jewish minds away and into the hands of the enemy.

Back to my point though, the typical american today isn't the undereducated buffon that for some reason seems to dominate all of our entertainment productions. Just as the typical German isn't an antisemitic bigot. Just as the typical isn't .

In fact, I'd prefer that the US kept it's nose out of the affairs of the rest of the world and that the rest of the world kept their noses out of ours.

LK

Re:One little question...

[paitre]

Nope, not Socket7.
Socket 7, fwiw, is pretty much officially dead by the end of this year/beginning of 2001. AMD will cut production on all non-mobile K6 line processors no later than Q1'01, if not earlier.

SOrry, you want a Duron, you need a SocketA board.
Ditto Thunderbird (unless you get -really- lucky)

Re:Where d'ya come up with dem names, dude?

[Spud Zeppelin]

Turns out Klamath and Willamette are rivers in Oregon... since Intel is largely in Oregon as well, that makes some sense. Besides, those are just codenames anyway (along the same line, IIRC, Katmai is a mountain).

And Pentium was their 5th generation chip, so it follows a standard branding formula -- take something descriptive, and add a futuristic, newspeak ending:

pent (5) + ium = pentium.

And so it came to pass that someone at Intel probably went back to the well, and came up with these lines of thought:

"Hey we have two new chips. Compared to the Pentium, we want to convey the image that one is a God and the other a vegetable...."

Zeus - us + X - Z + on = Xeon

Celery - y + on = Celeron

And then they had their 6th generation chip to contend with. Rather than name it Sexium ("It renders pr0n sites REALLY fast") they needed to jazz it up. So when they went to talk about it one of the marketroids probably asked his kid:

"It an -ium"

Not to be outdone, of course, our friends in San Jose or Dallas or wherever their headquarters is this week faced a similar problem wanting names for the chips after their K6. So they wanted something to indicate how fast the new chip runs:

Athlete -ete + on = Athlon

Then, they discovered, however, that their trusted business model of low-cost chips had started to leak away. "Quick, get something to keep this fluid from escaping!"

Durex -ex + on = Duron

And we are where we are today. At least hypothetically.... So what's next? Maybe our friends in Phoenix will unveil a stripped-down G4, and call it "McPPC" -- playing on the whole "tasty and inexpensive" identification. Or someone will come out with a low-power mobile chip called the "Volxon" (Volks -ks + x + on).


My opinion only, IANAL.

Re:One little question...

[Vanders]

The Duron will use Socket A. AMD have also anounced Socket A version of the Athlon too, iirc. But either way, you'll need a motherboard.

By the way, Socket 7? What where you thinking man?!