Slashdot Mirror

Some good links.

Looks like the Duron runs pretty hot too, http://www.aceshardware.com/articles/reviews/duron /duron_cpusink.jpg

Ace had some detail on overclocking the Duron, but I want to know, is it going to significantly reduce my heating bills.
I could get one for my granny to keep her warm during the winter. I could write it off on the tax too.

But seriously, is it not a case of too little too late??
$110-$130 is a good price but we've had the celeron for a long time, is there room for another low end chip in the arena, or should they be aiming higher with the performance. Perhaps as a cheap upgrade option for the celeron??

Search and destroy, lunch is served

IIRC x87 is the name for the floating point ISA originally presented in Intel's math coprocessor. And yes, the x87 ISA will be blown out of the water if you use the SSE2 floating point ISA in the upcoming Willamette (Ars as well as Ace's Hardware had pretty good articles about that a while back, too).

Has anyone looked at the code? I looked at some of his C code. Its in the directory with the makefile life.c, fib.c and fft.c

Take a look at life.c and draw your own conclusions.. For example

init0 = (int*)calloc(max_x,sizeof(int));
init1 = (int*)calloc(max_x,sizeof(int));
init2 = (int*)calloc(max_x,sizeof(int));
for (x=0; x&lt max_x; x++) {
init2[x] = 0;
init1[x] = 0;
init0[x] = 0;
}

I could at least double the speed of his code in about 15 minutes. I didn't look at his java code but I suspect there aren't nearly as many dumb things in it.

I might have just blown the whole thing off if it weren't for the fact that he claims MSVC with full optimization is slower than GCC in 'braindead' mode for a couple of tests. I've seen code GCC generates with -O, it makes me laugh.

Has anyone looked at the code? I looked at some of his C code. Its in the directory with the makefile life.c, fib.c and fft.c

Take a look at life.c and draw your own conclusions.. For example

init0 = (int*)calloc(max_x,sizeof(int));
init1 = (int*)calloc(max_x,sizeof(int));
init2 = (int*)calloc(max_x,sizeof(int));
for (x=0; x&lt max_x; x++) {
init2[x] = 0;
init1[x] = 0;
init0[x] = 0;
}

I could at least double the speed of his code in about 15 minutes. I didn't look at his java code but I suspect there aren't nearly as many dumb things in it.

I might have just blown the whole thing off if it weren't for the fact that he claims MSVC with full optimization is slower than GCC in 'braindead' mode for a couple of tests. I've seen code GCC generates with -O, it makes me laugh.

Has anyone looked at the code? I looked at some of his C code. Its in the directory with the makefile life.c, fib.c and fft.c

Take a look at life.c and draw your own conclusions.. For example

init0 = (int*)calloc(max_x,sizeof(int));
init1 = (int*)calloc(max_x,sizeof(int));
init2 = (int*)calloc(max_x,sizeof(int));
for (x=0; x&lt max_x; x++) {
init2[x] = 0;
init1[x] = 0;
init0[x] = 0;
}

I could at least double the speed of his code in about 15 minutes. I didn't look at his java code but I suspect there aren't nearly as many dumb things in it.

I might have just blown the whole thing off if it weren't for the fact that he claims MSVC with full optimization is slower than GCC in 'braindead' mode for a couple of tests. I've seen code GCC generates with -O, it makes me laugh.

There is a Linpack benchmark available http://www.aceshardware. com/Spades/read.php?article_id=156. Ace's hardware ran the test against a K7 athlon, a PIII, a celeron, and an UltraSparc II... Thunderbird kicked ass, of course, it's not exactly fair to match a 1 Gz thunderbird against a 733 PIII.


Zetetic
Seeking; proceeding by inquiry.

Elench
A specious but fallacious argument; a sophism.

According to Ace's hardware, AMD has announced the 760 chipset. A variant soon to follow, the 760MP, will have SMP support. All the AMD reps will say is that these chipsets are due in the second half of this year (thanks for narrowing it down). Tyan supposedly has a motherboard planned to be released before the new year that will have dual Athlon support. Head over to Ace's Hardware and search for "760MP".

My only issue is that this will allow two processors. We just got a server here at work with quad Pentium III Xeon 550MHz. Wow. I want that for my Athlon system.

Ace's has some benchmarks (under linux, no less) of the different compilers. They prove your point quite well.
-

The entire Itanium platform is basically nothing more than a test platform for the next IA-64 chip, McKinley.

This is because Merced/Itanium is 2.5 years late, and poorly designed. It is struggling to hit a meagre 800MHz (it was designed to debut at 800MHz, but on .25u process and in 1998), and appears to have only 96k L2 cache and 64k L1 cache, because rather than reaping the predicted simplifications of moving to a VLIW design, the Itanium core has fallen into a stew of complexity which leaves very little room for large cache size, even at .18u. This project has chewed up and spit out all of Intel's seasoned engineers on their Santa Clara design team, and was basically finished up by a huge committee of inexperienced kids fresh out of college. Every knowledgeble independent hardware analyst I've seen has said that, while IA-64 has some potentially promising ideas behind it, Itanium is going to be an ugly plodding beast of a chip, and its only chance at success is through miraculous marketing. Too bad for Intel's marketing machine that they're trying to move into a new space (high-end servers) with Itanium, and thus the people buying these machines are going to be well-informed and used to buying from vendors other than Intel.

The good news for IA-64 is that McKinley, the successor to Itanium, is much more on track (it should be out in volume by Q2 or Q3 2001), and appears to be performing quite nicely. Incidentally (or not), McKinley has been designed almost entirely by a much smaller (and more experienced) team over at HP (where EPIC, the philosophy behind IA-64, was designed as well); about all Intel will have to do with it is the fabbing and the Intel Inside logo. Also, McKinley will have the advantage of being fabbed with a brand-new .13u process, which ought to leave it enough room for a real L1 and L2 cache, and allow for presentable clock speeds. Thus, most organizations are looking at Itanium as a sort of public beta for McKinley--they'll buy a couple to get used to the platform, start to develop software for it, and have a core understanding in place in case they decide to upgrade all of their high-end servers to McKinley in a year or so. Almost no one is planning on deploying Itanium as a long term solution, because it will simply not be cost effective for the mediocre performance it will have.

And thus the "race" to get a good IA-64 OS up and running really isn't all that important. MS will be sure to have W2K-64 "gold" well before McKinley is released, and that's all that really counts. Getting Linux on Itanium solid before W2K-64 goes gold will be nice, but it most certainly isn't going to win Linux a beachhead in the high-end corporate market, simply because most high-end corporate stuff can afford to wait for W2K-64 to be ready before they switch to IA-64.

On the other hand, a rather peculiar consequence of the IA-64 design is that applications need to be recompiled for each new processor core in order to take advantage of the EPIC features. (This is because, in IA-64, the compiler does most of the instruction scheduling, and not a scheduler in the CPU's logic; thus, the compiler better know how many functional units the CPU has, for example, or else it will schedule things all wrong.) This might seem to give an advantage to OSS, since one can just get a new compiler and recompile all of one's applications if one has the source to them. However, considering the market all of this is directed towards, I'm quite sure no one will have any trouble getting seperate versions of W2K or Oracle or whathaveyou optimized for Itanium or McKinley or any subsequent IA-64 chips.

For anyone who was expecting IA-64 to show up on the desktop anytime soon: don't. The first IA-64 chip scheduled to have a shot at the high-end consumer market is Deerfield, planned for 2004, but expected to come out later if at all. It would take a little bit to explain why, but EPIC is very much better suited to single-tasking machines running streamlined, well-tuned code--things like databases. It's just a poor fit for consumer applications. And anyways, if you want a neat Intel chip to get excited about, Willamette--the next desktop chip--looks way sexier than Itanium, and has a lot of cool design features (eg. double-pumped ALU, trace cache) which are just as innovative as IA-64's VLIW stuff.

Just to close things off, here's a couple links for those looking to learn more about Itanium and IA-64. (Even if the design doesn't pan out, some of the principles behind it are very interesting.)

Hannibal at Ars Technica on IA-64: a bit old, but a very well written overview of the design theory behind EPIC.

Hannibal on IA-64 vs. Sun's MAJC: mostly about MAJC (very interesting in its own right!), but a good example of a different approach to VLIW than IA-64's. (Crusoe is yet another example of a completely different approach to VLIW.)

Paul DeMone's Itanium article on realworldtech.com (unfortunately, the site seems to be down at the moment, so I can't get a link): an impressive technical argument about why Itanium might not actually achieve higher ILP in actual conditions, along with an interesting historical parallel of why this wouldn't be the first time Intel pushed a radical new chip architecture and it completely flopped (I forget the name of the previous chip, but it's an interesting story).

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!

It's Ace's Hardware, and here's the article.

PPC is not the only one left sort of in the cold out there. Look at Alpha -- a great chip, but what about proper support with Linux and GNU tools? See this story and this review for details.

This whole situation can be summarized thus: GNU/Linux are dominantly Intel-based.

Yeah, well, but they didn't use the right compiler for Athlon either. Hint: egcs-1.1.2 is not particularry well optimized for Athlon either. Pgcc is a lot better. See the the thread here for some numbers.

Ace's hardware has an article on the newly announced Matrox G450. Unfortunately, it looks like a more evolutionary step from the G400, and not a true contender against the big guns from 3dfx and nVidia.

Dresden has been churning out 600 wafer starts per week of copper (i.e FAST!) Thunderbird all Q1..

That might seem a lot, until you consider that full capacity for Dresden is 5000 wafers per week. :-)

(Got this from Ace's hardware)
-

The thing that really confused me was the references in the article to this software being Itanium optimized. Fair enough then, Intel's motives could be seen as carrot dangling to persuade consumers to migrate more enthusiastically to a nascent technology platform. Then I was left wondering exactly how source code would be Itanium optimized. Surely it could be optimally tweaked and recompiled for any platform, even non-intel architectures

Possibly it is just a buzz-word. The Itanium is going to have to do good things for Intel otherwise they are going to be up the proverbial creek as far as 64 bit processors goes, and this is not a playing field they have to themselves (with 64 bit POWER, Alpha and AMD Sledgehammer processors also featuring). I suspect however that the source code makes copious use of things that the Itanium is supposed to do well - lots of use of 64 bit and longer integers in math processing, and accesses of memory in 64 bit lengths. Of course, that means that the current Athlon will also do well on the same code (see Aces hardware for an article on K7 memory access). So yes, you can write C code which favours a particular processor, as long as you understand it's strengths and weaknesses and have some appreciation of what the compiler does to your code. But I strongly suspect that knowing the most optimal set of compiler flags for a particular processor is also important in getting a given set of code to run as fast as possible.

Cheers,

Toby Haynes

Some more corrections/thoughts that seem to have gotten missed in this discussion so far:

1) It's not necessarily an AMD chip

Many of you seem to be under the impression that the X-Box will be using an Athlon variant (presumably a Spitfire), but the name of the CPU vendor was conspiciously left out of today's announcement. Indeed, according to this article at C|Net, MS has decided to go with Intel for the CPU instead of AMD as earlier rumored.

If I had to guess, I'd say this means a 600 MHz Coppermine modified to support Willamette's new SSE2 instructions, which look quite impressive. (Although the most impressive things I've read about them (see this article at Ace's) are in regards to their double-precision SIMD performance, and IIRC games almost always use single-precision floats.)

This makes sense because two of Willamette's other signature features--a 20-stage deep pipeline and a double-pumped ALU--don't make sense here; games don't need much in the way of integer performance, and the deep pipeline is only good for increasing clock speed (indeed, clockspeed being equal, it slows things down)--and is definitely not necessary to reach 600 MHz.

On the other hand, Willamette's "400 MHz" (really quad-pumped 100 MHz) bus might not be such a bad idea for a next-gen console. Indeed, it might be just the thing to keep the NV15 based graphics chip full of data. The problem, of course, is cost, cost, cost. Which leads me to my next point:

2) 600MHz isn't such a bad decision

Yeah, I know that by the time this thing comes out, new PC's will be sporting 2 GHz Willamettes and 1.8 GHz Athlons. However, there's one problem with all y'all going around saying that that means that the X-Box should have a much faster chip too; those 2 GHz chips are going to be selling for something like $800-$1000 a piece.

And then there's the problem of how chips are normally clocked versus how they need to be clocked for a fixed-spec market like a console. You see, when Intel (or AMD, or whoever) makes a chip, they don't stick a clock multiplier on it until it's done. They make the chip, then test it to see how fast it can reliably run (this depends on lots of factors, among them the quality of the particular piece of silicon; there's no way to definitively know this number without actually testing it), and then stick on a multiplier such that it runs at that speed (actually a speed bin or two lower, just to be safe). This means that some (very very very small) percentage of P3's ends up being smacked with a 10x multiplier and being sold as a 1GHz chip; some get an 8x multiplier and are sold at 800MHz; and some--but just a few--can't manage to run reliably at even 600 MHz (or whatever the lowest speed P3's are sold at these days is), and are tossed in the trash).

Now the thing is, all of this probability stuff is built into the price. You see, it costs Intel exactly the same--around $70, IIRC--to make that one chip that ends up being branded at 1 GHz as it does to make the one that gets sold at 600 MHz. The difference is, it takes a whole lot of chips before they make one that's good enough to run at 1 Ghz. And a bunch of them are lost to the trash bin along the way. That's why they charge different amounts for the faster chip--to make up for the fact that they're harder (but *not* more expensive) to make. And that's (partially) why even the cheapest P3's still cost about $200--far more than the cost to fab each particular one.

In the console market, though, that little trick just doesn't work. When you're fabbing CPU's for the X-Box, either it runs at 600MHz, or you throw it away. Furthermore, since the entire thing is only going to cost $300, the CPU better not cost more than, say, $35 or $40; after all, that $300 has to include 64 MB of (possibly Rambus??) RAM, the graphics chip you're buying from NVidia, which itself will have probably 32 MB and possible 64 MB of RAM (possibly DDR RAM); an 8 GB hard drive, a DVD drive, a motherboard, a stylish case, a controller, possibly a keyboard, probably pretty impressive sound support, and I'm sure a bunch of other stuff I'm forgetting. Point being, you want to make sure you can make these chips run at 600 MHz with *very high yields* in comparison to the yields that Intel and AMD normally achieve.

Furthermore, with a kickass graphics chip (and especially one that has hardware T&L like the GeForce does and the NV15 will) the speed of the CPU is much less important. Indeed, as Kyle over at HardOCP showed (check here and here), with today's fastest chips, in real-world conditions it is sometimes faster to run with a GeForce's Hardware T&L turned *off* (i.e. so the CPU calculates T&L) than with it on! On the other hand, that same GeForce, when paired with a mediocre CPU, speeds things up tremendously. Of course, the T&L in the NV15 will be considerably improved, such that it will no doubt be a great help when paired with that 600 MHz chip. But I wouldn't be surprised if it's a waste when paired with those 2 GHz Willamettes everyone wants in the X-Box instead.

3) The X-Box will perform identically to a 600 MHz / 64 MB RAM PC of today--i.e. worse than a PS2

Absolutely definitely maybe not.

First the absolutely not: the real guts of the X-Box is not its 600 MHz CPU, but rather its NVidia based graphics chip. Even today, a pretty slow Celeron with a kickass graphics card--i.e. a DDR GeForce--will be pretty competitive with the latest Ghz P3 with a very respectible graphics card, say a Matrox G400, when it comes to running games. Indeed, in many situations (i.e. at high resolutions), it will run just as fast as that Ghz P3 with the same kickass GeForce--and much faster than the P3 with the Matrox--because at high resolutions (i.e. 1280 and 1600), the limiting factor is always the fill-rate of the video card. Course, this doesn't help if you're running at TV resolution, but you get my point: for games, the video card is *more* important than the CPU--and the GPU in the X-Box will be much better than any graphics card on the market today.

Next, the definitely: the X-Box, like all consoles, will only come in one spec. That means game developers can program their games knowing exactly what they'll be running on--and taking full advantage of that as much as possible. This means, amongst other things, that they won't have to design their games to look adequate across a wide range of resolutions and graphical detail levels, but can instead concentrate on making it look good and run fast at the one graphical level it will be run on. Secondly, this means that, like on any other console, developers will be able to dip below the API level and reap the speed benefits that come from being able to program a much lower levels, including hand-tuning important graphical code at the register-level in the GPU. This can only be done when you know that the specs of the machines that will run your game are all identical.

Now for the maybe: one of the major "points" of the X-Box is that it will be nearly compatible with normal PCs, which of course come in all shapes and flavors. The difficulty here is that, in order to maintain this compatibility, developers would need to stay at the API level, and would need to design their games from a hardware-agnostic point of view, which would remove most of the benefits of uniformity I just mentioned. However, I'd guess that what will most likely happen is that developers will keep most of their code at the D3D level, but still optimize the most important routines for the X-Box's GPU. The end result will be that X-Box games *will not* run on PC's (although PC games might run on X-Box??), but that it will still be considerably easier to port PC games to X-Box than to any other console. On the other hand, it's reportedly very easy to port PC games to the PS2, so maybe this advantage isn't as great as MS banked on. In any case, it's important to note that it's this same loss of the benefits of uniformity which has lead to almost no Dreamcast games making use of the Dreamcast's ability to run WinCE and hence pseudo-D3D. Indeed, I believe that MS has officially withdrawn their WinCE support of Dreamcast due to a complete and total lack of interest from Dreamcast developers.

4) It's Windows, and it's a PC, so it will be confusing, take forever to boot, and crash like crazy

This is almost certainly wrong. For one thing, the X-Box will be running a version of what up to now has been called Embedded NT--which should be extremely stipped down and quite reliable, as well as offering very short boot times. (Reportedly the PS2's boot time is quite long for a console--on the order of 5 seconds or so.) Furthermore, probably most Windows crashes come as a result of either bad drivers--which should never happen on a standardized machine like the X-Box--or as a result of problems with memory management of legacy code--again, no problem since there will be none--or with multitasking apps not behaving themselves--which won't be a problem since the X-Box will only run one thing at a time. Furthermore, 64 MB of RAM should be more than adequate, considering the lack of multitasking and the fact that the OS will be much much leaner than normal Windows or NT.

On the other hand, I have to say that the prospect of an 8-gig hard drive scares me a bit, if nothing else than because it offers the possibility of quite a lot more complexity and variations in end-users' actual setups. I doubt MS will allow anything like DLL hell to manifest itself, though; I'm sure the X-Box OS will keep every program's DLLs seperate and well managed, especially since this is a (more like the) feature of MS's upcoming-and-stupidly-named Windows ME.

Phew. So--do I think the X-Box will be phenomenally successful? No, not really, I don't. While I do believe that it will be more powerful that the PS2 on a theoretical level, I don't know if the difference will shine through in the games. Basically, there are two possibilities: most X-Box developers will try to keep their games as trivial ports from their PC counterparts, in which case they won't be able to take advantage of the uniformity of having a single machine to develop for, and thus the PS2 will be more impressive, or X-Box developers will try to "program to the metal", in which case they will be a year behind on the learning curve of low level programming, and thus their games will probably never decisively beat what's coming out for PS2 at the same time.

On the other hand, I think that it just might be successful (depends on if the PS2 actually conquers the world beforehand, as many predict), and I'd give it about equal odds to succeed as, say, Nintendo's Dolphin.

qouting Ace's article, pg.2: Since they also wanted to get 1000+ CPUs in a single system, they also had to design the on-chip memory system and interfaces to be able to handle this. When you have a multi-processor system, you need to keep the data coherent in all caches (having the same data at the same time, or else you get bad results) one way or another, and this becomes increasingly difficult as more CPUs are added.

I guess there are alot of issues which are specific to Sun's hardware, so mabye my thought of learning from Sun's SMP code was a moot point. Moot is a fun word to say. Moot. Almost like saying Meept.

You know, sometimes people amaze me.

Exactly where are you getting these figures from? It's not Intel's or AMD's websites...
>To wit, SPECcpu95:
>Coppermine 800 MHz: SPECint - 38.9, SPECfp - 32.4

Taking a look at Intel's own posted benchmarks of the 800mhz Coppermine running on a 133mhz bus gives:

SPECINT - 38.4 SPECfp - 28.9
Source:
Intel's own website and benchmarks

Now, I'd compare these to AMD's benchmarks, but AMD hasn't published SPECINT results, and only publishes the base SPECfp results. (Which, by the way, show the Athlon soundly thrashing the Coppermine

In any event, comparing SPEC scores is a rather _bad_ way to judge system performance. If you know enough to extrapolate new benchmark scores from current ones, you should also know that there are much better real world tests available.

If you like, you can watch a 700mhz Athlon kick the snot out of a 733mhz PIII Coppermine running on a 133mhz bus over at Ace's Hardware

So next time, please take your results somwhere else or provide a real source for them.

• news.com
• spaceviews
• EETimes
• FiringSquad
• Ace's Hardware
• The Register

And if that post is worth Score:2 what is this post worth? Anyway, if this WAS a big story like the TPC stuff someone mentioned above that would be a newsworthy item (or a 2.4preXX release for that matter), but this isn't! This is another X-Box hype like trash! Why don't you post stuff about the Glaze3D Hemos if you really want vapourware and not-yet-maybe-never-out trash?

Come on now, guys.

Do you really think that if Intel could run Willamette at 1.5 GHz now with no special cooling that they would intro it in October at "only" 1.3?

This snippet at Aces Hardware kinda puts it into perspective:

In your opinion, could Willamette ALU run at 3GHz air-cooled on Intel's 0.18 transistors?

I am here at IDF. I saw the Albert's demo and Glen's presentation and demos. I was very impressed with the 1.5GHz. 3GHz ALU is even more impressive, although, it is only a tiny percent of the whole die.

Albert's presentation was designed for maximum effect. When a quick 1.5GHz frequency utility demo was run, Albert did not say anything about how the chip was cooled. Then he quickly switched to a different system and ran some 3D rendering. During this second demo Albert and his partner engaged in a lively conversation showing everybody that it was running air-cooled. Of course now it was not displaying the frequency.

The impression was created that the 1.5GHz system is running 3D rendering air-cooled. But for all we know, 1.5GHz system could have been floating in liquid helium, while the one running 3d rendering was running at 800MHz.

That said, Willamette could still be a very nice CPU, but Im not going to get it unless Intel has a change of heart. The point being that Intel says Willamette will only support RAMBUS memory, and Im not going to shell out 5 times as much $ fom my RAM with no advantage whatsoever (with RAMBUS yields only getting worse there is little hope in its price turning sane any time soon).

That is why Im going for AMD and DDR. In fact, my K7-500@700 should suffice for some time.

/Dervak

Man, me and my addled brain today...forgot to mention they've also got reviews of the new chips (800 MHz, Intel and AMD) up there too...

Ace's Hardware

Over at Ace's Hardware, they've got a news item about AMD's forthcoming announcement of an 800 MHz Athlon, supposedly today as well. And (AFAIK) unlike Intel, AMD's supposed to be able to be shipping them in volume very soon. Intel's still got volume problems, especially with the 800 MHz chip.

Check out these comparisons between the Coppermine and Athlon:

Ace's Hardware
Tom's Hardware

There is a good comparison on Ace Hardware between the Coppermine and the Athlon here.

I think it's funny that the AMD Athlon 700 processor beats out Intel's flagship 733Mhz beast in many tests. Check out Ace's Hardware's benchmarks.

GeForce / TNT2Ultra / Voodoo3 Roundup [ Shugashack]
Guillemot GeForce256 3D Prophet Review [Ace's Hardware]
Guillemot GeFroce256 3D Prophet Review [Puissance PC]
nVidia GeForce 256: To Buy or Not to Buy [AnandTech]
Guillemot GeForce256 3D Prophet Review [GA-Source]
nVidia GeForce256 DDR Review [3DGPU]
nVidia GeForce256 DDR Review [Riva Extreme]
nVidia GeForce256 DDR Preview [Thresh's FiringSquad]
nVidia GeForce256 DDR Review [Riva3D]
nVidia GeForce256 DDR Review [Planet Riva]
nVidia GeForce256 DDR Benchmarks [Bjorn3D]
Guillemot GeForce256 3D Prophet Review [CGO]
Guillemot GeForce256 3D Prophet Review [Fast Graphics]
Creative GeForce256 Annihilator Benchmarks [3DHardware]

Theoretically, you could do that but you would need a mighty powerful CPU to achieve the level of performance of the GeForce since CPUs aren't optimized for graphics processing (note: GPU stands for Graphics Processing Unit not Geometry Processing Unit as someone earlier posted.) The GeForce is a much more cost effective solution for graphics processing than getting another CPU.

According to Nvidia web page about the GPU, their technical definition of a GPU is:

"a single-chip processor with integrated transform, lighting, triangle setup/clipping and rendering engines that is capable of processing a minimum of 10 million polygons per second."

The review of the GeForce 256 at Ace's Hardware has good info on comparing CPUs to GPUs. As another poster mentioned, graphics processing exists in a limited form in CPUs (3DNow!,etc.). Possibily in the future CPUs will integrate more advanced graphic processing functions. But, even if you had a CPU with complex graphic processing functions you still need some sort of display adapter. Personally, I think that it makes more sense to have the display adapter and graphics processing integrated on one unit.

Theoretically, you could do that but you would need a mighty powerful CPU to achieve the level of performance of the GeForce since CPUs aren't optimized for graphics processing (note: GPU stands for Graphics Processing Unit not Geometry Processing Unit as someone earlier posted.) The GeForce is a much more cost effective solution for graphics processing than getting another CPU.

According to Nvidia web page about the GPU, their technical definition of a GPU is:

"a single-chip processor with integrated transform, lighting, triangle setup/clipping and rendering engines that is capable of processing a minimum of 10 million polygons per second."

The review of the GeForce 256 at Ace's Hardware has good info on comparing CPUs to GPUs. As another poster mentioned, graphics processing exists in a limited form in CPUs (3DNow!,etc.). Possibily in the future CPUs will integrate more advanced graphic processing functions. But, even if you had a CPU with complex graphic processing functions you still need some sort of display adapter. Personally, I think that it makes more sense to have the display adapter and graphics processing integrated on one unit.

Anandtech GeForce 256 Review
Ace's Hardware GEForce 256 Review
RivaExtreme GeForce 256 DDR Review
The FiringSquad GeForce 256 DDR Review
GA Source Guillemot 3D Prophet Review
3DGPU Geforce 256 DDR Review
Fast Graphics Guillemot 3D Prophet Review
CGO GeForce 256 Preview
Shugashack GeForce, V3 and TNT2 benchmark roundup
Riva3D Full GeForce 256 DDR Review
GeForce 256 DDR Review at Planet Riva

Did an NDA expire today or something?

Just a couple quick links:

Anandtech GeForce 256 Review
Ace's Hardware GEForce 256 Review
RivaExtreme GeForce 256 DDR Review
The FiringSquad GeForce 256 DDR Review
GA Source Guillemot 3D Prophet Review
3DGPU Geforce 256 DDR Review
Fast Graphics Guillemot 3D Prophet Review
CGO GeForce 256 Preview
Shugashack GeForce, V3 and TNT2 benchmark roundup
Riva3D Full GeForce 256 DDR Review
GeForce 256 DDR Review at Planet Riva

Any others?

• Sharky Extreme
• Ars Technica
• AnandTech
• Hard OCP
• Ace's Hardware
• CompHardware
• Tom's Hardware
• The Tech Zone
• Thresh's FiringSquad
• Review News

--

... are at Ace's Hardware.

(An anonymous coward posted the link, I am reposting for those filtering less then moderation level 1.)

Find them at www.aceshardware.com.

Ace's Hardware has something about the K7 optimization guide. Check http://www.aceshardware.com/ for more.

Check out Tom's Hardware and Ace's Hardware for TNT2 reviews. As far as I remember, the Hercules cards were the best performers.

> I'm a hardware research engineer from Acer Inc. (Singapore)

Do we have any proof of this? No.

> K7 test machines CPU K7 600MHz (FSB 200MHz x 3.0) 600MHz 128KB L1 Cache
> 2-way associative 200MHz 512KB L2 Cache 4-way associative EV6 BUS
> Controller L2 Cache Speed 1/3 FSB

Everyone on this message board has mentioned this part spec dozens of times, this could easily be expected.

>RAM TI Direct RDRAM 256MB at 600MHz

I believe Intel RDRAM spec is twin concurrent 16bit busses at 800mhz, not 600mhz. This would support the claims of 3.2GB/sec and multiples thereof with interleaving. Also, Sony has speced their PSX2 as having an 800mhz RDRAM bus. Nintendo RDRAM runs at 433mhz. I don't know of anyone making 600mhz rdram because nobody apparently intends to use it. I believe this is fake.

VGA
> Matrox Millennium G400MAX 32MB SGRAM 360MHz RAMDAC Resolution 1600x1200 32
> Bpp 85Hz

The card's been announced, everyone who read the press release saw it and full specs.

> Harddisk IBM Ultra3 SCSI 6ms 10.0 GB

Nobody makes a SCSI HD in 10.0GB. They're all multiples of 2.255GB. 4.55, 9.1, 18.2, 36.4 are all currently shipping capacities. I don't believe IBM has shown a U3W prototype yet, only Quantum has that I'm aware of. Also, there is no controller listed in this spec, and I haven't yet seen any manufacturer talking about and upcoming U3W controller for anytime this year. Not believable.

> Mainboard AMD Sample Chipset
> Northbridge AMD Irongate AGP4X SouthBridge AMD Cobra ATA66 USB 2.0

Chip names but no numbers. I believe the names were on AMD presentations or discussed at comp shows or shown on roadmaps. We obviously expect AGP 4x. Intel has not even finalized the USB 2.0 spec yet, so support for this in an already existing chipset is nonsense.

> OS-Windows 2000 Beta 3 WinHEC April 99 Release

Nobody in their right mind would run benchmark comparisons on a beta OS.

> P-III test machines CPU Pentium III Coppermine 600MHz (FSB 133MHzx4.5)
> 600MHz 32KB L1 Cache 4-way associative 600MHz 256KB L2 Cache on-die 4-way
> associative

We all expect this part eventually. Nothing new in the specs for it.

> Mainboard Intel Sample Chipset Northbridge
> FW82820 AGP4X Southbridge FW82801AA ATA66 USB 2.0

Chip numbers for this can probably be easily extrapolated from current intel partnumbers. We've been hearing for weeks about delays in the "intel 820" chipset. We know it will support agp4x and ata66 (not that anyone cares about ata66). I'm very confident it will NOT support USB 2.0 as intel has announced that will be coming up for early 2000.

> WinBench99 Version 1.1 ( 5 times measure ) CPUMark99 (CPU Integer
> Performance Not include MMX register) - K7@600MHz 68.4 Mark - Pentium
> III@600MHz 70.1 Mark

This is quite likely made up. The scores on the PIII which could be considered a "known" are way off from projections from standard clocks and results from overclocked systems. Sorry, this is most likely numbers drawn out of a hat and adjusted to make the K7 look bad.

> FPUWinMark( CPU Floating Point Performance Not Include
> MMX,SSE,3DNow!,FSTORE Extendtion register ) - K7@600MHz 2,819 Mark -
> Pentium III@600MHz 3,104 Mark

Again, probably random numbers.

Now I have to ask, how many people on here would not know where to find the press releases, product roadmaps, and rumored info to construct this kind of sham? The accuracy of the description is the only thing that makes it believable to any of us. The more you actually know about this upcoming hardware, the easier it is to fake, because you can get the audience's confidence with known facts and then slip the lies in at the end.

--

Check out Ace's Hardware , they have some good articles on the K7. I like that site a little better than other hardware sites, it seems more technically oriented in content. I remember reading an article on there about how K7's in an SMP setup use independent busses to the chipset, unlike the Intel SMP setups share busses (which is why SMP with more than four Intel CPUs doesn't work very well).