Even on the SpecInt benchmark, the earliest PPro benchmarks I could find on Spec's website show that while the PPro put in some respectible numbers, it was far from being the king of the SpecInt benchmark.
This is wrong. The original PPro launched just before the EV56 revision of the (at that time) DEC Alpha (i.e. the Alpha 21164). The PPro's SPECint was better than the previous Alpha's, and worse than the new Alpha which launched a few months down the road. Thus, the PPro was, at the time, the absolute SPECintCPU95 king. Of course it was later eclipsed in performance by better and more expensive chips; the point is, at the time it was an extraordinarily innovative design and a miraculous surprise to the entire MPU community.
For starters, SPEC is not a single benchmark, rather a consortium that comes up with benchmarks, the most well recognized being their CPU benchmarks (colloquially refered to as SPEC benchmarks). These benchmarks however, do not exclusively test a CPU, but rather a system as a whole, although they are designed to make the CPU the limiting factor (nonetheless using bucket loads of RAM, fast disk controllers, and a huge external memory cache can have wonderful impacts on SPEC benchmarks). Typically these benchmarks have been divided into those that stress the integer unit (SpecInt) and those that stress the floating point unit (SpecFP). The Pentium Pro was the first x86 CPU to post respectable SpecFP benchmarks, but it still got it's butt kicked all over the place compared to it's RISC competition.
Yes, the SPEC_CPU benchmarks test an entire platform--CPU, i/o, memory, and (very important) compiler--not just a CPU. Of course, this does nothing but put x86 chips at a major *disadvantage*, as they are only available in configurations which do not even approach the i/o and memory bandwidth available to the big RISC chips. (Of course, Intel has perhaps the best compiler group around, which mitigates this somewhat.) Still, the SPEC_CPU benchmarks are called SPEC_ CPU for a reason, as they make every attempt to, as you say, make the CPU (and its attendant cache hierarchy) the limiting factor as much as possible in a suite of varied, meaningful, non-synthetic benchmarks. As for the other SPEC benchmarks (i.e. not SPEC_CPU), they are of course designed to benchmark the, well, non-CPU parts of the computer, and are thus irrelevant here. Again, when one refers to "SPEC scores" it is assumed that they mean SPEC_CPU; thus the original poster's usage was quite correct.
As you point out, the PPro never challenged the Alpha's SPECfp lead, due to a major deficiency of the x86 architecture--backwards-compatability with the ass-backwards x87 fp instruction set. Luckily, this is finally being phased out with the P4's SSE2 instructions; thus if Compaq fails to execute with the Alpha in the coming months, and if Intel's SSE2 compilers are good enough, the P4 may just win both SPEC_CPU benchmarks outright for a time--perhaps as much as a year or more. That ought to shut up the ignorant anti-x86 FUD on/.
When the P6 was released, it was the fastest processor available in industry standard benchmarks (SPEC, including Alpha). Its design was highly original, and manages to keep the CISC nastiness contained to the first few stages of the pipe. Claiming that the P6 was not a world-class design when released is only a testament to your own ignorance.
Exactly correct. If I had moderator points, they'd be yours.
And indeed, the 1 GHz P3--on that same, 5 year old P6 core--is still tied with the moderately-vaunted brand-new mucho-expensive (not available until Q1) 900 MHz UltraSparcIII in SPECint2000. The 1.2 GHz Athlon would presumably perform even better (once they release SPEC scores from the new Compaq Fortran compilers), making it second only to the fastest (and also none-too-available) Alphas in terms of pure performance. The x86 ISA may be suboptimal, but Intel and now AMD have been able to keep up with the best--and most expensive--of the RISC world due to superior engineering (except when compared to the excellent Alpha team) and superior process technology. Sure they may not have the i/o bandwidth, RAS, or operating systems to compete in the big leagues, but anyone dissing today's x86 chips on account of their designs or engineering qualities is, as the poster said, demonstrating their ignorance.
And if Compaq doesn't hurry the EV68 (die-shrunk Alpha) to market, the P4 and perhaps Mustang as well will blow by even the mighty Alpha, in SPECint and possibly even SPECfp. (The last real knock against the x86 ISA is that it is saddled with the horrendous x87 fp architecture, which is why x86 SPECfp scores trail everyone else by so much. With the P4's upcoming SSE2 instructions, however, that problem may be in the past.) Aesthetics aside, there is no doubt that x86 processors, taken as a whole, are easily the best designed, highest performing MPU's around.
As much ire is directed at Rambus, I can't see the logic behind this. Ashton-Tate is relevant because they channeled resources into legal wrangling, rather than R&D and customer service. They failed, by putting all their eggs into one basket, a very wrong basket.
Check this out: in the last quarter, Rambus made $10 mil. on total revenues of $27 mil. Ok, so far so good (for a hideously overvalued speculative stock). The kicker is this (unfortunately, they took this little tidbit out of a previous version of this article on news.com): during the quarter, Rambus spent $1 million *per month* on legal fees. That's right--their legal fees were about a third of their profits and over 11% of their entire revenues!
Guess that's why they dropped their suit against Hitachi the day it was actually given a go-ahead by the courts. Of course, now that they're being countersued by not one but two industry giants (Micron and Infineon), with very deep pockets and very annoyed looks on their faces, those legal fees are just going to keep going higher and higher. And every single company in the industry--Intel included (if not Intel most of all!)--is going to enjoy watching them slowly bleed to death on legal fees.
That line first appeared with the 386. And it was wrong then, too.
The line with the 386 (and 486, and P5, and PPro) was that it was destined to remain a workstation chip for some time. This was Intel marketing bluster, yes, but it was moderately true--the initial versions of each of those chips was produced on an old process; once the chip moved to a new process, it became feasible for upper-end mainstream machines. In a two years, they were mainstream.
Right. But the difference here is, these chips are not intended for workstations. They're not intended for moderately sized servers. They're intended to replace mainframes, and to run high-high-end scientific code. In case you didn't read my other post, these things are going to cost *at least* $10,000 *apiece* to MAKE. Just for the MPU. Moreover, they will not work up to their full potential without *massive* bandwidth, which still costs mucho $, last time I checked.
What servers really need is multiple CPUs and huge I/O bandwidth, not faster individual CPUs.
Oh wait, you didn't even read the article. The POWER4 *is* multiple MPU's--it's 8 cores on 1 die. This thing doesn't come in anything less than 8-way configurations. As for I/O bandwidth, would 84 GB/s be enough for you?
You still think this thing is a desktop chip???
On the other hand, Apple can't afford to change CPUs again.
Or maybe they can't afford not to. Contrary to all those MacOS Rumors, there are no definite plans for Apple to move even to the upcoming G4+ MPU's, which are essentially another incarnation of the tired G4, just with a stretched out pipeline which will get it to 800MHz at the cost of lower IPC. There aren't definite plans yet because the G4+, like the G4 and the G3 before it, is not a desktop MPU but rather an embedded/DSP chip which Motorola happens to sell to Apple to use in Macs. The design, ramping, pricing, roadmap, are all influenced primarily by Moto's embedded customers first, not by Apple. Unfortunately, despite the fact that they are utterly dependent on them, Apple has decided to treat first IBM and then Motorola rather poorly, and thus haven't gotten much in the way of support when they decided that they may, perhaps, want to increase the speed of their top MPU more than 50 MHz in a year. Whoops. BTW, did you catch Apple's earnings today? Ouch.
All of this is too bad--OS X looks like perhaps the best thing going as far as operating systems goes. There are always rumors that Apple's going to finally make their surprise move to x86. Their experience with PPC the last year or so, and the accompanying beating their bottom line has taken, might be the thing to finally push them over. I personally think they might still be able to carry over enough incompatabilities to stay the sole supplier of MacOS hardware--after all, the XBox uses x86, and it will be plenty incompatible with PCs. Migrating software will be a gigantic pain...but on the other hand, it's not like the Mac has too much in the way of software anyways. (OS X, and any Cocoa programs, will port very very quickly.) Who knows?
I think the big advantage that VLIW instruction sets will have is strictly architectural, and I'm not sure how IBM's approach fits in yet, but it looks interesting. Throwing more chips at the problem is one approach, but remember that your competitors can do that too, *and* make the chips do more as well...
Not sure how IBM's approach fits in yet?? Read the article.
Amongst other things, the POWER4 is *not* VLIW, it's straight-ahead modern RISC at its finest. With massively gigantic buffers, bandwidth and execution resources (8 functional units/core * 8 cores = wow), this chip'll do quite nicely on IPC/core, not to mention combined IPC for all 8. While presumably not quite as elegant, the design for the individual cores bears a lot in common with the archetype of perfect RISC cores, the Alpha 21264, and it has even more aggressive resources.
Essentially what this means is, assuming this design is as good as it appears, the only way the competition will be able to catch up (without going the way IBM has and deciding on a prohibitively expensive 8-in-one design and packaging) will be through the use of innovative design tricks. The upcoming P4 has a few of those, incidentally, but the big one--and the one the P4 *doesn't* have--appears to be SMT, Simultaneous MultiThreading. Alpha has an 8-way SMT core coming out in a bit, and it ought to compete well with IBM's much more expensive 8-way SMP design here. And AMD appears ready to do 2-way SMT (or something similar) with the Sledgehammer in about 15-18 months. And Sun is rumored to have SMT in the USV design due in several years. But the POWER4 looks to lead in the "big bad" category for quite some time to come.
(As for Intel's EPIC, the VLIW-like design strategy for their IA-64 chips, at the moment it's looking like a rather poor competitor to SMT. A quick explanation of why:
There are exactly two ways to make an MPU run faster: 1) increase the clock speed, or 2) increase the IPC (instructions per clock). Unfortunately, the best we've been able to do so far in the IPC realm is about 1.4 IPC on SPEC benchmarks (Alpha EV6x). IPC on a P3 runs about 40% lower. Now, these IPC numbers are despite the fact that the Alpha can theoretically retire 8 instructions/clock, and the P3 5 (5 internal ops, not 5 x86 ops). Furthermore, simulations show that as far as attacking the IPC problem by adding more functional units, we're nearing the point of diminishing returns.
The problem is, in order to run lots of instructions in parallel, you have to be able to safely extract parallelism from your code. And the problem with this is, you can't run instructions in parallel if they have dependencies, etc. And furthermore, nowadays all this parallelism has to be safely extracted in real-time by special hardware in the MPU itself; this makes your chip more complicated, and means you need to build a big buffer to hold instructions in flight so you can pick and choose which ones you want to run each clock.
So many many years ago, HP had the idea, which it later sold to Intel (and which wasn't really there idea at all but has indeed been used in DSP chips for years and years), of getting rid of all that complex insruction-level parallel-finding logic on the MPU and doing it all at compile-time instead. This is the basic idea behind EPIC, the philosophy of Intel's IA-64 line.
It sounds very nice, especially because in theory it means simpler chips (no complicated control logic), and simpler chips means faster chips. Heh heh heh. See it turns out that the amount of instruction-level parallelism which can be safely discovered at compile-time is way way less than the amount that could be found in the chip at run-time (which, as we recall, is too small already). Thus EPIC was modified to allow the compiler to just place "hints" in the code. Well, this means you still need all that complicated control logic back in place, because you still don't have deterministically scheduled instructions. But following the "hints" and other changes to the ISA ends up making everything *more* complicated, not less. This, in a nutshell, is why Itanium is 3 years late, way over budget, unable to meet its very modest clock speed goal of 800 MHz, and fitted with a laugh-enducing 96kb of on-die cache, lower even than the lowliest Celeron: all this added complexity means bigger, slower, more complicated chips that don't have the room for cache or the elegance for high (or even adequate) clock speeds. Plus we have very strong evidence that compiler technology is still not nearly good enough to make the kinds of insightful IPC-giving "hints" which are necessary to even make the damn fool scheme work. Thus the only benchmark Intel has "released" for the Itanium is that of an RSA-encryption--a routine simple enough to be hand-tuned in assembly. Meanwhile they have made the patently ridiculous claim that the SPEC benchmarks--directed precisely at the mid-cost server/workstation market which Itanium is aiming for--are "not relevant" to Itanium's market.
A completely opposite approach is SMT, which uses a relatively small number of core changes to allow not just instruction-level parallelism to be gleaned, but also thread-level parallelism. In other words, the chip will run several threads in parallel, confident in the fact that their instructions will not have dependencies on each other, and thus be able to use much more of its full execution capabilities. Early indications are that SMT can improve IPC by remarkable amounts, like on the order of 2x the performance on otherwise similar cores!
Unfortunately, it is too early to tell whether SMT will be as easy a design enhancement as is being claimed. Furthermore I've heard tell that SMT on IA-64 will be a lot more difficult than on a RISC MPU, so Intel could be missing out on a huge speed-up with this technique.)
However, IBM will have to make sure people design their apps with more than one processor in mind, which will be a Good Idea for the future, since more people might have multiprocessor computers.
These chips are not to be confused with PowerPC chips. They are server chips only, intended for seriously expensive machines.
Heh! If only Apple would use these, the new iMacs wouldn't exactly be quite able to hit their price points. Paul (the author of the article) and some others were involved in a thread over on the tech forum at Ace's about (amongst other things) the expected cost of one of these puppies.
Maybe another way of looking at it is perhaps the price of four POWER4 known good die and the ceramic substrate and metal carrier totals $3000 (although I suspect that a tested and 100% functional ceramic substrate itself might approach or exceed $3000 in cost).
The real question is the cost of a fully assembled and tested, 100% functional, POWER4 8-way module? After all what are the chances one of these can be reworked if even just one of the 20,000+ solder ball joints was bad?
So for one of these 8-way on a chip jobs (unsure if they'll be offering 4-way configurations too or if those were just a prototype) it's looking like upwards of $10,000 just for IBM to fab, package, and test the darn things. Add in a system capable of feeding it the tremendous bandwidth it requires to run up to its full potential--8 GB/s to DRAM and a phenomenal 84 (!) GB/s I/O--and...ok, so I know Hemos was just joking when he made that comment about Apple, but you get the idea. These are MPUs you use to fold proteins and run gigantic dynamic-content websites, not surf the web and edit the home video of your kid's elementary school graduation.
On a related note, man these things oughtta show Intel a thing or two about how to marry clever instruction scheduling to brute-force functional units--forget about Itanium; it's gonna take a several-way McKinley system to even take a swing at this these. And it oughtta show Sun a thing or two about the dangers of resting on the laurels of your marketing success when designing new chips. And, as Paul notes in the article, it really oughtta make Alpha engineers worry that for the first time, having the most elegant design may not guarantee the best performance. Compaq has an 8-way SMT Alpha core on the way as well (EV8); too bad the Alpha group's customary position in the world--stepped on and neglected by their corporate masters--means they haven't got the money or manpower to bring it to market until well after POWER4.
Virtually any IC that you'll find anywhere can withstand voltage fluctuations, especially ones that small. If a processor rated for 1.5V couldn't run at 1.1V, you can bet they would write 1.8V or so for the recommended Vcc level. Even your TTL ICs can withstand that type of change.
This is nothing that should be seen as impressive, it's just PR that will fool the masses and make the engineers wonder why it would exist any other way.
Individual IC's can, yes. The point is that when you have many many IC's strung together to make a chip, the tolerances for the whole chip become much tighter. Perhaps you've heard of fan-out/fan-in? This is extremely impressive, dude. Don't believe me, try switching the voltage for your P3 from 1.6V to 1.2V and see how it does.
Seriously, all of the major recent Intel chipset problems were with RDRAM. There's the infamous i820E problem with the third RDRAM chip not getting registered (which was after the initial, pre-RDRAM i820 went bust). And now this, when the ICH2 is coupled with the i850 and i860 MTHs, which use... starts with R, you know this...... RDRAM! Right!
I think the time for the NVidia DDR chipset is NOW. Let's stop this half-assed hardware engineering and pre-alpha lithography which the Intel staff is undertaking.
It's not that simple. Both RDRAM and DDR require tighter tolerances on the chipset and motherboard levels for the same obvious reason--the more bandwidth you want to transfer, the less your tolerance for noise and defects, and the greater the danger of crosstalk. Period. This is simply a fact of life if you want the benefits of high-speed DRAM without taking the (in my opinion inevitable) step of ditching the current system of expandable commodity RAM on the motherboard in favor of a system which ties hardwired or embedded DRAM to the MPU.
Now, you can argue that RDRAM makes the problem worse by trying to cram the same amount of bandwidth as PC1600 DDR into a thinner bus. And you can argue that dealing with a new memory communications protocol has led to more bugs. I'm not going to argue with you there. On the other hand, RDRAM proponents would counter that RDRAM lessens these problems by switching to a packet-based protocol to cut down on interference.
And you can argue--as you did--that the fact that mighty Intel has run into myriad problems trying to implement RDRAM for their PC chipsets, and that their results to date--the i820 and i840--have been lackluster at best. But you can also point out that Intel has (for marketechture reasons) made this switch on a processor designed for use with SDRAM only. You can point out that they've (also foolishly) decided to have the switch coincide with a switch to hub-based memory management--which, incidentally, appears to be the source of the erratum in the i850, not the use of RDRAM. You might also want to note that Intel's first efforts with SDRAM chipsets, while not as starcrossed as the i820/i840, were nowhere near as efficient, stable or refined as their 4th-generation BX chipset or the 5th-gen i815.
Finally, it's worth noticing that there are exactly zero currently available DDR chipsets, bug-free or otherwise, with which to compare Intel's RDRAM record. Of course the main reason for this is politics--only after a year of Intel floundering with the RDRAM protocol did the industry finally coalesce behind DDR. And of course this fact will be changing quite soon, within the month in all likelihood. The fact that several working and apparently stable DDR chipsets are on the verge of being released, from chipset designers less accomplished than Intel, means that DDR cannot be as difficult to implement as RDRAM-backers have long argued it would be. Still--and this is very important--no one has claimed it was easy. Indeed, DDR chipsets were by all accounts much more difficult to get working than SDRAM chipsets, and even now many are reportedly quite finicky when working with DDR made by different manufacturers. Furthermore, all DDR motherboards due for release in the near future are, like all of Intel's RDRAM boards, six-layer. This improves stability and reduces crosstalk at the expense of extra engineering effort and manufacturing cost; SDRAM chipsets tend to be quite stable with just 4 layers.
I dunno where that leaves the ease-of-implementation balance. It appears that it's on the side of DDR, but it's still a bit premature to say so conclusively. In any case, the idea that whipping up a DDR chipset to replace RDRAM is child's play is absolutely false.
On the other hand, Intel already does have a DDR chipset for the P4--or rather, for Foster, the "P4 Xeon" due out in the beginning of the year. Now, this chipset could be modified for use with the normal P4 with little problem, but there are two big reasons it won't be:
1) Cost: In order to fill Foster's massive FSB, Intel's new chipset uses, IIRC, dual-channel double-wide DDR. Contrary to what you may have heard, this means a much much higher cost than the dual-channel RDRAM bus on the i850. The reason is that RDRAM's one unambiguous advantage over DDR is in using fewer pins; having two channels and doubling the bus width means multiplying the already quite large number of DDR pins by 4, which in turns means motherboards which are mucho expensive, even if two sticks of DDR is cheaper than two sticks of RDRAM.
2) Legal obligations. Intel is under contract with Rambus not to promote any other next-gen DRAM standard for its mainstream desktop line until 2003. That means that, unless they want their asses sued off (and we all know if there's one thing Rambus is good at, it's ass-sue-offing), the best Intel can do for the next couple years is license the P4 bus and allow 3rd-party chipset makers like VIA, ALi and, as you mentioned, perhaps even nvidia make DDR chipsets for the P4. Actually the contract specifically states only that Rambus has the option to revoke Intel's RDRAM license if they promote a DDR chipset for their desktop chips, so the big question is, would Intel risk losing their Rambus license, especially when there is no cheap DDR solution which can take full advantage of the P4's 3.2GB/s FSB? On the other hand, would Rambus risk revoking Intel's RDRAM license and thus taking themselves out of the PC DRAM industry possibly for good??
I dunno. Frankly I'm just hoping Infineon succeeds in overturning some of Rambus' RDRAM patents with prior art as they're seeking to do. (I'd be shocked if they didn't succeed in showing prior art for Rambus' "patents" on SDRAM and DDR.)
You also play down the FP performance, where the Alpha as is still kills the projected values for the (non-available) 1.5GHz P4. Intel's vaporware isn't even in the same league.
False.
Someone reported a slew of leaked benchmarks on a preproduction 1.4 GHz P4. While many of them were disappointing (more on this later), the SPEC scores reported were rather impressive. In particular, the SPECfp2000 score was 517 IIRC. That would put a 1.5 GHz P4 somewhere in the neighborhood of 550--not quite the 599 that Alpha can pull off, but certainly "in the same league", and certainly not "killed".
But there's more. The actual SPECfp score of the P4 will be much higher, for two reasons:
1) That preproduction 1.4 GHz P4 was almost certainly crippled in some way--i.e. one section of cache may be set to bypass, or several associativity ways turned off, or branch prediction could be turned off, or instruction paths could be routed very conservatively in microcode, or any number of dozens of things. This is almost always true of prerelease benchmarks of a new core; it happened before the Athlon release last year, and before the PPro release so long ago. Furthermore, there is good reason for prerelease chips to be crippled in this way--this is how engineers test finished cores to make sure they're ready for release. Essentially it's a lot like how coders test their code--turn something off and make sure everything works in the degenerate case; that way you not only know it will work in real operation, but if there is a bug you're much closer to isolating it. These settings are set in microcode and not changable by whichever two-bit employee at an OEM decides to leak benchmarks off a preproduction CPU sent there for validation (or whoever leaked the scores). Indeed, preventing accurate leaks is a reason to leave prerelease chips partially disabled like this.
In particular, the resulting slew of benchmarks which came from this particular preproduction P4 showed several very odd results almost certain not to represent the true performance of the fully functioning chip. It's not that the chip did poorly across the board, but rather that a couple particular benchmarks which stress particular aspects of the chip happened to be stupendously bad, and other benchmarks were mainly impressive. Paul DeMone, an EE who writes some excellent technical articles at realworldtech.com and posts often in places like Ace's Hardware, says that in his opinion these particular scores come from a P4 which has had its L1 associativity-way-prediction turned off, thus effectively increasing the latency of the L1 from 2 to 5 or so cycles. (Of course, there's no proof that the P4 even does way-prediction; the idea that it does is simply Paul's conjecture to explain how Intel got an L1 to run with 2-cycle latency at 1.5 GHz (and greater) on a.18um process. You can read his article on this here.) I'm not knowledgeable enough to comment, but Paul knows what he's talking about.
2) As I said in my original post (your accusation of "playing down fp performance" notwithstanding), the P4's SSE2 instruction set includes double-precision fp instructions for the first time. Thus we have an x86 chip which can finally run the double-precision operations necessary for SPECfp without relying on the horribly antiquated 8 register stack-based x87 fp implementation. While the 2-cycle latency L1 data cache will help the P4 run x87 code much faster, the reason all the major RISC chips have always creamed x86 chips in SPECfp is because they have sensible floating point implementations while the x86 chips are stuck with x87 for compatability reasons. But now that the P4 can use SSE2 instead of x87 instructions, it may have a real chance to compete or even win in SPECfp. We've already seen how the SPECfp performance of the leaked chip was remarkably better than that of a 1GHz P3 (517 vs. 327). What we don't know is if those numbers were made using a compiler which was fully optimized for SSE2. Indeed, it's likely that the SPECfp numbers which accompany the P4's release will later be improved upon by better SSE2 compilers.
Furthermore, classification of the P4 as "vaporware" is utterly uncalled for. This is a chip that's remained remarkably on schedule, with the only official delay being one month for a chipset (not CPU) issue. Believe it or not, Willamette is a chip that wasn't supposed to even exist, because by now the P3 was supposed to be replaced by a 2nd-generation, consumer oriented IA-64 chip. (Now *that's* slipping on your schedule!) Plus, Intel has been intentionally downplaying the performance of the P4, which is the exact opposite of what one does with vaporware. While some (who are not knowledgeable enough to study the actual P4 design) have taken this to mean that the P4 will be a disappointing performer, it in fact suggests just the opposite. Intel is not afraid to exaggerate (lie about) the performance of its upcoming processors in order to scare off competition--witness Merced/Itanium. Likewise, Intel is known to downplay the performance of upcoming MPUs in order to surprise the market when the chip finally debuts--they did this, very effectively, with the PPro, which surprised everyone in taking the SPECint lead away from Alpha when it was introduced. Intel leaked that the PPro core was supposed to be a dog; instead it's been perhaps the most impressive core design in history.
In any case, we may not have long to wait. While I'd bet we won't see any final SPEC numbers until the P4 release next month, Intel's presentation on the P4 at this week's MicroProcessor Forum is tomorrow. Will we see SPEC numbers? Who knows? But if we don't, it's likely only because Intel has a big surprise ready on Nov. 20.
Look at your history. Any time in the past 8 years that x86 has come close to Alpha, something happens where Alpha again leapfrogs in speed. That's not about to change.
Look at your history. x86 actually did briefly leapfrog the Alpha in SPECint95 when the PPro was first released; Alpha took the crown back with the EV6. In any case, the PPro was Intel's only new core introduction of the past 8 years! Thus, judging from our exactly and precisely 1 datapoint, we can conclusively predict that the P4 will indeed take the SPEC crown away from Alpha for a short while.
I mean, come on--this is engineering, not history. You're not going to get anywhere with a teleological theory of CPU performance over the last 8 years--that's just ridiculous. We don't need to guess or play "history" here; there is plenty of solid evidence about both Intel and Compaq's upcoming designs, how they will perform, and when and at what speeds they will be released. It is a known fact that the P4 is going to be released on November 20 at speeds of 1.4 and 1.5 GHz. It is also a known fact that the Alpha looks like it is stuck at 833 MHz until the release of the EV68 die shrink. It is moreover a known fact that the EV68 is behind schedule and that it will *not* be released before November 20.
Next, we can look at the again well known design specs of the P4 and EV68. The initial EV68s are essentially just a process shrink of the EV67 from.25um to a hybrid.18/.25um process. The only change we can expect there is a higher clock speed--around 1 GHz by January. A full.18um EV68 will ship by maybe March or April; that ought to hit 1.2-1.3 GHz and might include an on-die L2. Still, we're safe assuming that the EV68 will scale linearly with clock speed at best, with perhaps a slight bump from the on-die L2 in March or April.
The P4, on the other hand, is a completely new core, full of some really pretty impressive design features. From the point of view of SPEC, the most important are the 3.2GB/s FSB, trace cache, 2-cycle data L1, and larger reorder buffers. In addition, the half-clock-latency ADD will be a tremendous help for much of SPECint and the double-precision SSE2 instructions may allow the P4 to be the first x86 chip to compete or even win on SPECfp. (The much maligned 20-stage pipeline is largely compensated for by the improved larger branch predictors.) In other words, it is very likely that the P4 will achieve higher IPC on SPECint than did the P3. Oh, and it runs at 1.5 GHz.
So, assuming that the P4 will have the same SPECint IPC as P3 (an assumption generous to your argument) and the EV68 the same as the EV67, Compaq would need to release a 1.066 GHz Alpha before the 1.5 GHz P4 (i.e. in a month). Assuming the more likely occurrence that the P4 achieves ~20% better IPC on SPECint, Compaq will need a 1.266 or 1.3 GHz Alpha to win. Very very doubtful. Looking ahead, it doesn't appear that the EV68 will be able to keep up with the P4 in ramping clock speeds over the next 18 months, either; while the lead will likely go back and forth, it will probably be Intel, not Compaq, with the SPECint lead the lion's share of the time. This is all up until the release of the EV8, which looks to kick some serious butt. Note that I'm not claiming the P4 is a better design than the EV6x, just that Intel has access to better fab process than Alpha.
In any case, as you see, this is something that can be analyzed with known facts, engineering principles, and informed industry predictions, not with some blind appeal to (false) generalizations made over an entire 8 year period. In other words, "Alpha always wins SPEC" is not quite on a par with "never start a land war in Russia."
I'd like to see a P4 using LESS power, as small a power supply, and in the form factors that Alpha can get in today. Not to mention price. It'll be interesting to see all the same people who bitch about Alpha not having this or or that or being too expensive when they see P4 systems at $4000 using non-standard power supplies and cases.
LOL! The P4 *does* use less power than an Alpha; around 50W vs. an astounding 100W for an 833MHz EV67. And as for form factors, I truly have no idea what the hell you're talking about. Yes, the initial implementation of the P4 requires a somewhat ridiculous heatsink compared to the typical x86. That's alright, because the initial implementation of the P4 is being positioned essentially like the original PPro was. The P4 won't become Intel's mainstream chip until around 9 months from now, with the Northwood revision and die-shrink. Northwood, being made on a.13um process, will have much lower power, voltage and cooling requirements than Willamette. This sort of thing happens with every new Intel core: the first implementation is big, hot and power hungry and is aimed at a niche workstation market; then they do a die-shrink and move it into the mainstream. Thus the fact that P4 systems will cost around $2500 (*not* $4000) at introduction really doesn't mean anything. Even at $4000 they'd still be a full order of magnitude cheaper than an 833MHz Alpha system.
You think the Alpha can beat the P4 in form factor?? Uh...which do you think is going to come out first, an EV68-based laptop or a P4-based laptop?? Or howabout this: an EV68-based tablet computer or one with a P4?? Sun can make a stab at offering embedded CPUs, but I've never heard of anyone even considering an embedded Alpha. Why?? Well...power and form factor constraints, obviously. This is absurd.
The FUD is so thick, you can cut it with a knife.
You can say that again. Luckily, come Nov. 20 the FUD will be cleared away, for better or for worse. Now, I'm not arguing that the P4 or even Foster (the "P4 Xeon") will be able to replace Alpha for most of its markets. But it looks as if after years of failing to take advantage of their one clear selling point--unambiguous SPEC superiority--the Alpha is going to lose even that. Here's hoping Compaq finally decides to pour the resources necessary into making this excellent architecture prosper as it deserves. But let's not pretend that Intel is standing still just because the Alpha has been.
That's right, this site gives the impression
that P3 1.13 Ghz is the fastest CPU in the industry.
In fact, both Alpha and U-Sparc-III are TWICE faster.
Well, no. By far the most credible cross-platform CPU benchmark is SPEC; if you had bothered to check the SPEC CPU2000 scores you would find that a 1GHz P3 is precisely as fast as a 900MHz US-III on SPECint2000_base (438) and only 30% slower in SPECfp2000_base (327 vs. 427). Assuming linear scaling (not generally a great idea but close enough in this case), a 1.13GHz P3 will be 13% faster than the USIII-900 in SPECint, and 16% slower in SPECfp. In other words, they will be essentially equal.
Now, the Alpha actually is quite a bit faster than the P3, with SPECint/fp scores of 514/591 for an 833MHz chip. That makes it 17%/80% faster than our 1GHz P3, and an estimated 4%/60% faster than a 1.13GHz P3.
Now, on the other hand, both the Alpha and the US-III systems tested cost many times more than the i840 1GHz P3 system; not only do the chips cost a good deal more, but they get the benefit of much faster (and more expensive) buses to memory, etc. This makes quite a difference even in the SPEC CPU tests, and if the chips could somehow be placed on equivalent platforms, the P3 would easily win SPECint outright, and might be rather competitive even on SPECfp. (In case you were wondering, the reason the P3 sucks at SPECfp is because it is saddled with the register-starved x87 floating-point implementation for backwards compatability reasons; the P4 will go quite a ways towards solving this problem--at least as far as newly compiled code goes--with its SSE2 instructions.)
On the third hand, as has been pointed out, the 1.13 GHz P3 does not exist, and never did. (Intel "launched" what amounted to several engineering samples which turned out not to work properly anyways. Other than a couple dozen sent off to review sites and OEMs for validation, no 1.13 GHz P3's ever left the company.)
On the fourth hand, a chip which does currently exist in much higher quantities, the 1.5 GHz P4, looks like it will quite forcefully take the SPECint crown away from Alpha, and depending on Intel's progress in optimizing their compilers for SSE2, might even take the SPECfp crown as well. We'll get to find out when it is officially released in about a month or so.
On the fifth and final hand, though, the real advantage of the Alpha and US-III is their platforms, which give them much greater i/o throughput--often more important than CPU power for server applications anyways--and allow them to scale to configurations of 32 and 64 CPUs and beyond; Intel has a long way to go to compete on these measures.
8 months at the earliest--i.e. Q3 of next year. And that's for Tualatin, their.18->.13 die shrink of the P3. Northwood, the.13 P4, isn't due until Q4. (It'll be a lot easier for them to test and get the kinks out of the shrink with an old core that they understand very well than with a new.) Remember, they're already working on the shrink now.
The real story about Intel that their Pentium 4 is being manufactured here in Israel (in Kiryat Gat), and they got LOTS of problems with that (like low numbers of chips on 1 waffer, only 2 machines to product the P4), so until Intel gets more machines to produce those chips - this will take long time.
This story was posted here in the Israeli newspapers..
Well if it was a repeat of this piece at The Register, it may have misinterpreted things. For one thing, 70% yield on a new chip is quite impressive. For another, while a yield of 70,000 chips/week is not enough to substantially affect the x86 market, 70,000 represents more CPUs made in a week (and before launch no less) than there were 1 GHz P3s made for the first 6 months after its "launch"!
In any case, no OEM is going to buy many of the initial P4s, because Intel is planning a packaging change in March or so. That means new motherboards, new systems, and another entire validation process for OEM's--something that most of them are not going to want to waste time doing twice. Thus they'll only offer a couple models, in limited quantities, of the original P4, so it doesn't matter that only limited quantities will be fabbed.
In essence the first P4 will be positioned quite a bit like the original PPro. Whether the Willamette core ends up as successful as the P6 core has (remember, the PPro introduced the core now powering Celeron and P3) will be interesting to see.
I think the reason people get the impression that AMD chips are not top of the line, is because of the poor compatability with other components.
Exactly. While your issues may have been different, most of the major incompatability issues with the first Athlon chipsets involved AGP issues with various nvidia GeForce variants which were apparently drawing too much power or operating in some other non-standard-but-works-on-Intel way. Of course, the only reason these cards still worked-on-Intel is because nvidia and the card manufacturers worked hard to make sure their cards work with Intel hardware, and don't care nearly as much about compatability with the smaller AMD share of the market. Same as why more stuff is optimized for SSE than 3DNow even the 3DNow has been around for much longer. These issues are starting to go away now, as AMD gains strength, market-share and reputation in the consumer market; any issues would likewise go away quickly in the business market, given the chance.
Of course, it's worth noting that the components that gave AMD the most trouble, namely GeForce cards, don't belong anywhere near an enterprise server. Still, AMD needs to be very much more careful about such issues when launching into the business market than they were when launching into the consumer market. That's why they're taking their time introducing the 760MP and Mustang. Of course, your suggestion--having Compaq, IBM, etc. servers guaranteed to only contain validated configurations--will go a long way as well. (It'll be a cold day in hell before Dell offers an AMD enterprise server. I give them 6 months to a year.;)
As others have said, the first SMP chipset for the K7 family is the 760MP, due out in December or January. AMD has already released the single-processor 760 chipset--which looks to be the first DDR-capable PC chipset--to motherboard manufacturers; motherboards based on the 760 should be showing up at the end of this month.
It has been pointed out that one of the major reasons AMD has taken so long to get SMP going is that they already sell all of their processors anyways, and already have commitments for all the K7 chips they can make through the end of the year. This is true, but misses a more important point: with some few exceptions (i.e. nerds like us;) the market for SMP boxes is primarily business servers. This presents two issues:
1) Many if not most Intel SMP boxes are currently stuffed with Xeons. Thus, AMD has been perfectly content waiting for the release of their upcoming "Mustang" core tweak, featuring up to 1MB L2 cache--due out...you guessed it...in December or January--before rolling out the 760MP. Conversely, the big-L2 Mustangs without an SMP-capable chipset are dead-in-the-water.
2) By and large, business still clings to the notion of AMD as a cut-rate unreliable chip company. Despite the fact that knowledgable consumers have switched over to the cheaper faster Athlon in droves, Intel still has a nearly complete monopoly over the x86 business market. This impression of AMD as the "cheapo generic brand" persists despite Intel racking up delay after delay, errata after errata, recall after recall, embarrassment after embarrassment (i820, i840, 1.13GHz P3, Itanium) in the past year and a half. AMD knows that if they release the 760MP and it runs into one rumor of one conflict with one obscure 3D graphics card no business machine would ever contain anyways, their foray into the high-margin world of business computing is over before it began. (Never mind that Intel can release the i840, their new workstation-quality top-of-the-line chipset, with an "errata" which rendered it unusable with ECC memory!) Thus they are being very careful, and rightly so, with their validation process on this one.
Interestingly enough, if they get their act together (and purchasing departments take their heads out of their asses), AMD has a major market opportunity on their hands here. The Coppermine Xeon (i.e. a plain-old Coppermine P3 with $200 tacked on to the price) is incapable of scaling past 1GHz until Intel moves to a.13um process--in about 8 months. The 512kb, 1Mb and 2Mb L2 cache Xeons won't move above 800MHz or so in that time. Meanwhile, the new P4 chips are *not* SMP capable (or at least there will be no SMP chipsets available for them). Itanium is a joke and will likely never be launched. Now, Foster, the "P4 Xeon" will be released, possibly as soon as January, but the large-cache versions of Foster won't be out until Q2.
That leaves AMD with a quarter as the sole supplier of GHz+ large-cache multiprocessing x86 CPUs. Will that be enough to get them into the lucrative enterprise market?? Hard to say. After all, you never get fired for buying Intel...
Um, if you haven't noticed I *used* to be an insightful member of the/. community. Unfortunatley that community now sucks. I'm trying to blow all my karma on moronic trolls like the one I did above, though I'm not having much success losing all the karma.
Dude, I hadn't noticed. I mean, I remember your name vaguely from back in the day, but it's a bit narcissistic of you to think anyone is going to particularly notice the absence of your unique brand of irreplaceable insight. One particular poster's insightful comments are appreciated by many when they're there, but not missed when they're gone.
On the other hand, as a slashdotter for quite some time now I sympathize with the idea that the quality of debate around here has gone down a steep hill. Of course, when you and I were just coming around to slashdot, there was already the first wave of veterans leaving slashdot with complaints of how the community they used to appreciate was gone, how the new members (that's us!) were just a bunch of oafish ignorant trolling me-too waste-of-space first-posters who just didn't "get it". And yet we found slashdot to be a worthwhile place to offer our views and learn from others'.
So maybe we're right, and slashdot really has gone done the tubes. Or maybe that first generation of members was right, and slashdot was already worthless by the time we got here. Or maybe both groups are wrong: maybe it's not that slashdot's gotten worse, but just that we've gotten bored with it, grown out of it. That happens sometimes. In any case, even if they're "wrong" to derive worth from it (just as perhaps we were when we first started), obviously some people still like slashdot since its popularity continues to increase. Even if all those carpetbaggers are "incorrect" to want to join the slashdot community at a horrible time as this, they have a right to do so, and they have a right to expect that the existing community won't turn against them, take advantage of them, or abuse them.
Specifically, I mean that they have a right to be able to trust that someone who has earned a +1 karma bonus, ostensibly for reliably posting insightful statements, will use his amplified "more equal than others" voice in the community to inform rather than to mislead. By posting this troll at +2 by default, you abused the trust of the slashdot community, especially of those who were too ignorant to know better. The entire idea of the +1 karma bonus system is that those who have earned the bonus are insightful and honest enough to be trusted; you've subverted that assumption as well.
Now, I'm not trying to take away your right to troll. I firmly believe in that right, and indeed I'm finding more and more that I enjoy reading a well-done troll quite a bit more than the ho-hum repitition that constitutes the majority of "insightful" posts. Looking through your recent posts, I find that you've written several worthwhile and funny trolls lately, and have in some cases gotten rewarded for it (although it's a reward you claim not to want). But there are trolls and then there are trolls, and the line between the two is subtle. What I mean to say is that there are good trolls--where the effect is to trick some people not into believing what you say, but into believing that you believe what you say, and thus into humorously wasting their time trying to argue against your ridiculous point. And then there are posts like this--"misleading" might be a better description than "troll"--where the result is that at least some people are tricked into actually believing what you say is actually true. This is the sort of thing that has no place here, especially in a post with an automatic +2. Now, as I said the line is a subtle one, and on some boards this type of post would fall into the "funny troll" category, not the misleading one. But as you must know by now, slashdot is not known for its accurate knowledge of the x86 hardware scene. I'm sure it was inadvertant on your part, but your post had the potential to seriously misinform many readers who would assume it to be true, especially given the personal nature of your made-up anecdotes and your +2 score.
Besides, your stated reason for trolling in the first place is (sorry) pretty fucking lame. You don't feel slashdot is up to your standards anymore? Fine. Look around, find a new more sophisticated site to hang out on (there are plenty, trust me), or a better way to spend your time. (I'm sure there are many many better ways all of us could spend our time, but anyways.) What reason could you possibly have for trying to "blow" all your karma?? It's not like it's useful for anything. It's not like using up all your tokens on expensive snowboarding games when its time to leave the arcade. If I were you I'd keep it around--maybe that auto +2 will come in handy one day when you really want your voice heard on something. It's not like karma goes away over time or anything; it just means that, should you ever stop by slashdot, you'll get moderator access and your bonus. Nothing wrong with that.
Again, if you want to troll, go ahead; in fact, keep it up. But even if you're unhappy with it right now, you are part of the slashdot community, and with the recognition your karma bonus gives you comes responsibility. There are plenty of other options if you want to troll. Don't take your +1 bonus. Post anonymous, or create a new login and identity just for tongue-in-cheek trolling. Yes, your trolls will be seen by a lot fewer people if they don't start at +2, but that's the way the moderation system is designed to work--to give people who don't know any better an idea of what's accurate and what's not. Just because you can take advantage of it doesn't make doing so ok.
When I made my post, his was still at +2 (karma bonus), and had gotten 4 responses from people who obviously didn't realize he was trolling. So yes, the moderation system worked; whether moderators would have noticed had I not pointed it out is another question.
And I realize he's 15, and really actually enjoy many of his trolls and the responses they provoke. I wasn't trying to attack him, just point out that this time he made IMO a judgement in error, and to let everyone else know that what he said was made up.
All of my AMD CPUs have suffered what the technical community knows as Silicon Burn. Mind you, these are K6-2 and K6-3 CPUS, but all of them have failed due to Silicon Burn. I've heard several reports of the new Thunderbird Athlons failing due to Silicon Burn as well. Personally, the extry 200Mhz isn't worth it to me when I know that the CPU could fail tomorrow. I have had no problems with Intel CPUs, even when overclocked they do not experience Silicon Burn until you have been using them for several years.
LoC-
I really do enjoy most of your trolls--looking through your user info it's kind of humorous to see which happen to end up +5 funny (generally the heavyhanded ones) and which 0 troll and -1 flamebait (often the most subtle)--but I think this is a bit irresponsible. Believe it or not, there are many people on Slashdot dumb enough to swallow this. And while you (and I sometimes) might think they deserve to pay an extra $100 for an equivalant CPU, both they and the good folks at AMD would rather legitimately disagree.
Indeed, while I think it's a very disturbing sign of how much corporate power has usurped the 1st Ammendment, there have been companies who have sued posters for knowingly making similar false claims in online public forums--and won. Yes, they shouldn't win, and yes it's very doubtful that AMD would stoop that low, but that doesn't change the fact that you're purposely spreading ignorance and doing a lot of people a great disservice by posting shit like this here.
If you want to troll making fun of public misconceptions of AMD chips, save it for boards like JC's or Ace's or SI, where people actually know something about the CPU market, not Slashdot where the vast majority are completely ignorant on the subject. And think before you post, dude.
I guess you're going to be one of the few people who's going to rush out and jump on the bandwagon for this piece of trash who doesn't know that a PC by anyother name is *STILL* a PC, only in this case it a *CRIPPLED* PC...
First off, I doubt I'll buy the XBox. Don't have the cash, more of a PC gamer myself these days, and my dorm room'll probably go in for a PS2 this winter because it'll make a great cheap DVD player and it's coming when we're juniors, not seniors. In terms of who I want to win in this war, I'd have to go with Sony as well, because MS is MS, and while Sony is perhaps even worse of a corporate bully, they do a better job covering it up. Indeed, I was initially amazingly impressed with the PS2 when announced, and initially ready to write of the XBox as potential vapor.
What's changed on the PS2 front is that the PS2 has come out with a very disappointing (i.e. barely better than DC) first crop of games, and developers have gone on record in huge numbers decrying its difficulty of programming. It's still a great deal if you're looking for an awesome DVD player that also plays good games, but as a competitor in the video game wars it looks less and less appealing. (And incidentally Sony will lose mucho money if people like me buy it primarily for its DVD-playback.)
Meanwhile, the XBox has progressed far enough that it is absolutely certain to be a real product, not just a vaporware gambit. (155 developers with SDK's, detailed specs, and orders already in with component suppliers like Intel and nVidia.) And everywhere you go you hear about how MS is actually doing a good job listening to developers and giving them what they want.
As for the idea that the XBox is "a PC by anyother name" I'm not sure how exactly you get this. You are aware that Intel MPU's are actually capable of running the instructions necessary to play fighting games and platformers just as well as they can war simulations and FPS's, right? You do know that Direct X and OpenGL (both are offered as API options on XBox) are just as capable of rendering every display primative possible just as well as PS2 assembly code or arcade machines, don't you? That in fact several generations of arcade machines have run essentially off PC-ish hardware, like Voodoo-based Obsidian cards, and still managed to not be PC's??
Eh, I've had enough of this. Congrats on a successful trolling effort if that's what you're after (apparently, by your automatic 0 score); otherwise, please use a little self-reflection and skepticism before passing on nonsensical FUD you've happened to run across.
Not absurd at all. PC ports of NON-PC games has historically been awful.
Ok. Console ports of previous-generation console games are often fantastic.
Just take a look at any of the Arcade games that were ported over to the PC. Saying that most of them were awful would be an understatement.
Very true. However, let's look at the reasons for this:
1) PC controls are very different from console/arcade controls.
2) PC gamers belong to a different demographic than console/arcade gamers.
3) Arcade->PC ports are almost always done by 3rd-tier development houses and are thus of awful quality. The fact that there are no liscensing fees to develop for PC enhances this problem.
So, then, let's look at the XBox. It will have a console controllers, console-demographic buyers, and a cutoff of reputable 3rd-party developers because it follows the same economic model as other consoles. Crash X is being developed by Konami, one of the most respected development houses in the world, so that takes care of #3.
What am I missing? Or are you just trolling, repeating the inane non sequiter that because the XBox uses PC-ish hardware and modified PC API's it will inevitably be no different than a PC??
as much as we'd like to pretend, napster is _not_ under the GPL, and their fight is _not_ the same as Free Software's fight. while napster raises some interesting IP issues, they still want to make a lot of money with what they're doing, and for those looking to make napster into a shining cause of the New Internet, this was a train wreck waiting to happen.
Err, Red Hat and all the other distro makers are "under the GPL" but they still "want to make a lot of money with what they're doing". The difference is that while Linux is a great example of the power of the "Old" Internet--distributed programming and developer mailing lists and all that--Napster is indeed the quintessential example of one competing vision for the "New" Internet. (The other, of course, is an Internet in which only the big media companies have enough money to create/buy content, scare off most of their competition with lawsuits and choke off the rest through their control of the pipes into everyone's home.)
Now, of course, it can be argued that Napster has less of a right than Red Hat et. al. to make money off their insight, because whereas Red Hat provides tech support and reliability checks, and pays several programmers to write GPL'd software for the entire Linux base, Napster just provides an interface, some database servers, and, uh, pays for Limp Bizkit's promotional tour. Indeed, I have no problem paying for a distro but would strongly consider switching to a different peer-to-peer file network if Napster goes for-pay.
But that doesn't mean Napster's fight isn't worthwhile, and it doesn't make Napster's contribution any less revolutionary.
Atari and Commodore pulled this kind of stunt years ago with their "game machines/computers" Basically get a company to do a cheap port of old outdated games with retouched graphics. I'm *NOT* impressed.
Nintendo pulled this kind of stunt as well, launching the SNES with Super Mario World and the N64 with Mario 64. Each was hailed in its time as the greatest video game made to date.
I'm not following your point here. If you have some evidence that Crash X will be just a "cheap port with retouched graphics" please present it; otherwise, join the rest of the world in realizing that it will instead be an entirely new game which happens to be based around the Crash franchise. Not that that's such an amazing thing in itself--personally, I've found the Crash series to be underwhelming, although admittedly I've only played them a bit. Still, the idea that Konami would take a stellar franchise like Crash and a brand-new console of unprecedented power and ease of programming and produce nothing but a rehash of some old, linear-scrolling PS1 game is pretty absurd.
Umm sorry to burst your bubble but Metal Gear Solid X (the one being developed for the X-Box) is just a port of the original Metal Gear Solid with some enhancements. You won't be seeing "Sons of Liberty" on anything other than the PS2 (I doubt the X-Box could even handle it).
Wrong. "Konami has just confirmed that this title will be based on Metal Gear Solid 2, still coming to PS2, not the original Metal Gear Solid for PlayStation."
As for your misinformed comment on the power of the XBox, such a statement is laughable. On a purely technical level, the XBox is superior to the PS2 by a factor of 2 or 3 in every relevant category: GPU power, CPU power, graphics memory, main memory, and memory bus speed. As such, its theoretical performance is 2 to 3 times better as well, whether in terms of pure theoretical base polys pushed or polys pushed with effects. Of course, this is to be expected, as it will launch over a year after the PS2, not include the extra components needed to double as a rack-quality DVD player, and take advantage of the well-known benefits of using off-the-shelf hardware.
This is indisputible fact. Moreover, everything I've heard points to the conclusion that whereas most developers (console *and* PC) are finding the PS2 very difficult to program for, most (PC *and* console) find the XBox very easy on this measure. This adds extra weight behind the report (confirmed by Konami in any case) that MGS: Sons of Liberty will debut on the XBox concurrently or even prior to its release for PS2.
You may have been confused by the fact that the original MGS is indeed being ported to the PC (due out much earlier than MGS 2's Q4 2001, though). Or you may just be spreading FUD. Or perhaps just unable to consider the fact that as oafish, dimwitted, and ponderous as MS seems compared to the PC software industry, when compared to tyrannical giants like Sony and Nintendo (and to idiots like Sega) they are not at much disadvantage at all.
A version of Crash Bandicoot. Obviously it won't be Crash 1 or 2 or 3, but a new game. At least, I certainly hope that was already blindingly obvious to you. The point is that the XBox not only has some impressive games coming out for it, but some which were previously considered "Sony-only", i.e. Crash and MGS.
Another problem with this idea is that it (unlike Napster) is illegal under the 1992 Audio Home Recording Act and the DMCA. The AHRA specifically legalized the noncommercial sharing of all recorded music; the DMCA further "clarified" the definition of "noncommercial" to include only those instances of sharing in which nothing is gained in return. In other words, a service in which one person shares while expecting nothing in return from anyone who downloads from him or her is noncommercial, and thus legal, whereas one in which something is expected in return--i.e. a traditional ratio FTP site, and potentially the scheme you've laid out here--is not.
Of course, the only reason this would be necessary is if Napster is found to be illegal anyways, so I suppose the point is moot. On the other hand, as others have pointed out, this scheme seems to rely inherently on trusted clients, which is a bad idea in any system, much less an open-source mixed-client protocol like Gnutella.
Just because you don't like MS software doesn't mean they don't do anything. Since you asked, here is what I was able to turn up in a grand 45 seconds of research:
Konami announces 10 XBox titles a year. More importantly, the titles announced include Crash Bandicoot and Metal Gear Solid 2, which will be a launch title. Couldn't find it in 45 seconds, but I read last week an interview with someone from Konami in which it was confirmed that not only would MGS be a launch title for XBox but that it would be enhanced over the PS2 version and that the PS2 version would come out *later*, due to the difficulties of programming for the PS2.
Looking through the list of confirmed XBox developers, I see almost all the big console names (in addition to every important PC developer I can think of). Konami, Capcom, Midway, Rage, Hudson, Acclaim, Bandai, THQ, Ubi, lots lots more. The most notable exceptions from the list appear to be Square (rumored to be announcing support soon), Rare (Nintendo only, of course), and the big one, EA--or is EA subsumed under Eidos these days? And, of course, Nintendo and Sega.
In any case, I wasn't trying to argue that the XBox will be successful (I think it will, but it's much too early to tell), or that it will be the greatest console in history (probably not), but rather to refute the previous idiot who claimed that there was "no proof this thing actually existed". Being "in the game industry", you must agree that that's nothing but ridiculous FUD.
Slashdot Mirror
Even on the SpecInt benchmark, the earliest PPro benchmarks I could find on Spec's website show that while the PPro put in some respectible numbers, it was far from being the king of the SpecInt benchmark.
/.
This is wrong. The original PPro launched just before the EV56 revision of the (at that time) DEC Alpha (i.e. the Alpha 21164). The PPro's SPECint was better than the previous Alpha's, and worse than the new Alpha which launched a few months down the road. Thus, the PPro was, at the time, the absolute SPECintCPU95 king. Of course it was later eclipsed in performance by better and more expensive chips; the point is, at the time it was an extraordinarily innovative design and a miraculous surprise to the entire MPU community.
For starters, SPEC is not a single benchmark, rather a consortium that comes up with benchmarks, the most well recognized being their CPU benchmarks (colloquially refered to as SPEC benchmarks). These benchmarks however, do not exclusively test a CPU, but rather a system as a whole, although they are designed to make the CPU the limiting factor (nonetheless using bucket loads of RAM, fast disk controllers, and a huge external memory cache can have wonderful impacts on SPEC benchmarks). Typically these benchmarks have been divided into those that stress the integer unit (SpecInt) and those that stress the floating point unit (SpecFP). The Pentium Pro was the first x86 CPU to post respectable SpecFP benchmarks, but it still got it's butt kicked all over the place compared to it's RISC competition.
Yes, the SPEC_CPU benchmarks test an entire platform--CPU, i/o, memory, and (very important) compiler--not just a CPU. Of course, this does nothing but put x86 chips at a major *disadvantage*, as they are only available in configurations which do not even approach the i/o and memory bandwidth available to the big RISC chips. (Of course, Intel has perhaps the best compiler group around, which mitigates this somewhat.) Still, the SPEC_CPU benchmarks are called SPEC_ CPU for a reason, as they make every attempt to, as you say, make the CPU (and its attendant cache hierarchy) the limiting factor as much as possible in a suite of varied, meaningful, non-synthetic benchmarks. As for the other SPEC benchmarks (i.e. not SPEC_CPU), they are of course designed to benchmark the, well, non-CPU parts of the computer, and are thus irrelevant here. Again, when one refers to "SPEC scores" it is assumed that they mean SPEC_CPU; thus the original poster's usage was quite correct.
As you point out, the PPro never challenged the Alpha's SPECfp lead, due to a major deficiency of the x86 architecture--backwards-compatability with the ass-backwards x87 fp instruction set. Luckily, this is finally being phased out with the P4's SSE2 instructions; thus if Compaq fails to execute with the Alpha in the coming months, and if Intel's SSE2 compilers are good enough, the P4 may just win both SPEC_CPU benchmarks outright for a time--perhaps as much as a year or more. That ought to shut up the ignorant anti-x86 FUD on
Yeah, right.
When the P6 was released, it was the fastest processor available in industry standard benchmarks (SPEC, including Alpha). Its design was highly original, and manages to keep the CISC nastiness contained to the first few stages of the pipe. Claiming that the P6 was not a world-class design when released is only a testament to your own ignorance.
Exactly correct. If I had moderator points, they'd be yours.
And indeed, the 1 GHz P3--on that same, 5 year old P6 core--is still tied with the moderately-vaunted brand-new mucho-expensive (not available until Q1) 900 MHz UltraSparcIII in SPECint2000. The 1.2 GHz Athlon would presumably perform even better (once they release SPEC scores from the new Compaq Fortran compilers), making it second only to the fastest (and also none-too-available) Alphas in terms of pure performance. The x86 ISA may be suboptimal, but Intel and now AMD have been able to keep up with the best--and most expensive--of the RISC world due to superior engineering (except when compared to the excellent Alpha team) and superior process technology. Sure they may not have the i/o bandwidth, RAS, or operating systems to compete in the big leagues, but anyone dissing today's x86 chips on account of their designs or engineering qualities is, as the poster said, demonstrating their ignorance.
And if Compaq doesn't hurry the EV68 (die-shrunk Alpha) to market, the P4 and perhaps Mustang as well will blow by even the mighty Alpha, in SPECint and possibly even SPECfp. (The last real knock against the x86 ISA is that it is saddled with the horrendous x87 fp architecture, which is why x86 SPECfp scores trail everyone else by so much. With the P4's upcoming SSE2 instructions, however, that problem may be in the past.) Aesthetics aside, there is no doubt that x86 processors, taken as a whole, are easily the best designed, highest performing MPU's around.
As much ire is directed at Rambus, I can't see the logic behind this. Ashton-Tate is relevant because they channeled resources into legal wrangling, rather than R&D and customer service. They failed, by putting all their eggs into one basket, a very wrong basket.
Check this out: in the last quarter, Rambus made $10 mil. on total revenues of $27 mil. Ok, so far so good (for a hideously overvalued speculative stock). The kicker is this (unfortunately, they took this little tidbit out of a previous version of this article on news.com): during the quarter, Rambus spent $1 million *per month* on legal fees. That's right--their legal fees were about a third of their profits and over 11% of their entire revenues!
Guess that's why they dropped their suit against Hitachi the day it was actually given a go-ahead by the courts. Of course, now that they're being countersued by not one but two industry giants (Micron and Infineon), with very deep pockets and very annoyed looks on their faces, those legal fees are just going to keep going higher and higher. And every single company in the industry--Intel included (if not Intel most of all!)--is going to enjoy watching them slowly bleed to death on legal fees.
That line first appeared with the 386. And it was wrong then, too.
The line with the 386 (and 486, and P5, and PPro) was that it was destined to remain a workstation chip for some time. This was Intel marketing bluster, yes, but it was moderately true--the initial versions of each of those chips was produced on an old process; once the chip moved to a new process, it became feasible for upper-end mainstream machines. In a two years, they were mainstream.
Right. But the difference here is, these chips are not intended for workstations. They're not intended for moderately sized servers. They're intended to replace mainframes, and to run high-high-end scientific code. In case you didn't read my other post, these things are going to cost *at least* $10,000 *apiece* to MAKE. Just for the MPU. Moreover, they will not work up to their full potential without *massive* bandwidth, which still costs mucho $, last time I checked.
What servers really need is multiple CPUs and huge I/O bandwidth, not faster individual CPUs.
Oh wait, you didn't even read the article. The POWER4 *is* multiple MPU's--it's 8 cores on 1 die. This thing doesn't come in anything less than 8-way configurations. As for I/O bandwidth, would 84 GB/s be enough for you?
You still think this thing is a desktop chip???
On the other hand, Apple can't afford to change CPUs again.
Or maybe they can't afford not to. Contrary to all those MacOS Rumors, there are no definite plans for Apple to move even to the upcoming G4+ MPU's, which are essentially another incarnation of the tired G4, just with a stretched out pipeline which will get it to 800MHz at the cost of lower IPC. There aren't definite plans yet because the G4+, like the G4 and the G3 before it, is not a desktop MPU but rather an embedded/DSP chip which Motorola happens to sell to Apple to use in Macs. The design, ramping, pricing, roadmap, are all influenced primarily by Moto's embedded customers first, not by Apple. Unfortunately, despite the fact that they are utterly dependent on them, Apple has decided to treat first IBM and then Motorola rather poorly, and thus haven't gotten much in the way of support when they decided that they may, perhaps, want to increase the speed of their top MPU more than 50 MHz in a year. Whoops. BTW, did you catch Apple's earnings today? Ouch.
All of this is too bad--OS X looks like perhaps the best thing going as far as operating systems goes. There are always rumors that Apple's going to finally make their surprise move to x86. Their experience with PPC the last year or so, and the accompanying beating their bottom line has taken, might be the thing to finally push them over. I personally think they might still be able to carry over enough incompatabilities to stay the sole supplier of MacOS hardware--after all, the XBox uses x86, and it will be plenty incompatible with PCs. Migrating software will be a gigantic pain...but on the other hand, it's not like the Mac has too much in the way of software anyways. (OS X, and any Cocoa programs, will port very very quickly.) Who knows?
I think the big advantage that VLIW instruction sets will have is strictly architectural, and I'm not sure how IBM's approach fits in yet, but it looks interesting. Throwing more chips at the problem is one approach, but remember that your competitors can do that too, *and* make the chips do more as well...
Not sure how IBM's approach fits in yet?? Read the article.
Amongst other things, the POWER4 is *not* VLIW, it's straight-ahead modern RISC at its finest. With massively gigantic buffers, bandwidth and execution resources (8 functional units/core * 8 cores = wow), this chip'll do quite nicely on IPC/core, not to mention combined IPC for all 8. While presumably not quite as elegant, the design for the individual cores bears a lot in common with the archetype of perfect RISC cores, the Alpha 21264, and it has even more aggressive resources.
Essentially what this means is, assuming this design is as good as it appears, the only way the competition will be able to catch up (without going the way IBM has and deciding on a prohibitively expensive 8-in-one design and packaging) will be through the use of innovative design tricks. The upcoming P4 has a few of those, incidentally, but the big one--and the one the P4 *doesn't* have--appears to be SMT, Simultaneous MultiThreading. Alpha has an 8-way SMT core coming out in a bit, and it ought to compete well with IBM's much more expensive 8-way SMP design here. And AMD appears ready to do 2-way SMT (or something similar) with the Sledgehammer in about 15-18 months. And Sun is rumored to have SMT in the USV design due in several years. But the POWER4 looks to lead in the "big bad" category for quite some time to come.
(As for Intel's EPIC, the VLIW-like design strategy for their IA-64 chips, at the moment it's looking like a rather poor competitor to SMT. A quick explanation of why:
There are exactly two ways to make an MPU run faster: 1) increase the clock speed, or 2) increase the IPC (instructions per clock). Unfortunately, the best we've been able to do so far in the IPC realm is about 1.4 IPC on SPEC benchmarks (Alpha EV6x). IPC on a P3 runs about 40% lower. Now, these IPC numbers are despite the fact that the Alpha can theoretically retire 8 instructions/clock, and the P3 5 (5 internal ops, not 5 x86 ops). Furthermore, simulations show that as far as attacking the IPC problem by adding more functional units, we're nearing the point of diminishing returns.
The problem is, in order to run lots of instructions in parallel, you have to be able to safely extract parallelism from your code. And the problem with this is, you can't run instructions in parallel if they have dependencies, etc. And furthermore, nowadays all this parallelism has to be safely extracted in real-time by special hardware in the MPU itself; this makes your chip more complicated, and means you need to build a big buffer to hold instructions in flight so you can pick and choose which ones you want to run each clock.
So many many years ago, HP had the idea, which it later sold to Intel (and which wasn't really there idea at all but has indeed been used in DSP chips for years and years), of getting rid of all that complex insruction-level parallel-finding logic on the MPU and doing it all at compile-time instead. This is the basic idea behind EPIC, the philosophy of Intel's IA-64 line.
It sounds very nice, especially because in theory it means simpler chips (no complicated control logic), and simpler chips means faster chips. Heh heh heh. See it turns out that the amount of instruction-level parallelism which can be safely discovered at compile-time is way way less than the amount that could be found in the chip at run-time (which, as we recall, is too small already). Thus EPIC was modified to allow the compiler to just place "hints" in the code. Well, this means you still need all that complicated control logic back in place, because you still don't have deterministically scheduled instructions. But following the "hints" and other changes to the ISA ends up making everything *more* complicated, not less. This, in a nutshell, is why Itanium is 3 years late, way over budget, unable to meet its very modest clock speed goal of 800 MHz, and fitted with a laugh-enducing 96kb of on-die cache, lower even than the lowliest Celeron: all this added complexity means bigger, slower, more complicated chips that don't have the room for cache or the elegance for high (or even adequate) clock speeds. Plus we have very strong evidence that compiler technology is still not nearly good enough to make the kinds of insightful IPC-giving "hints" which are necessary to even make the damn fool scheme work. Thus the only benchmark Intel has "released" for the Itanium is that of an RSA-encryption--a routine simple enough to be hand-tuned in assembly. Meanwhile they have made the patently ridiculous claim that the SPEC benchmarks--directed precisely at the mid-cost server/workstation market which Itanium is aiming for--are "not relevant" to Itanium's market.
A completely opposite approach is SMT, which uses a relatively small number of core changes to allow not just instruction-level parallelism to be gleaned, but also thread-level parallelism. In other words, the chip will run several threads in parallel, confident in the fact that their instructions will not have dependencies on each other, and thus be able to use much more of its full execution capabilities. Early indications are that SMT can improve IPC by remarkable amounts, like on the order of 2x the performance on otherwise similar cores!
Unfortunately, it is too early to tell whether SMT will be as easy a design enhancement as is being claimed. Furthermore I've heard tell that SMT on IA-64 will be a lot more difficult than on a RISC MPU, so Intel could be missing out on a huge speed-up with this technique.)
However, IBM will have to make sure people design their apps with more than one processor in mind, which will be a Good Idea for the future, since more people might have multiprocessor computers.
These chips are not to be confused with PowerPC chips. They are server chips only, intended for seriously expensive machines.
Heh! If only Apple would use these, the new iMacs wouldn't exactly be quite able to hit their price points. Paul (the author of the article) and some others were involved in a thread over on the tech forum at Ace's about (amongst other things) the expected cost of one of these puppies.
To quote Paul's response:
Maybe another way of looking at it is perhaps the price of four POWER4 known good die and the ceramic substrate and metal carrier totals $3000 (although I suspect that a tested and 100% functional ceramic substrate itself might approach or exceed $3000 in cost).
The real question is the cost of a fully assembled and tested, 100% functional, POWER4 8-way module? After all what are the chances one of these can be reworked if even just one of the 20,000+ solder ball joints was bad?
So for one of these 8-way on a chip jobs (unsure if they'll be offering 4-way configurations too or if those were just a prototype) it's looking like upwards of $10,000 just for IBM to fab, package, and test the darn things. Add in a system capable of feeding it the tremendous bandwidth it requires to run up to its full potential--8 GB/s to DRAM and a phenomenal 84 (!) GB/s I/O--and...ok, so I know Hemos was just joking when he made that comment about Apple, but you get the idea. These are MPUs you use to fold proteins and run gigantic dynamic-content websites, not surf the web and edit the home video of your kid's elementary school graduation.
On a related note, man these things oughtta show Intel a thing or two about how to marry clever instruction scheduling to brute-force functional units--forget about Itanium; it's gonna take a several-way McKinley system to even take a swing at this these. And it oughtta show Sun a thing or two about the dangers of resting on the laurels of your marketing success when designing new chips. And, as Paul notes in the article, it really oughtta make Alpha engineers worry that for the first time, having the most elegant design may not guarantee the best performance. Compaq has an 8-way SMT Alpha core on the way as well (EV8); too bad the Alpha group's customary position in the world--stepped on and neglected by their corporate masters--means they haven't got the money or manpower to bring it to market until well after POWER4.
Virtually any IC that you'll find anywhere can withstand voltage fluctuations, especially ones that small. If a processor rated for 1.5V couldn't run at 1.1V, you can bet they would write 1.8V or so for the recommended Vcc level. Even your TTL ICs can withstand that type of change.
This is nothing that should be seen as impressive, it's just PR that will fool the masses and make the engineers wonder why it would exist any other way.
Individual IC's can, yes. The point is that when you have many many IC's strung together to make a chip, the tolerances for the whole chip become much tighter. Perhaps you've heard of fan-out/fan-in? This is extremely impressive, dude. Don't believe me, try switching the voltage for your P3 from 1.6V to 1.2V and see how it does.
Seriously, all of the major recent Intel chipset problems were with RDRAM. There's the infamous i820E problem with the third RDRAM chip not getting registered (which was after the initial, pre-RDRAM i820 went bust). And now this, when the ICH2 is coupled with the i850 and i860 MTHs, which use... starts with R, you know this...... RDRAM! Right!
I think the time for the NVidia DDR chipset is NOW. Let's stop this half-assed hardware engineering and pre-alpha lithography which the Intel staff is undertaking.
It's not that simple. Both RDRAM and DDR require tighter tolerances on the chipset and motherboard levels for the same obvious reason--the more bandwidth you want to transfer, the less your tolerance for noise and defects, and the greater the danger of crosstalk. Period. This is simply a fact of life if you want the benefits of high-speed DRAM without taking the (in my opinion inevitable) step of ditching the current system of expandable commodity RAM on the motherboard in favor of a system which ties hardwired or embedded DRAM to the MPU.
Now, you can argue that RDRAM makes the problem worse by trying to cram the same amount of bandwidth as PC1600 DDR into a thinner bus. And you can argue that dealing with a new memory communications protocol has led to more bugs. I'm not going to argue with you there. On the other hand, RDRAM proponents would counter that RDRAM lessens these problems by switching to a packet-based protocol to cut down on interference.
And you can argue--as you did--that the fact that mighty Intel has run into myriad problems trying to implement RDRAM for their PC chipsets, and that their results to date--the i820 and i840--have been lackluster at best. But you can also point out that Intel has (for marketechture reasons) made this switch on a processor designed for use with SDRAM only. You can point out that they've (also foolishly) decided to have the switch coincide with a switch to hub-based memory management--which, incidentally, appears to be the source of the erratum in the i850, not the use of RDRAM. You might also want to note that Intel's first efforts with SDRAM chipsets, while not as starcrossed as the i820/i840, were nowhere near as efficient, stable or refined as their 4th-generation BX chipset or the 5th-gen i815.
Finally, it's worth noticing that there are exactly zero currently available DDR chipsets, bug-free or otherwise, with which to compare Intel's RDRAM record. Of course the main reason for this is politics--only after a year of Intel floundering with the RDRAM protocol did the industry finally coalesce behind DDR. And of course this fact will be changing quite soon, within the month in all likelihood. The fact that several working and apparently stable DDR chipsets are on the verge of being released, from chipset designers less accomplished than Intel, means that DDR cannot be as difficult to implement as RDRAM-backers have long argued it would be. Still--and this is very important--no one has claimed it was easy. Indeed, DDR chipsets were by all accounts much more difficult to get working than SDRAM chipsets, and even now many are reportedly quite finicky when working with DDR made by different manufacturers. Furthermore, all DDR motherboards due for release in the near future are, like all of Intel's RDRAM boards, six-layer. This improves stability and reduces crosstalk at the expense of extra engineering effort and manufacturing cost; SDRAM chipsets tend to be quite stable with just 4 layers.
I dunno where that leaves the ease-of-implementation balance. It appears that it's on the side of DDR, but it's still a bit premature to say so conclusively. In any case, the idea that whipping up a DDR chipset to replace RDRAM is child's play is absolutely false.
On the other hand, Intel already does have a DDR chipset for the P4--or rather, for Foster, the "P4 Xeon" due out in the beginning of the year. Now, this chipset could be modified for use with the normal P4 with little problem, but there are two big reasons it won't be:
1) Cost: In order to fill Foster's massive FSB, Intel's new chipset uses, IIRC, dual-channel double-wide DDR. Contrary to what you may have heard, this means a much much higher cost than the dual-channel RDRAM bus on the i850. The reason is that RDRAM's one unambiguous advantage over DDR is in using fewer pins; having two channels and doubling the bus width means multiplying the already quite large number of DDR pins by 4, which in turns means motherboards which are mucho expensive, even if two sticks of DDR is cheaper than two sticks of RDRAM.
2) Legal obligations. Intel is under contract with Rambus not to promote any other next-gen DRAM standard for its mainstream desktop line until 2003. That means that, unless they want their asses sued off (and we all know if there's one thing Rambus is good at, it's ass-sue-offing), the best Intel can do for the next couple years is license the P4 bus and allow 3rd-party chipset makers like VIA, ALi and, as you mentioned, perhaps even nvidia make DDR chipsets for the P4. Actually the contract specifically states only that Rambus has the option to revoke Intel's RDRAM license if they promote a DDR chipset for their desktop chips, so the big question is, would Intel risk losing their Rambus license, especially when there is no cheap DDR solution which can take full advantage of the P4's 3.2GB/s FSB? On the other hand, would Rambus risk revoking Intel's RDRAM license and thus taking themselves out of the PC DRAM industry possibly for good??
I dunno. Frankly I'm just hoping Infineon succeeds in overturning some of Rambus' RDRAM patents with prior art as they're seeking to do. (I'd be shocked if they didn't succeed in showing prior art for Rambus' "patents" on SDRAM and DDR.)
You also play down the FP performance, where the Alpha as is still kills the projected values for the (non-available) 1.5GHz P4. Intel's vaporware isn't even in the same league.
.18um process. You can read his article on this here.) I'm not knowledgeable enough to comment, but Paul knows what he's talking about.
False.
Someone reported a slew of leaked benchmarks on a preproduction 1.4 GHz P4. While many of them were disappointing (more on this later), the SPEC scores reported were rather impressive. In particular, the SPECfp2000 score was 517 IIRC. That would put a 1.5 GHz P4 somewhere in the neighborhood of 550--not quite the 599 that Alpha can pull off, but certainly "in the same league", and certainly not "killed".
But there's more. The actual SPECfp score of the P4 will be much higher, for two reasons:
1) That preproduction 1.4 GHz P4 was almost certainly crippled in some way--i.e. one section of cache may be set to bypass, or several associativity ways turned off, or branch prediction could be turned off, or instruction paths could be routed very conservatively in microcode, or any number of dozens of things. This is almost always true of prerelease benchmarks of a new core; it happened before the Athlon release last year, and before the PPro release so long ago. Furthermore, there is good reason for prerelease chips to be crippled in this way--this is how engineers test finished cores to make sure they're ready for release. Essentially it's a lot like how coders test their code--turn something off and make sure everything works in the degenerate case; that way you not only know it will work in real operation, but if there is a bug you're much closer to isolating it. These settings are set in microcode and not changable by whichever two-bit employee at an OEM decides to leak benchmarks off a preproduction CPU sent there for validation (or whoever leaked the scores). Indeed, preventing accurate leaks is a reason to leave prerelease chips partially disabled like this.
In particular, the resulting slew of benchmarks which came from this particular preproduction P4 showed several very odd results almost certain not to represent the true performance of the fully functioning chip. It's not that the chip did poorly across the board, but rather that a couple particular benchmarks which stress particular aspects of the chip happened to be stupendously bad, and other benchmarks were mainly impressive. Paul DeMone, an EE who writes some excellent technical articles at realworldtech.com and posts often in places like Ace's Hardware, says that in his opinion these particular scores come from a P4 which has had its L1 associativity-way-prediction turned off, thus effectively increasing the latency of the L1 from 2 to 5 or so cycles. (Of course, there's no proof that the P4 even does way-prediction; the idea that it does is simply Paul's conjecture to explain how Intel got an L1 to run with 2-cycle latency at 1.5 GHz (and greater) on a
2) As I said in my original post (your accusation of "playing down fp performance" notwithstanding), the P4's SSE2 instruction set includes double-precision fp instructions for the first time. Thus we have an x86 chip which can finally run the double-precision operations necessary for SPECfp without relying on the horribly antiquated 8 register stack-based x87 fp implementation. While the 2-cycle latency L1 data cache will help the P4 run x87 code much faster, the reason all the major RISC chips have always creamed x86 chips in SPECfp is because they have sensible floating point implementations while the x86 chips are stuck with x87 for compatability reasons. But now that the P4 can use SSE2 instead of x87 instructions, it may have a real chance to compete or even win in SPECfp. We've already seen how the SPECfp performance of the leaked chip was remarkably better than that of a 1GHz P3 (517 vs. 327). What we don't know is if those numbers were made using a compiler which was fully optimized for SSE2. Indeed, it's likely that the SPECfp numbers which accompany the P4's release will later be improved upon by better SSE2 compilers.
Furthermore, classification of the P4 as "vaporware" is utterly uncalled for. This is a chip that's remained remarkably on schedule, with the only official delay being one month for a chipset (not CPU) issue. Believe it or not, Willamette is a chip that wasn't supposed to even exist, because by now the P3 was supposed to be replaced by a 2nd-generation, consumer oriented IA-64 chip. (Now *that's* slipping on your schedule!) Plus, Intel has been intentionally downplaying the performance of the P4, which is the exact opposite of what one does with vaporware. While some (who are not knowledgeable enough to study the actual P4 design) have taken this to mean that the P4 will be a disappointing performer, it in fact suggests just the opposite. Intel is not afraid to exaggerate (lie about) the performance of its upcoming processors in order to scare off competition--witness Merced/Itanium. Likewise, Intel is known to downplay the performance of upcoming MPUs in order to surprise the market when the chip finally debuts--they did this, very effectively, with the PPro, which surprised everyone in taking the SPECint lead away from Alpha when it was introduced. Intel leaked that the PPro core was supposed to be a dog; instead it's been perhaps the most impressive core design in history.
In any case, we may not have long to wait. While I'd bet we won't see any final SPEC numbers until the P4 release next month, Intel's presentation on the P4 at this week's MicroProcessor Forum is tomorrow. Will we see SPEC numbers? Who knows? But if we don't, it's likely only because Intel has a big surprise ready on Nov. 20.
Look at your history. Any time in the past 8 years that x86 has come close to Alpha, something happens where Alpha again leapfrogs in speed. That's not about to change.
.25um to a hybrid .18/.25um process. The only change we can expect there is a higher clock speed--around 1 GHz by January. A full .18um EV68 will ship by maybe March or April; that ought to hit 1.2-1.3 GHz and might include an on-die L2. Still, we're safe assuming that the EV68 will scale linearly with clock speed at best, with perhaps a slight bump from the on-die L2 in March or April.
.13um process, will have much lower power, voltage and cooling requirements than Willamette. This sort of thing happens with every new Intel core: the first implementation is big, hot and power hungry and is aimed at a niche workstation market; then they do a die-shrink and move it into the mainstream. Thus the fact that P4 systems will cost around $2500 (*not* $4000) at introduction really doesn't mean anything. Even at $4000 they'd still be a full order of magnitude cheaper than an 833MHz Alpha system.
Look at your history. x86 actually did briefly leapfrog the Alpha in SPECint95 when the PPro was first released; Alpha took the crown back with the EV6. In any case, the PPro was Intel's only new core introduction of the past 8 years! Thus, judging from our exactly and precisely 1 datapoint, we can conclusively predict that the P4 will indeed take the SPEC crown away from Alpha for a short while.
I mean, come on--this is engineering, not history. You're not going to get anywhere with a teleological theory of CPU performance over the last 8 years--that's just ridiculous. We don't need to guess or play "history" here; there is plenty of solid evidence about both Intel and Compaq's upcoming designs, how they will perform, and when and at what speeds they will be released. It is a known fact that the P4 is going to be released on November 20 at speeds of 1.4 and 1.5 GHz. It is also a known fact that the Alpha looks like it is stuck at 833 MHz until the release of the EV68 die shrink. It is moreover a known fact that the EV68 is behind schedule and that it will *not* be released before November 20.
Next, we can look at the again well known design specs of the P4 and EV68. The initial EV68s are essentially just a process shrink of the EV67 from
The P4, on the other hand, is a completely new core, full of some really pretty impressive design features. From the point of view of SPEC, the most important are the 3.2GB/s FSB, trace cache, 2-cycle data L1, and larger reorder buffers. In addition, the half-clock-latency ADD will be a tremendous help for much of SPECint and the double-precision SSE2 instructions may allow the P4 to be the first x86 chip to compete or even win on SPECfp. (The much maligned 20-stage pipeline is largely compensated for by the improved larger branch predictors.) In other words, it is very likely that the P4 will achieve higher IPC on SPECint than did the P3. Oh, and it runs at 1.5 GHz.
So, assuming that the P4 will have the same SPECint IPC as P3 (an assumption generous to your argument) and the EV68 the same as the EV67, Compaq would need to release a 1.066 GHz Alpha before the 1.5 GHz P4 (i.e. in a month). Assuming the more likely occurrence that the P4 achieves ~20% better IPC on SPECint, Compaq will need a 1.266 or 1.3 GHz Alpha to win. Very very doubtful. Looking ahead, it doesn't appear that the EV68 will be able to keep up with the P4 in ramping clock speeds over the next 18 months, either; while the lead will likely go back and forth, it will probably be Intel, not Compaq, with the SPECint lead the lion's share of the time. This is all up until the release of the EV8, which looks to kick some serious butt. Note that I'm not claiming the P4 is a better design than the EV6x, just that Intel has access to better fab process than Alpha.
In any case, as you see, this is something that can be analyzed with known facts, engineering principles, and informed industry predictions, not with some blind appeal to (false) generalizations made over an entire 8 year period. In other words, "Alpha always wins SPEC" is not quite on a par with "never start a land war in Russia."
I'd like to see a P4 using LESS power, as small a power supply, and in the form factors that Alpha can get in today. Not to mention price. It'll be interesting to see all the same people who bitch about Alpha not having this or or that or being too expensive when they see P4 systems at $4000 using non-standard power supplies and cases.
LOL! The P4 *does* use less power than an Alpha; around 50W vs. an astounding 100W for an 833MHz EV67. And as for form factors, I truly have no idea what the hell you're talking about. Yes, the initial implementation of the P4 requires a somewhat ridiculous heatsink compared to the typical x86. That's alright, because the initial implementation of the P4 is being positioned essentially like the original PPro was. The P4 won't become Intel's mainstream chip until around 9 months from now, with the Northwood revision and die-shrink. Northwood, being made on a
You think the Alpha can beat the P4 in form factor?? Uh...which do you think is going to come out first, an EV68-based laptop or a P4-based laptop?? Or howabout this: an EV68-based tablet computer or one with a P4?? Sun can make a stab at offering embedded CPUs, but I've never heard of anyone even considering an embedded Alpha. Why?? Well...power and form factor constraints, obviously. This is absurd.
The FUD is so thick, you can cut it with a knife.
You can say that again. Luckily, come Nov. 20 the FUD will be cleared away, for better or for worse. Now, I'm not arguing that the P4 or even Foster (the "P4 Xeon") will be able to replace Alpha for most of its markets. But it looks as if after years of failing to take advantage of their one clear selling point--unambiguous SPEC superiority--the Alpha is going to lose even that. Here's hoping Compaq finally decides to pour the resources necessary into making this excellent architecture prosper as it deserves. But let's not pretend that Intel is standing still just because the Alpha has been.
That's right, this site gives the impression
that P3 1.13 Ghz is the fastest CPU in the industry.
In fact, both Alpha and U-Sparc-III are TWICE faster.
Well, no. By far the most credible cross-platform CPU benchmark is SPEC; if you had bothered to check the SPEC CPU2000 scores you would find that a 1GHz P3 is precisely as fast as a 900MHz US-III on SPECint2000_base (438) and only 30% slower in SPECfp2000_base (327 vs. 427). Assuming linear scaling (not generally a great idea but close enough in this case), a 1.13GHz P3 will be 13% faster than the USIII-900 in SPECint, and 16% slower in SPECfp. In other words, they will be essentially equal.
Now, the Alpha actually is quite a bit faster than the P3, with SPECint/fp scores of 514/591 for an 833MHz chip. That makes it 17%/80% faster than our 1GHz P3, and an estimated 4%/60% faster than a 1.13GHz P3.
Now, on the other hand, both the Alpha and the US-III systems tested cost many times more than the i840 1GHz P3 system; not only do the chips cost a good deal more, but they get the benefit of much faster (and more expensive) buses to memory, etc. This makes quite a difference even in the SPEC CPU tests, and if the chips could somehow be placed on equivalent platforms, the P3 would easily win SPECint outright, and might be rather competitive even on SPECfp. (In case you were wondering, the reason the P3 sucks at SPECfp is because it is saddled with the register-starved x87 floating-point implementation for backwards compatability reasons; the P4 will go quite a ways towards solving this problem--at least as far as newly compiled code goes--with its SSE2 instructions.)
On the third hand, as has been pointed out, the 1.13 GHz P3 does not exist, and never did. (Intel "launched" what amounted to several engineering samples which turned out not to work properly anyways. Other than a couple dozen sent off to review sites and OEMs for validation, no 1.13 GHz P3's ever left the company.)
On the fourth hand, a chip which does currently exist in much higher quantities, the 1.5 GHz P4, looks like it will quite forcefully take the SPECint crown away from Alpha, and depending on Intel's progress in optimizing their compilers for SSE2, might even take the SPECfp crown as well. We'll get to find out when it is officially released in about a month or so.
On the fifth and final hand, though, the real advantage of the Alpha and US-III is their platforms, which give them much greater i/o throughput--often more important than CPU power for server applications anyways--and allow them to scale to configurations of 32 and 64 CPUs and beyond; Intel has a long way to go to compete on these measures.
Intel moves to .13 in 8 months? I would like to see that..
.18->.13 die shrink of the P3. Northwood, the .13 P4, isn't due until Q4. (It'll be a lot easier for them to test and get the kinks out of the shrink with an old core that they understand very well than with a new.) Remember, they're already working on the shrink now.
Here's the latest roadmap, in Japanese pastel no less.
8 months at the earliest--i.e. Q3 of next year. And that's for Tualatin, their
The real story about Intel that their Pentium 4 is being manufactured here in Israel (in Kiryat Gat), and they got LOTS of problems with that (like low numbers of chips on 1 waffer, only 2 machines to product the P4), so until Intel gets more machines to produce those chips - this will take long time.
This story was posted here in the Israeli newspapers..
Well if it was a repeat of this piece at The Register, it may have misinterpreted things. For one thing, 70% yield on a new chip is quite impressive. For another, while a yield of 70,000 chips/week is not enough to substantially affect the x86 market, 70,000 represents more CPUs made in a week (and before launch no less) than there were 1 GHz P3s made for the first 6 months after its "launch"!
In any case, no OEM is going to buy many of the initial P4s, because Intel is planning a packaging change in March or so. That means new motherboards, new systems, and another entire validation process for OEM's--something that most of them are not going to want to waste time doing twice. Thus they'll only offer a couple models, in limited quantities, of the original P4, so it doesn't matter that only limited quantities will be fabbed.
In essence the first P4 will be positioned quite a bit like the original PPro. Whether the Willamette core ends up as successful as the P6 core has (remember, the PPro introduced the core now powering Celeron and P3) will be interesting to see.
I think the reason people get the impression that AMD chips are not top of the line, is because of the poor compatability with other components.
;)
Exactly. While your issues may have been different, most of the major incompatability issues with the first Athlon chipsets involved AGP issues with various nvidia GeForce variants which were apparently drawing too much power or operating in some other non-standard-but-works-on-Intel way. Of course, the only reason these cards still worked-on-Intel is because nvidia and the card manufacturers worked hard to make sure their cards work with Intel hardware, and don't care nearly as much about compatability with the smaller AMD share of the market. Same as why more stuff is optimized for SSE than 3DNow even the 3DNow has been around for much longer. These issues are starting to go away now, as AMD gains strength, market-share and reputation in the consumer market; any issues would likewise go away quickly in the business market, given the chance.
Of course, it's worth noting that the components that gave AMD the most trouble, namely GeForce cards, don't belong anywhere near an enterprise server. Still, AMD needs to be very much more careful about such issues when launching into the business market than they were when launching into the consumer market. That's why they're taking their time introducing the 760MP and Mustang. Of course, your suggestion--having Compaq, IBM, etc. servers guaranteed to only contain validated configurations--will go a long way as well. (It'll be a cold day in hell before Dell offers an AMD enterprise server. I give them 6 months to a year.
As others have said, the first SMP chipset for the K7 family is the 760MP, due out in December or January. AMD has already released the single-processor 760 chipset--which looks to be the first DDR-capable PC chipset--to motherboard manufacturers; motherboards based on the 760 should be showing up at the end of this month.
;) the market for SMP boxes is primarily business servers. This presents two issues:
.13um process--in about 8 months. The 512kb, 1Mb and 2Mb L2 cache Xeons won't move above 800MHz or so in that time. Meanwhile, the new P4 chips are *not* SMP capable (or at least there will be no SMP chipsets available for them). Itanium is a joke and will likely never be launched. Now, Foster, the "P4 Xeon" will be released, possibly as soon as January, but the large-cache versions of Foster won't be out until Q2.
It has been pointed out that one of the major reasons AMD has taken so long to get SMP going is that they already sell all of their processors anyways, and already have commitments for all the K7 chips they can make through the end of the year. This is true, but misses a more important point: with some few exceptions (i.e. nerds like us
1) Many if not most Intel SMP boxes are currently stuffed with Xeons. Thus, AMD has been perfectly content waiting for the release of their upcoming "Mustang" core tweak, featuring up to 1MB L2 cache--due out...you guessed it...in December or January--before rolling out the 760MP. Conversely, the big-L2 Mustangs without an SMP-capable chipset are dead-in-the-water.
2) By and large, business still clings to the notion of AMD as a cut-rate unreliable chip company. Despite the fact that knowledgable consumers have switched over to the cheaper faster Athlon in droves, Intel still has a nearly complete monopoly over the x86 business market. This impression of AMD as the "cheapo generic brand" persists despite Intel racking up delay after delay, errata after errata, recall after recall, embarrassment after embarrassment (i820, i840, 1.13GHz P3, Itanium) in the past year and a half. AMD knows that if they release the 760MP and it runs into one rumor of one conflict with one obscure 3D graphics card no business machine would ever contain anyways, their foray into the high-margin world of business computing is over before it began. (Never mind that Intel can release the i840, their new workstation-quality top-of-the-line chipset, with an "errata" which rendered it unusable with ECC memory!) Thus they are being very careful, and rightly so, with their validation process on this one.
Interestingly enough, if they get their act together (and purchasing departments take their heads out of their asses), AMD has a major market opportunity on their hands here. The Coppermine Xeon (i.e. a plain-old Coppermine P3 with $200 tacked on to the price) is incapable of scaling past 1GHz until Intel moves to a
That leaves AMD with a quarter as the sole supplier of GHz+ large-cache multiprocessing x86 CPUs. Will that be enough to get them into the lucrative enterprise market?? Hard to say. After all, you never get fired for buying Intel...
Um, if you haven't noticed I *used* to be an insightful member of the /. community. Unfortunatley that community now sucks. I'm trying to blow all my karma on moronic trolls like the one I did above, though I'm not having much success losing all the karma.
Dude, I hadn't noticed. I mean, I remember your name vaguely from back in the day, but it's a bit narcissistic of you to think anyone is going to particularly notice the absence of your unique brand of irreplaceable insight. One particular poster's insightful comments are appreciated by many when they're there, but not missed when they're gone.
On the other hand, as a slashdotter for quite some time now I sympathize with the idea that the quality of debate around here has gone down a steep hill. Of course, when you and I were just coming around to slashdot, there was already the first wave of veterans leaving slashdot with complaints of how the community they used to appreciate was gone, how the new members (that's us!) were just a bunch of oafish ignorant trolling me-too waste-of-space first-posters who just didn't "get it". And yet we found slashdot to be a worthwhile place to offer our views and learn from others'.
So maybe we're right, and slashdot really has gone done the tubes. Or maybe that first generation of members was right, and slashdot was already worthless by the time we got here. Or maybe both groups are wrong: maybe it's not that slashdot's gotten worse, but just that we've gotten bored with it, grown out of it. That happens sometimes. In any case, even if they're "wrong" to derive worth from it (just as perhaps we were when we first started), obviously some people still like slashdot since its popularity continues to increase. Even if all those carpetbaggers are "incorrect" to want to join the slashdot community at a horrible time as this, they have a right to do so, and they have a right to expect that the existing community won't turn against them, take advantage of them, or abuse them.
Specifically, I mean that they have a right to be able to trust that someone who has earned a +1 karma bonus, ostensibly for reliably posting insightful statements, will use his amplified "more equal than others" voice in the community to inform rather than to mislead. By posting this troll at +2 by default, you abused the trust of the slashdot community, especially of those who were too ignorant to know better. The entire idea of the +1 karma bonus system is that those who have earned the bonus are insightful and honest enough to be trusted; you've subverted that assumption as well.
Now, I'm not trying to take away your right to troll. I firmly believe in that right, and indeed I'm finding more and more that I enjoy reading a well-done troll quite a bit more than the ho-hum repitition that constitutes the majority of "insightful" posts. Looking through your recent posts, I find that you've written several worthwhile and funny trolls lately, and have in some cases gotten rewarded for it (although it's a reward you claim not to want). But there are trolls and then there are trolls, and the line between the two is subtle. What I mean to say is that there are good trolls--where the effect is to trick some people not into believing what you say, but into believing that you believe what you say, and thus into humorously wasting their time trying to argue against your ridiculous point. And then there are posts like this--"misleading" might be a better description than "troll"--where the result is that at least some people are tricked into actually believing what you say is actually true. This is the sort of thing that has no place here, especially in a post with an automatic +2. Now, as I said the line is a subtle one, and on some boards this type of post would fall into the "funny troll" category, not the misleading one. But as you must know by now, slashdot is not known for its accurate knowledge of the x86 hardware scene. I'm sure it was inadvertant on your part, but your post had the potential to seriously misinform many readers who would assume it to be true, especially given the personal nature of your made-up anecdotes and your +2 score.
Besides, your stated reason for trolling in the first place is (sorry) pretty fucking lame. You don't feel slashdot is up to your standards anymore? Fine. Look around, find a new more sophisticated site to hang out on (there are plenty, trust me), or a better way to spend your time. (I'm sure there are many many better ways all of us could spend our time, but anyways.) What reason could you possibly have for trying to "blow" all your karma?? It's not like it's useful for anything. It's not like using up all your tokens on expensive snowboarding games when its time to leave the arcade. If I were you I'd keep it around--maybe that auto +2 will come in handy one day when you really want your voice heard on something. It's not like karma goes away over time or anything; it just means that, should you ever stop by slashdot, you'll get moderator access and your bonus. Nothing wrong with that.
Again, if you want to troll, go ahead; in fact, keep it up. But even if you're unhappy with it right now, you are part of the slashdot community, and with the recognition your karma bonus gives you comes responsibility. There are plenty of other options if you want to troll. Don't take your +1 bonus. Post anonymous, or create a new login and identity just for tongue-in-cheek trolling. Yes, your trolls will be seen by a lot fewer people if they don't start at +2, but that's the way the moderation system is designed to work--to give people who don't know any better an idea of what's accurate and what's not. Just because you can take advantage of it doesn't make doing so ok.
When I made my post, his was still at +2 (karma bonus), and had gotten 4 responses from people who obviously didn't realize he was trolling. So yes, the moderation system worked; whether moderators would have noticed had I not pointed it out is another question.
And I realize he's 15, and really actually enjoy many of his trolls and the responses they provoke. I wasn't trying to attack him, just point out that this time he made IMO a judgement in error, and to let everyone else know that what he said was made up.
All of my AMD CPUs have suffered what the technical community knows as Silicon Burn. Mind you, these are K6-2 and K6-3 CPUS, but all of them have failed due to Silicon Burn. I've heard several reports of the new Thunderbird Athlons failing due to Silicon Burn as well. Personally, the extry 200Mhz isn't worth it to me when I know that the CPU could fail tomorrow. I have had no problems with Intel CPUs, even when overclocked they do not experience Silicon Burn until you have been using them for several years.
LoC-
I really do enjoy most of your trolls--looking through your user info it's kind of humorous to see which happen to end up +5 funny (generally the heavyhanded ones) and which 0 troll and -1 flamebait (often the most subtle)--but I think this is a bit irresponsible. Believe it or not, there are many people on Slashdot dumb enough to swallow this. And while you (and I sometimes) might think they deserve to pay an extra $100 for an equivalant CPU, both they and the good folks at AMD would rather legitimately disagree.
Indeed, while I think it's a very disturbing sign of how much corporate power has usurped the 1st Ammendment, there have been companies who have sued posters for knowingly making similar false claims in online public forums--and won. Yes, they shouldn't win, and yes it's very doubtful that AMD would stoop that low, but that doesn't change the fact that you're purposely spreading ignorance and doing a lot of people a great disservice by posting shit like this here.
If you want to troll making fun of public misconceptions of AMD chips, save it for boards like JC's or Ace's or SI, where people actually know something about the CPU market, not Slashdot where the vast majority are completely ignorant on the subject. And think before you post, dude.
I guess you're going to be one of the few people who's going to rush out and jump on the bandwagon for this piece of trash who doesn't know that a PC by anyother name is *STILL* a PC, only in this case it a *CRIPPLED* PC...
First off, I doubt I'll buy the XBox. Don't have the cash, more of a PC gamer myself these days, and my dorm room'll probably go in for a PS2 this winter because it'll make a great cheap DVD player and it's coming when we're juniors, not seniors. In terms of who I want to win in this war, I'd have to go with Sony as well, because MS is MS, and while Sony is perhaps even worse of a corporate bully, they do a better job covering it up. Indeed, I was initially amazingly impressed with the PS2 when announced, and initially ready to write of the XBox as potential vapor.
What's changed on the PS2 front is that the PS2 has come out with a very disappointing (i.e. barely better than DC) first crop of games, and developers have gone on record in huge numbers decrying its difficulty of programming. It's still a great deal if you're looking for an awesome DVD player that also plays good games, but as a competitor in the video game wars it looks less and less appealing. (And incidentally Sony will lose mucho money if people like me buy it primarily for its DVD-playback.)
Meanwhile, the XBox has progressed far enough that it is absolutely certain to be a real product, not just a vaporware gambit. (155 developers with SDK's, detailed specs, and orders already in with component suppliers like Intel and nVidia.) And everywhere you go you hear about how MS is actually doing a good job listening to developers and giving them what they want.
As for the idea that the XBox is "a PC by anyother name" I'm not sure how exactly you get this. You are aware that Intel MPU's are actually capable of running the instructions necessary to play fighting games and platformers just as well as they can war simulations and FPS's, right? You do know that Direct X and OpenGL (both are offered as API options on XBox) are just as capable of rendering every display primative possible just as well as PS2 assembly code or arcade machines, don't you? That in fact several generations of arcade machines have run essentially off PC-ish hardware, like Voodoo-based Obsidian cards, and still managed to not be PC's??
Eh, I've had enough of this. Congrats on a successful trolling effort if that's what you're after (apparently, by your automatic 0 score); otherwise, please use a little self-reflection and skepticism before passing on nonsensical FUD you've happened to run across.
Not absurd at all. PC ports of NON-PC games has historically been awful.
Ok. Console ports of previous-generation console games are often fantastic.
Just take a look at any of the Arcade games that were ported over to the PC. Saying that most of them were awful would be an understatement.
Very true. However, let's look at the reasons for this:
1) PC controls are very different from console/arcade controls.
2) PC gamers belong to a different demographic than console/arcade gamers.
3) Arcade->PC ports are almost always done by 3rd-tier development houses and are thus of awful quality. The fact that there are no liscensing fees to develop for PC enhances this problem.
So, then, let's look at the XBox. It will have a console controllers, console-demographic buyers, and a cutoff of reputable 3rd-party developers because it follows the same economic model as other consoles. Crash X is being developed by Konami, one of the most respected development houses in the world, so that takes care of #3.
What am I missing? Or are you just trolling, repeating the inane non sequiter that because the XBox uses PC-ish hardware and modified PC API's it will inevitably be no different than a PC??
as much as we'd like to pretend, napster is _not_ under the GPL, and their fight is _not_ the same as Free Software's fight. while napster raises some interesting IP issues, they still want to make a lot of money with what they're doing, and for those looking to make napster into a shining cause of the New Internet, this was a train wreck waiting to happen.
Err, Red Hat and all the other distro makers are "under the GPL" but they still "want to make a lot of money with what they're doing". The difference is that while Linux is a great example of the power of the "Old" Internet--distributed programming and developer mailing lists and all that--Napster is indeed the quintessential example of one competing vision for the "New" Internet. (The other, of course, is an Internet in which only the big media companies have enough money to create/buy content, scare off most of their competition with lawsuits and choke off the rest through their control of the pipes into everyone's home.)
Now, of course, it can be argued that Napster has less of a right than Red Hat et. al. to make money off their insight, because whereas Red Hat provides tech support and reliability checks, and pays several programmers to write GPL'd software for the entire Linux base, Napster just provides an interface, some database servers, and, uh, pays for Limp Bizkit's promotional tour. Indeed, I have no problem paying for a distro but would strongly consider switching to a different peer-to-peer file network if Napster goes for-pay.
But that doesn't mean Napster's fight isn't worthwhile, and it doesn't make Napster's contribution any less revolutionary.
Atari and Commodore pulled this kind of stunt years ago with their "game machines/computers" Basically get a company to do a cheap port of old outdated games with retouched graphics. I'm *NOT* impressed.
Nintendo pulled this kind of stunt as well, launching the SNES with Super Mario World and the N64 with Mario 64. Each was hailed in its time as the greatest video game made to date.
I'm not following your point here. If you have some evidence that Crash X will be just a "cheap port with retouched graphics" please present it; otherwise, join the rest of the world in realizing that it will instead be an entirely new game which happens to be based around the Crash franchise. Not that that's such an amazing thing in itself--personally, I've found the Crash series to be underwhelming, although admittedly I've only played them a bit. Still, the idea that Konami would take a stellar franchise like Crash and a brand-new console of unprecedented power and ease of programming and produce nothing but a rehash of some old, linear-scrolling PS1 game is pretty absurd.
Umm sorry to burst your bubble but Metal Gear Solid X (the one being developed for the X-Box) is just a port of the original Metal Gear Solid with some enhancements. You won't be seeing "Sons of Liberty" on anything other than the PS2 (I doubt the X-Box could even handle it).
Wrong. "Konami has just confirmed that this title will be based on Metal Gear Solid 2, still coming to PS2, not the original Metal Gear Solid for PlayStation."
As for your misinformed comment on the power of the XBox, such a statement is laughable. On a purely technical level, the XBox is superior to the PS2 by a factor of 2 or 3 in every relevant category: GPU power, CPU power, graphics memory, main memory, and memory bus speed. As such, its theoretical performance is 2 to 3 times better as well, whether in terms of pure theoretical base polys pushed or polys pushed with effects. Of course, this is to be expected, as it will launch over a year after the PS2, not include the extra components needed to double as a rack-quality DVD player, and take advantage of the well-known benefits of using off-the-shelf hardware.
This is indisputible fact. Moreover, everything I've heard points to the conclusion that whereas most developers (console *and* PC) are finding the PS2 very difficult to program for, most (PC *and* console) find the XBox very easy on this measure. This adds extra weight behind the report (confirmed by Konami in any case) that MGS: Sons of Liberty will debut on the XBox concurrently or even prior to its release for PS2.
You may have been confused by the fact that the original MGS is indeed being ported to the PC (due out much earlier than MGS 2's Q4 2001, though). Or you may just be spreading FUD. Or perhaps just unable to consider the fact that as oafish, dimwitted, and ponderous as MS seems compared to the PC software industry, when compared to tyrannical giants like Sony and Nintendo (and to idiots like Sega) they are not at much disadvantage at all.
A version of Crash Bandicoot. Obviously it won't be Crash 1 or 2 or 3, but a new game. At least, I certainly hope that was already blindingly obvious to you. The point is that the XBox not only has some impressive games coming out for it, but some which were previously considered "Sony-only", i.e. Crash and MGS.
Another problem with this idea is that it (unlike Napster) is illegal under the 1992 Audio Home Recording Act and the DMCA. The AHRA specifically legalized the noncommercial sharing of all recorded music; the DMCA further "clarified" the definition of "noncommercial" to include only those instances of sharing in which nothing is gained in return. In other words, a service in which one person shares while expecting nothing in return from anyone who downloads from him or her is noncommercial, and thus legal, whereas one in which something is expected in return--i.e. a traditional ratio FTP site, and potentially the scheme you've laid out here--is not.
Of course, the only reason this would be necessary is if Napster is found to be illegal anyways, so I suppose the point is moot. On the other hand, as others have pointed out, this scheme seems to rely inherently on trusted clients, which is a bad idea in any system, much less an open-source mixed-client protocol like Gnutella.
Just because you don't like MS software doesn't mean they don't do anything. Since you asked, here is what I was able to turn up in a grand 45 seconds of research:
Konami announces 10 XBox titles a year. More importantly, the titles announced include Crash Bandicoot and Metal Gear Solid 2, which will be a launch title. Couldn't find it in 45 seconds, but I read last week an interview with someone from Konami in which it was confirmed that not only would MGS be a launch title for XBox but that it would be enhanced over the PS2 version and that the PS2 version would come out *later*, due to the difficulties of programming for the PS2.
Looking through the list of confirmed XBox developers, I see almost all the big console names (in addition to every important PC developer I can think of). Konami, Capcom, Midway, Rage, Hudson, Acclaim, Bandai, THQ, Ubi, lots lots more. The most notable exceptions from the list appear to be Square (rumored to be announcing support soon), Rare (Nintendo only, of course), and the big one, EA--or is EA subsumed under Eidos these days? And, of course, Nintendo and Sega.
In any case, I wasn't trying to argue that the XBox will be successful (I think it will, but it's much too early to tell), or that it will be the greatest console in history (probably not), but rather to refute the previous idiot who claimed that there was "no proof this thing actually existed". Being "in the game industry", you must agree that that's nothing but ridiculous FUD.