Slashdot Mirror

If you want people to click on your links, make them links. Put "<URL:" in front and ">" in back. Six extra characters. Not much work. Then "<URL:http://www.emulators.com/docs/nx03_10fixes.htm>" becomes "http://www.emulators.com/docs/nx03_10fixes.htm". Easy, no?

Did you write this? http://emulators.com/docs/pentium_1.htm :)

Some good reading here on what people are doing with these machines. http://emulators.com/docs/nx32_powerpc_g5.htm For the most part you can still run all the modern OS X software you want, just not Snow Leopard (10.6).

I dont buy either argument.

Has anyone profiled the code. Actually looked at where the slowdowns ARE. You know 'function x is called 18% of the time'. Then reorganizing the code to make it faster. You know actually code changes things like changing n^2 scans to log(n) scans. That sort of thing. Computer sciency work. All the arguments come down to 'i havent looked but i *GUESS* it is this'.

Has anyone run the code thru something like boundschecker? Memory leaks/overruns/underruns?! Good lord those are dead easy to find. There are tools out there that do it for you. Hell they even find subtle ones.

Compiling to machine code is a HACK. It will yeild some nice speedups. But does not address the core problem. No one is spending time actually restructuring the code to make it perform well.

I have worked on a couple of projects with 'many cooks'. You get to the point where you are afraid to change things as it will break a whole unrelated set of things. Firefox is almost there, but it is not too late!

Its is time to refactor!
see http://www.emulators.com/docs/nx01_intro.htm to see what you can do with just interperted code
see http://www.joelonsoftware.com/ about refactoring and what it can do for your code.

The Atari 400 was my first computer as well. I recall punching in BASIC programs from books and playing jumpman, miner2049r, star raiders and pacman with colored ghosts. I also remember calling POKE and seeing junk came up (not knowing what it was at the time). good times. http://www.emulators.com/xformer.htm

Then again, so does Linux and MacOS under emulation.

The smart thing is for Apple to write or hire someone to write an emulator tailored to "Classic" apps. Heck, if they play do the initial work and make it FOSS (not including MacOS9 and the Mac ROMs of course) it will get done faster and be better quality, assuming a lot of people make good contributions.

Of course, FOSS means run-on-any-Intel or even run-anywhere, and this might mean a bunch of Legacy-only Mac users who have no interest in MacOSX would just buy a cheap PC instead of a new Mac, Intel or otherwise.

It's not FOSS, but SoftMac may do the trick, but it requires specific versions of the MacOS ROM either as hardware or as a properly-licensed (i.e. you own a Mac and aren't using it) image file.

Some net.stores, including Other World Computing, sell older versions of Mac OS X for under $10. Ebay is also a likely source for OX X and 9.x and earlier.

You can get Mac OS X 7.5.3 + 7.5.5 updates straight from Apple. You might aslso check out Emulators.com for some emulators that let you run vintage Mac stuff on your PC.

Oh, yeah, I'm thinking this is PearPC under the hood. Someday a run-of-the-mill PC will be able to emulate a G4, but not today, certaintly not at less cost than a real G4 box from Apple.

never read Pentium 4: In Depth? Sure, Intel fixed some the blatent design flaws eventualy, but it still only works as good as it does becuase of huge-Mhz inflating-heat producing pipeline The P4 is a poor design from the start. P3-t's OTOH...

Take a look at http://www.emulators.com/softmac.htm
though it only goes up to Mac OS 8.1

And yours, of course, is based on the Truth.

There's a lot of interest in PPC emulators but the proposition is not feasible! Go read up on PowerPC assembly for a little while, then take a look at some x86 assembly. It is not a simple task to emulate one on the other. Several groups have been trying for years to build a PPC emulator on the x86 architecture, it just isn't in the cards.

OK, after a brief Google search, I found SoftMac From the site:

"All of our emulators run Mac software very quickly, typically delivering 50% to 70% of the clock speed in emulation. In other words, a 500 MHz PC running Windows 98 can emulate the equivalent of a 300 MHz Macintosh Quadra running Mac OS 8. Your results will vary depending on the exact type of PC you are using, the amount of RAM present, the version of Windows being used, video drivers, etc. For a good explanation on exactly what type of PC to use and how fast your results can be, see our Hardware Requirements and Benchmarks page."

This happens to be much more in line with the performance I would expect than your (unsubstantiated) claims would suggest. It is also a concrete example of a 100% complete PPC emulator running on x86.

You need to stop pulling things out of your ass and research them. Apple's AVERAGE margin on hardware is about 25%, read their quarterly statements for that. They make beaucoup cach off their hardware because they're the only ones that sell it and have tight control over their production process. Their high end equipment like PowerBooks have really nice margins reaching towards 30%.

Fine, I'll grant you the point regarding Apple's margins. With the new figures it will require 3-4 copies of MacOS X to equal one Powerbook.

I'm going to cherry pick the rest of you're article since we've already argued many of these points at least once.

OSX would need to support everything Windows does to be a viable alternative to Windows.

Not at all, as long as there is a logo program and an easy location to find the hardware compatibility list. The important thing is that high quality hardware is supported, not junk. And guess what, as Apple marketshare grows, more and more OEMs will support Apple.

Did you eat paint chips when you were young or do you not understand the pond concept? Apple makes very good money selling products to a niche market.

Surely you see that it does not follow that Apple wouldn't make even more money in a larger market. Especially if Apple creates additional market segments (say, the white box Mac crowd). Those who buy white boxes would likely never buy a Mac, and vice versa.

Apple's is a big fish swiming around in a little pond. They cater to the people other computer companies either miss entirely with their marketing or ignore because they desire selling to a lower common denominator.

No, in many markets (like education) Apple is head-to-head with the likes of Dell, Gateway, and Sony.

Dell is a little fish in a very big pond.

Dell is the biggest whale in the ocean.

They have to fight all the other little fish, even the fish building PCs in their garages out of white box parts, to survive.

So does Apple. If you think there aren't plenty of (for instance) students who look at iBook then decide a Compaq laptop is a better value for a variety of reasons, you're delusional.

If you dumped Apple into the big pond they would cease being a big fish and would also find they weren't equipped to survive in the big pond.

Apple could keep all of it's current branding, style and so on. The group that finds that appealing (about 3% of new computer buyers, IIRC) will continue to buy those boxes at Apple's rather obscene margins.

For the other 97%, something different is needed.

Thanks for an interesting discussion, and have a great day!

I'm curious about his technique and where the big holdup is. This page emulated an Atari 130XE, a very similar beast to the C64, and in their admittingly simplistic benchmark achieved 4x the speed running the emulator on a 486/66 (9x the speed on a Pentium 90). Put in the numerous architectural enhancement in today's processors, and I find it hard to believe that an 800Mhz processor cannot emulate a C64 at a 1:1 speed.

I see this alot nowadays - people saying that AMD have "lost their edge", or "been taking it easy for a while" ... that is simply not true. An AMD Athlon XP 2800+ _will_ beat an Intel Pentium IV at 2.8 GHz in most benchmarks (and the 3.06 GHz P4 in quite a few - see the latest ones at THG or AT if you don't trust me), just as it is supposed to. And you can still practically get two Athlons (not 2800+'s mind you) for the same price as one high-end Pentium IV. Surely no-one here thinks that a single P4, HT or no HT, stands a chance against a true SMP system (given apps that take advantage of both CPU's)?

Furthermore, there's no app or game available on this earth, and there probably won't be for at least two years to come, where the speed difference between an AXP/2800+ and a P4/3GHz is big enough to really mean anything to anyone other than the fanatical overclocking crowd, who will spend any amount of money just to have the fastest stuff on the market, only to use it for stuff like playing Counter-Strike, which uses perhaps 20% of the total CPU and graphics card capacity. Well, if you're into that sort of stuff, sure. Get a P4 and enjoy having the fastest CPU there is .. until the next model P4/AXP is out, that is.

For the rest of us, who base our computer purchases on common sense, for speed, stability and price, the obvious choice is still the Athlon XP.

Besides, the Pentium IV still has a pretty fucked up design. See this page if you don't know what I'm talking about. I always laugh at people who whine that Windows is poorly designed, only to praise Intel CPU's in the next breath.

Anyone care to disagree? Remember, modding me down is so much easier than posting an intelligent reply.

The P4's x87 FPU and x86 ALU are just plain slow compared to P3s and Athlons. Though I am surprised your code is running 82x slower. I'd expect more like 2-8x slower for compute bound code. You can get a somewhat sensationalistic overview of why it's so slow at this link.
If you want more in-depth numbers you can compare appendix C of the Intel Pentium 4 Optimaztion Manual with chapter 29 of Agner Fog's Pentium/II/III Optimization Manual. You can see the Athlon numbers in Appendix F of AMD's Athlon Optimization Manual.
If you want to do number crunching with Pentium 4s your best bet is to use the SSE2 instructions/registers. You should be able to get a noticable speedup by using the Intel C++ compiler and telling it to use SSE2 instructions. If you want to eek out max performance you'll have to use assembly language. Though you can probably get most of the way there using the Intel C++ Compiler's SSE2 intrinsics.
I'm curious as to why your code is so much slower on a P4 than on an Athlon. The best way to find out would be to look at the assembly code that gcc is producing. You can do that by using gcc's -S option. If you'd like send me the C code and the output from -S and I'll see if I see anything obvious.
I'm somewhat paranoid about posting my email address. My paranoia seems to work, as I've received no more than the occasional spam in the last few years. My email address is my slashdot user name at woh.rr.com.

Any place I can look for some doc on that issue ?

Darek Mihocka of emulators.com has written a whole bunch of stuff about the Pentium 4. He has examples of code that performs badly on Pentium 4, although I'm not sure how the most recent versions of the P4 would work on his code samples.

http://www.emulators.com/pentium4.htm

steveha

Hello.. please stop staring at your cock, and please in future read posts before you attack the poster.

Furthermore, your assumption that PPC is automagically more powerful than Intel architectures is a clear indication that you are severiously under-informed.

Note that the poster you responded to never said the PPC was more powerful than intel. They referenced the fact that when apple changed from 68k to PPC hardware, they included an emulator so that legacy apps could be run on PowerPC computers. The emulation he referred to was for third party apps which have yet to be recompiled, not for the ported OS.

All the original poster said was that while it was no big deal to emulate the 68k on the vastly more powerful PPC, emulating a PPC on an x86 would be not so easy, as x86 and PPC are roughly equal. I am not able to see where your rediculous ad hominem attack comes from. They did not even advocate PPC as more powerful than x86.

That being said, it would indeed be extremely difficult to emulate PPC on the x86! This is simply because of the way the chips are designed. The PPC is RISC; it has simple instructions and lots of registers; the x86 is CISC; has few registers and complex instructions. RISC is not necessarily better or worse than CISC, and the x86 is not necessarily better or worse than the PPC. However, it is generally well-known and accepted fact that it is easier to write an emulator that runs on a RISC machine than a CISC one, and it is quite obvious to anyone who is familiar with the emulation scene that the PPC and x86 are good at different things, and one of the things that the PPC really shines at is emulation.

This will become blatantly obvious if you consider that there are multiple, at least three, separately developed programs-- one of which is open source-- which emulate an x86 PC on a PPC Macintosh. There are, however, no extant PPC Macintosh emulators for the x86 PC. None. And it isn't for want of trying; you can see here that there have been a number of macintosh emulators for the PC, just that none of them have done PPC emulation, only 68k. There have been many attempts to emulate the PPC on the x86, it is just that they have all come to nothing-- becuase the architecture of the two machines is simply such that it is relatively easy to emulate x86 on PPC and relatively extremely difficult to emulate PPC on x86.

I suspect i am responding to a troll. I really ought to submit this as AC. Oh well.. ;;winces, hits submit;;

General awareness of Intel's problems started with this article.

There were a series of follow-ups until Intel regained the speed crown. (And now looks to hold it indefinitely.)

General awareness of Intel's problems started with this article.

There were a series of follow-ups until Intel regained the speed crown. (And now looks to hold it indefinitely.)

General awareness of Intel's problems started with this article.

There were a series of follow-ups until Intel regained the speed crown. (And now looks to hold it indefinitely.)

General awareness of Intel's problems started with this article.

There were a series of follow-ups until Intel regained the speed crown. (And now looks to hold it indefinitely.)

Apparently he's decided not to fight any war at all. This was flat out just a remarkably naive and stupid article. Why settle for only keeping coders unaware and docile?

You missed the point. What does "coders fighting a war" mean? Usually it means they rant and rave and send a lot of emails and post hotheaded articles to osopinion.com. This doesn't help anything. Most geeks don't realize it, but their rants often come across as shallow and misguided. I mean, we're talking about people who write long, rambling pieces about AMD vs. Intel (case in point).

If you want to rant, fine, but get a non-techie to proofread your stuff, and try to keep it low key. Make sure you're not just echoing the same sentiments you read on Slashdot or some other geek-news site. Don't write unless you have some good insight, and have some broad experiences that keep you from looking like someone who doesn't get out much (hint, and I mean this in all sincerity: if you're 18, and just want to get free music because you don't have a job, then you're the wrong person to be writing on the subject).

Hm.. Looks more like a paper written by an Intel fanatic by me. :)

Have you read from the beginning? I've been reading his articles year ago and he was fiercing bashing Intel, especially first Pentium 4. I was surprise way back ago when he changed his stand, but with good reasons.

This is just another proof of the author's lack of insight in processor architecture.

I'm not really sure if you've read his previous article explaining why and how Intel attempted to kill off Pentium III. Judging from the speed of your reply you obviously haven't read them all before reaching your conclusion.

Also, Pentium III is NOT equal to P3 you idiot. While you bash people lacking insight in processor architecture, you must do more homework in this area and learn what P3 architecture refers to.(or read the article and learn about P3-P7, but I'm sure the hell you won't)

My first Athlon I've installed worked instantly. The first Athlon a friend installed worked instantly. And it was even one of those super hot Thunderbirds. Cool huh? I'm successfully running an (non-replaced!) Athlon at work. Are we super lucky then?

You intentionally extract part of the text and make the worse comment to shine your point. He is not just simply sitting back and installing a Windows/Linux on his computers to browse webs and do games like you, he's writing emulators on all platforms. What are you? You are, in my opinion, the worse kind of cluebie critics.

"Intel caught up as expected"
Actually, you only need to read the first paragraph to see where the entire article is heading (and he get to show off his bias too):

He has been spending a whole year and a half bashing Intel and Pentium before reaching this conclusion. If you could read(obvious you don't) than copy and paste, you will see he's always been the faithful supporter of AMD until recently.

Rating: -1 Troll

The author switches side with reasons. Unlike you, who can only copy and paste part of the text to make your point.

Hm.. Looks more like a paper written by an Intel fanatic by me. :)

Have you read from the beginning? I've been reading his articles year ago and he was fiercing bashing Intel, especially first Pentium 4. I was surprise way back ago when he changed his stand, but with good reasons.

This is just another proof of the author's lack of insight in processor architecture.

I'm not really sure if you've read his previous article explaining why and how Intel attempted to kill off Pentium III. Judging from the speed of your reply you obviously haven't read them all before reaching your conclusion.

Also, Pentium III is NOT equal to P3 you idiot. While you bash people lacking insight in processor architecture, you must do more homework in this area and learn what P3 architecture refers to.(or read the article and learn about P3-P7, but I'm sure the hell you won't)

My first Athlon I've installed worked instantly. The first Athlon a friend installed worked instantly. And it was even one of those super hot Thunderbirds. Cool huh? I'm successfully running an (non-replaced!) Athlon at work. Are we super lucky then?

You intentionally extract part of the text and make the worse comment to shine your point. He is not just simply sitting back and installing a Windows/Linux on his computers to browse webs and do games like you, he's writing emulators on all platforms. What are you? You are, in my opinion, the worse kind of cluebie critics.

"Intel caught up as expected"
Actually, you only need to read the first paragraph to see where the entire article is heading (and he get to show off his bias too):

He has been spending a whole year and a half bashing Intel and Pentium before reaching this conclusion. If you could read(obvious you don't) than copy and paste, you will see he's always been the faithful supporter of AMD until recently.

Rating: -1 Troll

The author switches side with reasons. Unlike you, who can only copy and paste part of the text to make your point.

Hm.. Looks more like a paper written by an Intel fanatic by me. :)

Have you read from the beginning? I've been reading his articles year ago and he was fiercing bashing Intel, especially first Pentium 4. I was surprise way back ago when he changed his stand, but with good reasons.

This is just another proof of the author's lack of insight in processor architecture.

I'm not really sure if you've read his previous article explaining why and how Intel attempted to kill off Pentium III. Judging from the speed of your reply you obviously haven't read them all before reaching your conclusion.

Also, Pentium III is NOT equal to P3 you idiot. While you bash people lacking insight in processor architecture, you must do more homework in this area and learn what P3 architecture refers to.(or read the article and learn about P3-P7, but I'm sure the hell you won't)

My first Athlon I've installed worked instantly. The first Athlon a friend installed worked instantly. And it was even one of those super hot Thunderbirds. Cool huh? I'm successfully running an (non-replaced!) Athlon at work. Are we super lucky then?

You intentionally extract part of the text and make the worse comment to shine your point. He is not just simply sitting back and installing a Windows/Linux on his computers to browse webs and do games like you, he's writing emulators on all platforms. What are you? You are, in my opinion, the worse kind of cluebie critics.

"Intel caught up as expected"
Actually, you only need to read the first paragraph to see where the entire article is heading (and he get to show off his bias too):

He has been spending a whole year and a half bashing Intel and Pentium before reaching this conclusion. If you could read(obvious you don't) than copy and paste, you will see he's always been the faithful supporter of AMD until recently.

Rating: -1 Troll

The author switches side with reasons. Unlike you, who can only copy and paste part of the text to make your point.

Read here and the author tells you how The Pentium 4 finally takes the speed crown as AMD falls asleep at the wheel

These are really great read.

Read here and the author tells you how The Pentium 4 finally takes the speed crown as AMD falls asleep at the wheel

These are really great read.

Emulating a 68K Amiga or Mac (at like 8Mhz) only takes a 400Mhz processor to perform decently

Second generation emulators (Optimized, dynamic recompilation, all other buzz words, etc) require something in the rage of 10 times more powerful

Nothing at all, but then that's not too surprising considering the company's rather entertaining approach to life, the universe and everything. A google search on Darek Mihocka, the company's owner, is always good for a laugh...

So surely the Athlon's inefficient design, as compared with the G4, is holding it back as well?

Yes. The G4 is by many measures designed far better than the Athlon, P4, P3, or any x86 CPU. Unfortunately, it was not designed to scale particularly well, something that could likely be improved by adding a couple more prefetch stages to the beginning of the pipeline. That would, of course, come at a small IPC cost.

Overall, the G4's design is vastly more elegant and intelligent than any x86 cpu, which has been the case with Motorola vs. x86 since the 68000CPU.

That's not really true. What are your sources? A lot of the design of the P4 looks weird at first, but was designed to allow for more efficient SMT.

Err, actually I did word-o (a superset of typo) in mentioning the x86 decoder as it is the trace cache (the small cache that stores instructions that have already been decoded so that they needn't be decoded again--which is important due to the P4's insanely long pipeline). Here's a reference, though it can be derived from the design and optimization docs. To quote:

Intel put 3 integer ALUs in the core, two of which operate at double the chip speed. So between them, the three ALUs can accept up to 5 micro-ops per clock cycle. But we've already learned that the trace cache can provide at most 3. So one or more integer ALUs sit idle each clock cycle. It is impossible to even feed 4 micro-ops into the two double-speed units. So why did Intel waste transistors to implement a redundant ALU, but then cut corners by eliminating a much more needed second floating point unit? (Sorry for the bold font, wanted to differentiate this quote from a quote of your message)

Huh? Where do you get that? The FSB needs to scale with the memory linearly -- if the memory bandwidth exceeds the FSB bandwidth, the extra bandwidth is wasted as can be seen in Nforce motherboards (dual channel DDR RAM) on the Athlon, whos FSB's bandwidth is identical to single channel DDR at a given clockspeed. (This is assuming that you aren't using the onboard video, in which the extra bandwidth would be utilized and then some, but that doesn't relate) Admittedly I did blip on the increased RDRAM clock though. My mistake.

Yes, the P4 is less "efficient" than the Athlon. Yes, the Athlon is less "efficient" than the G4. Does that make the G4 better than the Athlon? I'm gettign sick of typing these HTML tags all the time. Anyway, as you said, it depends. If you're Transmeta, 'better' is speed per watt, which is perfectly valid as a measurement in some situations. If you're the vast majority of computer users, better is (among other things) the price/performance ratio, which AMD wins by a country mile. If you're a professional in which a 5% speed increase can reduce costs by $10,000(such as with 3D rendering), then speed is all that matters and at the moment, for many applications one would choose the Pentium 4. (actually, they're probably choose an EV6, hundred processor US3, IBM Power4 or the like, but those aren't part of the topic). I originally just wanted to point out the huge theoretical speed gap vs. the real-world speed gap, which is pretty slim. Admittedly, I am a bit biased in that I resent Intel designing the chip around marketing its clockspeed as most people equate that directly with performance, which is a believe reinforced by advertising (which is reinforced by user belief. A vicious circle.)

You mean E-Maculation makes PPC emulators now? Jesus W. Christ! I thought all E-Maculation did was make fun of Emulators, Inc.'s Darek Mihocka, who claims that he's going to debut a PPC emulator at MacWorld. Hath Hell freez-ed over?

The Itanium itself is dog-slow. However the original poster's claim:

Did someone mention "clockspeed is not everything"? It you look at (non x86-based MPP) super-computers, I'd bet none of them runs with CPUS faster than 1 GHz, yet each processor is often an order (or two) of magnitude faster.

Check the floating point numbers - A 1GHz Alpha (EV68) outperforms a 2.2GHz Xeon in both SPECfp and SPECfp Rate:

SPECfp2000:
AlphaServer ES45 Model 68/1000: 960
Dell Precision WorkStation 530 (2.2 GHz Xeon): 802

SPECfp2000 Rate:
AlphaServer ES45 Model 68/1000: 21.1
Dell Precision WorkStation 530 (2.2 GHz Xeon): 13.6

Both of these are quad-processor machines, yet the Alpha's floating-point rate is 155% that of the Xeon at less than half the clock speed.

Clockspeed is irrelevant. Especially when the architecture is flawed.

This is because the P4 is a very peculiar beast that needs many optimisations for the code to run fast. Indeed, when Intel first shipped it they had to ship a specially optimised MPEG decoder for it to appear any faster than the PIII on benchmarks. For more info, check this out:

Compilers are notoriously slow at catching up with the latest processor design, and you can probably expect gcc to catch up with the P4 around the time it's superseded by the 64-bit babies.

This is not to slur gcc - M$'s Visual Studio compiler suite hasn't yet been optimised for the P4 as far as I know (although I expect the .NET version will be) despite the vastly greater resources they have to throw at it...

Take a look at Gemulator 2000. It emulates either a 68000-based Macintosh, Atari ST or Amiga - and it is a free download. There's another company that makes an excellent 68000-based Macintosh emulator for x86 - Ardi. These emulators show what is truly possible with JIT-based emulation. UAE is dreadfully slow by comparison.
The 68000 chip would be a wonderful virtual machine - GCC already targets the chip so you would not need to write a new compiler suite from scratch.
As a side note, I wish people would divorce the notion of "Abstract Virtual Machines = Garbage Collection." An architectural neutral binary distribution format is the most important characteristic of a VM. Garbage collection just slows down the runtime.

The question of porting a non-unix MacOS X application to Linux makes me wonder what the current state of MacOS emulation under Linux is. I see that Basilisk is apparently a GPL'ed 68k Mac emulator under relatively active development, and that the proprietary executor is still available, along with Carbonless Copies from the same company. Also, a couple of others are discussed on emulators.com.

You do pay for quality

Like Microsoft products? I do think that the Pentium III was a very good chip (certainly better than the old AMD k6), but right now, I would take an AMD processor over any Intel. Check out this article for a good read on the subject.

Ok... so, now I apparently have to stop using AMD processors after my athlon 1.4, because I won't be able to determine the true mHz that my processor is running. I don't necessarily see the reasons why this rating is masked on the XP processors... its probably ok for the average home user, but I'm not average. MhZ ratings mean something to me, because I enjoy tweaking the most from my system.

I stopped using Intel processors a while ago, after learning that AMD's chip architecture was superior to Intel's, the choice was obvious. If you haven't read this document, please do. It'll give you a good technical understanding of performance issues with Pentium processors compared to AMD processors.

So, now what? I guess I'm forced into some hard choices over the specs of my next machine. It may be time to consider Intel again... I just don't know. AMD's new CPU scheme sounds really sketchy to me.

One of the recurring themes in the above URL is, "Clock speed is not everything."

This article says it all, really. AMD only have to compete on marketing grounds now. They'll have no problem winning a technical shootout.

There's a few articles out there that say a P3 is faster than a P4 (in some aspects) just because most software hasn't been written in favor of the newer chips."In terms of speed and running existing Windows code, the Pentium 4 is as slow or slower than existing Pentium III and AMD Athlon processors."

There is a good page about wh the Pentium 4 sucks. It's written by an assembly-level programmer, so he know quite a bit about processors.

The Pentium 4 has several glaring faults that cripple it.

the level 1 cache is way too small

it can only pass the decoded micro-ops to 3 of its internal execution units per clock, so it can only execute 3 micro-ops per clock (compare to the Athlon, with up to 9 micro-ops executed per clock)

instructions that execute very quickly on other Pentium chips now execute slowly (in particular, anything involving bit-shifting)

These faults and more are discussed here.

Unlike the Pentium 4, the Athlon executes exisiting x86 code very quickly. You don't need fancy optimization tricks to get code to run fast on an Athlon; it has no major faults to work around.

A Pentium 4 system, with its expensive high-speed RDRAM, will be very fast for certain uses. And it has the lead in raw clock speed. If Intel can crank the clock speed way up, say to double what AMD can do, it won't matter that the Pentium 4 is broken; it will still be the fastest chip you can get. I predict this will not happen; AMD will continue to make ever-faster Athlon chips, which will remain competitive with anything Intel can make. (And of course if you look at the performance-over-price ratio, the AMD chips totally crush the Intel chips.)

Of course, it must be said that the chips are so fast these days that few people will really notice any difference between a good AMD system and a good Intel system. The AMD may out-benchmark the P4, but if both of them can run Quake 3 nice and fast, few people will actually care about the differences.

steveha

Historically, if you took code for one processor and ran it on a later processor, the later processor would always do a better job of running it than the original. (The major, glaring exception to this was the Pentium Pro, which really sucked unless you optimized the code for it.) This is why Linux distributions such as Debian just optimize for the 386 and call it good -- most of the time, for most of the applications, you won't pick up very much performance by optimizing for a specific chip architecture. (By the way, you should rebuild your kernel with chip-specific optimizations. Your kernel is running all the time, and any savings will add up quickly. Of course, all the CPUs are so fast these days that few of us will really notice any difference even with the kernel.)

But now the Pentium4 has so much wrong with it, that unless you rearrange the code specially, it chokes and underperforms. The Level 1 cache is actually a cache for decoded instructions, which is cool... but it is only 8K, which is insane! Sure, since the instructions were already decoded, the 8K cache is probably worth a bit more than a simple 8K instruction cache, but the Athlon has a 64K instruction cache! The Pentium4 has all these internal execution units, but it can only feed three of them per clock cycle from the cache, so most of them will be idle in any given clock cycle. And while earlier chips introduced cool features that would make code run really fast (bit-shifting was really fast, and there were special instructions like CMOVE) these all run dog-slow on the Pentium4.

So, the Pentium4 runs really hot, and needs special cooling and a special power supply. Right now it needs expensive RDRAM. And it needs special optimizations to allow it to run at full speed. Summary: unless you really need its special features, buy an Athlon.

When does a P4 beat an Athlon? Some specific situations where RDRAM is really appropriate, some specific situations where the SSE features really work (and assuming the code is optimized for it), and that's about it.

Can a future P4 dethrone the Athlon? Maybe. Intel claims that the P4 is slower, clock-for-clock, than the Athlon for a good reason: because the P4 will reach really high clock speeds really fast. Some breathless press release I read said something about a 10 GHz version of the P4 within four years or so. Let's face it, the P4 can stay as broken as it is and still stomp the Athlon if Intel can really get the P4 going twice as fast or more than the Athlon! But I'll believe it when I see it. The current P4 goes into thermal overload and slows to half-speed if you work it really hard, and dissipates 73 Watts at 1.5 GHz; even with a die shrink I'll bet a 10 GHz P4 would melt itself into a puddle.

Because the Athlon gets more work done per clock, and is available at clock speeds nearly as high as the P4, the Athlon is better than the P4 across the board. There are a few narrow situations where the P4 is better than the Athlon, but if you check the price/performance ratio the Athlon still wins.

steveha

Can the P4 dethrone the Athlon

No.

Let me explain this way: Pentium III has 6 10-stage pipelines for out-of-order superscaler execution, while Pentium 4(avoid using short-form P4 - Pentium 4 is in P6 family) has 9 20-stage pipelines.

More pipelines more stages sounds good huh? Unfortunately, in some benchmark tests Pentium III beats Pentium 4, it's due to the fact that Pentium will flush the entire pipelines during branch-misprediction/pipelines stall. As a result Pentium III would out-perform Pentium 4 in some occasion, as the latter tends to lose more instructions when branch-misprediction rate is too high.

Althon, on the other hand, only flush 1/2 pipelines on averages. They really need to fix this fundamental design glitch before they could beat Althon.

If you are very interested in this subject you can read this article. You can understand why Intel cannot giveup Pentium III in favour of the market of Pentium 4.
&nbsp_
/. / &nbsp&nbsp |\/| |\/| |\/| / Run, Bill!

And, low low prices won't change the fact that the p4 is the worst engineered chip in 5 (or was it 10?) years by Intel..

see http://www.emulators.com/pentium4.htm

For those who didn't click the link above, let me iterate it here: [why the P4 is a dog]. It's a well-written, accessible technical article that documents the development of x86 generations, and describes why the design choices in the P4 are sub-optimal (and why design choices in the K7 family have been optimal).

It's a pretty factual examination of things. Worth the read -- it might save you from pissing your money away on something that's not any good.


--

By an amazing coincidence, I was just reading this article when this story came up on /.

I hope they arn't doing what I think they're doing, That is: To make a "cool" looking interface for marketing reasons. Instead of practical reasons.

Linux just dosn't have the apps I need at the moment. And Apple is lagging a bit too behide (and too expensive) for me. Which is why I'm stuck with win98 Until someone wakes up.

Which brings me to another issue. Someone mentioned that there might be a Mac OS X for Intel.
I really hope apple go ahead with this, as I think it would be a big help to them. One of the main resons I don't use a Mac instead of my wintel box is 'cause Macs are so damn expensive. Sure, they're high quality. But I'm not that rich. For the price of a G4, at the very lest, I could get a 1gighz Athlon (i hope i'm right there, but of not, u get the drift).
And I'm sure I'm not the only who's been put off a Mac because of the price.

If they do make an intel version. I really hope that thay also have the brains to make it possable to run it from within windows. If that happened... My problems would be solved.

BTW, Emulators.com Have SoftMac, which can emulate up to a Mac Quadra with Mac OS 8.1
The clame that they can get ~50-60% clock speed on a pentium or celeron. And ~80-90% on a Athlon k7 And up to a gig of RAM. And It can run in a window, or full screen.
They also talk about possable linux versions, and a PowerMac emulator.