Over that there a good point.Intel have major probleme like AMD have many for many decate.Intel have lose is crown and reputation of ultimate cpu.AMD shyne like gold.The question is does intel can take the pressure.Personnaly yes if craig barret can just let R&D make is job.Intel as allwyse offer good produce.Some mistake PPro P3 1.13 and others that can happen.If northwood and tuatulin fail there will be some change in Intel.I think they have learn something 1 do not try to fold us 2 do not rush a release.In TU/NW there have delay du to small probleme.AMD must continiue is work a trillion compagny will not die easy.For now P4 is not slower that t-bird last bench prove it.PALAMINO is a transition cpu to (AMD 0.13 micron).Willimete was a beta test still he work well not as good as they want or expected.Intel is in shape change stage.
That is the problem of long pipeline. A long pipeline is good at pumping up clock speed, but it is inherently inefficient. If software can be completely rewritten to suit the long pipeline architecture, the problem may be alleviated. But, that is a BIG if.
This is a classic example of Intel trading practicality (performance) for marketing (hype of clock speed). The irony is that Intel does not have the exclusive right to the long pipeline architecture. So, if and when software ever become friendly to the long pipeline, there is no preventing AMD or other companies to use a similar architecture. The same is true for SSE/SSE2 as well. That is why I never believe the spin -- just wait till software companies...
**Spin all you want, but we the paying consumers will have the final word**
I'm not an intel or amd fanatic (don't know about the author), but I have to say that was a hell of a cpu history lesson and explanation of cpu architecture. I learned a <i>lot</i>! Thanks for posting that link. I always wondered where I could find that kind of info, as most sites that I've found are either for those who know absolutely nothing about computers, or for real experts. I'm kinda in between, so that was real helpful.
btw-Did I say thanks?!
My brain has performed an illegal operation and will be shut down<P ID="edit"><FONT SIZE=-1><EM>Edited by lamer_gamer on 06/20/01 02:23 AM.</EM></FONT></P>
Intel CPUs can never beat AMD for now and that's it!
Do you even know how to read a benchmark?
Pentium4 can only beat Athlon on a few programs only and Pentium4 suck at others ALL!
<font color=green>THG's reviews are the key to become a computer hardware pro!!!</font color=green>
<font color=blue>"Willimete was a beta test still he work well not as good as they want or expected."</font color=blue>
I try to stay out of threads like this, but Juin, I cannot believe the things you say!
Willamette a beta test? Why are they selling hundreds of thousands (millions?) of them? How could you possibly ever buy another product from a company that is selling you a "beta test" product that doesn't work as advertised, especially at the introductory price of $700, when the competition was selling a comparable product for $200!
I would certainly look forward to being ripped off in the future by even more "beta products". How many beta products does a company have to sell in order to figure out that the product sucks? Over a million? Where in the Intel advertisments does it say "help us beta this new processor, and pay our development costs as well"? We would all jump at that, wouldn't we?
The P4 is no beta test. It is simply a very poor decision by a company trying to get back the processor speed crown by pushing a product to market too soon.
Juin, I know English is not your first language and will never fault you for that. But your thought process......
<font color=blue>This is a Forum, not a playground. Treat it with Respect.</font color=blue>
oh, I can read benchmark tests, there is a AMD computer in the mail with my name on it. I have done my research. I just thought I would share it with someone else
Who ever can make the best chip thats what I buy, Amd or Intel.
<P ID="edit"><FONT SIZE=-1><EM>Edited by Joeyman101 on 06/20/01 10:40 AM.</EM></FONT></P>
expand the decoder to handle 3 instructions at once. This places a burden on the programmer because instructions now have to be grouped in sets of 3, not just in pairs. Potential 50% speed increase if the code is written properly.
Talking about AMD:
Quote :
in the AMD Athlon, add additional floating point units to allow up to 3 floating point instructions to execute at once. Big speed win for the Athlon, allowing it to trounce the Intel chips on 3-D and math intensive tasks
He didn't sa;y that code had to be optimized correctly for the Athlon. It might seem like a small difference (and it is), but it says a lot about the author. And simply the fact that the whole thing is about Intel's mistakes in the P4.
BTW, I've got a T-bird.
------------------------------
Apple? Macintosh? What are these strange words you speak?
this is an old link.
I'd noticed his bias as well, but he does have a lot of good information in there.
hell, just look at the design of the processors, compare it to engines, we have the V8(Athlon core) that makes lots of power across the board, it just takes more to rev it up, and we have this nice import V6(or something, P4)that can make the same power, has to rev a lot higher to get it.
you can make the same comparison, torque x rpm vs ipc x clock.
the Athlon can get more done per clock cycle due to an inherently stronger system, the P4 can get the same work done in the same amount of time, but at higher clock cycle. there are some things that require all that torque, which the current P4 just doesn't have.
300 hp V8 vs 300hp V6, which do you want to use to pull a heavy load?
----------------------
Independant thought is good.
It won't hurt for long.
Um... FatBurger.. you say he's biased against Intel, but didn't you READ his 25 page essay? I did... and he provided hard cold reason for every single one of his complaints against the P4. His argument, simply put, was that the P4 sucks, not that Intel sucks. If you had noticed, he spent 25 pages saying WHY the P4 sucks, and it is one of the most comprehensive arguments against the P4 that I've run across to date.
He gave a tremendous ammount of credit to Intel for previous innovations, as well as discrediting them for some of their worst stunts, such as the celly with no on die cache and of course the current P4 disaster. You'll notice he also gave due credit/discredit to AMD in their line of processors and REASONS for each(such as the inferiority of the K6 and clone processors)
I can't believe you'd shrug off his essay simply by saying "Oh, he's obviously biased against Intel", especially when you seem to miss his entire of point of WHY people should be biased against the P4. Regarding your quotation of his remarks about the decoder, you seem to have failed to notice that he explains why this is a burden and how it could have been avoided, as in AMD's solution.
Cries of "bias, bias!" are completely misdirected when trying to discredit this guy. Maybe when you can put together a similar essay of why his arguments are worthless, and include the kind of documentation and exhaustive testing that he has put into HIS article, I'll listen to your rebuttal.
Bummer for us Intel people since he does not want to update his code for the P4. I guess we will have to miss out or run those elite emulated atari apps at only 190FPS instead of 191FPS.
Damn!! I never should have thrown my atari 800 away. I could have saved a ton of money if I wanted to play space mines or donkey king!
As for all those expensive atari apps we spent millions on development, I guess we will have to BUY a AMD based machine to get the most performance. I sure hate to wait 1.2 seconds for the emulator itself to open on my P4. That damn AMD can open in in 1.19 seconds.
You apparently missed the point of the article. The guy was describing why the P4 does so badly in everyday code, not just emulators. The word "emulator" occurs maybe twice in the entire 25 pages, and once in the domain name of the URL.
<HUMOR CLASS="sarcasm">Sounds like the guy just won't shut up about emulators!</HUMOR>
Oh, and Adobe <i>did</i> recompile their code for the P4. The P4 still got whipped in Photoshop.
Kelledin
"/join #hackerz. See the Web. DoS interesting people."
