Anand: "Intel has performance crown in all measurable categories"

[Red Dawn]

<< First I just want to compliment you on your trolling. I mean this not at all in a pejorative sense; it takes a great deal of talent to mix facts and opinion in such a way as to get so many people fired up like this >>

There's no reason for anybody to get upset. Besides PM none of them really have a stake in the Pee 4's Future so if I state that in my opinion that CPU isn't that impressive it's my opinion nothing more.

As far as encoding goes, that's a load of BS. Not the fact that the Pee 4 encodes faster, the fact the encoding DIVX and SVCD is something most people would bother to do. Why, to get a crappy looking Movie that you've already seen? Hell if you are going to steal it you might as well download it of the net and let some other fool do all the work for you.

 

[Texmaster]

<<

<< The P4 is not an impressive CPU. >>

This statement just irritates me - and I'm not that easily irritated by comments like this. I'm tempted to say, "compared to what?" but I guess it doesn't matter. Pretty much all of the major CPU's on the market today impress me - compared to what I used several years ago... or even last year.

My only question is, Red Dawn, are you purposely trying to provoke people? It seems like you are. >>



Dont worry about it pm.

 

[ssanches]

IMHO benchmarking is more an art and less a science. A certain benchmark may portray a particular CPU as "faster", another benchmark may show it otherwise. I think you folks need to read this excellent article taken from MSDN :

http://msdn.microsoft.com/msdnnews/2001/july/perf/perf.asp

The author correctly states "Computer benchmarks are revered by marketers (at least those that make their product look good), and discounted by those who recognize the limitations of benchmarks". The folly of taking synthetic benchmarks at face value has been proved time and time again by many matured people. Also recently ZDNet critiqued PCMark 2002 and proved that it's results should be taken with a grain of salt.

At the moment i feel the P-4 NW has taken the speed crown, and it definitely deserves it. Intel and AMD have been seriously competing since the launch of the K7; sometimes AMD has taken the speed crown and sometimes Intel... This thing is pretty cyclic...

I bought my Athlon XP 1600 in late November 2001. At the time it had the best price/performance one could have. Things apparently have changed since the arrival of the P4 NW and things may change again when the T-bred comes out..This thing is cyclic. I feel folks should buy what is the best AT THE CURRENT MOMENT and forget what may happen / or happened up or down the lane.

 

[dullard]

Red Dawn: "I upgraded my 1.2 Tbird to a P4 1.6a...While it's faster...The P4 is not an impressive CPU?and I do spend most of my time surfing the net, using Word and posting on this fourm"

You could have upgraded to a 10,000 GHz Hammer and been unimpressed with Word and web browsing performance since these don't really need a fast processor for typical usage.

Assuming the Tbird is 15% faster than an equally clocked P4, you made a jump from 1.2GHz*1.15 = 1.4 GHz to 1.6 GHz. That isn't much difference; in fact, the difference should be barely noticeable. You could have upgraded to a 2.4 GHz P4 and saw a much larger difference in speed. However, you instead upgraded to the slowest P4 that Intel currently sells in bulk! You are correct that there isn't much difference especially since you don't do anything that the P4 is good at.

I could argue that I upgraded from a 1.4 GHz Athlon XP to a 1.6 GHz Athlon XP and was unimpressed. Following Red Dawn's argument, I could conclude that the Athlon XP sucks since my upgrade had disappointing improvements. Of course I'd be correct that the speed difference is minor. But my conclusion would be dead wrong.

No one can argue against your results, since they are correct. It is your faulty conclusion that angers people. Until you realize that, this thread will go nowhere.

Note: for those who want to confuse the matter with overclocking, the 2.4 GHz P4 overclocks a lot higher than the 1.6 GHz P4 and Red Dawn never stated if the T-bird was overclocked.

 

[gregor7777]

As far as encoding goes, that's a load of BS. Not the fact that the Pee 4 encodes faster, the fact the encoding DIVX and SVCD is something most people would bother to do. Why, to get a crappy looking Movie that you've already seen? Hell if you are going to steal it you might as well download it of the net and let some other fool do all the work for you.


Now you're just trying to piss everyone off.

You're completely full of it. I use divx as a way to put home movies of my baby daughter to a small file that can be burned and stored permantly. And making backups of movies I have that can be watched on the road without taking a dvd that costs $20.

Crappy? I think not. Not if you're good at it. Mine come out fantastic, and I'm a videophile.

Why would you assume that it's something to be put to illegal use?

BTW, my grandma's P2 300 or whatever the heck it is runs the internet and Word just fine too. She probably wouldn't notice the upgrade to a P4 either, so I think I see where you're coming from now.

 

[WarCon]

Just one little comment and suggestion. If you really want it "look" faster, you might try changing the window animation speed......... Sorry couldn't help myself.

Since my upgrade from my 1ghz@1.4ghz, I have only been able to notice slight differences in how Win98 runs, but some of those differences are really kinda cool. I change forums through the bar and sit there for a sec while it changes and wonder why it isn't going only to notice upon closer examination it has already changed. There is no little blink that I have grown used to. My word opens faster, (but I actually think only barely and that is from faster burst speed on my 845's IDE controller most likely).

I guess basically all I am saying is that most applications have been processor saturated since the 1gig chips came out. Now the only speed increases we are going to really notice are in rendering, ripping, gaming, and benchmarks. What I would like for my next upgrade is a stable well written OS. Hard to compensate for lack of that with faster hardware.

 

[Red Dawn]

I can see that most here haven't much of a sense of Humor. Like I said earlier, it's like I'm insulting their manhood.

Dullard

<< Assuming the Tbird is 15% faster than an equally clocked P4, you made a jump from 1.2GHz*1.15 = 1.4 GHz to 1.6 GHz. That isn't much difference; in fact, the difference should be barely noticeable >>

Wait a second the Pal, I did say I had my Pee4 OC'd to 2.24 ghz (down to 2.13 GHZ now)



<< You could have upgraded to a 10,000 GHz Hammer and been unimpressed with Word and web browsing performance since these don't really need a fast processor for typical usage. >>

OK , according to your logic 99% of those out there are wasting their money buying anything more than a P3 1000. Most people don't spend their time stealing Movies, running RC5 or what ever it is the Dweebs do these days with the extra Clock cycles, playing Kiddie Games and running "Gee Whiz" benchmarks so they can post here about how fast their system is making them l33t among the pocket protector crowd.



<< . It is your faulty conclusion that angers people. >>

How can it be faulty when I conclude that the Pee 4 compared to the XP Mhz per Mhz is a weaker CPU? Yeah it's cooler and you can OC the hell out of it but to make it competive against the XP you have to OC the hell out of it. If you couldn't many of these Bandwagon jumping Fanboys wouldn't be Pee 4 Fanboys. They might be impressionable but they also are tightwads and most of them aren't stupid enough to spend the kind of coin Intel would want for a few extra FPS over the half the cost XP.

gregor7777, you are an exception, not the rule. No amount of money spent to preserve your daughters childhood on video is a waste.

Warcon

<< Just one little comment and suggestion. If you really want it "look" faster, you might try changing the window animation speed......... Sorry couldn't help myself >>

Why apologize, you make a good point.

 

[majewski9]

Yeah yeah the pentium4 2.4ghz northwood is slighty at best faster than Athlon XP 2100+!

It only took a 670mhz lead! Hmm lets see what Barton or Thouroughbred has to offer in the coming months. Lets all remember that the CPU performance crown switchs hands often. AMD has always had the price/performance crown though and that is the only thing I care about.

 

[DeathByDuke]

INTEL DIE! INTEL DIE! INTEL DIE! INTEL DIE!
Damn! Wrong company! VIA DIE! VIA DIE! VIA DIE!
Hmm, still not right - doh! ah yes!
AMD RULEZ JOO! AMD RULEZ JOO! AMD RULEZ JOO!

Got a Duron 700 and sticking with it an me Geforce3 til i see how Unreal2 an Unreal Tourny2 perform then maybe get a XP 2100+

I'm running EVERYTHING with 4XAA and 1280x1024 res with 8tap Anistropic filtering with no drop beneath 35fps on Comanche4 or 120fps on MOH:AA (erm, Ti200 is clocked at 250 core and 550 memory) OK, it faster than it should be Intel lost the plot with P4 long ago - i wish they give up now.

 

[dullard]

"Wait a second the Pal, I did say I had my Pee4 OC'd to 2.24 ghz (down to 2.13 GHZ now)"
Is that with RDRAM, DDR, SDRAM, fast motherboard or slow, different OS or same, more or less memory, did you have the t-bird overclocked, etc. My point is that you don't provide enough information to us. The fact is you bought a processor that is on average about 10% faster than your old t-bird. The fact that you overclocked is meaningless - you still only bought a 10% faster processor when there were much faster ones available.

"OK , according to your logic 99% of those out there are wasting their money buying anything more than a P3 1000. Most people don't ... playing Kiddie Games"
Hmm, did you notice a huge slump in computer sales after the 1.0 GHz mark was reached (the first slump in computer history)? Maybe that is since most people are satisfied with their 1.0 GHz computers... My home computer is 300 MHz and can run all but one program I've ever wanted just fine - so I have no need to upgrade yet. (The one failed game had graphics errors due to my integrated graphics.) I think you will find that the majority of home computers play "kiddie games", not 1% like you claim. These games are a multi-billion dollar industry since many of us play them.

"Yeah it's cooler and you can OC the hell out of it but to make it competive against the XP you have to OC the hell out of it"
You needed to overclock since you bought the slowest P4 that Intel sells in bulk. There are much much faster P4s that don't need overclocking to be competitive. Or with overclocking will blow your 1.6a GHz away.

 

[gregor7777]

gregor7777, you are an exception, not the rule. No amount of money spent to preserve your daughters childhood on video is a waste.

You're completely right.

Actually, that being said, I wouldn't waste my money for an upgrade from my Athlon for a quarter of an hour less encoding time either.

 

[Red Dawn]

OK Dullard, you are right and I am wrong. None of my points are valid and everybody should go out and spend $500.00 to get the top of the line Pee 4 because frankly the extra FPS you get over an AMD 2100+XP is worth the extra $250.00. What was I thinking?

BTW, don't take what I said here personally. Some of you made good points and I hope that those who might have been on the fence about getting a P4 will be able to use what was said here to make an informed purchase.

One thing that PM said that really hit home with me

<< Pretty much all of the major CPU's on the market today impress me - compared to what I used several years ago... or even last year. >>

It didn't seem that long ago to me when I was plugging away with a P233 and a Voodoo1 and I was damn happy to have it. Todays CPU's, Memory, Hard Drives etc. really have come a long way and us arguing over which one is the best is, as my old Granny would say, just chipping teeth.

 

[dullard]

<< OK Dullard, you are right and I am wrong. None of my points are valid and everybody should go out and spend $500.00 to get the top of the line Pee 4 because frankly the extra FPS you get over an AMD 2100+XP is worth the extra $250.00. What was I thinking?

One thing that PM said that really hit home with me "Pretty much all of the major CPU's on the market today impress me - compared to what I used several years ago... or even last year." It didn't seem that long ago to me when I was plugging away with a P233 and a Voodoo1 and I was damn happy to have it. Todays CPU's, Memory, Hard Drives etc. really have come a long way and us arguing over which one is the best is, as my old Granny would say, just chipping teeth. >>



Actually I agree with lots of your points. A 1.6 GHz P4 isn't much faster than a 1.2 GHz T-bird. You are completely right in that respect. I just disagree when you say the P4 is a poor CPU just because it barely edged out your T-bird. If the P4 is a poor CPU and Anandtech shows it winning or tied in all the benchmarks - then the Athlon XP must also be a poor CPU!

Money is a different thing altogether. So far, I hardly saw anyone talking about money in this thread. I also agree with you that the price premium isn't worth buying any top of the line Intel product produced to date. But you didn't argue about price - you just said the P4 is an unimpressive CPU.

I agree with your last paragraph there. BOTH are great CPUs since they both perform nearly identically (or within a few percentage points). For typical usage, there is no reason to get one over the other. If I were to buy a computer today, I don't think I could choose between an Athlon XP 1900+ or a 1.8A P4. In my opinion their price and performance is nearly identical - thus I'm not getting a computer today. A price of full computer system is within 2% when comparing these chips. Overclocked and non-overclocked performance are the same within a few percentage here and there (some favor the Athlon and some favor the P4). Since in my opinion they are about the same, if you say one sucks, then the other must suck as well... I perfer to say they are both great.

 

[pm]

<< It didn't seem that long ago to me when I was plugging away with a P233 and a Voodoo1 and I was damn happy to have it. Todays CPU's, Memory, Hard Drives etc. really have come a long way and us arguing over which one is the best is, as my old Granny would say, just chipping teeth. >>


That's probably why I posted. It's not that you said specifically that the Pentium 4 isn't impressive - if you had used the Athlon, or the G4, or even Transmeta as your example I would have felt the same... although to be honest, I might not have posted a comment.

To say that a CPU that has millions of transistors, occupies the space of a thumbnail, completes billions of transactions per second, runs at a clock frequency of greater than 2GHz, and contains 8x more cache than my first computer had as total memory isn't impressive just irritates me because it seems SO wrong. It's like saying "the athletes at this year's Winter Olympic Games were all pathetic." Perhaps if it were in comparison to something more impressive then it might be defensible, but just to say it seems wrong.

I don't stray into OT, and actually don't read that much in GH, so I was unfamiliar with your style, Red Dawn, and was honestly unsure if you believed everything you said or if some things were tongue-in-cheek/sarcastic. It's hard to tell in posts. That's why I asked.

 

[grant2]

The author correctly states "Computer benchmarks are revered by marketers (at least those that make their product look good), and discounted by those who recognize the limitations of benchmarks".

Can you suggest an alternative method to benchmarks to compare relative or absolute processor performance? Maybe by having a panel of users perform various tasks on platforms and give subjective reviews? Wouldn't that lead to the same problems that benchmarks have? (eg., the person running photoshop will think the macintosh is way faster and the guy playing quake will think the p4 is faster)

It seems to me that benchmarks would still be a much more effective way to evaluate a processor... as long as the benchmarks reflect the intended use of the machine.

 

[Diable]

From Red Dawn:


<< OK , according to your logic 99% of those out there are wasting their money buying anything more than a P3 1000. Most people don't spend their time stealing Movies, running RC5 or what ever it is the Dweebs do these days with the extra Clock cycles, playing Kiddie Games and running "Gee Whiz" benchmarks so they can post here about how fast their system is making them l33t among the pocket protector crowd. >>



If your not doing anything that taxes your cpu like the things you maligned above then you don't need a +1GHz CPU.

 

[Red Dawn]

Diable


<< If your not doing anything that taxes your cpu like the things you maligned above then you don't need a +1GHz CPU. >>

That's true for many of us I believe I use my system for Multitasking for the most part (8 IE windows open at once)

PM



<< I don't stray into OT, and actually don't read that much in GH, so I was unfamiliar with your style, Red Dawn, and was honestly unsure if you believed everything you said or if some things were tongue-in-cheek/sarcastic. It's hard to tell in posts. That's why I asked. >>

Well I don't mind being proven wrong. I guess my main bone of contention is with those who act like this system or that system is like a Super Computer or an Intel or AMD CPU is vastly superior because they aren't, not that anybody who posted in this thread said they were. Almost everyone who disgreed with me in this had a good point even if I don't agree with it.

Right now you really can't go wrong choosing either one and I beleive for the average user a High End P3 or a low end XP would be more than enough processor for them. For those who have special needs such as Encoding DIVX or 3D Modeling then only the fastest will do and there are those here who fit that criteria. For the rest of us it's like souping up our Cars. It isn't necessary but we do it because we enjoy it.

 

[ssanches]

<<
It seems to me that benchmarks would still be a much more effective way to evaluate a processor... as long as the benchmarks reflect the intended use of the machine. >>



Apparently you haven't understood the crux of the statement. Benchmarks are the best way to review performance. Infact the author himself goes on to create a way to benchmark software under different systems. Nobody has ever said to perform a subjective review of different platforms or CPU's!. The problem lies when people take benchmarks as the gospel truth without knowing the limitations of certain benchmarks themselves. I feel the author has expounded that well. Here's another article by Extremetech : Linky

 

[BFG10K]

It only took a 670mhz lead!

This tired old crap is really getting irritating. It's the total performance that matters, not the clockspeed necessary to attain it.

 

[Barrei]

Boys boys can't we all just get along?

 

[Duvie]

I will give you some real data....

encoded divx movie...gone in 60 seconds..."stealing" RED dingDawn can kiss my arse....


With athlon 1.4ghz at 1.5ghz...took 4:10 minutes 2 pass, 2cd rip, audio done separately....

p41.8@2.4ghz did it in 2:36 minutes....no "quarter of an hour here...the movie is 2 hours long and I ran 2 pass and alomost did it real time....

P4 also multitask far better...I can run many more things and not feel the system lag...I lok running a dvd movie, internet, and cadd operations simultaneously...p4 does it with minimal cpu cycles...

My cadd is faster in my test rendering file, though not stunningly...but everyone knows the amdexcels at those fpu intensive apps....

My photo editing and home movie editing is sizeably faster and much more responsive using same harddrives, same amount of ram, and same programs....


Everybody just buy what the f^ck you want and shut-up....

 

[Red Dawn]

Duvie

<< RED dingDawn can kiss my arse >>

You da man Duvie. Your stats seem to prove prove me wrong. Wait, I didn't say that the P4 wasn't faster at 2.2 over a 1.2 Tbird, I just said that it didn't seem like that big of a boost for the piddly everday apps I run.



<< Everybody just buy what the f^ck you want and shut-up.... >>

But if we did that how would those looking to make an informed decision be able to tell if they should spend the bucks for a P4 or stick with their current setup? Guys like you know that the P4 would show a significant boost in productivity for the CPU intensive apps you run where as someone who doesn't run CAD or Encode DIVX might not realize that the 5 bones they shell out won't really give them a $500.00 boost in productivity running Word, Winamp, IE5 etc. They might be better off shelling the money out for a faster drive, more memory, a faster Video Card or a high priced Hooker..

Finally I apologize for using the term stealing. Frankly I don't care what you do (in fact I would hope you didn't pay for "Gone in 60 seconds")

 

[gregor7777]

RoundUp for the kiddie's: (I think we've covered just about everything)

1)Intel and AMD Marketing Statigum
2)Morality of Digital Video Encoding
3)Price/performance Ratio
4)Modern CPU performance in comparison of yesteryear's CPUs
5)Grandmothers (Several Times)
6)Process of coming to conclusions
7)Arse Kissing
8)Benchmarking, and its effect in the real world.
9)Need for today's processing power
10)3 Blind Chimpanzees (can you find it?)
11)Role of video games in today's society

 

[Priit]

<< With athlon 1.4ghz at 1.5ghz...took 4:10 minutes 2 pass, 2cd rip, audio done separately....

p41.8@2.4ghz did it in 2:36 minutes....no "quarter of an hour here...the movie is 2 hours long and I ran 2 pass and alomost did it real time.... >>



OK... my Athlon 1Ghz finishes kernel compilation with 5min16sec, my P4/1.5Ghz does the same with 6min20sec. Athlon puts out 3.56Mkeys/s in RC-5 cracking, 1,5Ghz P4 is capable of 2.1Mkeys/s. Athlon can decode any divx movie under linux with maximum quality, P4 can't, etc. Point is, there are thing that P4 does wery well and there are things that Athlon does very well. Depending of what you do with your computer, you may even see a performance loss with your system when "upgrading" from 1.4Ghz Athlon to 2.2Ghz P4.



<< P4 also multitask far better...I can run many more things and not feel the system lag... >>



Umm, I have exactly opposite feeling... my P4 is noticeably laggier than my Athlon when doing sth under more heavier load (like burning CD-ROM with 24X speed, cracking RC-5, running bunch of web browsers and other smaller crap and playing UT in the same time).

 

[grant2]

Apparently you haven't understood the crux of the statement.

I understood perfectly. My message was seeking superior alternatives to benchmarks, ideally, forms of evaluation that needn't be discounted.

It seems that you, reasonably, mistook my honest question & dialogue as rhetoric.