Tom Hardware's Power Box (AMD thermally cooled 1.6 GHz vs P4 1.5 GHz)

[fkloster]

...my point is, in order for the P4 to best that AMD concoction, it would need to be running @ 2Ghz.

Amd zealot: "Haaa Haaaa (like Nelson) , stupid P4 has to run @ least 400mhz faster to outperform AMD's best, Haaa haaaa, such terrible clock for clock performance"

...that is the entire point about P4's. They must run faster to get good performance. A 20 stage pipeline comes @ a heavy price in which the P4 pays in advance @ higher speeds (i.e. 1800mhz ++) Comparing it to a similarily clocked AMD proc is stupid and proves nothing.

Apple zealot: "Haaa Haaaa, my G4 700mhz motorola absolutely spanks & bludgeons any AMD processor hands down in Photoshop rendering in clock for clock performance. Haaa Haaaa AMD sucks Motorola rules..."

Get it? Tom has put togethor a stealthy clock for clock comparison that is FULLY LOADED to push AMD. What a joke...

I also must call attention to a gripe of mine that remains unsettling. Why doesn't Tom list his Bios for the Asus P4T? Could it be Tom is using a legacy bios that I had which crippled that mainboard substantially?

 

[Dulanic]

You call people zealots yet you where one of the first to turn this into a arguement fkloster! You are one of the biggest zealots on this board, and you seem to feel a need to defend your P4 everytime someone doesn't like it. Tom didn't make this a AMD vs. P4 article.... he made it the best box he could make vs. what you could buy article. Yet there you went getting all mad because he went and made a KICK arse AMD box. But you didn't seem to read that part, you just got all pissy that he compared it to a P4. Well what good would have this dream box article be without comparing it to the best you can buy normally? You are by far the zealot fkloster.

 

[fkloster]

Who said I wasn't a zealot? Who's mad?

 

[Wingznut]

"you where one of the first to turn this into a arguement fkloster"

Actually, the title of this thread, along with the first three posts are what turned this into a AMD vs P4 thing.

 

[NFS4]

<< ...my point is, in order for the P4 to best that AMD concoction, it would need to be running @ 2Ghz.

Amd zealot: &quot;Haaa Haaaa (like Nelson) , stupid P4 has to run @ least 400mhz faster to outperform AMD's best, Haaa haaaa, such terrible clock for clock performance&quot; >>


And all that you stated their is true. It's a damn shame that Intel must be clocked so high to beat AMD. That's what happens when MARKETING is dictating your processor releases.It may take a 2GHz P4 now to get equal performance, but it is gonna get even worse when Palomino gets here and consistently outperforms the P$ and for a much cheaper price.

I don't even see how fkloster could possibly have a leg to stand on concerning the P4. I mean, a 1.5GHz P4 is 2 times the price of the 1.2GHz Athlon...that alone should tell you 99% of the story.


<< ...that is the entire point about P4's. They must run faster to get good performance. A 20 stage pipeline comes @ a heavy price in which the P4 pays in advance @ higher speeds (i.e. 1800mhz ++) Comparing it to a similarily clocked AMD proc is stupid and proves nothing. >>


&quot;They must run faster to get good performance&quot; I can't stop laughing!!!!!!!! You have to buy a $586 P4 1.5GHz chip and then overclock it to make it perform better than a $255 1.2GHz Athlon. How retarded is that?

Give me a break...this discussion should have been over a long time ago, b/c the P4 advocates don't have a peg leg to stand on.

 

[fkloster]

Price/performance: AMD WINS!! I have never denied it. For the love of God: PRICE/PERFORMANCE IS NOT WHAT THIS THREAD IS ABOUT!!!!!

 

[NFS4]

Then what is the thread about? Bashing Tom for making a comparison that he EXPLICITLY explained in his opening? Hmmmmmm...

Anyway, I question the motives of people that buy Pentium 4's anyway, but again it's your money

 

[fkloster]

Its simple NFSF, I wanted memory performance 3 to 4 times faster than anything AMD will come up with in the distant future (i.e. 4.2 GB/s SSE2 optimized memory bandwidth ala Sisoft). And I did it with memory I have owned for 14 months! For the $600 I spent, 1800mhz and 10K+ 3dMarks was well worth it &amp; much better price/performance upgrade than AMD would have been for me. I'm sure I could show Tom a thing or two about tweaking that P4. Starting with a Bios upgrade?

 

[Wingznut]

I'm starting to sound like a broken record... But think back a couple years.

Intel releases the initial Pentium. AMD releases the 5x86-133. AMD seems much faster than the Pentium, but was still running on the 486 platform. Soon thereafter, the P5 takes off. It didn't take long at all for the (once standard) 486 platform to be an antique.

Again, do you expect Intel to stay with the Status Quo and only speed up the current architecture, bit by bit? How fast do you think it can go? What happens when they hit the wall (which isn't too far off)?

486's hit the wall. Intel designed something new.

 

[MichaelD]

FLAME ON! POOOOOFFFFFFF!!! ....sniffle.....can't we all just....sniffle....git' along?.....

 

[NFS4]

<< I wanted memory performance 3 to 4 times faster than anything AMD will come up with in the distant future >>


Besides SiSoft Sandra and Quake III 640x480, where does that increased memory bandwidth help you out?? I have yet to see any real world impact. Can you tell me what you do that makes memory bandwidth that important?

I can see where it made sense for you to upgrade fkloster, but you are not everyone else here. You had connections

 

[fkloster]

<< ...You had connections >>



...not connections, subsidizations (sp)

Understand this too gang, there is more than one game base on the Quake 3 engine no?

1) Alice
2) F.A.K.K.2
3) Startrek Elite Force
4) Klingon Acadamy
5) Team Arena
...damn it, I know there is more than that but whatever. My D3D performance is much higher than what Tom is getting. Its o.k for a guy like me to be biased but he runs a freakin' hardware review site that influences lots of people...

 

[PlatinumGold]

All right all you TOM haters, what is so wrong about seeing how fast you can get the t-bird to go?? comon give the guy a break, so what if he hates americans.

To all you intel lovers, comon how can you say that Tom is biased (of course he is) but this particular article should be seen as a compliment to intel not the other way around. Tom had to use an additional $400+ US dollars to put T-Bird in the same class as the PIV 1.5. Ghz. how is that insulting??

 

[sharkeeper]

You act like you're getting paid for higher SISOFT memory scores and 3D Mark Scores. (which in the real world mean NOTHING!)

Talk about being brainwashed!

How many RC5 Keys can your Pee Fore crack? MHz for MHz, the Pee Fore looses bigtime.

Here is a good read regarding the Pentium Fore:

How Intel Blew It!

Cheers!

 

[NFS4]

You still didn't answer my question

 

[MadRat]

The 486 became obsolete due to the lack of an FPU, which newer software was dependent upon, not due to its inability to do most normal work. The AMD 5x86, Cyrix 586, and Nexgen 6x86 were ALL very good at what they did. Intel pushed for the fpu to be a requirement which made these competing designs obsolete. People were coaxed into believing the fpu was necessary when most of the software they did use didn't need pentium hardware to run. Nobody could even tell you what an fpu was back then.

 

[fkloster]

Actually NFS4, I'm still waiting for that question to be answered by the likes of many who thought AMD's DDR memory bandwidth prior to its release would be the cats meow. Turned out to be the biggest flop in recent memory but know one talks about it now. To answer your question, I take pride in knowing that when optimized code is present, the P4 will scale at a much steeper rate in higher performance than any other cpu present period. We proved that (remnant2 &amp; pm &amp; I) using Povray 3d pawn test. The P4 I am using scored 70 seconds. U no think people will ever optimize code for P4?

 

[Wingznut]

Shark, even the AMD fans acknowledge that the article is horribly inaccurate. But that's been discussed in previous threads.

Also, RC5 is optimized for 6x86 architecture. When RC5 gets optimized for the Pentium4's pipeline... Well, you'll see.

MadRat, and just how advanced would software be today if cpu's still didn't utilize a FPU? Intel didn't push the Pentium to make other platforms obsolete. But they did it to advance the PC. Not too dissimilar than today.

People then didn't need a FPU. People today don't necessarily need a long pipeline (NOTE: An oversimplification for the example). Right?

 

[PlatinumGold]

MadRat

Am I the only crazy one on this board? the 486 did have an FPU. Wasn't that the main difference between the 486 and the 386?

 

[NFS4]

<< Actually NFS4, I'm still waiting for that question to be answered by the likes of many who thought AMD's DDR memory bandwidth prior to its release would be the cats meow. Turned out to be the biggest flop in recent memory but know one talks about it now. >>


I don't even use DDR, I use PC133...works just fine


<< To answer your question, I take pride in knowing that when optimized code is present, the P4 will scale at a much steeper rate in higher performance than any other cpu present period. We proved that (remnant2 &amp; pm &amp; I) using Povray 3d pawn test. The P4 I am using scored 70 seconds. U no think people will ever optimize code for P4? >>


And that translates into the real world how????...................

You keep saying future this, future that...future optimizations yadda yadda. The &quot;future&quot; moves very quickly and by the time your &quot;goods&quot; get here (SSE2), the new P4 core will be here and AMD will have a new core as well.

The P4 is just a pawn in the whole scheme of things.

 

[Shagga]

Albeit I understand fkloster's point, I believe that this is just a PR exploitation exercise. Don't get me wrong I really like Intel. I strayed to Cyrix once, never again! Anyway, what I'm trying to say is that Intel have released a P4 that could go to 2GHz under current architecture. They will more than likely be first get to the 2 Gig barrier. Now, when Jo public see's a 2GHz P4 and the closest contender being a 1.6GHz AMD they will automatically think the P4 is faster! Do you not think?

It's a clever game if you ask me. Once the new Intel core is released, perhaps then we'll see better performance clock for clock when compared to the AMD offering and also better memory bandwidth. It seems to me in a nutshell that Intel have rushed this P4 into circulation, to pretend they are still in front. This might be their way of buying time.

 

[NFS4]

It's PURELY PR. That's all it is right now, just a fat 'ol smoke screen for gullible buyers who think more MHz = more performance.

 

[Adul]

<< and as a month of close study of the chip reveals, the Pentium 4 does not live up to speed claims, loses to months old AMD Athlon processors, and lacks some of the crucial features originally designed into the Pentium 4 spec. >>



You can blame intel's management for that one, has well has AMD. Competition prob forcd intel to move it up and get it out the door. I definitely think intel has to rev their P4 core.

 

[sharkeeper]

<< , I take pride in knowing that when optimized code is present >>



By the time that code IS present, your CURRENT system will be a doorstop!

I *used* to be pro Intel. I just can't see throwing money away like that--anymore.

This is a subject (like SCSI vs. IDE) that you can talk about until your weiner looks like a prune and falls off and nobody will settle on it.

Cheers!

 

[Adul]

486DX had a FPU on it, the 486SX had a disable fpu.


Fkloster, the athlons/durons still hada better clock for clock performance then the P3. Even with optimized code.