Dr. Pabst says the P4 is finally faster?

[fkloster]

What is the world coming to...has the Dr. finally become fatigued?



<< ...As you will see in the benchmarks however, at 2 GHz Pentium 4 finally manages to beat AMD's fastest processor, the Athlon 1400, in virtually any application... >>



Check it here.



<< I wish AMD would take an example and come up with a better solution for their processors... >>


(Tom on Intel cooling...)

Holy smoke!

 

[CZroe]

It's true. Like Intel said, in the end a higher clock speed will be able to beat the optimizations it breaks. It's the same reason a P4 beats a Coppermine PIII 500. Despite the fact that the 500 is more optimized, the P4 threw out all those optimizations to achieve a higher clock speed & made new ways to optimize so as to gain more efficiency in the future so that you can realize a "true" relative 2Ghz performance.

Now they just have to beat AMDs 1.5Ghz pricing w/ the 2.0Ghz P4 NOT likley! AMD still has a crown for sure!

 

[fkloster]

Why is it Toms benchmarks depict 2ghz P4's own in benchmarks when Annands seem to show a lackluster performance edge?



<< Now they just have to beat AMDs 1.5Ghz pricing w/ the 2.0Ghz P4 >>



What is the pricing on AMD 1.5's today?

 

[RGN]

If you look at Toms P4 numbers, they are very close to Anands. It's Toms AMD numbers that are lackluster.


But I don't care, I'm<U> NOT</U> buying AMD anytime soon. I'm actually looking at the i845 as a <U>short</U> term cheap upgrade.


/me backs away from the firestarta's

 

[fkloster]

<< The processor intensive NV15-demo used to give Athlon a better chance against Pentium 4, but at 2 GHz Pentium 4 cannot be beat. >>





There is a doctor in the house



<< Intel's new Pentium 4 at 2 GHz however is beating AMD's fastest Athlon in the majority of frequently used applications and deserves therefore the title 'fastest PC processor' >>

 

[Insane3D]

"Why is it Toms benchmarks depict 2ghz P4's own in benchmarks when Annands seem to show a lackluster performance edge? "

I think they used different mobo's. Tom's used the Asus P4T I think and Anand used the Abit TH7-Raid. Also, Anand used the ECS board with the Athlon and Tom used the slower MSI 6341 motherboard. The Asus is obviously faster for the P4 and the ECS (Sis735) is obviously faster for the Athlon...it's still a dead heat in my eyes. Not that impressive in my opinion that the P4 needs 600mhz clockspeed advantage, along with faster memory and a faster FSB to just break even with a 1.4ghz Athlon that costs over five times less. It's still damn fast though...

 

[Insane3D]

"But I don't care, I'm NOT buying AMD anytime soon. "

That's cool...more for me.

 

[Adul]

<< If you look at Toms P4 numbers, they are very close to Anands. It's Toms AMD numbers that are lackluster. But I don't care, I'm NOT buying AMD anytime soon. I'm actually looking at the i845 as a [<U>U]short[/u]</U> term cheap upgrade. /me backs away from the firestarta's >>



So you are going to buy a P4 and castrate its performance buy using sdram? Are you mad man?

 

[Insane3D]

If the differences in test can be achieved with simply changing mobo's, then that shows that the results are still so close that most users will not notice the difference.

 

[OneOfTheseDays]

yeah no sh*t a fscking 2000 mhz processor would beat a 1400mhz processor. That's a 600mhz gap between the two. I don't care too much about performance if the price is as ridiculous as the 2ghz pentiums are gonna cost.

 

[Athlon4all]

I posted something about this here. I disagree that the P4 results are similar. Look at SYSMark 2001 OFfice Apps, Tom gets a score of 167, Anand gets a score of 151. Both running in Win2k, Tom used a GF2 Ultra, Anand a GF3, and while Tom used Detonator4, that wouldn;t make too big of a difference (And make certainly no difference in Office Apps), because he's using an Ultra.

Now next big difference I saw was in AquaMark(at 640x480x16), accroding to Tom he got 29.2 fps but still beating the Athlon 1.4, but Anand he got 42.4 but was slower than the MP 1.2 and A 1.4. Tom here did use 98SE, while Anand used 2K, but would that seriously make this big of a difference? The odder thing is that according to Tom, AquaMark is P4 Optimized, but Anand said


<< ?in spite of the aforementioned conclusions, AquaMark does not seem to favor the Pentium 4 at all. The Athlon MP even comes out on top of the Pentium 4 2.0. >>

i don't understand I guess why Anand gets these results.

Finally and this is really the shocker is in iDCT Flask. According to Tom "The heavily P4-optimized iDCT of FlasK has always ensured that Athlon gets badly beaten by Pentium 4", but now Anand says quite the contradictory "We used the MMX iDCT algorithm for all of the platforms (on the Pentium 4 the SSE2 iDCT algorithm did not perform any better). " How is that explained? Tom says Flask is P4 optimized, Anand says that it may be but performance is no better. Now I admit that Tom's test was a res of 720x576, and Anand's was 352x288, but seriously would a resolution increase do this much to the standings.

The lower Athlon numbers for Tom are easily explained. Tom used AMD 760 based MSI K7 Master, Anand used SiS 735 based K7S5A. at most, Anand's Athlon numbers were 10% better than Tom's, that's right in with the difference between 760 and SiS 735.

The last point that I'm gonna make which just adds to the confusion. Tom used Intel's own 850 Reference board, Anand used TH7II-RAID. Now One could say well the motherboards are different, that's why the P4 numbers are different, but realize what that is saying, that would be saying that an Intel mobo beat a Enthusiast aimed Abit TH7II-RAID mobo. Does that make sense? I don't think so.

So I think that there's something fishy going on. I'm not sure what but I think somebody should ask Tom or Anand himself about it.

EDIT: Insane I thought that at first (Tom using Asus P4T), but it's a Socket 423 mobo. He most likely had a S478 CPu, plus he showed us Intel's reference board so it makes sense that he used Intel's reference board for the 2Ghz, but the P4T for the other's. That as I said doesn't make any sense that Intel's reference board supposedly beat TH7II-RAID.

 

[Guilty]

The stupidity of Intel vs. AMD never ceases to amaze me. And I'm surprised how immature fkloster is acting, like an NFL player dancing in the inzone after scoring a 1 yard rush for a touchdown. Lets get something straight, AMD and Intel go back and forth, it's always been that way. Sometime in the next few weeks the 1.53GHz Palomino will be released, then we get to see AMD zealots dance around in the inzone until the 2.2GHz Northwood is out. NEITHER PROCESSOR IS A BREAK AWAY. The only great difference in this "war" is price.

 

[Insane3D]

Athlon4all -

You're right about the reference board. It seems Anand used a faster Athlon board (K7S5A) and a slower P4 board (TH7-RAID), while Tom used a faster P4 board (Intel Reference) and a slower Athlon board (MSI K7Master).

 

[Adul]

<< The stupidity of Intel vs. AMD never ceases to amaze me. And I'm surprised how immature fkloster is acting, like an NFL player dancing in the inzone after scoring a 1 yard rush for a touchdown. Lets get something straight, AMD and Intel go back and forth, it's always been that way. Sometime in the next few weeks the 1.53GHz Palomino will be released, then we get to see AMD zealots dance around in the inzone until the 2.2GHz Northwood is out. NEITHER PROCESSOR IS A BREAK AWAY. The only great difference in this "war" is price. >>




Well said.

 

[Athlon4all]

<< faster Athlon board (K7S5A) and a slower P4 board (TH7-RAID), while Tom used a faster P4 board (Intel Reference) and a slower Athlon board (MSI K7Master). >>


Yes, but it just seems odd that a Intel reference board was faster than TH7II-RAID. I could be wrong it just seems odd that an Intel beat an Abit board.

 

[Insane3D]

"<< The stupidity of Intel vs. AMD never ceases to amaze me. And I'm surprised how immature fkloster is acting, like an NFL player dancing in the inzone after scoring a 1 yard rush for a touchdown. Lets get something straight, AMD and Intel go back and forth, it's always been that way. Sometime in the next few weeks the 1.53GHz Palomino will be released, then we get to see AMD zealots dance around in the inzone until the 2.2GHz Northwood is out. NEITHER PROCESSOR IS A BREAK AWAY. The only great difference in this "war" is price. >>




Well said. "

Ditto.

Both CPU's are uber fast and side by side, there would probably be almost no difference in performance. The only difference is really price.

 

[Insane3D]

"Yes, but it just seems odd that a Intel reference board was faster than TH7II-RAID. I could be wrong it just seems odd that an Intel beat an Abit board. "

It doesn't surprise me that much...Abit has had their problems recently. Their KG7 board is reportedly having AGP slot issues fairly commonly and it's just been released, and they supposedly are just coming to market with a AMD761 board now so they could "take their time and get it right".

 

[NFS4]

<< yeah no sh*t a fscking 2000 mhz processor would beat a 1400mhz processor. That's a 600mhz gap between the two. I don't care too much about performance if the price is as ridiculous as the 2ghz pentiums are gonna cost. >>


LOL, thanks for the laugh. I'm surprised that fkloster is bragging b/c a 600MHz difference FINALLY allows the Pentium 4 to pull ahead in SOME benchmarks other than Quake 3 (at least by Anand's numbers).

Now if we look at the same situation with Tom's scores, I dunno. The price still doesn't justify the cost difference.

SOOOOOOOOOOOOOO, if Tom used a slow Athlon board and a fast Pentium 4 board, while Anand used a fast Athlon board and slow Pentium 4 board, then the numbers averaged out between the two should make the performance difference the two even smaller.

I say that you all need to quit this mine is better crap. As for fkloster, I commend you on your NOW justified victory (in some but not all benches) with a 600MHz clock advantage:disgust:

But the real winner here for "US" is AMD. $107 for a 1.4GHz Athlon vs $562 for a 2.0GHz Pentium 4 is all you need to say to me. You must be very screwed up on the head to choose a 2.0GHz P4 over a 1.4GHz Athlon.

 

[Athlon4all]

<< It doesn't surprise me that much...Abit has had their problems recently. Their KG7 board is reportedly having AGP slot issues fairly commonly and it's just been released, and they supposedly are just coming to market with a AMD761 board now so they could "take their time and get it right". >>

Yeah no argument about Abit. I'm just used to seeing Intel boards perform slower than big name mobo maker's boards like Asus Abit, MSI etc.


<< But the real winner her for US is AMD. $107 for a 1.4GHz vs $562 for a 2.0GHz Pentium 4 is all you need to say to me. You must be very screwed up on the head to choose a 2.0GHz P4 over a 1.4GHz Athlon. >>

Yeah, the bottom line is that the difference is small enough between the 2 to still AMD be the better option.

 

[NesuD]

<< "<< The stupidity of Intel vs. AMD never ceases to amaze me. And I'm surprised how immature fkloster is acting, like an NFL player dancing in the inzone after scoring a 1 yard rush for a touchdown. Lets get something straight, AMD and Intel go back and forth, it's always been that way. Sometime in the next few weeks the 1.53GHz Palomino will be released, then we get to see AMD zealots dance around in the inzone until the 2.2GHz Northwood is out. NEITHER PROCESSOR IS A BREAK AWAY. The only great difference in this "war" is price. >> >>



Direct hit. It comes down to pay less or more for the same performance for all practical purposes.

Another positve side to the cpu wars. Yuo now have the freedom to spend your money foolishly or wisely as you see fit without taking a performance penalty or gaining a significant performance advantage. God I love this country.

 

[Insane3D]

2 x 1.4ghz Thunderbird - $214
2 x ECS K7S5A - $126
4 x 256mb Crucial PC2100 - $148

Total - $488

1 2.0ghz P4 - $562

I could build two 1.4ghz systems with 512mb of DDR for $74 less than it would cost for 1 P4 2.0ghz CPU alone....that is incredible to me.

 

[fkloster]

<< And I'm surprised how immature fkloster is acting >>



I guess the reason I'm celebrating in confusion is Tom normally will slant most tests toward AMD because he is pissed @ Intel. This test seems to be unfairly waited in Intel's favor.

 

[NesuD]

toms broker must of just bought a good deal on Intel stock.

 

[NoReMoRsE]

<< The stupidity of Intel vs. AMD never ceases to amaze me. And I'm surprised how immature fkloster is acting, like an NFL player dancing in the inzone after scoring a 1 yard rush for a touchdown. Lets get something straight, AMD and Intel go back and forth, it's always been that way. Sometime in the next few weeks the 1.53GHz Palomino will be released, then we get to see AMD zealots dance around in the inzone until the 2.2GHz Northwood is out. NEITHER PROCESSOR IS A BREAK AWAY. The only great difference in this "war" is price. >>

Well said.

Exactly. One minute nForce will reign, then Northwood, then Crush, then.....

It doesn't matter if you get 120FPS in one and 125FPS in the other, it won't make a difference in the real world! Just buy whichever processor you want!

 

[NFS4]

OMG, ROTFLMFAO!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

http://forums.anandtech.com/messageview.cfm?catid=38&threadid=546253