AMD Introduces World?s Highest Performing Desktop PC Processor, the AMD Athlon? XP Processor 3200+

[oldfart]

Originally posted by: Smilin

Don't pop a vein mr intel guy. AMD won't hold the crown for long. It's not THAT much faster. Besides if they used more modern benchmarks you'd probably gripe about them being unfair as well. I'm kinda interested though: Please name **6** chipsets for the p4 that are better than the i845g.

I'll give it a try.

850E
845PE
E7205
865P
865PE
875P

 

[Acanthus]

Originally posted by: oldfart

Originally posted by: Smilin

Don't pop a vein mr intel guy. AMD won't hold the crown for long. It's not THAT much faster. Besides if they used more modern benchmarks you'd probably gripe about them being unfair as well. I'm kinda interested though: Please name **6** chipsets for the p4 that are better than the i845g.

I'll give it a try.

850E
845PE
E7205
865P
865PE
875P

SiS 655
SiS 655DX
VIA P4X400

 

[oldfart]

One of those I would not agree with.

 

[Acanthus]

P4X400 < I845G?

last i knew they were close.

 

[MikeMike]

Read this

looks like the end of the athlon XP

 

[Pariah]

It's a new/higher graded processor release in this period and shouldn't it make more sense to see how it performs now with today's software?. Consider XP's rating are suppose to show an accurate, up-to-date representation of performance as of now (ie, 'World?s Highest Performing' at this moment, until something faster comes along)

The whole concept of a ratings system is flawed, and this is one of the reasons. AMD can't change the application suite because it might (likely will actually) change the ratings of the CPU's. According to THG, the 3200+ rates closer to a 2800+ than 3200+. Ignoring whether or not you agree with that, lets say AMD changes their suite and they find that to be the case. Now they give the chip a 2800+ rating. Wait a minute though, how can a 2200MHz chip with a 200MHz bus have a lower rating than a 2166MHz with 166MHz bus and identical cache and is obviously slower? The rating of the 3000+ would have to change. You can't sell the current 2800+ at that rating since the 3200+ is a decent bit faster and rated at 2800+. See the problem? AMD would have to retroactively change the ratings of all current selling chips to avoid complete confusion. In essence, AMD has screwed themselves with the ratings systems that was supposed to help them.

Considering AMD is likely to use this PR system for some time, dare I say, it's misleading customers by not reflecting today's performance with today's software.

The rating system was misleading from day one. It's too vague and simplistic to ever be accurate.

 

[AndyHui]

Hhmm....why did AMD run their benchmarks with HyperThreading disabled on the Pentium 4?

 

[Confused]

Originally posted by: AndyHui
Hhmm....why did AMD run their benchmarks with HyperThreading disabled on the Pentium 4?

Good catch, Andy!

My guess is that they would say that it's making it a "dual CPU" type machine, which you cannot compare a single to dual-cpu system.

Of course, it could be that they know that their crappy PR Rating will not hold up against the P4 with HT enabled, so disabled it on the tests!


Confused

 

[Whitedog]

The 3000+ should be a 2900+
The 3200+ should be a 3000+

nuff said...

Bad Bad AMD :brokenheart:


Please change the title of this thread... it's in no way correct. NO WAY.

*note* I am an AMD CPU user since the t-bird came out.

 

[SteelCityFan]

Originally posted by: AndyHui
Hhmm....why did AMD run their benchmarks with HyperThreading disabled on the Pentium 4?

Because they wanted to put their best up against a crippled P4 HT CPU... they figured most people would just see the pretty graphs. IMO, this is a new Low for AMD. Thats like disabling the cache on one of the CPU's and then running the benchmark tests.

 

[SteelCityFan]

Originally posted by: Whitedog
The 3000+ should be a 2900+
The 3200+ should be a 3000+

nuff said...

Bad Bad AMD :brokenheart:


Please change the title of this thread... it's in no way correct. NO WAY.

*note* I am an AMD CPU user since the t-bird came out.

3200+ should not even be a 3000+. From looking at the tomshardware review, they didn't take a single benchmark and in many cases were beaten pretty badly. Tom is right... they should have called it a 2800+.

 

[oldfart]

Originally posted by: AndyHui
Hhmm....why did AMD run their benchmarks with HyperThreading disabled on the Pentium 4?

At least they used an 875 mobo for the testing instead of a 845G like F.S.

I do also notice that they use a SBLive on the Intel platform and the nforce MCP for the AMD platform. They also use driver
"Sound Card: Microsoft, v5.1.2535.0, 7/1/2001" Old plug and play Windows drivers?

I don't know the details of how they benched, but when benching video games, the sound card should be disabled. Different sound cards can have a large impact on FPS benches. You are usually testing for video card or CPU performance, not FPS impact of a soundcard. The SBLive is not a very efficient card compared to the nforce. They should have used same sound card/drivers in both platforms, or had sound disabled.

And I think HT should have been enabled.

 

[dexvx]

Originally posted by: Acanthus

Originally posted by: oldfart

Originally posted by: Smilin

Don't pop a vein mr intel guy. AMD won't hold the crown for long. It's not THAT much faster. Besides if they used more modern benchmarks you'd probably gripe about them being unfair as well. I'm kinda interested though: Please name **6** chipsets for the p4 that are better than the i845g.

I'll give it a try.

850E
845PE
E7205
865P
865PE
875P

SiS 655
SiS 655DX
VIA P4X400

845GE
850

Actually lets try this in reverse mode. The only chipsets slower than the 845G are the 845E, 845D, 845 (SDRam).

 

[oldfart]

Actually, I'd take an 845D/E over an 845G.

 

[DaveSimmons]

I remember the "good old days" when AMD PR ratings were conservative and an Athlon at a given rating would soundly beat the P4 at that clockspeed. It was sad to see the rating becoming more and more of a marketing lie after the P4 moved to 533 FSB, and with barton chips reality went completely out the window.

It's too bad AMD doesn't have the guts or integrity to do something radical like come out with a PR'03 rating an re-rate their chips at more honest levels.

 

[T9D]

..

 

[Wingznut]

Originally posted by: AndyHui
Hhmm....why did AMD run their benchmarks with HyperThreading disabled on the Pentium 4?

Well, at least they updated *some* of their benchmarks and the P4's hardware. It would've been nice had they ran both systems with the same DDR timings, though. (2-2-2-5 for the Athlon, 2-3-3-6 for the P4... Although I'm not sure you can adjust the timings on the P4 board they chose.)

Originally posted by: Smilin
Don't pop a vein mr intel guy. AMD won't hold the crown for long. It's not THAT much faster. Besides if they used more modern benchmarks you'd probably gripe about them being unfair as well. I'm kinda interested though: Please name **6** chipsets for the p4 that are better than the i845g.

oldfart and Acanthus took care of the chipsets question, so I'll just address the rest...

"Pop a vein"? If you think I might get upset about such (relatively) trivial things, then you surely don't know me very well. I was just pointing out how silly the whole PR thing has become. I don't gripe about AMD... I've said on numorous occasions that they make an excellent product. It's just sad how it has to be overshadowed by the BS that is PR.

 

[dullard]

Originally posted by: xerosleep
Wow it's already up on pricewatch.

Its been up on pricewatch for at least 2 weeks.

 

[dexvx]

Originally posted by: DaveSimmons
I remember the "good old days" when AMD PR ratings were conservative and an Athlon at a given rating would soundly beat the P4 at that clockspeed. It was sad to see the rating becoming more and more of a marketing lie after the P4 moved to 533 FSB, and with barton chips reality went completely out the window.

It's too bad AMD doesn't have the guts or integrity to do something radical like come out with a PR'03 rating an re-rate their chips at more honest levels.

See thats the fundamental flaw of PR ratings. They are only useful when they come out. But then again you still have people living in caves and believing a 3Ghz P4-C/875 to be only 50% faster than a willamette 2Ghz/850 PC800.

Actually, I'd take an 845D/E over an 845G.

845G was faster than D/E because of memory tweaks being utilized for the onboard graphics. There were versions without AGP slots, but most had AGP slots. The most noticeable difference was that G/D boards had no HT support whereas E boards do.

 

[Wurrmm]

AMD Introduces World?s Highest Performing Desktop PC Processor, the AMD Athlon? XP Processor 3200+

Not so much. Intel has the fastest right now.

 

[oldfart]

Originally posted by: dexvx

Originally posted by: DaveSimmons
I remember the "good old days" when AMD PR ratings were conservative and an Athlon at a given rating would soundly beat the P4 at that clockspeed. It was sad to see the rating becoming more and more of a marketing lie after the P4 moved to 533 FSB, and with barton chips reality went completely out the window.

It's too bad AMD doesn't have the guts or integrity to do something radical like come out with a PR'03 rating an re-rate their chips at more honest levels.

See thats the fundamental flaw of PR ratings. They are only useful when they come out. But then again you still have people living in caves and believing a 3Ghz P4-C/875 to be only 50% faster than a willamette 2Ghz/850 PC800.

Actually, I'd take an 845D/E over an 845G.

845G was faster than D/E because of memory tweaks being utilized for the onboard graphics. There were versions without AGP slots, but most had AGP slots. The most noticeable difference was that G/D boards had no HT support whereas E boards do.

845G was too buggy for me. They had all kinds of ram compatibility issues, and conflicts with Intel infs and nVidia drivers. They did not overclock as well as 845D/E either. 845D/E was a more solid chipset.

 

[Wingznut]

Originally posted by: Pariah
And you wonder why you catch flak and suspicion from some members here when you take unnecessary and unwarranted potshots at AMD like this.

I guess I'm confused... I point out one (of the many) fallicies of this PR "system", and you jump on me about it. ("Unwarranted"?)

But then in your next post, you pretty much validate my point, even better than I did. (Apparently my point wasn't "unwarranted"?)

Should the point not have been made? Or was it just a problem that I made the point?

 

[PrinceXizor]

Of course Wingz, you're from the "evil" company....haha

Anyhow, what's all this nonsense from THG about it should be labeled a 2800+? I've read about a dozen reviews so far today (none from Tom's, that should tell you what I think about his reviews) and even the one that had the AMD in the worst light still had it keeping pace with the older P4 3.06Ghz w/ HT. Most reviews I've seen show the XP 3200+ about even to slightly slower overall then a P4 3.0C at 800Mhz FSB (meaning it beat the P4 3.0C on some benches and was slower on others, maybe slower on a few more than it was faster). The PR rating is inherantly and obviously flawed, but saying this chip should be called a 2800+ isn't really improving my opinion of THG.

P-X

 

[Pariah]

I guess I'm confused... I point out one (of the many) fallicies of this PR "system", and you jump on me about it. ("Unwarranted"?)

Your childish mocking of AMD in the post lent nothing to the credibility of what you were trying to say. You made no attempt to surmise why they were using those benchmarks when there are multiple logical reasons for doing it beyond just trying to BS the public.

But then in your next post, you pretty much validate my point, even better than I did. (Apparently my point wasn't "unwarranted"?)

There are definite problems with the rating system, but they are much deeper than the banchmarks being used. You made no attempt to make an actual argument beyond laughing at AMD for the benchmarks they were using. There was no reason for you make that post about AMD. And I would have said the same thing to anyone who posted the same immature drivel about Intel.

Should the point not have been made? Or was it just a problem that I made the point?

You didn't make a point. Ripping AMD's shameful PR tactics? Glass houses

 

[jaeger66]

Originally posted by: oldfart

845G was too buggy for me. They had all kinds of ram compatibility issues, and conflicts with Intel infs and nVidia drivers. They did not overclock as well as 845D/E either. 845D/E was a more solid chipset.

Hm, been using an Asus 845G for 11 months without a trace of a problem. The Epox one was sort of iffy, but that was Epox's fault.