Athlon @ 2.485 at HardOCP

[Snoop]

From HardOCP:

We have tested the card on a Pentium 4 2.53GHz system and also an AMD 2.485GHz system. The AMD numbers were done on a yet-to-be-announced CPU based on the Thoroughbred core.

Whaaaa, 2.485ghz???? AMD must have performed a miracle core stepping to take the 2200+ ie 1.8 ghz to 2.485, if the pr scale remains constant this thing would be a ~3200+ ??????? Possibly this is a 2400+ or 2600+ that we have heard rumours about?


I have yet to finish reading the ocp article (their server is funked) so maybe their is a simple explanation (liquid nitrogen??? )

I cannot wait, Nforce 2, dual channel PC3200, 333 fsb...... maybe a Radeon 9700 to boot

edit: Link to Hard OCP article.

 

[Czar]

wow, thats amazing

wonder if those new chips work on old motherboards

 

[bjc112]

linky?

2.4 ahhhhhhhh that would smack a 2.8

 

[Mingon]

repost It seems like its a 2ghz part (xp2400) running the 15 x 133 bus speed which has then been overclocked by kyle.

 

[bjc112]

Originally posted by: Mingon
repost It seems like its a 2ghz part (xp2400) running the 15 x 133 bus speed which has then been overclocked by kyle.

May-b AMD got it together... They are hopefully getting some killer Yeilds...

Hope all these bad boys overclock

 

[imgod2u]

I'm reading that review right now and the benchmarks disturbs me. That can't be right. The P4 at 2.53 stays within roughly 5% of the Athlon with the same 9700 Pro vid card. Hell, it even won a Q3A bench. Something has to be wrong. I don't care how much you optimize, a P4 with less than 100MHz advantage beating an Athlon? WTF.

 

[ravedave]

looks like the rating is really starting to fade... mostly the 2.5 and 2.5 were neck n neck...
but thats cool that AMD is finally coming out with new speed grades!

 

[CubicZirconia]

Hope all these bad boys overclock

Me too. If thats the case then a lot of people will again be jumping ship. This time it will be back to good ol' amd.

The performace rating thing has to be going out the window. Correct me if I'm wrong, but a chip @2.485 would have a performace rating of 3237+.

 

[RaynorWolfcastle]

Originally posted by: CubicZirconia

Hope all these bad boys overclock

Me too. If thats the case then a lot of people will again be jumping ship. This time it will be back to good ol' amd.

not really, what's the point of jumping back to AMD if you have a Northwood @ 2.4 GHz or better? Judging by the article, you're only a few benchmark points behind, probably nothing noticeable in the real world.

-Ice

 

[CubicZirconia]

not really, what's the point of jumping back to AMD if you have a Northwood @ 2.4 GHz or better? Judging by the article, you're only a few benchmark points behind, probably nothing noticeable in the real world.

Actually, I bet a lot of people will switch back. It might take awhile, but some people will switch. Granted, the performace difference might not be that great. That hasn't stopped people in the past though. And who knows, maybe people will be able to take the new tbreds well beyond what hardocp has done.

I wish I Hardocp would work, I still can't read the article.

 

[Athlon4all]

:Q

 

[kuk]

2845 mhz is very close to 2490, which corresponds to a 15x multiplier on a 166mhz bus. The 15x multiplyer on a current 133FSB will get you 2.0Ghz ... which is the megaherts rating of the rumored 2400+.

My guess is that these figures are from an overclocked 2400+ AthlonXP, on a 166FSB.


But that's just my opinion. I could be wrong.
Kuk

 

[SexyK]

Originally posted by: CubicZirconia

not really, what's the point of jumping back to AMD if you have a Northwood @ 2.4 GHz or better? Judging by the article, you're only a few benchmark points behind, probably nothing noticeable in the real world.

Actually, I bet a lot of people will switch back. It might take awhile, but some people will switch. Granted, the performace difference might not be that great. That hasn't stopped people in the past though. And who knows, maybe people will be able to take the new tbreds well beyond what hardocp has done.

I wish I Hardocp would work, I still can't read the article.

I don't see why anyone would swich back now. Who wants to go out and buy a motherboard for Socket A that's pretty much at the end of it's life, when they could just keep their current P4 board and most likely be able to throw in a 4+GHz CPU in the not distant future. Doesnt seem like much of an arguement to ever go back to the Socket A platform, espicially if you have a 2+GHz P4 already. Now, Hammer, that may draw some converts, but it's way to early to tell about that.

Kramer

 

[microAmp]

Originally posted by: Mingon
repost It seems like its a 2ghz part (xp2400) running the 15 x 133 bus speed which has then been overclocked by kyle.

No it's an Athlon running at 2.48 GHz, see here in 2nd paragraph. 15 x 165, AGP 82 MHz, PCI 41 Mhz.

The Athlon has help from the faster FSB which helps it in benchmarks. Haven't seen them yet at , haven't gone that far in the article.

*posting isn't working for me well today

 

[Chronic321]

What the hell? I thought my 1.8 ghz Athlon was suppose to best the 2.26 P4? Now these new T-bred's are barley besting the P4's clock for clock, and this is with a non-stock FSB. What's going on?

 

[Dulanic]

He needs to kick that baby up to a nice 200Mhz FSB the 8K3A should have no problem doing that, mine will do 215.

 

[Mingon]

No it's an Athlon running at 2.48 GHz, see here in 2nd paragraph. 15 x 165, AGP 82 MHz, PCI 41 Mhz.

I meant the stock part is a 2ghz amd chip (2400xp) which has been overclocked

 

[natopotato]

i think some of you forget how much 5% really is when it is the same vid card. those benches rely more on the vid card than CPU. 5% is actually quite a bit in vid card intensive apps with the exact same card. i think at these speeds the t-bred could use more cache, then it would really fly.

 

[dexvx]

Originally posted by: Chronic321
What the hell? I thought my 1.8 ghz Athlon was suppose to best the 2.26 P4? Now these new T-bred's are barley besting the P4's clock for clock, and this is with a non-stock FSB. What's going on?

If you mean 1800+, you're badly mistaken, but if you mean 1.8Ghz (2200+), they are roughly even give or take.

The lowest res benchmarked is 1024x768, certainly not CPU limited. You need to bench at 640x480 to give a CPU limited bench. Although it is interesting to note that P4's scale better in clock speeds (1.6Ghz p4 is way weaker than a 1.6Ghz XP where as at about 2.5ghz they are much closer). I'd expect the 15x166 XP to be on par with a 3Ghz P4 (depending on a lot of factors like FSB and ram).

 

[Chronic321]

1.8 GHz does not refer to the AMD PR Rating. The AMD PR Rating is not intended to mean "GHz" or "Mhz" instead it is a comparison to T-bird CPU's. So by saying 1.8 GHz Athlon it would mean that I am reffering to the AMD 2200+.

 

[Anubis]

i think they ahve a unreleased cersion of the athlonxp that i ahve heard rumors about. suposedly AMD is moving to a 166mhz FSB so it makes sense that the procsessor runs taht fast... and if it si stock i wonder how far it Overclocks.... i mean thats like 3500+ levels

 

[MadRat]

The Athlon core over 2GHz is probably so cache starved that it runs cooler than the same core at 2200+ speeds because the core is waiting for memory requests to come back...

 

[Jmmsbnd007]

Originally posted by: MadRat
The Athlon core over 2GHz is probably so cache starved that it runs cooler than the same core at 2200+ speeds because the core is waiting for memory requests to come back...

How so? Sure, the cache bus could be a bit wider, but the cache is already on-chip and full speed.

 

[BD231]

Originally posted by: imgod2u
I'm reading that review right now and the benchmarks disturbs me. That can't be right. The P4 at 2.53 stays within roughly 5% of the Athlon with the same 9700 Pro vid card. Hell, it even won a Q3A bench. Something has to be wrong. I don't care how much you optimize, a P4 with less than 100MHz advantage beating an Athlon? WTF.

Dont forget its on a 133mhz fsb, alot of the reviews of OC'd Athys you see have a higher FSB which adds a good deal of mem bandwidth to the picture.

 

[imgod2u]

Yes but the Athlon core is less memory dependent than the P4. It shouldn't perform less even at 2.48 GHz. I mean, hell, a P4 with 266MHz memory drops maybe 10% in performance, the Athlon shouldn't drop even that. We're talking less than 100 MHz difference here, even less percentage-wise and an Athlon with such a higher average IPC being only around 5% faster than a similarly clocked P4?