How much better is Athlon that Duron for Folding

[Clocker]

Any idea? Would a 1Ghz Duron be any slower than a 1Ghz Athlon (non-XP) at folding? Any idea how much if yes?

C

 

[Buck]

Any idea? Would a 1Ghz Duron be any slower than a 1Ghz Athlon (non-XP) at folding? Any idea how much if yes?

C

Having F@H power improve with XP processors would imply that this software is somewhat dependent on cache. As far as electrical speed goes, 1 Ghz is 1 Ghz. Perhaps someone with a better CPU knowledge can help.

 

[Handruin]

Any idea? Would a 1Ghz Duron be any slower than a 1Ghz Athlon (non-XP) at folding? Any idea how much if yes?

C

Having F@H power improve with XP processors would imply that this software is somewhat dependent on cache. As far as electrical speed goes, 1 Ghz is 1 Ghz. Perhaps someone with a better CPU knowledge can help.

Are the cores fundamentally the same with one lacking cache?

 

[Clocker]

Also, for the sake of the comparison...assume both cpus are 200mhz fsb.

Thanks,
C

 

[Cliptin]

Any idea? Would a 1Ghz Duron be any slower than a 1Ghz Athlon (non-XP) at folding? Any idea how much if yes?

C

Having F@H power improve with XP processors would imply that this software is somewhat dependent on cache. As far as electrical speed goes, 1 Ghz is 1 Ghz. Perhaps someone with a better CPU knowledge can help.

It makes sense to me that something this CPU intensive is dependent on cache for for the final analysis I will defer.

I will do some examination later on my machines to see if I can come up with something approximate.

 

[Mercutio]

It'd be interesting to know if F@H is memory-bandwidth intensive. I don't think it is but it might also be an interesting comparison.

AFAIK there is no difference between the Duron and the Athlon save the artificial limitation to 200MHz bus and the amount/quality of cache on the Duron.

 

[NRG = mc²]

Don't all Duron 1000s use the Morgan core which is the Duron equivalent of the XP? Perhaps this will affect results and actually end up being faster (if the cache is of no significance)?

 

[Tea]

Just so, NRG.

Athlon Classic was one of a kind.

Athlon Thunderbird = Duron 600, 650, 700, 750, 800, 850, 900 & 950.

Athlon MP = Athlon XP = Duron 100, 1100, 1200 & 1300.

As Mercutio says, cache and FSB aside, the Athlons and the Durons are exactly the same.

 

[Clocker]

So...is the Athlon faster? Do you think the cache makes a difference in folding?
C

 

[Tea]

I don't know, Clocker. I'd have to benchmark it. Tannin might know, I'll ask him.

 

[Cliptin]

I ran SuperPI(1M) on my machines:

Duron 950 @107Mhz FSB ->1018Mhz =2m04s
&
XP 1800+ @107Mhz FSB ->1231Mhz =1m31s

Be my guest on making something lively out of it.

 

[Cliptin]

I ran SuperPI(1M) on my machines:

Duron 950 @107Mhz FSB ->1018Mhz =2m04s
&
XP 1800+ @107Mhz FSB ->1231Mhz =1m31s

Be my guest on making something lively out of it.

If I did my calculations* right and who know 'cause I was basicly guessing:
The XP is better by a factor of .6.



*

1231Mhz        1018Mhz
----------  =  ----------     X
91s              124s

X works out to be 1.6

 

[Pradeep]

Now we just need someone with a 1GHz+ Duron to give some results. The 950 would be the old core.