• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

How does CPU cache impact SMP PPD?

J

jebo_4jc

[H]ard|DCer of the Month - April 2011
2FA
Joined
Apr 8, 2005
Messages
14,572
I've searched around, but not sure if I've found the definitive answer. I'm curious if anybody has seen direct comparisons of PPD comparing CPUs with different cache sizes.
I did see this on another forum, one guy claimed:
QX9650 (12MB) at 9x400mhz (3.6) = 3500 ppd
Q6600 (8MB) at 9x400mhz = 3200 ppd

So, about a 8-9% difference from a 50% cache increase. While that is significant, it might not always be cost effective to purchase a CPU with bigger cache if the return is going to be that small. Any input?
 
It depends on the work unit. Differences were more pronounced with the 2609/2610 work units but we don't get them any more. In dual-cores, the drop from 2MB to 1MB seems to be less severe than the drop from 4MB to 2MB. In quads, I've yet to see any comprehensive numbers from the new 12MB chips. Your numbers look about right though.


 
I agree it definitely plays a part. I can see a big difference going from my old 6600 to this temp E2140. Hopefully in a month or so I see the opposite when I get a Q9450 :) But overall, I think with the dual core chips, if you can clock the chip with the smaller cache 100mhz or so higher, you'd most likely make up for the lack of cache.
 
Look up the SMP protiens at Fah Info.
You'll be able to see the difference the cache makes in terms of PpDpGhz there.
Its anything from 0 -> 50% slower.
Luck ............ :D
 
A comparison between the Conroe quads and the Penryn quads isn't completely accurate in regards to differences in cache amounts. The cache setup is a little different I believe as well as some other changes in the architecture which gives the Penryn based quads a higher IPC than Conroe quads. Just take that into account when looking at this. This could have more to do with the IPC differences rather than cache amounts.

 
You must log in or register to reply here.
Back
Top