What kind of SETI times should I expect with this setup?

[CraigRT]

P4 1.8@2.4 on an EPoX 4PEA+ 845pe chipset.
using 512 DDR333 @ 354MHz - all memory timings on fast.
running win2k with sp3 and the CLI client.

I have a feeling it won't be any better than my Athlon XP1800+@2000+ which commonly did WU's in the 3:05 range.

anyone???
thx alot. 🙂

 

[Zim Hosein]

Yield, I would venture a guess that your WU times should be around 2.5-2.8 hours per WU 🙂

 

[CraigRT]

you seriously think less than 3 hours?
kind of shocks me..

 

[CrazySaint]

Originally posted by: Yield
you seriously think less than 3 hours?
kind of shocks me..

I experimented a bit with SETI on my 1.8A@2.7GHz DDR400 and I was getting around 2.5 hours per WU.

 

[CraigRT]

Originally posted by: CrazySaint

Originally posted by: Yield
you seriously think less than 3 hours?
kind of shocks me..

I experimented a bit with SETI on my 1.8A@2.7GHz DDR400 and I was getting around 2.5 hours per WU.

Heheh, excellent.. well i ended up around 2.8 (can push further but doesn't seem to be solid) so hopefully i can get those times. 😀
i am running 1:1 MEM:FSB because it wasn't stable on 3:4 (strange) so it may be a bit different.

 

[BadThad]

2.7-3.2 hr/wu

 

[sean2002]

I had my P4 2.26 (northwood) with 256MB of PC-1066 RDRAM running on a 154Mhz FSB (616 QDR) which puyt the CPU @ 2618Mhz, this gave me time on WU's over 1.0AR of under 2 hours. like 1:57-1:59, and then the normal 0.417 AR WU would get done in around 2:22. my average was about 2:12



Just ask Greg, a.k.a Smokeball

 

[smp]

Maybe I should run SETI again.
I stopped because i suspected that it was really echelon and carnivore TYPE operations that were using my spare cycles and not little green men as I had originally thought (and been excited about).

 

[CraigRT]

Originally posted by: sean2002
I had my P4 2.26 (northwood) with 256MB of PC-1066 RDRAM running on a 154Mhz FSB (616 QDR) which puyt the CPU @ 2618Mhz, this gave me time on WU's over 1.0AR of under 2 hours. like 1:57-1:59, and then the normal 0.417 AR WU would get done in around 2:22. my average was about 2:12



Just ask Greg, a.k.a Smokeball

oh my gawd those times are good. :Q
i'm running 155 FSB on my 1.8A for about 2.8GHz
but I have DDR333. guessing slightly slower than RAMBUS times. but the 845pe chipset is no slouch either 😉
i'll find out tonight what the times are like for sure.

 

[sean2002]

Originally posted by: Yield

Originally posted by: sean2002
I had my P4 2.26 (northwood) with 256MB of PC-1066 RDRAM running on a 154Mhz FSB (616 QDR) which puyt the CPU @ 2618Mhz, this gave me time on WU's over 1.0AR of under 2 hours. like 1:57-1:59, and then the normal 0.417 AR WU would get done in around 2:22. my average was about 2:12



Just ask Greg, a.k.a Smokeball

oh my gawd those times are good. :Q
i'm running 155 FSB on my 1.8A for about 2.8GHz
but I have DDR333. guessing slightly slower than RAMBUS times. but the 845pe chipset is no slouch either 😉
i'll find out tonight what the times are like for sure.

the 845 can only supply half of the bandwith needed for the P4, Thats why my dual channel RDRAM crunches so fast

 

[Assimilator1]

Originally posted by: smp
Maybe I should run SETI again.
I stopped because i suspected that it was really echelon and carnivore TYPE operations that were using my spare cycles and not little green men as I had originally thought (and been excited about).

Yeah I think you should run SETI again too🙂 ,btw they are little purple men not green😉

 

[CraigRT]

Originally posted by: sean2002

Originally posted by: Yield

Originally posted by: sean2002
I had my P4 2.26 (northwood) with 256MB of PC-1066 RDRAM running on a 154Mhz FSB (616 QDR) which puyt the CPU @ 2618Mhz, this gave me time on WU's over 1.0AR of under 2 hours. like 1:57-1:59, and then the normal 0.417 AR WU would get done in around 2:22. my average was about 2:12



Just ask Greg, a.k.a Smokeball

oh my gawd those times are good. :Q
i'm running 155 FSB on my 1.8A for about 2.8GHz
but I have DDR333. guessing slightly slower than RAMBUS times. but the 845pe chipset is no slouch either 😉
i'll find out tonight what the times are like for sure.

the 845 can only supply half of the bandwith needed for the P4, Thats why my dual channel RDRAM crunches so fast

HALF?
there's no way.. i've seen some DDR boards scoring higher than RDRAM when overclocked... :Q
but that would be bad... RDRAM might actually be worth the price if it was true.. 😉

 

[sean2002]

Originally posted by: Yield

Originally posted by: sean2002

Originally posted by: Yield

Originally posted by: sean2002
I had my P4 2.26 (northwood) with 256MB of PC-1066 RDRAM running on a 154Mhz FSB (616 QDR) which puyt the CPU @ 2618Mhz, this gave me time on WU's over 1.0AR of under 2 hours. like 1:57-1:59, and then the normal 0.417 AR WU would get done in around 2:22. my average was about 2:12



Just ask Greg, a.k.a Smokeball

oh my gawd those times are good. :Q
i'm running 155 FSB on my 1.8A for about 2.8GHz
but I have DDR333. guessing slightly slower than RAMBUS times. but the 845pe chipset is no slouch either 😉
i'll find out tonight what the times are like for sure.

the 845 can only supply half of the bandwith needed for the P4, Thats why my dual channel RDRAM crunches so fast

HALF?
there's no way.. i've seen some DDR boards scoring higher than RDRAM when overclocked... :Q
but that would be bad... RDRAM might actually be worth the price if it was true.. 😉

clock for clock, FSB to FSB RAMBUS will always beat the 845 and DDR. PC-2100= 2.1GBps PC1066 dual channel= 4.2GBps, and when you start to overclock the RDRAM increases it's lead as it's latency drops

 

[Assimilator1]

That's GB's not GBps😉

 

[Zim Hosein]

Originally posted by: Yield
you seriously think less than 3 hours?
kind of shocks me..

Consider yourself shocked Yield 😉 😀

I can't wait to see your daily WU times go up 🙂

 

[CrazySaint]

Originally posted by: sean2002

Originally posted by: Yield

Originally posted by: sean2002

Originally posted by: Yield

Originally posted by: sean2002
I had my P4 2.26 (northwood) with 256MB of PC-1066 RDRAM running on a 154Mhz FSB (616 QDR) which puyt the CPU @ 2618Mhz, this gave me time on WU's over 1.0AR of under 2 hours. like 1:57-1:59, and then the normal 0.417 AR WU would get done in around 2:22. my average was about 2:12



Just ask Greg, a.k.a Smokeball

oh my gawd those times are good. :Q
i'm running 155 FSB on my 1.8A for about 2.8GHz
but I have DDR333. guessing slightly slower than RAMBUS times. but the 845pe chipset is no slouch either 😉
i'll find out tonight what the times are like for sure.

the 845 can only supply half of the bandwith needed for the P4, Thats why my dual channel RDRAM crunches so fast

HALF?
there's no way.. i've seen some DDR boards scoring higher than RDRAM when overclocked... :Q
but that would be bad... RDRAM might actually be worth the price if it was true.. 😉

clock for clock, FSB to FSB RAMBUS will always beat the 845 and DDR. PC-2100= 2.1GBps PC1066 dual channel= 4.2GBps, and when you start to overclock the RDRAM increases it's lead as it's latency drops

Well, my P4 OC'd to 2.7GHz with RAM@DDR400 on a i845G chipset scores pretty darn close to PC1066 RDRAM in Sandra 😉

 

[CraigRT]

Originally posted by: CrazySaint

Originally posted by: sean2002

Originally posted by: Yield

Originally posted by: sean2002

Originally posted by: Yield

Originally posted by: sean2002
I had my P4 2.26 (northwood) with 256MB of PC-1066 RDRAM running on a 154Mhz FSB (616 QDR) which puyt the CPU @ 2618Mhz, this gave me time on WU's over 1.0AR of under 2 hours. like 1:57-1:59, and then the normal 0.417 AR WU would get done in around 2:22. my average was about 2:12



Just ask Greg, a.k.a Smokeball

oh my gawd those times are good. :Q
i'm running 155 FSB on my 1.8A for about 2.8GHz
but I have DDR333. guessing slightly slower than RAMBUS times. but the 845pe chipset is no slouch either 😉
i'll find out tonight what the times are like for sure.

the 845 can only supply half of the bandwith needed for the P4, Thats why my dual channel RDRAM crunches so fast

HALF?
there's no way.. i've seen some DDR boards scoring higher than RDRAM when overclocked... :Q
but that would be bad... RDRAM might actually be worth the price if it was true.. 😉

clock for clock, FSB to FSB RAMBUS will always beat the 845 and DDR. PC-2100= 2.1GBps PC1066 dual channel= 4.2GBps, and when you start to overclock the RDRAM increases it's lead as it's latency drops

Well, my P4 OC'd to 2.7GHz with RAM@DDR400 on a i845G chipset scores pretty darn close to PC1066 RDRAM in Sandra 😉

That's what I thought.. nice!
P.S. I am back on my XP rig as something happened to the board... it wouldn't boot on the 2nd day (at all) so it's sitting on my desk until I can get out, and get a stick of memory so I can plug all this crap in again. ugh.

 

[RickD]

This little old 1600+ that I'm running has an average of 2:39 and thats on 2 of them

 

[sean2002]

Originally posted by: CrazySaint

Originally posted by: sean2002

Originally posted by: Yield

Originally posted by: sean2002

Originally posted by: Yield

Originally posted by: sean2002
I had my P4 2.26 (northwood) with 256MB of PC-1066 RDRAM running on a 154Mhz FSB (616 QDR) which puyt the CPU @ 2618Mhz, this gave me time on WU's over 1.0AR of under 2 hours. like 1:57-1:59, and then the normal 0.417 AR WU would get done in around 2:22. my average was about 2:12



Just ask Greg, a.k.a Smokeball

oh my gawd those times are good. :Q
i'm running 155 FSB on my 1.8A for about 2.8GHz
but I have DDR333. guessing slightly slower than RAMBUS times. but the 845pe chipset is no slouch either 😉
i'll find out tonight what the times are like for sure.

the 845 can only supply half of the bandwith needed for the P4, Thats why my dual channel RDRAM crunches so fast

HALF?
there's no way.. i've seen some DDR boards scoring higher than RDRAM when overclocked... :Q
but that would be bad... RDRAM might actually be worth the price if it was true.. 😉

clock for clock, FSB to FSB RAMBUS will always beat the 845 and DDR. PC-2100= 2.1GBps PC1066 dual channel= 4.2GBps, and when you start to overclock the RDRAM increases it's lead as it's latency drops

Well, my P4 OC'd to 2.7GHz with RAM@DDR400 on a i845G chipset scores pretty darn close to PC1066 RDRAM in Sandra 😉

If your going to compare overclocked RAM why you forgot to mention the SANDRA does not provide benches for PC-1200 RAMBUS, which in my case it would be PC1218 RABUS, which would widen the gap and lower the latency in the RAMBUS