Hmm... I like this benchmark. :up: I'm only running a dual Xeon setup so nothing to spectacular from me.
Read: 4497 MB/s
Write: 1415 MB/s
Latency: 84.0 ns
Memory is only running at 2-2-2-6, 207MHz dual channel.
Printable View
Hmm... I like this benchmark. :up: I'm only running a dual Xeon setup so nothing to spectacular from me.
Read: 4497 MB/s
Write: 1415 MB/s
Latency: 84.0 ns
Memory is only running at 2-2-2-6, 207MHz dual channel.
Thanks.Quote:
Originally Posted by agenda2005
You can my settings right here:
http://www.xtremesystems.org/forums/...ad.php?t=47018
:)
Right back at you mate.Quote:
Originally Posted by Zeus
Down to 36.2, this is at 260X10 with the below timings.
I did have 265 stable but ive tweaked my 3.3v line down to test this OCZ booster, so im Vdimm is very low ATM (stuck at 3.3)
260 at this low Vdimm isnt propperly stable either.. but once I get the booster working ill be able to get 265+ stable.
Nice going Holst! :toast:Quote:
Originally Posted by Holst
I gave my system hell but 36.5 was all she wrote. :D
Oh well, not too shabby for 2,5-3-3-7
another one from Polish OCer :)
MSI K8N Neo2 Platinum - Q47353.1 0440A2 S Taiwan
A64 3200+ 90nm "Winchester" - CBBFD 0441UPCW
Thermalright XP-90 w/ YTech 120mm
CLICK for Full Size Screenshot - 2048x1536 384KB
http://www.xtremesystems.org/forums/...chmentid=20513
EDIT: oh, it looks like it's the best score so far :) let's wait until someone with FX shows up :stick:
Man! that's a sick latency! That's the kind of latency I'm dreaming about (34-35ns) in dual channel with >8000MB/s Bandwith. Good job, I'm sure that's the lowest latency I've seen so far.Quote:
Originally Posted by bachus_anonym
thanx agenda2005 :) i can do SuperPi 1M at 273x10, but not much changes in Latency department :(Quote:
Originally Posted by agenda2005
bah i had the crown for a bit :D
I got 39'3ns at 250 2-2-2.
Some number, from my FX-55 @ 3.330 in the Prometeia with 2 TCCD stick at 290Mhz, 2.5-3-3-6 with just 2.8v ..
http://pedrorocha.planetaclix.pt/Eve...4600_Write.jpg
http://pedrorocha.planetaclix.pt/EverestPC4600_Read.jpg
http://pedrorocha.planetaclix.pt/Eve...0_Latencia.jpg
http://pedrorocha.planetaclix.pt/EverestPC4600_FSB.jpg
:D
nice going Pedro, do you have any BH-5 > 270MHz ??? i'm curious what it looks like with FX-55.... i mean latency.
BTW, i just saw a whopping 32.8ns @ 277x10 !!!!!
Blimey Bachus, thats super low latency.
Is that the same timings as in the pic above?
pretty much.... just TRTW lowered to 1 and Refresh Rate at 15.6ns :DQuote:
Originally Posted by Holst
EDIT we need to invite Onepagebook here to see what his BH-5 @ 287 can do :D
Nice numbers Bachus! :up:
This once again proves how much faster the winnies are over the socket 754 winchesters.
I need 2830MHz for 31s SuperPi, but 32m takes me a little over 28 mins to finish, your 26 min 39sec is a lot better.
However latency doesn't suprise me, your timings are a lot better than mine.
For now you seem to have the best latency score. :cool:
and all the glory for Bachus...
Bachus, remember which whisper give You that :)
I have 35.1 in single channel - so Poland wins in Latencies! (for a moment)
:thumbsup:Quote:
Originally Posted by rapra
38.0ns latancy
1572mb/swrite
3828MB/s read
I am more Interested in Latency values. Here you go.
A64 3500+ DDR400 200MHz 2-2-2-10 Latency = 46.5ns
A64 3500+ DDR434 217MHz 2-2-2-10 Latency = 42.9ns
My BH-5 will be on ground tommorrow. I will play with higher Mem speed to see what happen. DDR533 2-2-2-10 will be interesting to compare with Intel DDR2. From the latency value, DDR2 sucks big time.
Here is what I got...havent tweaked it yet...
Latency @ 46.7ns
250x10.0 1:1 @ 2.5-3-3-0
Memory Write @ 1633MB/s
Memory Read @ 3849MB/s
Not sure if computer being rebooted fresh start would make a difference...my PC has been on for 3 days straight at these results...does anyone know if that makes a difference or not?
some, but note a huge one.
Yeah after rebooting my Latency is 47.3ns...not really a big difference.
That latency is not good at DDR500, even with cas 2.5. I guess rebooting your computer makes no difference either.Quote:
Originally Posted by ugp
Well what I am trying to go for from my memory is bandwidth:D And I am assuming the higher the bandwidth the higher the latency goes up with it alittle right...well at CL2.5 anyway...Quote:
Originally Posted by agenda2005
Wrong assumption mate.Quote:
Originally Posted by ugp
Check my bandwidth, it's over 5K in Sandra, my latency is as low as 36,5ns.
It has to do with memory settings.