Printable View
Yesterday I received my OCZ 4000EL 2x512 sticks and put them to the test. I got 292fsb 1:1 at 2.85v. Here's the benchies:
CPU-Z
http://radelon.home.mchsi.com/el4000/292cpuz.gif
Sandra Unbuffered
http://radelon.home.mchsi.com/el4000/292unbuff.gif
Sandra Buffered
http://radelon.home.mchsi.com/el4000/292buff.gif
PCMark2002: 12,591
Wahoo!!!!
:cool: and at 2.85v too !
Holy crap!!! I just got the same kit in today. Just awesome!
that stuff screams for a vdimm mod:D
what about buffered?
it can run sandra but is it STABLE?
It ran PCMark2002, as you see the link down there. And it's runnin' burn-in tests now as we speak, 2 hours no problems.
Buffered is there, it's a little low as I've had a 7k run in that.Quote:
Originally posted by r00t3d
what about buffered?
Nice Radelon :toast:
Vdimm shouldnt help it much.
Mine loves 2.85 as well anything higher it starts getting errors
Cept mine does 290 @ 3-3-4-7 you may want to try those timings btw
You are on a P4C800 also right DJ? If so, I'm beginning to think it doesn't like higher vdimm because it doesn't have more vdd. Any modded voltage fluctuates from IDLE to LOAD......Quote:
Originally posted by SpicyHuevos
Vdimm shouldnt help it much.
Mine loves 2.85 as well anything higher it starts getting errors
Cept mine does 290 @ 3-3-4-7 you may want to try those timings btw
Reason is, yes, DJ is right, anything over 2.85v for me gives errors. And my chips don't like those timings DJ..... I'll still give it a shot later.
Would 2x256mb of the same sticks perform the same I just got the MAX3 but going to run stock speed for 48hrs before I start feeding:slobber: :banana:
Some very nice benches there!
Though assuming you've got a decent graphics card, what i'd really like to see is some rounds of 3dmark2k1.. with the only difference being your cpu:memory ratio.. i want to see what actual score differences you get just by having the ram at 1:1, 5:4 and 3:2.. same cpu fsb, and same timings. I'm very curious how significant it is.
Nice scores, but Sandra and PCmark 2002 are really reliant on mem bandwidth... not neccesarily real performance. What are your scores in 2dm2001 ?
Holy crap:slobber:
That's cool as for the benchies and for the high frequency..........Now,RUN a Pi-fast V.4.1 (hexus pi-fast) with these timings and see how it does........ ;)
*spills coffee* holly cr*p! :eek:
Well hipro5, I did more than you asked, how about a 12x294 (3.52ghz) pifast:
http://radelon.home.mchsi.com/el4000/pifast2.gif
Sickly amazing. Like a fatal car crash. I just can't look away. Simply awesome!
Thx dogman....... the biggest problem right now is that the northbridge needs more voltage..... and I need to cool it. I'm only running 2.85v as a v-modded vdimm causes errors.Quote:
Originally posted by DoGMaN
Sickly amazing. Like a fatal car crash. I just can't look away. Simply awesome!
That's COOL......You've got one of the highest Pi-fast benchies comparing to your speed of CPU........ :D :toast:
hmm, I wonder how 294 fsb 1:1 would be on IC7-MAX3
I tried it on IC7-G and got terrible scores comparing to
Asus P4C800-E
at 275 fsb 1:1 2.5 4 4 7 I got about 2800 Unbuffered in Sandra,
and thats bad, and all gaming benchmarks and pifast, superpi and all were not as good as if I ran 5:4 with my 3700 gold at 2 6 3 3...
It seems to be different on Asus though...
maybe i try ocz 4000 gold on this MAX3 also just to know... :)
Ya, I have noticed the P4C seems to score better than IC7 at 1:1, and the IC7 seems to score better than P4C at 5:4Quote:
Originally posted by WizardB
hmm, I wonder how 294 fsb 1:1 would be on IC7-MAX3
I tried it on IC7-G and got terrible scores comparing to
Asus P4C800-E
at 275 fsb 1:1 2.5 4 4 7 I got about 2800 Unbuffered in Sandra,
and thats bad, and all gaming benchmarks and pifast, superpi and all were not as good as if I ran 5:4 with my 3700 gold at 2 6 3 3...
It seems to be different on Asus though...
maybe i try ocz 4000 gold on this MAX3 also just to know... :)
... This is, of course, assuming you are lucky enough to get memory that is compatible with either of them :rolleyes:
:D
lol noone listens too hipro :D
everyones is so hyped up by this Mhz-race they dont care about results anymore. Like bodybuilders, real musclestrength is worth nothing, volume is everything o_O
:D :toast:Quote:
Originally posted by karmakid
lol noone listens too hipro :D
everyones is so hyped up by this Mhz-race they dont care about results anymore. Like bodybuilders, real musclestrength is worth nothing, volume is everything o_O
Hey now... not "no one" , I'm with yoos toos ;)Quote:
Originally posted by karmakid
lol noone listens too hipro :D
everyones is so hyped up by this Mhz-race they dont care about results anymore. Like bodybuilders, real musclestrength is worth nothing, volume is everything o_O
:toast:
true...Quote:
Originally posted by karmakid
lol noone listens too hipro :D
everyones is so hyped up by this Mhz-race they dont care about results anymore. Like bodybuilders, real musclestrength is worth nothing, volume is everything o_O
Hehe,
I like the bodybuilder analogy.
Its not like anybody needs to have 25k+ 3dmarks, just as its pretty unusual for somebody to have to pick up a small family car on there own....
but if its just the thrill of the chase when wh dont we just OC GF4MX440 to save money.....
Although some people dont like to admit it there is a certain pride in being the best, or even being better than the guy sitting next to you.
Nice clock radelon :D :D
:eek: :eek: :eek:
woah....
I know this may sound stupid, but are there any tricks involved in getting the ram that high, mine doesn't really like to do more then 275@1:1 and it gave me lots of errors on test 5 of memtest86 when at that speed. (this is at 2.8) Any extra cooling on those things? :confused: