Corsair 800ul vs 8500

[proth]

Have been running this ul since it first came out, but looking for a little more performance, possibly the new 8500. Does anyone know if the 8500 OCs as well as the 8000ul or if the same D9s are used.

[metro.cl]

wait a bit more new ics are coming then you'll see a nice improovement

[Steensen]

I'll recieve 4*512 8500's in a week or so. gonna clock a lot, so ill come with some results then.

[boostedevo]

proth, I'd hang on to those 8000s for the time being. Until we start seeing some results from the stuff coming out, it may still not be as good as the D9s for tight(er) timings even though the default speeds are increasing.

It seems 8000ULs are starting to pop up here and there again...I ordered a pair and they should be here today. Anyone have any idea if they are still D9s or have they switched to some "comparable" ICs? My Mushkin XP6400s are D9s, so I was hoping they would play nice together for 2GB total with pretty decent timings, but now I am not sure what the 8000s are going to have and I really don't feel like pulling the spreaders on new sticks to find out.

wait a bit more new ics are coming then you'll see a nice improovement

I am really kinda anxious to see the new stuff too...particularly what the Corsair 8500s will do in "real life". If they can do like DDR1100+ @ CAS4 with upped voltage, that would be nice. OCZ is listing some DDR1000 CAS4 now and I saw DDR1100 CAS5 from them, but no word on their site about it yet. Mushkin seems to really be dragging their feet with their new stuff and they're really the ones I was waiting for.

I gave the XP6400s a test to see how they are holding up after 8+ months of abuse from me and they seem just as strong or stronger than when I got them. I didn't try max, so that's still pretty much what is in my sig, but this was what I got just today. I was too lazy to bump the voltage further to see how far they would go at these timings, but this was it at 2.60V...definitely the tightest it's ever run at DDR800 (does up to DDR1066 4-3-2-4 @ 2.65V). I am dying to see if the new stuff bumps this up performance up...



Looking forward to seeing the results of the 8500s!

[proth]

I'll recieve 4*512 8500's in a week or so. gonna clock a lot, so ill come with some results then.

Good luck with them Steensen. Hoping for you @ DDR1100 4,3,2,8

@boostedevo (LOL). Nice timings man.

[boostedevo]

Hoping for you @ DDR1100 4,3,2,8

Don't need 8500s for that, but that would be SUPER impressive on 2GB and 4 slots to boot. I'd sell all my D9s for 2GB that would do that I finally had some time to test out my new 8000ULs:



@ 2.55Vddr DDR1100 4-3-3-4 benches SPi's w/o errors. As soon as I get my DI tube back, I'll do some 5.7-5.8GHz SPi's, naturally

[Hazzan]

Don't need 8500s for that, but that would be SUPER impressive on 2GB and 4 slots to boot. I'd sell all my D9s for 2GB that would do that I finally had some time to test out my new 8000ULs:



@ 2.55Vddr DDR1100 4-3-3-4 benches SPi's w/o errors. As soon as I get my DI tube back, I'll do some 5.7-5.8GHz SPi's, naturally

This your 8000UL ? ehmm Fantastic Versi ? 512 x2 or 1G x2

[Gautam]

Don't need 8500s for that, but that would be SUPER impressive on 2GB and 4 slots to boot. I'd sell all my D9s for 2GB that would do that I finally had some time to test out my new 8000ULs:



@ 2.55Vddr DDR1100 4-3-3-4 benches SPi's w/o errors. As soon as I get my DI tube back, I'll do some 5.7-5.8GHz SPi's, naturally

Fantastic work, my friend!

[althes]

amazing

[boostedevo]

Thanks everyone!

@Hazzan, yes they are mine...just got them at the beginning of the week (brand new) and they are v1.1.

[b1tterman]

performing pretty well:beer:

[boostedevo]

I am happy with them

@ Hazzan, missed it: 2x512MB. I don't think they have any 2x1GB ULs

[proth]

Damn Boostdevo, reminds me of my old drag racing days, always getting beat by a fender, or maybe a door with that 570 @ 4-3-3. Super timings there man. Time to pull out the tube and trim off another 30 seconds of SPi32m for OCF.

[Steensen]

This was what I was able to score (with minimal tweaking) with my system Monday night GMT+1
The 2001Se I got tweaked up to 32290.

Enough OT.
These 8500 are no good.
I cant eaven get them to run 1066 5-5-5-15 with the 0101 BIOS, but they are 3D stable and 32M stable at DDR1000 5-5-5-15 2.3Vdimm (like in pics attached). They freese memtest86 in test 5 around 25-40%, if they are clocket any further than DDR2-1000.
Perhaps it's just me that's no good tweaking my memmory, but these 8500 RAM are no good IMO. Haven't tried with only 2*512MB yet (haven't got thetime).

[kone]

Steensen : Give it more voltage and test 512X2 ?

[boostedevo]

[quoteDamn Boostdevo, reminds me of my old drag racing days, always getting beat by a fender, or maybe a door with that 570 @ 4-3-3. Super timings there man. Time to pull out the tube and trim off another 30 seconds of SPi32m for OCF.[/quote]Run bracket races if it's always that close Those are the best races though...nothing fun about winning or loosing by a lot. Close competition keeps you focused I might run the tube this weekend and stick the 930 (15x) or 550 (14x). Looking forward to your results on voltage!

@ Steenson, sounds like a bad pair maybe if they won't even run at specs? How much voltage are you running on them?

[proth]

@ Steensen.. Problems here also with 8500. Maxes out @ around 1180 5-5-5, 2.7v doesn't help. Only SPi stable so far.
@ Ross..Got ya' by 2. Still 8000ul. 8500 seems to ...uh...SUCK. Ran these @ 2.6v


http://valid.x86-secret.com/show_oc?id=84309

[boostedevo]

Very nice proth! That's actually really excellent for 2.6V. I'll have to see if I can squeeze a couple more FSB more out of it or maybe try lowering .1V.

That stinks that much extra voltage doesn't do anything on the 8500s...was hoping for some big speeds out of them considering CAS4 isn't an option.

[proth]

Just be careful with them D9s there buddy. I don't think you will find another pair. I may have a few more MHz if I get some cold air back on them.

[boostedevo]

I meant trying a lower Vddr and see if that's all it needs there. It was at 2.7V for the 1140 I had. I have the XP6400s as a back up, but I was thinking about selling them so I'll make sure not to kill the 8000s

[Steensen]

Steensen : Give it more voltage and test 512X2 ?

Haven's moded My MoBo, so can't give any more than 2.4Vdimm, wich makes no difference.
Try out 2*512MB another day, since I haven't got enough time currently.

Maybe I should start getting my MoBo moded, so the Droop, TT and VDimm is no problem any more.

[proth]

I didn't see any differency in the 8500 going from 2.3v to 2.6v, sorry. But I understand ever pair is different so maybe you will have better luck. If you have the P5WD2-E, you MUST doo the droop mod above 4.5GHz. It works like magic, but is a little tricky because of small SOT parts. The Vdimm mod is quite a lot easier. Good Luck with it. You have a very nice system

[boostedevo]

Steensen, that's just not right. I mean they are not even running at rated specs (DDR2-1066 5-5-5-15 @ 2.2V). It should not be up to you to add more voltage and mod the board on top of that to do it. I'd mod it anyway since it will be beneficial down the road, but it sounds to me like more Vddr isn't going to help in this case.

If anything is a legitimate RMA, those sticks are. There's no telling if you'll get sticks that are any better just swapping them under RMA. You might want to see if you can do a return RMA (or return to where you purchased as defective) and look for something else?

[crackhead2k]

Lets see the Benchmarks for both these chips so we can actually compare.

[Steensen]

Proth THX m8, But I haven't got the nerves to do the mod myself.
But yet again I'm only running 4.2, so it's close, but no cigar.


boostedevo That's what I thougt too, but yet again, the Dimm mosfet can be a bottleneck, and not capable of supplying enough juice to the dimm's, and therefore they cannot run the specs.
Try out two dimms this weekend I think, otherwise, I'm going to RMA the dimms (all four of them), to see if that helps.
Just weird that the dims runs super stable at 1000, but 1010, then they want complete a memtest86.