OCZ 8500 Low-Voltage Blade

[Shahryar_NEO]

why not too high ?! 54 is too high mate . don't believe test with that value and for example 32 and say your results !

of course FSB is so important . i think FSB is bottleneck at that test .

[Eagle_Watch]

a bit of OFF-TOPIC (im sure this is not related to the performance of ocz blade)

My results, same fsb, same strap, same memory speed than mkk. Higer latencys, 5-5-5-15 vs 4-4-4-15. TRFC is the same, 54, in my pl is higher, 6 vs 5 (see post 141)


Much better results for all memory tests, see now why i dont understand?

[18 is # 1]

Much better results for all memory tests, see now why i dont understand?

You changed the multi.
To compare fairly, same FSB & multi, different ratio and latency.
I usually see about 10% increase with faster RAM vs. tighter timings.

P.S. I don't think your tRFC is too high either
P.S.S. Don't forget this RAM stuff is fun and interesting, but nothing boosts performance better than over all CPU clock.

[Eagle_Watch]

You changed the multi.
To compare fairly, same FSB & multi, different ratio and latency.
I usually see about 10% increase with faster RAM vs. tighter timings.

P.S. I don't think your tRFC is too high either
P.S.S. Don't forget this RAM stuff is fun and interesting, but nothing boosts performance better than over all CPU clock.

thanks, i didnt knew cpu could have that impact on everests memory benchmark

[clo007]

Ok, as my motherboard will only run one stick of any ram, I am going to buy a set of these just so I can rip one of the modules apart to see what ICs they are.

EDIT: Ordered from newegg for $79 shipped

[zsamz_]

Ok, as my motherboard will only run one stick of any ram, I am going to buy a set of these just so I can rip one of the modules apart to see what ICs they are.

EDIT: Ordered from newegg for $79 shipped

put em in freezer first
why not rma your board lol
1 stick of ram bandwith sucks real bad

[CobraXP]

I got mine in today... going to try it out tonite.

[paul_]

Couldn't you just use the CPU IHS method of getting the heatspreaders off - that is, heat it up and suspend the module to fall off with gravity?

Anyhow, my eye is here squarely, and I am kicking myself for getting a new set of Reaper 1150 (my old set got hosed for no real reason grrr). I may be doing a quick RMA, if Newegg has the 92/9600s available..

[Eagle_Watch]

Couldn't you just use the CPU IHS method of getting the heatspreaders off - that is, heat it up and suspend the module to fall off with gravity?

Anyhow, my eye is here squarely, and I am kicking myself for getting a new set of Reaper 1150 (my old set got hosed for no real reason grrr). I may be doing a quick RMA, if Newegg has the 92/9600s available..

Bad luck, i used my reapers 2x2gb 1066 @ 2,3v 24/7 for a long time and 2,5v for benchs, and they doesnt even have signs of electromigration

[paul_]

2.28 is as high as they've ever been, and I can't seem to run them at even 1000... Does OCZ RMA?

[mkk]

I'm not sure I will use 2,15V anymore. Blades hate high temp. I noticed their oc power drops down fast when they become hot. Never saw max MHz to drop so fast before.

There was no problem with counting 4M at 1300 5-5-5-15 @1,9.
After 5 minutes needed to go down to 1280. Voltage was low but MHz made it hot.

Hope OCZ will use Reaper's cooler if the future. It may be necessary for 1250 / 1300 series. If they ever appear.

paul_ If I were you, I'd choose Reapers 1066 instead of 1150. All 1066 i had in my hands were better than 1150's.

[clo007]

put em in freezer first
why not rma your board lol
1 stick of ram bandwith sucks real bad

I don't have time to have my computer down for 1-2weeks waiting for an RMA right now.

BTW: 1 stick of 8800s I would not saying sucking at all.

[paul_]

paul_ If I were you, I'd choose Reapers 1066 instead of 1150. All 1066 i had in my hands were better than 1150's.

Now that I've seen these, they are what I'll be using, definitely. Any idea whether the 9200s will be better OCers?

[Eagle_Watch]

I'm not sure I will use 2,15V anymore. Blades hate high temp. I noticed their oc power drops down fast when they become hot. Never saw max MHz to drop so fast before.

There was no problem with counting 4M at 1300 5-5-5-15 @1,9.
After 5 minutes needed to go down to 1280. Voltage was low but MHz made it hot.

Hope OCZ will use Reaper's cooler if the future. It may be necessary for 1250 / 1300 series. If they ever appear.

paul_ If I were you, I'd choose Reapers 1066 instead of 1150. All 1066 i had in my hands were better than 1150's.

Not mines, they do 1066 with 1,96V, but only 1100mhz with 2,3v. And believe me, i tried everything timmings, straps, rcomps configurations, high and low vdimm, subtimmings, etc. I would recomend to choose corsair dominator instead of reapers in 2x2gb kits.

[kagula323]

anyone tested how they scale with high fsb? I think a lot of people is interested in knowing this

[railmeat]

Ok, as my motherboard will only run one stick of any ram, I am going to buy a set of these just so I can rip one of the modules apart to see what ICs they are.

EDIT: Ordered from newegg for $79 shipped

"I am going to buy a set of these just so I can rip one of the modules apart to see what ICs they are."

i ripped my spreaders off my older 1066 sli dimms apart for WAY better cooling with my fans on em..the spreaders themslves had a good design with holes but never worked good because of the THICK layer of clear tape completely covering the ics holding the spreaders on.ocz tony might argue the clear tape was thermal conductive etc...i guess.spreaders off and naked ics with fans on em is best.warranty voided but best performance acros the board...


i still have these sticks in my closet and work FLAWLESS.best runs were dfi 680i 4-3-4-10 1t 2.25 volts ddr 890

WHATEVER HAPPENED TO 1T???






few pics.

last year...







please let us know when u take them off....

just so u know ovoiusly u are voiding the warranty and the heatsink glue can rip your ics right off the pcb!!!!!u have been warned!!!!!

also it sounds like tony made sure the ics are blank so even with the heatspreaders off u may have nothing to look at except size of ics....waiting on your find.good clocks,reviews etc by all in here.



***********AND HAS ANYONE RAN 1T on these,and im not sure what chipsets these days offer 1t in BIOS NOT MEMSET windows software based crap ??????***********

[clo007]

I am willing to waste a stick to find out for you guys whether the ICs have info on them or not. I just think it is bad taste how Tony presented himself here.

This is for you guys.

[Mescalamba]

I am willing to waste a stick to find out for you guys whether the ICs have info on them or not. I just think it is bad taste how Tony presented himself here.

This is for you guys.

Use razor and carefully cut it out.. then iso-propyl or just pure alcohol (be careful to not touch PCB much). If you are careful enough, there will be perfectly fine..

[LuKe88]

i'm wondering just one thing .. the ocz platinum low voltage pc8500 kit mounts same chips as this blade kit ?

[HousERaT]

Just got mine from the Egg. Platinum with the black pcb. They look nice. Now to test.

i'm wondering just one thing .. the ocz platinum low voltage pc8500 kit mounts same chips as this blade kit ?

yes

[LuKe88]

thank you.
so i m gonna buy the platinum kit..
black pcb ?
niiice

[mkk]

why not too high ?! 54 is too high mate . don't believe test with that value and for example 32 and say your results !

of course FSB is so important . i think FSB is bottleneck at that test .

Especially for You I tried one more time with lower tRFC. Almost none change.
BTW. 1101 4-4-4-15 @ 2,15
One stick only second one didn't get it.

Need to use higher FSB for better bandwith. tRCF doesn't matter.

[railmeat]

[QUOTE=mkk;3817237]Especially for You I tried one more time with lower tRFC. Almost none change.
BTW. 1101 4-4-4-15 @ 2,15
One stick only second one didn't get it.

Need to use higher FSB for better bandwith. tRCF doesn't matter.




"1101 4-4-4-15 @ 2,15"


2,15 is gonna kill these?lol..thought i read above 1.9 does nothing for them?nice results tho...!

[mkk]

yes it can, so bench @2,15 and than back to 1,9

[tool_462]

yes it can, so bench @2,15 and than back to 1,9

The consensus so far is that nothing is gained other than heat with over ~1.9v.