Cas 2.5-3-2-1-1T, Stable... at DDR508... at 2.2 Volts.

Quote:

---

Originally Posted by oohms

I have 2x512m corsair 3200 cas 2 ram rated at 2-3-3-8 that has Promos -5 Rev A chips on it.
They also work best at 2.5 volts but i will try and get lower voltages :D

This is pretty much what they are capable of (test #5)
2-3-3-8 215
2.5-3-3-8 255
2.5-4-4-8 255
3-3-3-8 280
3-4-4-8 300

They are stable for about 20 minutes, until they heat up (a LOT) and then they start producing 1 or 2 errors every pass or so...
Is it worth removing the corsair heatspreaders (the standard platinum ones) or are they actually helping in this case?
The rams are also right next to each other on the mobo but not a lot i can do about that :(

---

Wow awesome results from those Elcheapo Promos IC's :slobber: Do you put a fan on top of the memory's ? And removing the heatspreaders helps most of the time a little bit and sometimes it doesn't matter. You want to keep them on for the warranty, that's a personal choice.

My 2 x 1Gb Infineon BE-5 @ 2.5volts lowest DFI will go for me 510mhz @ 2.5-3-2-0:

http://valid.x86-secret.com/show_oc?id=75197

This is 32m Super PI stable and gaming stable.

Would it be possible to mod A DFI Ultra-D bios to give lower voltages than 2.5?

Everything is possible in a certain way, but I doubt that it'll be easy.

Perhaps DFI would add these options down to 2.2V or even lower, if only enough people would ask the support to do so. :) Could be worth a try.

Edit: Just checked the latest BIOS file for the NF4 Ultra-D and thee's no hidden option for a lower Vdimm than the defualt 2.5V. :( So, if you don't find a BIOS guru to fix that, the only way will be to really ask the DFI support for help (or do a Vdimm Mod :D ).

Quote:

---

Originally Posted by oohms

I have 2x512m corsair 3200 cas 2 ram rated at 2-3-3-8 that has Promos -5 Rev A chips on it.
They also work best at 2.5 volts but i will try and get lower voltages :D

This is pretty much what they are capable of (test #5)
2-3-3-8 215
2.5-3-3-8 255
2.5-4-4-8 255
3-3-3-8 280
3-4-4-8 300

They are stable for about 20 minutes, until they heat up (a LOT) and then they start producing 1 or 2 errors every pass or so...
Is it worth removing the corsair heatspreaders (the standard platinum ones) or are they actually helping in this case?
The rams are also right next to each other on the mobo but not a lot i can do about that :(

---

That sounds reaalllyyyy interesting! :eek:

BE-5 :up: but super hard to find.....
and BE-5 looks like no 3D problem

does anyone know what ic's rev 1.3 is. as it needs volts to go higher. and fails if i lower the volts to low.
2.5-3-2-5 245 2.6v fails
2.5-3-2-5 245 2.71v success

Quote:

---

Originally Posted by Redbeard

I work for Corsair, this is exactly right. We initially used the Infineon revCs, but when they were in short supply, we used the Infineon revB ICs.

Right now it really depends on what we have in stock, we're backlogged on the 3500LL because it's such a popular part, luckily we have been able to buy a buttload (technical term) of ICs over the past week or so and it's made a huge dent in our backlog. Keep patient, the modules are shipping out every day by the hundreds.

As for the revB vs. revC, I know revision 1.0 is supposed to be revC, but if you post your revision number in this thread I will let you know for sure which IC is on them. (To the original poster, not to everybody else, that will be a huge headache since I would have to look them up each time...heh.)

---

seems like this is what the ic' are.
rev 1.0 Infineon -5 Rev C
rev 1.1 Infineon -5 Rev B
rev 1.2 Infineon -5 Rev C
rev 1.3 Infineon -5 Rev C

I have rev 1.3 Infineon -5 Rev C

fall back to CAS 3 and press on.

Quote:

---

Originally Posted by Tulatin

fall back to CAS 3 and press on.

---

i will see what i can do. thanks .

See you @ 280 :toast:

testing at 275mhz 3-3-2-8 7-14 1T 2.87vDIMM TREF-3120 all other setting at auto.

any hints if i hit a wall?

Quote:

---

Originally Posted by M_com

seems like this is what the ic' are.
rev 1.0 Infineon -5 Rev C
rev 1.1 Infineon -5 Rev B
rev 1.2 Infineon -5 Rev C
rev 1.3 Infineon -5 Rev C

I have rev 1.3 Infineon -5 Rev C

---

So is there any other dif. than IC's from rev. 1.2 to 1.3? Why did they change the revision nr?


Mine are rev 1.2 and @ 2,6V, I can get 229mhz@2,3,2,6 memtest stable and 227mhz@same in the memtest #5 test. (havn't tried higher@#5 but I rekon 227 is as high as the can go)



Don't stop posting this thread is really something. And where's that rep guy from the pirate boat company, where U at? ;)


@Kunaak. How you tried to find out the ICs and revision nr with help from the Corsair rep?

Thanks...

v1.2 has to be Infineon BE-. most 3500LL Pro review samples were v1.2 and they were all topping around 250MHz with no gains in MHz when going from Cas 2.5 to 3. that's typical for rev B.

and BTW i don't think v1.0 exists.

Huh, those sticks are freakish .. I have a 2x1gb set of Aeneon (infineon) sitting here ;-) maybe I'll see what's up *cross fingers*

Hi Kunaak,

so far I've managed this, running ram @ 216mhz, 225htt, 2.5-3-2-6 with my 4800X2 cpu,


running ram @ 216mhz gives an effective memory speed of 245.5mhz, this is @ 2.7v. I have not tried higher.

I don't think my ram has the ability to go as high and tight as yours though, I am using DFI CFX3200. 2.7 ghz is completley stable, I can get upto 2.8 and run super pi 1M, but it is not satble in S&M stress test with a 100% load, as I am only using air cooling.


Raja

edit I'll try knocking down a bit, as this board allows right down to 2.01v for the ram..

well give it a shot...
I been wanting to see what happens on that board.
unfortunatly, theres about 3 revisions of this ram, and I have no idea what mine is, cause the paint on the sticker wore off a long time ago, as I said a few posts ago, and Bachus showed as well.

so don't be suprised if your ram doesn't do low voltage overclocking...
so far I have only seen 2-3 other people get better OC's with lowering thier volts instead of going higher.

but it's certainly worth a shot.

ive got the rev2 3500 pro 2gig sticks

im 3d mark and super pi stable at

260 2.5,3,2,0 1T 2.6v

if that helps anybody

Hi,

It's revision 1.3, I bought it this year, from what I hear, it probably won't overclock as well as some of the older stuff.


ATM I've found that I can use it @2.3v with the same settings.

Raja

Tryin to get rev1.3 of these sticks stable at 2,3,2,0 at 223 mhz and cant. Board is dfi expert. Seems to need more volts rather than less to help stability and have followed bios settings as in 1st page screen shots. Any ideas to help stability. Am using orange slots on board.

very unique

Quote:

---

Originally Posted by CandymanCan

From what im getting @ Infineon chips like low voltage. Maybe i should try this.

Setting youre refresh period to 4708 will lower youre memory bandwidth btw. You may get a higher overclock but itll just contridict itself

---

Isnt it the other way round: 4708 makes it faster than a lower refresh rate. I have got faster memory speed in memtest by raising this value - only lower it if having problems getting stable.


Finally got my rev1.3 sticks stable. Couldnt get them stable at 223 2,3,2,0 no matter what i did. I copied original bios settings in first page and didnt help - lowered volts (made it worse but higher volts improved it but didnt make it stable). Now running with with 2.6v stable after setting drive strength to 6 - made a massive difference to stability with my sticks so might be something for others to try if having probs with rev1.3 sticks