Need info on old TwinMOS Winbond CH-5 chips

Recently I got a Dual Kit of 2x512MB TwinMOS PC3200 RAM with Winbond CH-5 chips (production week 44, year 03). A week ago I changed my setup to DFI nF4 Ulta-D
and 3000+ Venice running the same RAM sticks.

I would like to know, are there any reviews on the web of
these old CD-5 chips (I fould only one at http://www.vr-zone.com/reviews/TwinM...3200/page4.htm)

Is these RAM suitable for good overclocking (it has similar markings on the chips as the one in the above review). My
chips also have 2 dimples on the center of the chip and on
dimple on the lover corner - these sould mark good overclockable chips.

Also, regarding the voltage, is it safe to put them on 3.2v (DFI supports also voltage up to 4v - however I do not intend to run that high, max 3.5v) or could I destroy the
sticks with such high voltage?

Welcome to xtremesystems.org.

How about using the search function? If you would have taken the time to use the integrated search function of these forums, you would've found out, that these CH-5 Chips do about the same as BH-5 (for example 250Mhz 2-2-2-x with 3.3V), when fed with enough voltage (as already said, above 3V).

So, these chips have great overclockability, but you have got to actively cool them when running high voltage through them, and I would advise you not to go any higher then 3.3V Vdimm. The DFI board seems to have some problems with vdimm BTW. I don't know if it's really DFI's fault, or if other manufacturers or even the users are to blame, but I would be careful with this board. ;)

Thanx for the welcome :)

I tried using seach option however, everything I found was
related to the new UTT CH-5. Maybe I was not looking hard enough. BTW, the old CH-5 and new UTT CH-5 are they related in any way?

As for the board, I intend to run max 3.2v for the time being .
I need to buy a better PSU (now I have Chieftec 420W) and
DFI needs to fix the 5v jumper problem.

Any other (older) reviews for older CH-5s?

The reviews of old CH-5, which is IMHO the same as new CH-UTT, won't help you much, as the reviewers never used more than 3V, as far as I know. So they didn't achieve great results and everyone thought that 2-2-2-x timings on CH-5 was not possible. :D
With 3.2V you should get something like 233-250MHz out of your CH-5. ;)

I would be Extremelly :D happy with 250@2-2-2-5.
Will try it - but first I need to put some ram cooling
into the box (the wife will be extremlly happy when
I will be openin the computer again) :banana:

The problem of CH-5 has been the magical 2 (2-2-2) I dont think you wouldnt archive 250MHz but using 2-3-2-x and a loose tRC + tRFC.

Quote:

---

Originally Posted by Aleman

The problem of CH-5 has been the magical 2 (2-2-2) I dont think you wouldnt archive 250MHz but using 2-3-2-x and a loose tRC + tRFC.

---

Had some Twinmos CH-5 here, some time ago and it did 250MHz at 3.3V with the exact same tight timings as any BH-X RAM that I had [CAS Latency: 2, RAS to CAS: 2, RAS Precharge: 2, Cycle Time (Tras): 10, Bank Cycle Time (TRC): 12] . But we'll see. :) That was on a Shuttle AN50R BTW.

I've got 4 sticks of TwinMOS UTT with the CH-die..... all four can do 260 with 3.4v 2-2-2-x timings. The BH-die stuff is a little harder to get that high.....

no he asked about OLD CH5....

Do a search for Kingston KHX3500 alot of that was old CH5, and the results/settings will help you out.

I got 2x256 CH5 to hit 245 11,3,3,2.5 on NF7-S @ 3.45Vdimm But 2x512M of it wouldn't go past 220-225. 2x512M also likes CPC OFF.