• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

TCCD Memory and High Voltages, a question

Joined
Mar 13, 2005
Messages
8
Harro!

My first [H] post and I'm looking forward to many more :D

Anyways I have a question reguarding most TCCD memory's voltage tolerances. I've recently ordered an OCZ DDR Booster to go with my MSI Neo 2 Platinum mobo (vdimm limited to 2.85V and I dont wanna hard-mod it) and my Kingston 2x512 PC3200 Samsung TCCD dimms.

I know Winbond chips (and OCZ's new VX line of RAM) are notorious for taking lots of juice, but what about the TCCD's? I have not found a review or post anywhere showing anybody running these memory types past 3.0 V and I was wondering if this is the maximum tolerance or if things just get messy past 3.0 V ?

If you have any personal experience on say a DFI board, or better yet a set up using the DDR Booster or a hard-mod, taking this ram beyond 3.0 V I would love to hear your experiences. What FSB did you reach at what voltage and what timings coudl you run? I'm interested in two areas here : 1) Highest obtainable FSB with tight 2-2-2-(?) timings and 2) highest possible FSB with loosened timings.

Or if you could tell me what NOT to take this ram too that would be great as well ;)

PS - I will have air flowing directly onto the dimms in case that helps. Thx :D
 
i have no actual experience, but i've seen time and time again that tccd doesn't like voltage.
with two exceptions. there was one week (437 i think?) that liked a bit more voltage.. 3v would still be around the max though
and i read something from an ocz rep about how the newest batch of tccd seems to do really well over 3.1v. the only problem is that if you don't know if you have this batch or not, you may kill your ram in the process of finding out.
 
Thx Eclipse that does help.

I think I'll see what I can get at 3.0 V before I try pushing further. If I'm satisfied with what I get at 3.0 V I'll put of quenching my OC curiosity until I'm ready to buy new RAM just in case I nuke my current dimms in the process :D

Given what I've read so far, I really wish I had some OCZ VX memory hehe. Always the same thing in this PC hardware world we live in. You buy something awesome and 2 weeks later something that much awesome'er is out :rolleyes:
 
well when i mean sensitive to voltage, i mean really sensitive. like you go .1v over the ideal voltage, and your max clock will be reduced.
 
Buy Twinmos Speed Premiums with AA4T serial number. They use the same chips as OCZ VX but half the cost and I've seen people use 3.6+ volts on them. I use 3.3.
 
TCCD (as the rest of Samsung's DRAM chip lineup) has integrated current-regulation circuits, called stabilitrons. A stabilitron is a very simple device; similar to a diode. What it does is let voltage of only a certain range pass. Too low, it won't work and too high, it will just dissipate the extra as heat.

The way it is now, TCCD is very "picky" about the voltage. For different latencies, and different bus speeds, there is a different optimal voltage, it seems.

Rumor has it that the mobile TCCD (revision "F", such as 503F as opposed to 503) does not have current regulation for some reason or another, so these sticks should respond to increase in voltage as well as our beloved Winbond chips do. Mushkin supposedly uses rev."F" chips for their pc3200 level 2 v2 sticks, and i've seen some people getting unusually high 2-2-2 clocks (>250MHz) with added voltage (3.3 I think?), but both were reviwers, so they might have got binned chips instead.
Spare-Flair said:
Buy Twinmos Speed Premiums with AA4T serial number. They use the same chips as OCZ VX but half the cost and I've seen people use 3.6+ volts on them. I use 3.3.
Of course, you can save a lot, but I wouldn't expect the same kind of performance as you can get from OCZ's VX. The thing is, TwinMOS machine-tests their chips for 200MHz, while OCZ hand-tests them for 250MHz (impossible to machine-test). A small fraction of chips actually pass the test for 250MHz... so you'd typically go through several sticks of TwinMOS ram to find ones that overclock well. Don't know if it's worth the while to you... Personally, my time is more expensive.
 
excellent input gentlemen but he asked the same thing i have asked over and over.

what are the max clocks regardless of timings.

i just got a gig kit of this and max overclocking is what i am looking for. even if i have to run 2.5-3-3-X no one seems to be wnating to answer that. if all im gunna get is 250-260 then this ram is absolutely worthless. regardless if its 2-2-2- or 2-3-2 i need high mhz
 
My gskills are RATED 2.5-3-3-7 @ 275MHz. They do about 282 and 1T on my setup. So many reviews on all this stuff. On the good stuff 2.5-3-3 is good for 265-290 normal range and 4's can often get you into the 300's but I don't think it's worth it. Also, some manufacturers of this stuff, sell high speed bins for a premium. While others just market pc3200 at 2-2-2. If you are after fsb at ok timings I might go for one of the binners (OCZ, GSkill all have DDR550+ with TCCD)
 
Sorry to interrupt, but I am working on finalizing my OC, so is 278 memory speed @ 3-4-4 timings with a 1:1 divider or 250 memory speed @ 2.5-3-3 with a 9:10 divider better? My ram goes up to 280 mhz at the highest, but it needs some pretty lax timings, so is faster speed and 1:1 ratio going to outweigh the gain in latency?
 
maybe, maybe not. it really depends on the things you do.
my advice? run some benchmarks you thik are worthwhile. do some things you normally do that you can time accurately. see which one is really faster. ;)
 
Back
Top