UTT does only 2-3-2 - who's to blame?

[Lifthanger]

all right, I guess I found the main reason for all this problems.

One stick produces about 5 errors in test 6 and 7 at two diffenrent positions on the stick it self. at 158MB and 180MB. Seems as if this is one damaged chip on this stick, which caused all the trouble.

Does anyone know, why only 6 and 7 fails?

 

[g0dM@n]

That's a tough question to answer. You'd either be lucky enough to have someone who has a good amount of memtest knowledge to answer it here, or read up on it (a.k.a. Google.com).

I don't think that the fact that it's failing at the same MB necessarily means that it's a specific chip that's "damaged." I think it just shows you where the weakest link is. Have you tried underclocking it? I had UTT CH5 from the TwinMOS TMII sets and one of them always failed at the same MB, but not when I clocked it a bit lower than my high OC.

I would just RMA it man. Do it while the TwinMOS SP are still winbond.

 

[Moto7451]

My bro's AA4T chips do 250MHz stable after being burned in. Previously they could only do 247MHz fully stable He got a 1A4T 512MB stick which couldn't do over 228MHz at 2-2-2-5 which wasn't exactly optimal. I think having two 512MB sticks could be part of your problem. See what each stick can do individually. You may have one stick which can do more than the other.

 

[g0dM@n]

I already mentioned that to him.

I also told him to RMA it if it can't do stock, obviously

 

[Lifthanger]

yeah, thanks for the advice .

the memtest errors occured at 2.5-3-3-8 200MHz 2.8V - SPD settings.

 

[g0dM@n]

Send it back. I just got 4 more 1A4Ts in yesterday...

The 1A4Ts that are in my sig (in my current rig) actually were able to do 257fsb memtest stable at 3.56v, but I couldn't stabilize in windows at 250+. These 1A4Ts did 49 passes on memtest (full test for 20 hours) until I stopped it... that was at optimized defaults in bios, but with 2-2-2-6 timing at 1T. I could do 2-2-2 at stock voltage!!!

These things are the best bang for your buck. It cost me $198 to have 4 more shipped (4x512), and that's why I bought more. Cheap ram is like 75-80 for 2x512, where this stuff is only 20-25 bucks more... definitely worth it. Other companies, like Corsair, PDP Patriot, PQI, sell 2-3-3-6 2x512 kits at $120+. The 2-2-2 2x512 sets are usually $130-150.

 

[Lifthanger]

this shouldn't be a bump.
I just wanted to thank everyone who participated in this thread. It seems as if this stick was damaged all the time and they've already sent me a replacement, which should be here tomorrow.

Maybe I'll bother one of you in the next few days, because I intend to really overclock my system now.

 

[g0dM@n]

You sent it back to the egg, right? I have 6 of these, and haven't even opened two of them. They've been sitting in my computer drawer for over a week now. Man, I just don't have time for anything anymore

 

[Lifthanger]

I'm from germany, so I sent the bad stick back to cheeep.de .
I hope you'll find some time soon, to test your sticks.
price/value ratio is the best you could get - at least in germany.

 

[g0dM@n]

I paid $198.XX for 2 gigs shipped. I was watching the prices for weeks until I saw it go down to $47.00... they actually did a shipping combo too. Thank God newegg has that option instead of paying like $5 for each shipped. I still paid $11 to have 4 ram sticks shipped (ripoff), but still... altogether it was a good deal.

Let's just see when my life has the time to play with ram. =/

 

[Lifthanger]

so the replacement stick is here and I got 6 errors in test 7 in 15 passes, which is about 150 times less than the replaced stick. both at stock.

I guess this is something that could be helped by finding the right settings.

At the moment I'm trying them with your timings HouseRat. Let's see what happens.
g0dM@n, your timings wouldn't even get into windows, I guess it's because you've got bh5-die UTT.

I'll do some testing and if nothing works, I'll keep on whining in this thread .
Maybe I can find out what test 7 is meant to test.

 

[el]

post your settings in this format.

==========


Genie BIOS Settings:

FSB Bus Frequency - 311
LDT/FSB Frequency Ration - 3
CPU/FSB Frequency Ratio - 9
PCI eXpress Frequency - 100Mhz

CPU VID StartUp Value - 1.425

CPU VID Control - 1.300v
CPU VID Special Control - Avove VID * 123%
LDT Voltage Control - 1.40v
Chip Set Voltage Control - 1.70v
DRAM Voltage Control - 3.20v

DRAM Configuration Settings:

DRAM Frequency Set - 140 (DRAM/FSB:7/10)
Command Per Clock (CPC) - Enable
CAS Latency Control (Tcl) - 2.0
RAS# to CAS# delay (Trcd) - 02 Bus Clocks
Min RAS# active time (Tras) - 05 Bus Clocks
Row precharge time (Trp) - 02 Bus Clocks
Row Cycle time (Trc) - 07 Bus Clocks
Row refresh cyc time (Trfc) - 13 Bus Clocks was at 14
Row to Row delay (Trrd) - 05 Bus Clocks
Write recovery time (Twr) - 02 Bus Clocks was at 3
Write to Read delay (Twtr) - 02 Bus Clocks
Read to Write delay (Trwt) - 02 Bus Clocks was at 5
Refresh Period (Tref) - 1816 Cycles
Write CAS Latency (Twcl) - 1 was at 5
DRAM Bank Interleave - Enabled



DQS Skew Control - INCREASE
DQS Skew Value - 0
DRAM Drive Strength - Level 6
DRAM Data Drive Strength - Level 3
Max Async Latency - 7 nano
Read Preamble Time - 5nano
IdleCycle Limit - 256 Cycles
Dynamic Counter - Disable
R/W Queue Bypass - 16 x
Bypass Max - 07 x
32 Byte Granularity - Disable(8 Bursts)



==========

 

[g0dM@n]

so the replacement stick is here and I got 6 errors in test 7 in 15 passes, which is about 150 times less than the replaced stick. both at stock.

I guess this is something that could be helped by finding the right settings.

At the moment I'm trying them with your timings HouseRat. Let's see what happens.
g0dM@n, your timings wouldn't even get into windows, I guess it's because you've got bh5-die UTT.

I'll do some testing and if nothing works, I'll keep on whining in this thread .
Maybe I can find out what test 7 is meant to test.

In my opinion, if you run everything on OPTIMIZED DEFAULTS, then you shouldn't have to tweak a single thing to pass memtest. Every ram stick has to run at its rated settings with stock bios settings, or else RMA... that is what I believe is the situation with almost every computer part.

Keep RMAing if it's doing that, that is if you have other ram that doesn't have the same problem. If you can stabilize other ram in your rig, then that should show that the rest of your rig is fine... unless it's not liking a CERTAIN kind of chip?

Try all stock, with defaults loaded, and if that still doesn't work, then try one of DFI's official bioses... good luck, and let us know.

 

[Lifthanger]

I try chaging slots and things, maybe this caused the few errors.
As soon as I've got some useable results, I'll post back. This will take while.
Thanks for your interest godm@n and el!

btw. my first run two days ago was with 2.5-3-3-8 everything else on optimized on oscar wu's 5/10-2 200Mhz 2.8V Vdimm.
15 complete runs ended in 6 errors in test 7. all at the same adress in the same pass, if this is of any use.

 

[Lifthanger]

Finally I found a stable setting - if 7 hours/ 15 passes of memtest is considered stable:

Everthing on auto except the 3 main timings which I set to 2.
I tried to manually set settings which are working for others, but errors occured every time. Sometimes even in test 2.

FSB Bus Frequency - 245
LDT/FSB Frequency Ration - 4
CPU/FSB Frequency Ratio - 9 = 2.2GHz Memory runs on 5/6 = 200MHz
PCI eXpress Frequency - 100Mhz

CPU VID StartUp Value - 1.5

CPU VID Control - 1.425V
CPU VID Special Control - Avove VID * 110% or 113%? it's at 1.55V
LDT Voltage Control - 1.30v
Chip Set Voltage Control - 1.50v
DRAM Voltage Control - 3.20v

Here's the A64 tweaker shot since everything was on Auto:

I'll try to push higher and report back.

edit: would it be better to start a new thread about this matter, or is it all right to use this one, el?

 

[g0dM@n]

Isn't this your thread? Why would you need to start a new one when it's already yours.

 

[Lifthanger]

update:
all settings like in the last post, except now running at 9x240 1:1

here's the A64 tweaker shot with timings. This is 11 hours blend stable and 25 passes test 5 in memtest.

screenshot

I can't believe it's stable at these settings.Now I want to tighten those timings.
My question is: is there a way to test stability without letting prime run for 12 hours? Is test 5 memtest really reliable?
Overclocking requires so much patience.. if only I had enough .
It seems as if I'll finally get happy with my system. Great!
If all goes well, I'll start overclocking my cpu seriously (did stop at a stable 2.6GHz) and of course I'll bother you even more .
Too many smileys, but I just can't hide it.

edit: forgot to add, that I'm using the yellow slots now. Read everywhere that they're electrically superior to the orange ones. bios is 5/10-2.
I've also read that you should leave 0.2V difference between your 3.3V rail and vdimm setting, which I don't. Read this in mushkins recommandations for their redlines IIRC. What do you know about this?

 

[AMD-Me]

I am in the same boat, my redlines refuse to do 2-2-2, they only do ras to cas of 3

 

[g0dM@n]

Have you tried a different CPU or different MOBO? If not, have someone you TRUST try out your ram. If you want me to try it out I can send you collateral of some sort. I almost had someone to test out my 2x6600gt when my SLI was freakin' out... instead I had Krag test my SLI bridge... which he did and it came out alright.

See if you can find a friend to test out your ram to its potential; otherwise, see if someone here can do it (I would if you really wanted me to).