• 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.

Twinmos 3500 & XP2500-m

S

slunge

Limp Gawd
Joined
Apr 1, 2003
Messages
216
Ok, the reason im putting this thread up is for 2 reasons, one to record whats going on with my system as i find / fix problems, and two so anyone with any experience who fancies contributing can chip in anything they like... (thx :) )

Basics

Upgraded from:
xp1700 (JIUHB) which did 200*12 @ 1.8 V
2x256 3200 twinmos 2.5 cas (2 years old which turned out to be bh5) which did 200 2-2-2-5 @ 2.8V | chip serial: 3125E W942508BH-5


Upgraded to:
xp2500m (axmh2500fqq4c aqyha0430)
2x512 3500 twinmos speed premium | chip serial: 05154 BH-5 TM07608F8E44B


Other Gear
MB : NF7-S rev2.0
Gfx : 9500np > Soft9700
HDD : 1x160GB SATA WD 7200 8mb 1x80GB IDE WD 7200 8mb
Cooling : Heatsink SLK800 w/ Sfan2, Fans 2x front 80mm in, 1x 80mm blowhole, 1x 80mm rear, vga silencer on gfx card.
PSU : 400W X-Pro Super Silent PSU (with 120mm fan, output specs available if need be)

The problem is, i was running my xp1700 day in day out at 2.4 ghz, old ram at 200 2,2,2,5 the whole thing was stable as can be... I was then given the chance to upgrade to a 2500-m and 2x512 ram so did hoping that the same speeds shouldnt be too big a problem... wrong

Will post as testing is done.
 
Running memtest x86 on the memory from bootable cd.

Initally failing all memtest runs with memory in slots 1,2 (asuming slots are numbered 1-3 from psu-cdrom), after changing to slots 1,3:

Inital tests at 2.9V 200 2.5-3-3-8 (SPD) pass 10x run of test no.5

2.9V 200 2.2.2.11 fail test no.5 with 10 errors+ - allowing to burn in during the day at 3.0V
 
having gone home at lunch there was a shed load of errors but i think it was because it was still up at 11.5x200

Having turned it down to 5x200 2,2,2,5 @ 3V, 2 errors a pass but at 2,2,2,11 @ 3V it seems to be passing, im starting to think the mobile barton is going to be worse than the class xp1700 i had before :(
 
why did you choose to upgrade the cpu at all if your old one was hitting 2.4??

do you have that 2700+ still.. you could pop it back in for testing..
 
well i passed my memory ram and an old hdd on to a m8 for a pc i built for him, i just figured that seen as the mobile 2500's were clocking so well... i could pop one of those in and just benifit from the extra cache on it.. most ive seen are able to do 2.4 with 200 fsb same as my 1700 does.. my plan of action is to grab the memory out of my old one if need be to test this cpu, then if i really have to.. grab the 1700 back in and test on that.

back on track, just got in from work.. and at 5x200 2,2,2,11 3V the ram passed 129 passes off memtest #5... gonna start slowly ramping up the clock speed with these timings now.

The question is how big an effect will the clock speed have on the errors produced in memtest? (i know they would have a big effect on prime but am fairly new to MT)

thx.
 
well having passed at 2,2,2,11 i tested briefly at 2,2,2,5 and a day or so's burning in seems to have helped memtest slightly, passed a few tests before i got impatient.

so in to windows at 5x200 2,2,2,11 (that its just passed an entire days worth of memtest), a brief try of prime... and fails within a minute or two :(

Ive now swapped over my ram, the 2x256's have been put in to it, as i know for sure their limits, time to spank this cpu some for a little while. Testing at 5x200 2,2,2,11 with the 2x256 atm.
 
having swapped back to my old ram, the mobile xp seems to be working fine...

just primed for 24 hours at 2.3 ghz 11.5x200 and managed it in to windows at 2.6 without any trouble, so the problems down to the ram, going to test it out at spd settings but i dont think it was making it even at that.. also this afternoon there is a strong possibility of a Vdimm mod being done to rape the ram alittle.
 
after wasting my weekend checking this and that, im 100% certain the ram is the problem now, voltmodded my motherboard over the weekend to see if it would help slightly which didnt (now capable of putting 3.25 Vdimm through).

checking the sticks individually at 217 2.5-3-3-8 (spd) both pass memtest but in dual channel they fail with 20000+ errors.. any suggestions where i can go with this now??
 
You must log in or register to reply here.
Back
Top