**Official DFI LanParty UT X48-T2R(and LT) Discussion

[MomijiTMO]

I'm planing to use the BP Compression on them, what problems you anticipate?

I will use the Loctite Blue Thread locker do just to be safe...

I used the normal BP fittings but because it's really wide the o-ring did not form a complete seal and I had a leak from the get go. So if the fittings don't work, you'll see it straight away if you leak test. Hopefully yours will work well though

[Seanie's Show]

Now im getting somewhere with this ram, although memset states its running PF lvl 7, I have the pull ins enabled or what ever they are pulling it down to PF lvl 6, had to raise the NB slightly, set 1.57v in the bios and the bios giving me a reading of 1.58v for the NB, SG is reporting 1.53v but as you see from the pict, temps are still good, and this is with the stock cooling, and a 60x60x10mm fan on the NB radiator, no other cooling on the board at all, water cooling the CPU, overnight 12hr run of memtest pro.

Would like to know though if its possible to hard set the tREF Refresh Period in the bios anywhere, as each time I reboot this defaults itself back to 2600.

[x0vash0x]

So, got a problem.

I can do on an 8600 475 x 10 perfectly stable, but if 500 x 9.5 is completely unstable. Any reason why the multiplier would make the system unstable?

[zsamz_]

So, got a problem.

I can do on an 8600 475 x 10 perfectly stable, but if 500 x 9.5 is completely unstable. Any reason why the multiplier would make the system unstable?

give it more nb volts 1.5-1.6
n chek your memory
try updating your sig too

[x0vash0x]

give it more nb volts 1.5-1.6
n chek your memory
try updating your sig too

Yeah.... I'll update it soon enough.

NB 1.65 Memory is 2.1 & DDR2-8000. I can't get anymore out of the NB, anything over 1.65 just freezes the system.

[Beefy22]

give it more nb volts 1.5-1.6
n chek your memory
try updating your sig too

Does it really need that much for 500FSB? I know it's not an apples to apples comparison, but IIRC I can get my Q6700 bench stable @ 450FSB x9 with just ~1.35 vNB. Though I might be mistaken, I'll see if I can run that setting later on.

[zsamz_]

Yeah.... I'll update it soon enough.

NB 1.65 Memory is 2.1 & DDR2-8000. I can't get anymore out of the NB, anything over 1.65 just freezes the system.

you should be able to do 500-540fsb with 1.52v on nb
just tweak the gtls to 61-63%
and use 333 memory divider
lower memory volts to 2.02v
n dissable legacy usb support
+dissable cpu features but not the cores

[x0vash0x]

you should be able to do 500-540fsb with 1.52v on nb
just tweak the gtls to 61-63%
and use 333 memory divider
lower memory volts to 2.02v
n dissable legacy usb support
+dissable cpu features but not the cores

I have 8GB RAM, so it needs that much on the NB.

[zsamz_]

I have 8GB RAM, so it needs that much on the NB.

you wanna run stock you dont need much nb volts lol
i ran 1,62v for 4 sticks but i was doin 500fsb n memory runnin @ 1200mhz
i wouldnt use less than 1.5v for 8 gigs unless i runnin stock

[adya]

you should be able to do 500-540fsb with 1.52v on nb
just tweak the gtls to 61-63%
and use 333 memory divider
lower memory volts to 2.02v
n dissable legacy usb support
+dissable cpu features but not the cores

zsamz, whats the go with dissabling legacy usb support? how does it effect overclocking? if i dissable it will it effect usb functions?

[zsamz_]

zsamz, whats the go with dissabling legacy usb support? how does it effect overclocking? if i dissable it will it effect usb functions?

dissable usb support =dissabling booting from usb drive
if you dont dissable it you gonna get errors in memtest where you shouldnt
so i think its more stable

i was doin some quick testing on my gskills pc2-9600 2x2 set
with it enabled i got errors
i dissabled it n i got them to 1300mhz with 2.1v
i wanted to test more but i got no time
i moded an accelero v2 for the nb n the nb hit 60c
soon i gonna mod a xigmatek 1283 n slap it on the nb
it performs almost as good as a true
i gonna start usin my x38 soonto see how good it is with the gskills 9600

[x0vash0x]

dissable usb support =dissabling booting from usb drive
if you dont dissable it you gonna get errors in memtest where you shouldnt
so i think its more stable

i was doin some quick testing on my gskills pc2-9600 2x2 set
with it enabled i got errors
i dissabled it n i got them to 1300mhz with 2.1v
i wanted to test more but i got no time
i moded an accelero v2 for the nb n the nb hit 60c
soon i gonna mod a xigmatek 1283 n slap it on the nb
it performs almost as good as a true
i gonna start usin my x38 soonto see how good it is with the gskills 9600

Well, as it stands right now, I'm running trying to run FSB 495 x 9.5 (4.702) with a 1:1 divider 333/667 strap.

I can do 470 x 10 (4.7) 1.66 NB 1.44 vCore 2.1vDIMM 1.00 vtt perfectly fine, but when I use a 9.5 multiplier the computer locks up. I've increased the VTT (1.00 -> 1.04) and it helps to be able to use another divider besides 10. However, I'm having trouble getting 9.5 x 495 stable. 495 x 9.5 is doable, I just need to figure out how to get my system stable, and it's driving me crazy figuring out what I need to do.

[CERO]

which bios u guys have found out to be the best for E8600?

[adya]

hey i finally found some time to play with my E8500.
im going for a 24/7 clock first so i can save it for later.
i managed to get past 4ghz ,but not stable. not sure if it wants more volts or less.
the more i give it vcore the more unstable it gets same with vtt?
what do E8500s like at this stage?
i feel like ive tried almost everything lol

vcore 1.4v
vtt 1.350 120 110 90
1.250 110 105 85
vnb tried between 1.5 - 1.6v
vdim 2.1v-2.3v
sbv 1.6v

[zsamz_]

i started playin with x38 and e8600
the .5 divider is a verry big pain anyone else?

[x0vash0x]

i started playin with x38 and e8600
the .5 divider is a verry big pain anyone else?

Trying to get any divider other then 10 to be stable with x48 and e8600 was a complete pain in the ass. What worked for me was raising the VTT. I needed vtt 1.114 to get 9.5 x 495 = 4.702

What's interesting though, is that 10 x 470 was completely stable with vtt of 1.00

[Praz]

Trying to get any divider other then 10 to be stable with x48 and e8600 was a complete pain in the ass. What worked for me was raising the VTT. I needed vtt 1.114 to get 9.5 x 495 = 4.702

What's interesting though, is that 10 x 470 was completely stable with vtt of 1.00

Going from 470 FSB to 490 requiring an increase of VTT seems obvious to me. Don't really see where the issue is.

[x0vash0x]

Going from 470 FSB to 490 requiring an increase of VTT seems obvious to me. Don't really see where the issue is.

Not everyone has the experience that you have.

[zsamz_]

board is acting weird with e8600
but it needs less voltage that ut p35
i had slapped on an accelero but now i put a xigmantek 1283
1.36v cpu n 2.06v ram "gskills 9600 2X2" 1.62v nb got me this

[zsamz_]

Going from 470 FSB to 490 requiring an increase of VTT seems obvious to me. Don't really see where the issue is.

no need for more vtt to goto 500fsb
i'm havin hard to=ime to pass 525 and me e6850 can do 555 on the board
the board just actin verry weird with e8600
i gonna try out latest bios
i;m on 602 n as soon as i use .5 on cpu board goes berzerk took me 15 min to get it to start again i had to take out bios chip n put spare chip to get it to boot even led went dead no code

[Praz]

I needed vtt 1.114 to get 9.5 x 495 = 4.702

no need for more vtt to goto 500fsb

As an increase of VTT is what allowed 495 FSB to work evidently is was needed. Users would have a much easier time not thinking all components react the same and overclock settings are universal.

i;m on 602 n as soon as i use .5 on cpu board goes berzerk took me 15 min to get it to start again i had to take out bios chip n put spare chip to get it to boot even led went dead no code

Post back if a different version BIOS makes a difference.

[zsamz_]

i updated to latest bios .5 is ok now
but board still verry weird need bios update for e8600
i cant get 525fsb stable with e8600
with e6850 555fsb is ok tested 540 with e6400
with e6400 i got gskills 9600 to 1333 stable n with e8600 cant go past 1270 lol
with ut p35 never went above 1.22 vtt i even went as far as 1.4 n lol

tired settled on this 4 now
correction :515fsb is max for e8600 with latest bios
this is bull doudou better cpu should be better not worse
should i smash x38 go back to p35 or smash e8600 n put on e6400 ridiculous

[adya]

Trying to get any divider other then 10 to be stable with x48 and e8600 was a complete pain in the ass. What worked for me was raising the VTT. I needed vtt 1.114 to get 9.5 x 495 = 4.702

What's interesting though, is that 10 x 470 was completely stable with vtt of 1.00

1.114 vtt is pretty low though isnt it? for 495fsb?
i need 1.35 vtt for 475fsb + if i lower it i loose stability straight away.

edit *sorry , misread ur post. im amazed then that u managed to reach that with such low vtt.
im going back to the drawing board lol i want less vtt and vcore. not sure why my board requires so much volts.
on my biostar , i use half the volts with the same chips and ram.

[zsamz_]

i went back to 111 bios
i can clock my ram to 1300 @ low fsb
no problem with .5 divider
and using 1,23 vtt for this
copy way up with older bios lol

[eva2000]

Probably the limit of the board itself as my DFI UT P35-T2R, LT X38-T2R and UT X48-T3RS all topped out around 500-510FSB for 32M pi and around 510-525FSB for validation. But been playing with my DFI DK X48-T2RBS Plus and fsb wise it's much better than previous DFI boards I have played with! How's E8600 @563FSB 32M Pi http://i4memory.com/showthread.php?t=10937

Memory clocking ain't half bad either

tRD/Performance level = 7

2x1GB Teamgroup PC2-10400 CAS6 @701Mhz 5-5-5-18 at 2.675v





2x1GB Crucial Ballistix PC2-8500 Tracer Micron D9GMH @698Mhz 5-5-5-18 at 2.65v for CPUZ Validation