**Official DFI LanParty UT P35-T2R Review/Overclock/Guide Thread**

[Jupiler]

Any 65nm owners tried the new beta...?
Seems like i'm a minority here all of a sudden with 65nm...

If I could, I would have flashed to the new beta.
For now, I'll have to do it with the 11/07 bios.
Running my Q6600 at 3800, 1.392V ATM (prime since 2.5 hours)

[C-N]

Any 65nm owners tried the new beta...?
Seems like i'm a minority here all of a sudden with 65nm...

You are not alone I am back on my Q6600 & I think I'll drop my E6850 in tonight 4 some fun.

CN

[mad_skills]

@Jupiler Well, not bad...you are just 2 cores ahead of me at the same frequency, 3800MHz...
@C-N So i guess i'll be the ginny pig for this new bios 65nm testing..
Gonna report later.

[Jupiler]

Yeah,
My quad runs pretty good.
Does 3800 / 1.376V and 3900 / 1.456V on the P35-DQ6.
But I wanted to try another board so, after the DK, I bought the UT. (buying 2 boards within 2 weeks, I never did that before )

I think that once I have those small problems sorted, I'll be able to see if this one is better than the DQ6 (which I love BTW, and never had any problems with)

[mad_skills]

The thing that annoyed me about the DQ6 series, both P35 and X38, is that because of the bloody crazy cool heatsink you can't mount anything decent which goes with a backplate (and all decent tower-like heatsinks have backplates), without having to remove the crazy cool thingy. Could be a problem or perhaps not but i find it very irritating.

[mrcape]

can someone clarify what are the differences with beta 09/01 ??

"Added VTT Voltage control for Intel 45nm CPUs."

I'm curious too. Can someone do the favor of posting the voltage page and the available vtt, pll values?

[Smy]

looks like nice mb from dfi

[C-N]

looks like nice mb from dfi

Awesome board

CN.

[Mav451]

The funny thing about that - even on my 12/24 BIOS, HW Monitor (and SmartGuardian) were showing 1.1VTT in Windows, even though checking in BIOS - it definitely was set at the default of 1.2VTT.

I guess the DFI board was 'aware' of the 45nm CPU.

[mad_skills]

I guess the DFI board was 'aware' of the 45nm CPU.

It has a mind of it's own, that's why it overclocks so good...

[ether.real]

Is there a consensus on what BIOS is best? Thinking I may need to flash this weekend.

[MikeMK]

Depends on your chip really. 1107 worked great on my Q6600, but with my QX9650 I think both the 1224 and 0109 have some issues. The 0109 seems the better of the two though imo.

BTW, thanks for posting the new beta. Definately gonna give that ago.

[C-N]

I am not 100% sure as I didn't try that hard but I think the newest BETA 0214 has a high fsb problem on the 266 strap, it dident like 480~500fsb. Apart from my/the fsb problem it ran flawlessly on the 333 strap right round to 550fsb +

Q : Is anyone else seeing this or was it just me missing a trick last night?

There is nothing cosmetically wrong with this BETA (unlike 0109) that I noticed anyway.

I don't see any changes to the voltages with the 65nm either so all the enhancements must be for 45nm as the change log suggested.

This was tested using a 65nm e6850 btw.

CN

[cantankerous]

Thanks for the update C-N. I take it there is no real reason to update from the 12/24 bios if using 65nm CPU's such as the Q6600? I am glad the cosmetic bug is fixed, but if nothing else has changed or been added then it may only be geared toward 45nm CPU's like mentioned.

The 1/09 bios is no longer listed on DFI's site and has since been replaced with the 2/14.

BTW, did you see any new listed options?

[C-N]

Thanks for the update C-N. I take it there is no real reason to update from the 12/24 bios if using 65nm CPU's such as the Q6600?

I haven't seriously compared any benching numbers properly but from my quick play last night it didn't look any better or worse than 1224.

I cant see any advantage running the BETA over the very good 1224 official with the 65nm e6850 I tested.

I will be sure to let you guys know if I find any benefit with the 65nm's

The 1/09 bios is no longer listed on DFI's site and has since been replaced with the 2/14.

Never liked it anyway so no big deal 4 me.

BTW, did you see any new listed options?

Nothing I noticed with the 65nm, & all the voltage options etc looked the same as 1224 too.

CN

P.S I would be interested to know what vtt scale is available with the 45nm CPU's if anyone can help me out.

[vortex240]

I would recomend the 1107, I get the highest fsb with my lowly e2180 with it, no other bios after that one can post at that same fsb. Also it seems like my corsair 5400c4 can do 1021 @5-5-5-15 2.3v with this bios and cant with the ones that followed.

e2180 405x9=3648 @1.55v
max stable fsb 425

I do have a question about the gtl voltages for this processor. Does anyone have any recomendations about what to use? After I removed the heatspreader the temps dropped by a massive 20 degrees C, so I am once again interested in oc it. It can boot into xp at 3800 at 1.66v so I think with some proper gtl voltages I might get it stable at 3700.
thanks for the help
Wojtek

[mad_skills]

@MikeMK 12/24 does ok with me, i don't think it's any worse than the 01/09 beta. Can't go wrong with that bios.

I am not 100% sure as I didn't try that hard but I think the newest BETA 0214 has a high fsb problem on the 266 strap, it dident like 480~500fsb. Apart from my/the fsb problem it ran flawlessly on the 333 strap right round to 550fsb +

Does okey with 333 straps? Guess i found myself a new bios already.. Tnx C-N. Even if there isn't anything new for 65nm it's still a newer bios..had the same effect when gogin from 11/07 to 12/24..

[kberguy1]

I keep racking my brains on how to maximize my system overclock. the apparent voltage limitation of 1.39125 max is frustrating, particularly when it is not topping out because of heat issues/stability. I can run Prime 95 for 8 hrs straight at 3820 Mhz, but only IF the system allows me past the C1 display on LED. If I could push the volts up to perhaps 1.425 with water cooling, I am sure I could meet or exceed 4000 MHz with a 9 x multiplier without exceeding safe temperature limits, but I have my doubts this will work. The reason is I can dial all the way back to 266 MHz x 9, and as soon as I get to 1.39125 vCore, boot up becomes erratic (like 1-2x out of 10).

I was going to water cool this anyway, so ordered all the kit earlier in week. I am hoping that will somehow help. Temps should drop a bit, and the noise levels will definitely improve, which was one of the factors that pushed me to water cooling. I HOPE this drop in temperatures will allow a boost in vCore, but am not counting on it.

Does anyone know if the current beta BIOS/next official BIOS will support half multis? I can get up to 468 FSB stable on my Q6600 GO stepping, but only with an 8 multiplier (or less). I can run a 9 x 430 FSB, but this doesn't push my memory very hard.

I believe I could do 450-465 x 8.5x (3825-3950 MHz) and stay within the apparent voltage limits for vCore. With a 4:5 CPU:RAM ratio, this would put my memory at between 562 and 582 MHz (DDR2 1125-1165), which is well within its limits. While this probably wouldn't get me to the 4 GHz point we all want to brag about, the extra memory bandwidth in this situation would certainly help boost general performance.

Stay tuned while I try to sort things out. If you have any ideas or feedback, please post it.

[mad_skills]

I think 4GHz 24/7 stable is a bit in the range of liquid cooling and not so easy achievable on air cooling, even it be TR..

[The Dome]

Anyone know how many fan headers including the cpu header this mobo has, mine is in a place where I can't look at it to count.

[Sparda]

The Type 2 is the one you need. The thermal tape can't be used as shipped. The SMD caps sit higher then the PWM slave chips. The thermal tape needs to be trimmed so that it doesn't cover the caps. Then the tape will also act as a spacer so that the heatsink clears the caps.

Which Type 2 Thermalright did you meant ? This one HR-09S Type 2 or this one HR-09U Type 2. Would you kindly please post a picture of your mobo if you have use one.
I think I want to buy one.

[mad_skills]

@TheDome

Number of Fan Connectors: 6

http://computers.pricegrabber.com/mo...56375/details/

[Praz]

Which Type 2 Thermalright did you meant ? This one HR-09S Type 2 or this one HR-09U Type 2. Would you kindly please post a picture of your mobo if you have use one.
I think I want to buy one.

Depending on CPU heatsink the U version may create a clearance issue. As long as there is no case interference the S version works well slanted towards the rear of the board.

For best results the length of the base needs to be trimmed slightly. Also as I have already wrote the SMD caps sit higher off the board then the PWM slave chips that you will be cooling. The thermal tape will have to be trimmed so as to act as a spacer to hold the cooler off the caps.

[mrcape]

Don't forget to consider the exhaust fan if you will have it in a case. I would have loved to use something like that but I have a 38mm thick San Ace that hangs over the area.

[gsan]

does this mobo have serious vdroop with quadcore even the vdroop control disabled in bios?