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

[luihed]

Hmmm.... This is sad cos Im getting the same C1 error now and my comp wont boot .... Seems like this last 3 weeks there is something in my computer thats killing motherboards cos Ive gone through 3 mobos in the last 3 weeks and one E6700.... This is getting ridiculous.....

Ive never broken any motherboards in my life and Id like to believe that I know what Im doing ( got my A+ 5 yrs ago ) but with this chipset Ive gone through about 8 of them now .... No clue whats killing them, Im using basically the same hardware Ive been using since P965... Got this mobo yesterday, set it up this morning and it ran fine, flash to the 823 bios, no prob .... I shut it down for about 7 hrs and now I get this C1 postcode ( same one I got from my last IP35pro)....

Whats weird is none of the boards died during overclocking, they just wont boot after I shut them down .... with this one, I didnt even get a chance to push it, last and most I pushed this board was running spi32M at 514x7 with most voltage still fairly low.... I just dont get it....

I have 2 comp and Ive checked each HW to make sure they work and thyre fine... I cant believe my cheapo Giga 965 S3 ( not DS3 ) has outlasted all these expensive boards while running 3700mhz 24/7 for god knows how long now....

Here's my spec..

This mobo
crucial ballistix 8500 and 512 kingston value ram
Silverstone decathlon 750
WD raptors x2 ( no raid )
ATI 9250 pci card
mcw5002 waterblock

Like Brother Esau I really feel like giving up... I do play with ln2 but more than half of the boards that died never saw any ln2 action... and the ones that did didnt die while benching.... Im really gonna feel like an idiot tommorrow when I RMA this...

Im thinking my biggest mistake now is not buying Asus.... If Im gonna kill a bunch of boards, they might as well be Asus...

[pippo369]

Anyone know if there's problems with Topower 750W psu?
TKS

I don't think, i have this mainboard with Topower PT700 and no problem.

[luihed]

Anybody know where the measuring point for vdimm is? also when you power up, should the Dram Power led turn red? mine stays yellow and the manual says if its red, theres power in dimm socket...

[jimmyz]

Anybody know where the measuring point for vdimm is? also when you power up, should the Dram Power led turn red? mine stays yellow and the manual says if its red, theres power in dimm socket...

mine stays yellow, don't think it can turn red? c1 error won't go away with different ram or a cmos clear?
my board got here saturday and i have had a lot of issues with my d9gmh (4 different sets of ram) they all will do ddr2 1000 with 4-4-4-9 but won't boot with 5-5-5-15 @ 1067 ? weird stuff.

[luihed]

I would imagine most people that runs this board are using D9GMH including me but I also have a stick of value rams for troubleshooting....

Tried everything I can think off, I plugged in the extra floppy connector eventhough Im only using a pci card ( lol ) and i actually got passed C1.... it went C1, 26, .. and I thought I got though but it rebooted and stuck at C1 again.... if this board had 5 led Im sure it would have read P.S.Y.C.H .....

I was reading the manual and it said that if the dram led was red theres power in the dimm socket so I thought yellow means it doesnt have power....

Dont have a case so it cant be grounded and my waterblock was tighened lightly so Im lost to what can be causing this.... It ran without any problems today, even after I flashed the bios everything was fine then I shut it down.... Never fired again... didnt even abused it yet ...

This is the same crap I went through with the Ip35 and I thought it was the board thats why I swithch to DFI... Now it has to be something else, it could be me but Ive been doing this a long time ( see my joined date ) and Ive never killed a board till this chipset... Looking at Brother Esau's HW, were both using Silverstone PSU and ballistix 8500s, but Ive used ballistix and this PSU on my commando and 965-s without any problems...

[jimmyz]

i also think this board has some quirks with the n/b -ram settings i had everything running higher on my p5k than i can get this board to do.
c1 seems related to n/b instability either ram voltage or timings off or not enough n/b voltage. each time i had it happen it would recover though.

btw for an instant yesterday while i was trying to get past a c1 i swear the led said FU as a post code.lol i keep my camera ready after that...

[C-N]

i also think this board has some quirks with the n/b -ram settings i had everything running higher on my p5k than i can get this board to do.
c1 seems related to n/b instability either ram voltage or timings off or not enough n/b voltage. each time i had it happen it would recover though

I have read quite a few posts now complaining of RAM/C1 errors etc coupled with my own experience of this board I would say 1:1 has problems running CL5.

CL4 on the other hand has never given me a C1 error & I have run 1140MHz + where as CL5 throws errors frequently & randomly @ at timings/speeds I know are good for 1200MHz +

I too had a P5K & could get my RAM running higher & tighter speeds on benching voltage. The only problem for me with the Asus was the vdimm needed to run tweaked/tight 24/7 stable over say 1100MHz CL5 was on or very close to maximum warranted voltage.

The DFI on the other hand will make higher RAM speed with much lower voltages however the interesting thing in all this is the P5K's bandwidth & latency was higher @ lower fsb than the DFI so what we are looking at IMO is a kind of trade off between speed/fsb & efficiency/work done due to latency.

My DFI will make very high RAM speed @ low voltage simply trading off performance for speed. I have been able to out perform my P5K in terms of RAM FSB at the expense of performance.

I cant really offer any opinion as to which board/way of doing things is best as I think the net effect is about the same just achieved in a different way. This different way of doing the same thing kind of makes me wonder if this DFI P35 is giving us a glimpse of the way the up coming X38 may offer its power. I believe from what I have seen & read, the X38 makes its power @ high FSB/MHz rather than tight timings eg CL6/8 DDR 1600-2000.

btw for an instant yesterday while i was trying to get past a c1 i swear the led said FU as a post code.lol i keep my camera ready after that...

I know I shouldn't laugh at someone struggling but bhhh he he FU that would be priceless if you caught that on camera.

CN

[Praz]

I have read quite a few posts now complaining of RAM/C1 errors etc coupled with my own experience of this board I would say 1:1 has problems running CL5.

I have not seen this issue involving 1:1 myself. Although it may come down to how the combination of the individual components work together. I have noticed that how the SPD is programmed will play a big part in the stability of the ram on this board.

5-5-5 1:1

[C-N]

I have not seen this issue involving 1:1 myself. Although it may come down to how the combination of the individual components work together. I have noticed that how the SPD is programmed will play a big part in the stability of the ram on this board

It very well could be the component combinations causing issues it wouldn't be the 1st time a DFI has been picky with RAM & PSU's but thats all just the nature of the beast.

I am seeing is a fair amount of people having a similar apparently PSU/RAM related issue & the more people that confirm or deny observations the better IMO. There must be a reason so many people are apparently having issues with this board.

I have thrown a few different sticks through this board & not had any issues that I didn't create & although the SPD may be hindering some tightly set modules I cant see the SPD causing a problem with any of the RAM I have tried. IMO as long as you have put in sensible primary timings the board will set perfectly reasonable secondaries on auto so there should not be such a problem.

For comparisons sake my OCZ PC2-8500 SLI's will happily do 1200MHz CL5 with fairly slack timings they will also pass any test you can throw at them & have run for at least a week around 1200MHz without a single C1, yet the other night when I messed about with 1:1 I had numerous C1's from the board that required an EZ Clear to allow me to boot. I later played for 4-5 hours 1:1 CL4 & had timings so tight it crashed going into windows sometimes yet I still didn't get one single C1 this is what I am basing my earlier observation on.

I don't have an issue as such with 1:1 as I don't need to use it & if & when I want too its slow enough to run CL4 anyway.

I would need to do a bit more testing to try to qualify what I said earlier rather than just casually mentioning it but one way or the other there are a lot of people out there having C1 issues & from what I have seen the majority of them are 1:1.

CN

[Praz]

Some of the C1 errors seem to be related to DFI's attempt to minimize the double boot we all seem to dislike. Even when the double boot takes place it seems once in a while the clock generator isn't being written to correctly or the initial strap settings aren't being applied properly.

To get around this the system needs to be powered down. Our first thought when getting stuck on C1 is to reach for the reset button. If getting C1 because of the above a reset not only will not allow it to boot but usually results in having to do a CMOS clear. Instead of reset pull the power to the psu. Upon powering back up during the first boot the strap settings will be properly applied and the system will boot. This is assuming that none of the components are being pushed beyond their capability and that all other settings have been properly set by the user.

[cantankerous]

My board is due here Tomorrow and guess what? I too am using a Silverstone DA750 and have 4GB of Crucial Ballistix Tracer coming in along with the board. I really do hope I don't have these issues. I think I am going to run stock for a good week to be sure things are good before I start pushing. That way I will know if it is the board acting up for my doing. I want the board to have some burn in time before I push it, should be safer this way.

Praz brings up a very interesting case regarding the C1 and a power down instead of a reboot. If anyone gets a C1 error in the near future try this and see how it fares.

[Brother Esau]

i also think this board has some quirks with the n/b -ram settings i had everything running higher on my p5k than i can get this board to do.
c1 seems related to n/b instability either ram voltage or timings off or not enough n/b voltage. each time i had it happen it would recover though.

btw for an instant yesterday while i was trying to get past a c1 i swear the led said "FU" as a post code.lol i keep my camera ready after that...

Heavens To Mergatroid Already! What have I gotten myself into? You, know its Pretty Bad when the boards acting Funky and you get the "FU" post code on the LED .........Ummmm.....hey, Oskar whats up with that?


Yet another incomplete Intel Board at a premium price! ......I think all of us on this Forum should be getting PAID some form of Royalties for being Beta Testers and the Research and development Team. I hardly see how this is Fair that companies like Intel are making Trillions off of our hard work providing solutions and feedback to ongoing issues for their products! You just know the Intel Croanies are Trolling this Forum and Capitalizing off of our expertise and hard work!!!!

[RLM]

Hmmm.... This is sad cos Im getting the same C1 error now and my comp wont boot .... Seems like this last 3 weeks there is something in my computer thats killing motherboards cos Ive gone through 3 mobos in the last 3 weeks and one E6700.... This is getting ridiculous.....

Ive never broken any motherboards in my life and Id like to believe that I know what Im doing ( got my A+ 5 yrs ago ) but with this chipset Ive gone through about 8 of them now .... No clue whats killing them, Im using basically the same hardware Ive been using since P965... Got this mobo yesterday, set it up this morning and it ran fine, flash to the 823 bios, no prob .... I shut it down for about 7 hrs and now I get this C1 postcode ( same one I got from my last IP35pro)....

Whats weird is none of the boards died during overclocking, they just wont boot after I shut them down .... with this one, I didnt even get a chance to push it, last and most I pushed this board was running spi32M at 514x7 with most voltage still fairly low.... I just dont get it....

I have 2 comp and Ive checked each HW to make sure they work and thyre fine... I cant believe my cheapo Giga 965 S3 ( not DS3 ) has outlasted all these expensive boards while running 3700mhz 24/7 for god knows how long now....

Here's my spec..

This mobo
crucial ballistix 8500 and 512 kingston value ram
Silverstone decathlon 750
WD raptors x2 ( no raid )
ATI 9250 pci card
mcw5002 waterblock

Like Brother Esau I really feel like giving up... I do play with ln2 but more than half of the boards that died never saw any ln2 action... and the ones that did didnt die while benching.... Im really gonna feel like an idiot tommorrow when I RMA this...

Im thinking my biggest mistake now is not buying Asus.... If Im gonna kill a bunch of boards, they might as well be Asus...

Had the same issues with my geil 8500s

Occasionally I was able to get out of the C1 error, with flushing cmos and using 1 stick. Eventually though I ended up with a C1 error I couldn't get out of, even with a long cmos flush and battery removal.

I then stuck in a Gskill 6400 GBHZ stick and was able to get into bios. I changed the CLK values up a notch or 2, put back the Geils and was up and running again.

Seems to me maybe rightly or wrongly that it has something to do with these CLK values.

Interesting comment about the P35s as well. Does raise a question or 2.

Regards measuring Vdimm. I did post a few pages back.

My findings

Just taken some readings for VDIMM with a multimeter.

Smartguardian does seem to be more accurate than bios from what I've found.

Bios set to 2.49v vdimm

Readings in windows xp taken from a free slot.

1 memory stick in reads 2.46v
2 memory sticks in read 2.44v

Reading taken while in bios

2 memory sticks in read 2.44v

So according to my multimeter -0.05v from what is set in bios.

I did a blend run, while taking a reading. SMG reports a droop of about -0.04v. Multimeter reports no droop at all.

Not sure how these things work, and whether that's accurate for the slots the memory sticks are occupying at the time.

RLM

This is where I took the readings from



RLM

[RLM]

btw for an instant yesterday while i was trying to get past a c1 i swear the led said FU as a post code.lol i keep my camera ready after that...

Love it

RLM

By the way. Anyone in the know here (hint hint) know when a new bios update can be expected?

[Brother Esau]

I have my brand new set of G-Skill HZ's here that I know are good and am using right now in the IP35 Pro and I will use them vs the Crucial Tracers when the replacement board shows up in like a week.

[luihed]

Its alive........ but still very quirky cos I get stuck at 8.8 after double booting, but I get through usually after that.....

I thought Ive seen everything but it seems this board doesnt like my value rams and my E6700 .... Got it running with my E6600 and ballistix, so I tried my E6700 with the ballistix and back to C1 I go.... got a scare that my 6700 is toast so I put it in my other rig and shes fine....

This is too weird....

[PcCI2iminal]

Thanks for the reply. Do you think the lower PWM temp is due to the copper plate or the transpiper? I was thinking of using the transpiper on the PWM but without the copper plate. Do you still think this will lower temps the 2-3c you mentioned or is the copper plate needed as well to achieve this?

Thanks.

I think both are helping to get better temps

the heatpiper system(all together sb/nb/pwm->cpu/transpiper) works great

406FSB x 9 room temp 27C for all




1st screen vcore 1.61v and 1.71v on NB -> PWM 47C & NB 51C

i will remove the transpiper and i will re-check.

and i did a little mod ,2 cooper sinks on the cooper bridge

[emoners]

@PcCI2iminal: Did those 2 cooper sinks on the cooper bridge improve your temps?

[Gorod]

Played with my Q6600 G0 and P35-T2R for a couple hours today and looks like CPU FSB Wall's @ 505-511MHz , prety sure it's not mobos fault and seems like that CPU is very hungry for VTT voltage , for ex. with 1.55 VTT it was able to pass super pi 1M @ 500MHz and with only 0.05V increase (1.60V) it went all the way up @ 511MHz . Wonder if there is hard any mod for VTT ?

Q6600 G0 @ LN2 -120C @ 1.69V :

[Brother Esau]

146c OMFG

[orkan]

146c OMFG

Stands out a bit... doesn't it. hehe

[Solarfall]

nice clocks there Gorod ..
getting this board later on this week, cant wait.

[M3kk]

146c OMFG

Thats mean 146 - 255 = -109c.

[Gorod]

getting this board later on this week, cant wait.

Thats mean 146 - 255 = -109c.

Yup ... exactly

[jimmyz]

finally had some success last night. i set the memory to auto and gave it 2.3v and viola.
[IMG][/IMG]

cpuz :http://valid.x86-secret.com/show_oc.php?id=241171


also got in some 3-d benching results later.