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

That Bios setting sounds corrupt-re save a freshly set group of BIOS settings.That may help.Also Have You tried the clock setting fine delay? That can be a great stabilizer for elusive stability issues as long as the bios isn't corrupt:yepp: .This setting seems to manually jump start/set up specific hidden Northbridge Latencies and it does a FRIGGIN SWEET job keeping data running smooth too:D I can offer a tutorial on my personal experiences/findings on the particular settings for the Clock Setting Fine Delay group settings if nothing else seems to work:shrug:
Best Wishes:yepp:

Quote:

---

Originally Posted by Thrmionicshaman

That Bios setting sounds corrupt-re save a freshly set group of BIOS settings.That may help.Also Have You tried the clock setting fine delay? That can be a great stabilizer for elusive stability issues as long as the bios isn't corrupt:yepp: .This setting seems to manually jump start/set up specific hidden Northbridge Latencies and it does a FRIGGIN SWEET job keeping data running smooth too:D I can offer a tutorial on my personal experiences/findings on the particular settings for the Clock Setting Fine Delay group settings if nothing else seems to work:shrug:
Best Wishes:yepp:

---

that sounds really good. thanks for the help. im using a reconfigured bios after i flashed. had to set manually all over again. u got me interested on that clock fine delay setting. keep me informed on that setting there. as i can tell u that even resetting the bios manually hasnt fully cured the problem. it hasnt happend again but im waiting for it to act up again. and to think all this started with just swapping a harddrive. how weird. anyway thanks for your input
i look foward to wat your experience on CLOCK FINE DELAY. thanks again

Quote:

---

Originally Posted by tonytouch

that sounds really good. thanks for the help. im using a reconfigured bios after i flashed. had to set manually all over again. u got me interested on that clock fine delay setting. keep me informed on that setting there. as i can tell u that even resetting the bios manually hasnt fully cured the problem. it hasnt happend again but im waiting for it to act up again. and to think all this started with just swapping a harddrive. how weird. anyway thanks for your input
i look foward to wat your experience on CLOCK FINE DELAY. thanks again

---

You're welcome:)
That is so strange how it all started with a hard drive:confused:
BTW have you set all of your Timing and Subtimings from auto to the values shown on your second screenshot?
Perhaps it might also be that your CPU is throwing a tantrum for whatever reason at those high speeds(How are your CPU/NB temps?).Getting those sorted out along with any CPU clock based instabilities first is part of the path to fine delay adjustment:) Once that is ruled out, clock setting fine delay can be used in conjunction with Memtest 86+ to find the right delay values. Sometimes what shows up as an immediate CPU/cache/FSB error in Memtests 1-4) is due to a misaligned the fine delay-that is of course if everything else is ROCK STABLE and it is now the fine delay that is being adjusted.If tests 1 through 4 pass then tests 5 through 8 are testing memory and therefore this is now where you will see the fine delay make or break the stability of the system.
Also since you changed hard drives which SATA controller are you using the intel or the JMB? Green ports or the two yellow ports?

Honestly I really dig this board on so many levels:shocked: but alas no one really gives a review on Newegg even these days:( Every time I crash the bios or turn off the mains by unplugging(which seems to sometimes corrupt my CMOS:eek:) I just use the jumpers for resetting everything and BAM!!! -The Phoenix BIOS Rises from the ashes and serves up,yet again, a whopping good time:cool: There is just something about this board....:yepp:

More to come on fine delay, just drop a line as to what has been ruled out as far as instability goes.Most def disable T2 Dispach and set Enhance Address to Normal and Enhance DATA to Fast or even normal. Set tRRD(not tRD which can be set at 6) to 4 or 5 for now, and put on your favorite overclocking music and have a blast:cool:

And I must say very nice clocks on that Quad!:yepp:
Best Wishes in your clocking quests:up:

Hey Tonytouch how's it going with your overclocking journey?
I just got an E8500 E0 on the 12th:yepp::D and now I am overclocking/testing just the memory controller(stock CPU or up to 12% below stock FSB on CPU) to see how much of bandwidth I can haul as well as to either confirm or deny my previous understanding of the CLOCK SETTING FINE DELAY.
More to come.

In two weeks I will begin my OC of the CPU:D

Hi guys, I have little problems to get my memories stable at 1:1 @ 500FSB. Sticks are A-Data Vitesta Extreme Edition DDR2-800+ CL4. I've put 5-5-5-15 2.3v and got errors. Whats the key to stability?

Those sticks are rated for 400MHz and you're trying to run them 500MHz, they may simply not be able to do it.

Quote:

---

Originally Posted by Shiranui Gen-An

Those sticks are rated for 400MHz and you're trying to run them 500MHz, they may simply not be able to do it.

---

But they are 800+ and CL4. I know one guy who runs them 5-5-5-15 500mhz stable, why can't I do it. Should I change fine delay values? Add voltages? Slower timings?

Quote:

---

Originally Posted by J0kke

Hi guys, I have little problems to get my memories stable at 1:1 @ 500FSB. Sticks are A-Data Vitesta Extreme Edition DDR2-800+ CL4. I've put 5-5-5-15 2.3v and got errors. Whats the key to stability?

---

you got the ones with promos
run them 5-5-5-15-5-50 2.24v"they gonna get hot so run a fan"
i bet you can hit 1050;)
i got a few sets of promos and the crap ones hit 990 the good ones hit 1100:D
dont go over 2.26v no need n they dont like too much juice

Thanks zsamz_ i'll try that out. They are quite close to my Xigmatek, you think I need an additional fan for memory?

Quote:

---

Originally Posted by J0kke

Thanks zsamz_ i'll try that out. They are quite close to my Xigmatek, you think I need an additional fan for memory?

---

promos chips get hot after 2.1v
just put a fan over em or on the side to get some airflow
got 4 stix of promos in sons pc runnin @ 1020mhz for almost a year with 2 80mm fans
mind you his gigabut board killed 1 set@ 2.24v:rofl:

Quote:

---

Originally Posted by zsamz_

promos chips get hot after 2.1v
just put a fan over em or on the side to get some airflow
got 4 stix of promos in sons pc runnin @ 1020mhz for almost a year with 2 80mm fans
mind you his gigabut board killed 1 set@ 2.24v:rofl:

---

I think there is something wrong with my sticks. They can't boot at their specsed 4-4-4-12 2.1v. At 5-5-5-15 1.9v they seem to be stable. You think I should RMA them?

Quote:

---

Originally Posted by J0kke

I think there is something wrong with my sticks. They can't boot at their specsed 4-4-4-12 2.1v. At 5-5-5-15 1.9v they seem to be stable. You think I should RMA them?

---

I did more research and found out that the other stick is broken, I RMAd them.

Quote:

---

Originally Posted by J0kke

I did more research and found out that the other stick is broken, I RMAd them.

---

just remember 1 error in memtest @ rated speed is rma time

Can anyone direct me towards some results with E6600's that I could glance at?

Much appreciated :up:

Quote:

---

Originally Posted by Lord AnthraX

Can anyone direct me towards some results with E6600's that I could glance at?

Much appreciated :up:

---

Have a look at this, all the clock info is there along with 3d mark screenies.

I'm still so impressed with this board. I set up two e86 machines, one based on this board and another with the Rampage Extreme. Both are running Vista 64 ultimate with 4gb ram.

The DFi is 24/7 stable (orthos large and blend pr9 & memtest) with a PP of 45968 @ 4.420ghz. Freakin amazing for a 4gb ram socket 775 24/7 setup. The Rampage can't get to that efficiency at that clock and stability. Not to mention the DFi needs WAY LESS nb and dram voltage.

I just can't let go of this board.

Quote:

---

Originally Posted by mrcape

I'm still so impressed with this board. I set up two e86 machines, one based on this board and another with the Rampage Extreme. Both are running Vista 64 ultimate with 4gb ram.

The DFi is 24/7 stable (orthos large and blend pr9 & memtest) with a PP of 45968 @ 4.420ghz. Freakin amazing for a 4gb ram socket 775 24/7 setup. The Rampage can't get to that efficiency at that clock and stability. Not to mention the DFi needs WAY LESS nb and dram voltage.

I just can't let go of this board.

---

:yepp: same here
its actually more efficient than my x38

Quote:

---

Originally Posted by zsamz_

:yepp: same here
its actually more efficient than my x38

---

I just picked one of these up and plan on firing it up tomorrow, just wanted to thank everyone for the wealth of info on this board. I should have no problems with all these reference points to start me off.

Has anyone installed the new EK-Mosfet DFI-UT? Id love to see some pics if anyone has.

Hi everyone :)
I recently bought a DFI UT P35-T2R, the version with heatpipe. Some people told me that I exagerated because such a board exceeds by far my knowledge regarding overclcoking. Indeed the myriads of settings I found in bios kinda left me speechless. Being my first steps on this platform I won't know for sure what specifically to ask. I have an E8600 Q820A671 VID 1.2500, 2 pairs of Crucial D9GMH and a Tuniq Tower120 for cooling. The PSU is a Corsair520HX. I try to reach the highest freq for 24/7 use and I thought about 4,4 GHZ but I guess I am a little bit too optimistic about that. So far I managed to reach 4,2Ghz using the default multiplier (420x10) and of course I increased some of the voltages in bios NB Core, CPU VTT, CPU PLL. I noticed that the VID decreased at some moment to 1,1125V then it came back to 1,2500V. Which way should I start to search for the limits of stability, by increasing the FSB with all voltages stock, decreasing the multiplier, thus trying to find the highest FSB attainable or should I leave the multiplier at 10? Which is the highest v core I should not exceed in order not to deteriorate (start the degradation of) the cpu? Regarding the NB Core, CPU VTT and CPU PLL is there a reasonable voltage to use and yet to avoid damaging anything? I guess for Crucial's the best would be some where in between 2,1-2,3V. Should I disable PPM Eist, CPUID maxval, C1E, Execute disable bit, and virtualization technology? Is there any point to increase SB 1,05V voltage, clockgen voltage? What about that thing Clock CPU divider (i have it on auto, yet it provides 3 options 2, 3, 4). Lots of questions...:) Thanx

Quote:

---

Originally Posted by ferdasyn

Hi everyone :)
I recently bought a DFI UT P35-T2R, the version with heatpipe. Some people told me that I exagerated because such a board exceeds by far my knowledge regarding overclcoking. Indeed the myriads of settings I found in bios kinda left me speechless. Being my first steps on this platform I won't know for sure what specifically to ask. I have an E8600 Q820A671 VID 1.2500, 2 pairs of Crucial D9GMH and a Tuniq Tower120 for cooling. The PSU is a Corsair520HX. I try to reach the highest freq for 24/7 use and I thought about 4,4 GHZ but I guess I am a little bit too optimistic about that. So far I managed to reach 4,2Ghz using the default multiplier (420x10) and of course I increased some of the voltages in bios NB Core, CPU VTT, CPU PLL. I noticed that the VID decreased at some moment to 1,1125V then it came back to 1,2500V. Which way should I start to search for the limits of stability, by increasing the FSB with all voltages stock, decreasing the multiplier, thus trying to find the highest FSB attainable or should I leave the multiplier at 10? Which is the highest v core I should not exceed in order not to deteriorate (start the degradation of) the cpu? Regarding the NB Core, CPU VTT and CPU PLL is there a reasonable voltage to use and yet to avoid damaging anything? I guess for Crucial's the best would be some where in between 2,1-2,3V. Should I disable PPM Eist, CPUID maxval, C1E, Execute disable bit, and virtualization technology? Is there any point to increase SB 1,05V voltage, clockgen voltage? What about that thing Clock CPU divider (i have it on auto, yet it provides 3 options 2, 3, 4). Lots of questions...:) Thanx

---

I can tell you that from my experience with DFI bioses it is best to set the CPU VID to a base value of 1.25v and then use the VID special add percentages to increase voltage. It requires that you use a calculator to set a few times but gives you finer increments and generally gives a more stable voltage :)

Quote:

---

Originally Posted by SaiNRuB

I just picked one of these up and plan on firing it up tomorrow, just wanted to thank everyone for the wealth of info on this board. I should have no problems with all these reference points to start me off.

Has anyone installed the new EK-Mosfet DFI-UT? Id love to see some pics if anyone has.

---

I don't have the EK block but I do have a mosfet block on. You want a picture of it?

Quote:

---

Originally Posted by SaiNRuB

I just picked one of these up and plan on firing it up tomorrow, just wanted to thank everyone for the wealth of info on this board. I should have no problems with all these reference points to start me off.

Has anyone installed the new EK-Mosfet DFI-UT? Id love to see some pics if anyone has.

---

check my worklog, I have pics of it installed on the P45

Quote:

---

Originally Posted by saint-francis

I don't have the EK block but I do have a mosfet block on. You want a picture of it?

---

Yeah that be awesome, lets see it bro.

Quote:

---

Originally Posted by SNiiPE_DoGG

check my worklog, I have pics of it installed on the P45

---

I just read the whole worklog on your build, that is looking real nice. I saw the mosfet block installed, im definitely wanting one.

Quote:

---

Originally Posted by SaiNRuB

Yeah that be awesome, lets see it bro.





I just read the whole worklog on your build, that is looking real nice. I saw the mosfet block installed, im definitely wanting one.

---

thanks :). I'll let you know how my mosfets temps turn out in the woklog. As it is I only used a thermal pad on the small mosfets and left the CPL chips without contact. there is some speculation that if you attach the cpl chips to the block they only worsen cooling performance by dumping a ton of heat onto the waterblock. they certainly dont need to be actively cooled but hopefully I can resolve the debate.

***Update to Clock setting fine delay***
This goes out to all curious fellow readers who don't already know about this setting.
I am OC'ing my E8500 and I am still getting the feel of what my board wants to be set to with this setting but I can say this much so far..........
The general rule of thumb is to set Control Fine Delay to a value=to half of
Command Fine Delay. Command fine delay is determined by what the board initially gives you for a particular memory speed then set to the nearest division using a combination of Tfine item and offset in degrees from "current" setting. This is done on both channels and usually involves setting both channels to the same entries.

Dimm1 fine delay is determined by the strap and actual FSB used.
Regarding the DIMM1 Fine Delay-DO NOT REPEAT DO NOT stray too far when using the Tfine and degrees especially do not go higher or you will be visited by a NO BOOT and CMOS clear that REQUIRES using the jumpers:rofl:
.........................REBOOTING to gather numbers :p: