DFI LT X38-T2R LanParty - Living Review.

[supershanks]

Try this FSB Overvoltage Control (Vtt) and this Overclocking with the DFI UT P35-T2R Motherboard

luck

[Clunk]

Originally Posted by bustamove

Anyone here have any idea what the cpu vtt voltage and the clockgen voltage refer to?
Im unsure on those two, Ive searched google/forums etc for a definitive answer but have found a lot of contradicting info on them.
Clunk? anyone?

The VTT is the main voltage for the FSB.

I will find out more on the Clockgen voltage as soon as I can

[Ace-a-Rue]

there was no difference in attaining a higher overclock when i increased my clockgen voltage on my board....i tried a Q6600 'G0' and a QX9650 on this board.

[supershanks]

Looks like there may be some stock ogf these boards coming back into suppliers Shipping time: small amounts possibly avail. has just been out of stock for long enough. Seen nothing in uk outlets.

[bustamove]

Very impressed withis board so far, I hit 3.5ghz with a e2140 on it today and stable with 1.5v it all seemed too easy.

On a Sadder note had to return a qx9650 after running at 3.9 ghz at 1.375vcore stable. for day, it refused to run at anything other than default values. so it has gone back for replacement. nothing to do with the board though, e8400 returned to sender too

on the bright side the e2140 is a true warrior!

[Clunk]

Good going on the 2140

What happened with the E8400?

[Clunk]

Just spoke to DFI, they didn't get my email last week, but anyway.

Regarding the SB Core / CPU PLL voltage.

From what I can gather, the design of their current board made it difficult to split the two circuits so they kept them together, and the two settings are indeed the SB voltage and the CPU PLL voltage.

It seems to work well anyway

I forgot to ask about the clockgen setting, sorry. I will ask next time I speak to them

[bustamove]

Originally Posted by Clunk

Good going on the 2140

What happened with the E8400?

Basically with the e8400, performance gradually detereroated, overclockability got lower and lower, untill it wouldnt post at all.
The post diagnostic lcd on the board just showed a squiggle.
put the e2140 in and it booted straight away.

Im begining to think the board itself might have some problems with 45nm chips, is that possible?

yours seem to run ok though, so its doubtfull. Either I got really unlucky with hardware or something else is wrong, but as I said all 65nm chips ive put in the board have had no problems at all.

so im pointing at the cpu's at the moment. kind of feeling a bit nervous around 45nms now.
Thing is I have a reasonable wc set up and didnt throw big volts into them. temps were out of whack on both 45's I had.

I received a phase from LD last friday so will be setting that up to test the replacement chips.

any thoughts or ideas on my 45nm woes?

[Clunk]

Unfortunately, that seems to be the nature of the 45nm CPUs.

They don't like a lot of voltage, and like the 65nm CPUs, high PLL voltage will degrade them fairly quickly. PLL can be a killer

I've seen different recommendations for max vcore on 45nm CPUs to avoid fast deterioration, and 1.45v seems to be the common figure, but how true that is, remains to be seen.

It could also just have been a faulty CPU.

I didn't keep my 8400, I didn't like the sensor problem. Maybe they will fix it? I don't know

[bustamove]

Well I disnt raise the qx9650 over 1.395 so it SHOULD be able to handle that on a decent wc set up surely? pll being the southbridge values?, I never went over 1.65, second option in bios.
But the fsb was set to like for example vcore being 1.375, fsb vtt being set at 1.3.
could that have been an issue?

I still get a bit confused tbh, whjen i get the 9650 replacement I think I will just aim for 4ghz with as many stock settings as possible keeping s/b and vcore as low as possible.

northbridge values im told are dodgy over 1.5 on 45nms so ill be keeping that in line with vcore.

does that sound about right or am I missing something?