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

Although it is not long since Intel launched its 45nm E8000 series CPUs, shortages have already been observed in the Taiwan channel. The lack of supply has spurred prices to increase by over 20%, prompting many consumers to hold off their consumption, according to sources in the channel.

Intel introduced the E8000 desktop CPU series, which feature enhanced performance and power efficiency, on January 20. Despite that the launch drew attention, Intel's strategically planned shipments have led to a shortage, the sources said. Driven by the attempt to lessen impact to sales of E6000 CPUs before inventory is completely sold, Intel is cautiously monitoring shipments of the 45nm-based CPUs, they explained.

In light of the shortages, some channel distributors have raised the price of E8000 CPUs by over 20%. The E8400, for example, is priced above NT$8,300 (US$257), up from the original NT$6,800. Some channels are only willing to sell E8000 series CPUs at a cheaper price if customers buy them bundled with specific hardware, the sources observed. Despite the CPUs being short of demand, sales of E6000 series models have not shown any improvement, the sources said. They attributed the noticeable performance leap that E8000 carries as the major reason for consumers waiting to buy. The sources further noted that E8000 CPU supply will only resume in small volumes after the Lunar New Year in early February. However, the key criteria to judge supply will be the status of E6000 inventory, they noted.

How is the above post related to anything specific concerning the DFI P35 board?

Quote:

---

Originally Posted by palese

Although it is not long since Intel launched its 45nm E8000 series CPUs, shortages have already been observed in the Taiwan channel. The lack of supply has spurred prices to increase by over 20%, prompting many consumers to hold off their consumption, according to sources in the channel.

Intel introduced the E8000 desktop CPU series, which feature enhanced performance and power efficiency, on January 20. Despite that the launch drew attention, Intel's strategically planned shipments have led to a shortage, the sources said. Driven by the attempt to lessen impact to sales of E6000 CPUs before inventory is completely sold, Intel is cautiously monitoring shipments of the 45nm-based CPUs, they explained.

In light of the shortages, some channel distributors have raised the price of E8000 CPUs by over 20%. The E8400, for example, is priced above NT$8,300 (US$257), up from the original NT$6,800. Some channels are only willing to sell E8000 series CPUs at a cheaper price if customers buy them bundled with specific hardware, the sources observed. Despite the CPUs being short of demand, sales of E6000 series models have not shown any improvement, the sources said. They attributed the noticeable performance leap that E8000 carries as the major reason for consumers waiting to buy. The sources further noted that E8000 CPU supply will only resume in small volumes after the Lunar New Year in early February. However, the key criteria to judge supply will be the status of E6000 inventory, they noted.

---

try posting it here: http://www.xtremesystems.org/forums/...d.php?t=173999

mrcape,
Great job there on the 4.0Ghz 32M run, very impressive! Is the VCORE reading in CPU close to what you set in BIOS. On my system the CPU-Z VCORE reading tends to be pretty accurate with BIOS settings!

Quote:

---

Originally Posted by Praz

How is the above post related to anything specific concerning the DFI P35 board?

---

Oh wrong thread :rolleyes:

Quote:

---

Originally Posted by Roger_D25

mrcape,
Great job there on the 4.0Ghz 32M run, very impressive! Is the VCORE reading in CPU close to what you set in BIOS. On my system the CPU-Z VCORE reading tends to be pretty accurate with BIOS settings!

---

I had set only 1.16875v in bios! Also passed 3dmark05 and 06 at that low voltage 4ghz setting.

I have yet to get as high as 1.225v with this new chip. Just for kicks I'm going to try and get as much as I can out of this chip with the max v listed on the box.

Announcement: ......All Bios Chip Orders have been Filled and are in Transit to those who purchased Pre-Programmed D.F.I P35 T2R Bios Chips:up: ...Doors are now closed for Business until after move Approximately 6-8 weeks from Wednesday Febuary 6th !

Campbell ....AKA...Naughty Monkey;)

Ok, I finally got my board installed, with the following components...


- Q6600 G0 1.25vid
- G.SKILL 4GB(2 x 2GB) 240-Pin DDR2 SDRAM DDR2 1000 (PC2 8000)
- Thermalright Ultima 90 HSF (core temps are 40ish under CinebenchR10 load)

...So far, after disabling EIST, C1E, etc. with most everything else on auto and auto core voltage, at 334x9 and mem at 500 (1000), with 2.0 vdimm it was running fine. I then tried 400x8 with 1:1 memory (3.2GHz - DDR2 800) and Cinebench exited when it got to the 4-core render test.

Anyway, I'm not a terribly experianced overcloker (and neophyte when it comes to this board), but/so I'm just looking for some dirt-simple 3.0 - 3.4GHz suggestions. I'm not looking to break any records... just moderate 24/7 air overclocking on this work machine (mostly work in Cinema4D).

Thanks for any suggestions.

guys have you seen the 0109 beta bios? there's an option there to Enable/Disable CPU N/2 Ratio...what does it do?

Half multi's,8.5x,7.5x etc

Quote:

---

Originally Posted by quake6

Half multi's,8.5x,7.5x etc

---

which, it doesn't seem to work in my case (E8400)

thanks quake6 for the fast reply...

karateo: so it doesn't work on the e8400? coz i saw this post & got me thinking about e8400s doing a half-multi http://www.xtremesystems.org/forums/...&postcount=813

in my case,
when i enable it and go to adjust multi, it just reverts to disabled again.

I have got the latest beta 09/01 bios which has the known problems with overlapping texts.
There is a photo earlier in this thread I think.

Quote:

---

Originally Posted by karateo

in my case,
when i enable it and go to adjust multi, it just reverts to disabled again.

I have got the latest beta 09/01 bios which has the known problems with overlapping texts.
There is a photo earlier in this thread I think.

---

For an 8400 you need to change the multiplier first. Enabling N/2 while still using the x9 multiplier results in x9 1/2 which is not valid.

Quote:

---

Originally Posted by Praz

For an 8400 you need to change the multiplier first. Enabling N/2 while still using the x9 multiplier results in x9 1/2 which is not valid.

---

thanks.
i will try it.

i was using multi 9, enabling N/2 option and then try to lower multi.
obviously this was my mistake.

Someone can help me?

My Q6600 runs @ 475FSB stable on P5K deluxe and Giga X38-DQ6. But it seems impossible to do that with this motherboard. I tried every GTL from 110 to 130 and nothing, VTT 1.60, vMCH 1.65...all :(

Help?
Thanks

Rbs.
:wave:

Here is a screen shot of what I did with it yesterday...

http://img441.imageshack.us/img441/4...ewq6fv0.th.jpg

Today, I hit a new personal record for overclocking. I am into 2nd hr of Prime95 at 3852 MHz (428 MHz x 9 multi) using 1.375 v in BIOS (1.376 showing in CPU-z).

My temperatures are 68, 65, 65, 68 C, and if this proves Prime 95 stable for 8 hours I will be VERY happy. If I knew there were such good chips out there, I would have gotten rid of my Q6600 B3 a long time ago.

Note that my current performance is with an unlapped CPU and a lapped Thermalright 120 Extreme heat sink with one Scythe SFF-21F fan.

Given this chips current performance, it seems likely that I will be able to break the 4000 MHz mmark once I water cool it.

Quote:

---

Originally Posted by kberguy1

Given this chips current performance, it seems likely that I will be able to break the 4000 MHz mmark once I water cool it.

---

Get that puppy under a Fuzion ASAP. That is a really nice G0 there :)

Hey guys, well after a fair amount of time spent tweaking this board, im finally getting some where. 4ghz is done, and VERY stable. Pretty pleased with the timings I have out of the 4x1gb flex too -

http://img266.imageshack.us/img266/4...5stablefi8.jpg

Already working on 4.2ghz. Its bootable, and will run a few benchies (like superpi) but not prime stable atm. Getting the hang of GTL's takes ALOT of patience! LOL!

Dang good work. Kinda disappointing that the e8400 does not seem to have much of a realistic clock advantage compared to the 45nm quads though. Would you mind telling us what you've learned about the GTL settings?

And what your GTL settings are?

Great work Mike, all that time really pays off! Is Smartguardian reading your actual VCORE accurately at 1.32v, if so even more impressive! Last but not least, at those settings can you quickly run Everest Ultimate's Memory benchmark and post your latency and read results?

a quick question on GTL's...

how do I know when its time to tune them? :D This thing is making me nuts

Quote:

---

Originally Posted by Mav451

Dang good work. Kinda disappointing that the e8400 does not seem to have much of a realistic clock advantage compared to the 45nm quads though. Would you mind telling us what you've learned about the GTL settings?

And what your GTL settings are?

---

My GTL's are 95/94/96 @ 1.31v VTT. As far as I can tell, they starting point seems similar to what anandtech state in their GTL tables. The other thing I have noticed is that my Yorkie is FAR more sensitive to GTL's than my Q6600 - it makes the deference between it actually getting into windows or not LOL!

I was actually going to ask a quesiton myself regarding GTL's and the amount of difference they make. Now im pushing the chip further (4.2ghz as I said) it seeming to require alot more voltage, and I'm not sure if its GTL's or just doesnt have enough vcore.

At 1.32v (same as 4ghz) it boots into windows, will run everest benchies, and superpi, but fails prime instantly. I played with the GTL's for a couple hours the other nite, after upping the VTT to 1.4v to account for the extra FSB. Didnt make a huge amount of difference. So I upped the vcore.

At each step Ive played with the GTL's (all around 100 somewhere) but doesnt seem to be playing ball. Makes the difference between failing instantly and up to 15mins stable, but nothing more.

Anyhow, vcore is now @ 1.39v which is a fair old jump from 1.32v for 4ghz, and im just wondering - will GTL's really make the difference between prime failing with in a few mins & being 100% stable? Have I been a bit impatient? The last thing I want to do is just plough this chip full of voltage unneccessarily, but at the same time, im concious I may have hit a voltage wall?

Quote:

---

Originally Posted by Roger_D25

Great work Mike, all that time really pays off! Is Smartguardian reading your actual VCORE accurately at 1.32v, if so even more impressive! Last but not least, at those settings can you quickly run Everest Ultimate's Memory benchmark and post your latency and read results?

---

Roger - Yes, thats reading it correctly. 1.32v vcore. As for everest, ive run it at 4.2ghz for a m8 of mine asking the exact same questions. Therefore mem is running at marginly over 1000mhz @ the same settings -

http://img82.imageshack.us/img82/141...stbenchtp7.jpg

Remember I'm on Vista X64, so XP may be a little faster too :)

Quote:

---

Originally Posted by MikeMK

At each step Ive played with the GTL's (all around 100 somewhere) but doesnt seem to be playing ball. Makes the difference between failing instantly and up to 15mins stable, but nothing more.

---

Same question. The most I got was upto 45mins of priming. but nothing more.

Quote:

---

Originally Posted by fireice2

a quick question on GTL's...

how do I know when its time to tune them? :D This thing is making me nuts

---

I can't say for sure, but i started with mine when extra vcore made no difference. I notice that a lot of peeps are using VTT up to 1.3v/1.4v. A look at my settings here may well surprise a few peeps. While i was trying to find the right settings, i noticed, the more i raised VTT, the faster Prime failed. Since i posted the OCDB entry my ram has gone belly up, if it was dying while i was going for 4.6ghz, then it might explain my settings. Once a replacement arrives i'll know one way or another.