DFI 965-S is here! For sale @ ewiz

[ziddey]

the layer I placed full covers the chip and is thick enough to cover the marking...the amount of TIM on this NB was nothing like the RD600 but it was like gum paste...literally sounded like I was tearing something when i remove the HS

I can go back and place more TIM..but always thought to much was no good

Mind there is heat transfer occuring as the HS get hot...again this is at stock voltage

blow a fan and see if temps drop

[memory]

Any body here experienced whenever you up the vdimm to 2.4volt, the rig wil auto shut down during bencing?

450mhz FSB with memory timings as loose as 5-5-5-15 on D9. It is like some kinda 'over voltage protection'. My cpu and memory is capable of doing more than 450mhz(FSB) on Infinity 975X/G and Asus P5B dlx. But no luck on this Dark-S=(

Please help me on this auto shut down. Thx

[ziddey]

hmm maybe fiddle with dq calibration / fine clk tuning? I've also heard to bump vtt a notch, although I haven't confirmed

[XtremeTiramisu]

vtt defintely has an important role in overclockig my RAM to be stable.
I had to bump it up to 1.36v, otherwise my system would freeze during orthose blend that stresses both CPU & Ram.

It's almost certain that other people and including myself had to change DDR configuration to mode 3 and possibly change the GTL driving strength to mode 2 for tighter timing stability.

I also noticed that for my ram to run stable over 450FSB, I had to increase my voltage from 2.2 to 2.225. whereas I used to be able to bring my rams up to 500FSB 1:1 with only 2.2v on the DS3 Rev3.3

I'm somewhat dissapointed with this board, however I'll be getting a pair of PC-8500 D9 rams and try again to go over 500FSB threadshold.

This board is known to be pickier in RAMS than other boards.

You definetely need a lot of time tweaking (took me exactly 7 days) from setting up the board to finalize my OC.

[Bail_w]

@ XtremeTiramisu - do you experiencing any squeezing sounds from the motherboard while you running blend stress test after you bump up the VTT voltage?

[ziddey]

i get that squealing even with 1.2vtt

[XtremeTiramisu]

@ XtremeTiramisu - do you experiencing any squeezing sounds from the motherboard while you running blend stress test after you bump up the VTT voltage?

i get that squealing even with 1.2vtt

you've drawed a very good question!

I did in fact started noticing the squealing noise comming from the mobo when the ram is under stressing (Orthos Blend) after I raised my VTT above 1.28v. I hardly notice any squealing sound when I didn't touch VTT adjustment. However, when I eventually moves on to higher clock speed vs FSB vs RAM clock speed, I had to bump the voltage somewhere and VTT does seems to be the source of the squealing noise I'm getting.

Like ziddey mentioned, he even noticed squealing noise even at stock 1.2v
That doesn't mean my ears are less sensitive than his but many each mobo with each diff ram under each diff condition may turn out a bit differently. But all in all, it's confirmed that this mobo does like to squeal when your ram is under stressing or even just under regular accessing.

The squealing noise is audible but not extremely annoying.
However it differs from cases to cases. If you're under an extremely quiet environment ~10-20db and you've a water cooling setup that's less noise generated than air cooled system, you may find the noise more noticeable.

sine the noise is at around 16k range so when you notice the noise and you started paying attention to it, it's quite piercing to the ears...

This is ANOTHER downside I found with this board, too bad we all rushed in buying this mobo like we're all xxxxing beta testers on this board.

Who would buy P965 chipset mobo in late april-may 2007 when this tech was released over a year ago....why?! cos we want to take the OC to the next level with absolute stability over 500FSB...and although there are some succesful cases with ppl bringing over 500FSB but there are also many cases where ppl would face problems when OC this board. This is certainly not a OC board for OC noob to start with...and I only consider I'm a OC intermeidate to be conservative.

If you check DFI club forum, you'd notice that there's already a thread existing with over 5 pages in regards to "squealing noise" and mobo killed their rams or mobos is dead along with squealing noise..etc.

Once again, I learned the lesson the hard way like I feel myself as a complete noob once again. I think I'll never trust ANY reviews from any reviewers from the even most reputable and legit site. I'll ONLY TRUST private/personally/non-commercial/non-advertising reviews and OC results (at least reported form >50 people) on FORUMS not paid/sponsored sites....

DFI P965-S Dark potential buyers, if you're reading this...don't you think I've exaggearated any of my words/describtion..take my words very literally.

However, that still doesn't conclude this board is simply bad. There is no simple GOOD or BAD mobo. The best rated board WILL has its flaw(s) and the only solution/remedy to that is...of course...a newer mobo...a new revision..a new tech to take over...

[Bail_w]

Well, if i bump up the VTT voltages up once. i will be experincing squeezing sounds out of the motherboard. look like the motherboard is trying to sing to me which is kind of annoying.

Well, although this motherboard couldnt be flawless as the lanparty series but the "best" p965 motherboard asus p5b deluxe have lots of problems. So i am pretty happy with this motherboard right now.

[MADMAX22]

true true, ive got the squeling sometimes, i mean running blend it will do it off and on for a while then it wont do it for several hours then itll come back again for a little while,

but i must say that this board for this chipset gets to 500fsb very easy, out of all my 965 boards this has been the easiest to oc to high fsb, thats out of the p5b-d,biostar 965pt, quadgt, and a S3,

still its the last 965 chipset ill get, good thing those p35 are looking good for ocing so far

[Bail_w]

Well, i tired to bump up my VTT voltages up to 1.32v, and the squeezing sound still does exists but not as bad as 1.27v.

[Apollonas]

Once again, I learned the lesson the hard way like I feel myself as a complete noob once again. I think I'll never trust ANY reviews from any reviewers from the even most reputable and legit site. I'll ONLY TRUST private/personally/non-commercial/non-advertising reviews and OC results (at least reported form >50 people) on FORUMS not paid/sponsored sites....

Even those 50 people are sponsored! Think more

[The Nemesis]

The 500FSB stock volts was true. Just anything above it is hit or miss. My second board went up in smoke for no reason @ all. Powered up & it started smoking between the dimm slots & the 24 pin connector. Fortunately all the hardware was not damged. The board goes back to Newegg today

[XtremeTiramisu]

The 500FSB stock volts was true. Just anything above it is hit or miss. My second board went up in smoke for no reason @ all. Powered up & it started smoking between the dimm slots & the 24 pin connector. Fortunately all the hardware was not damged. The board goes back to Newegg today

HOLY * S M O K E S *

[awhir]

Try my bios settings a few posts back. Our rams may be somewhat similar. I really don't think you need to adjust your GTl, auto should be fine. My memory settings may be a bit tight so try 4-4-4-12-3-21-2-11-10-2-8-6-6-5-5-3

hey i it my ram thats holding me back me thinks.

i have tried 4-4-4-15-4-25-3-11-12-3-8-6-6-5-6-4

and the ram still fails at 465.should i try cooling it or uping vtt to 1.3 ??

also this ram should do 1000 shouldent it ?

[The Nemesis]

hey i it my ram thats holding me back me thinks.

i have tried 4-4-4-15-4-25-3-11-12-3-8-6-6-5-6-4

and the ram still fails at 465.should i try cooling it or uping vtt to 1.3 ??

also this ram should do 1000 shouldent it ?

Are you still running 2.2v on your ram? Perhaps 2.25-2.3v will stabalize the oc. I suggest you download memtest 1.70 & loop test #5 for 2 hours to see if you get errors. You need to determine if your memory oc is unstable.

[Bail_w]

hey i it my ram thats holding me back me thinks.

i have tried 4-4-4-15-4-25-3-11-12-3-8-6-6-5-6-4

and the ram still fails at 465.should i try cooling it or uping vtt to 1.3 ??

also this ram should do 1000 shouldent it ?

Yes you should up the vtt to 1.32v, because i was not able to pass 467 FSB without increasing the vtt voltages.

[ziddey]

This is kind of bumming news. After thinking I finally got stability, after a week of usage, I noticed about 25% of the time, when I would boot into Windows, explorer would crash once. Other than that, nothing else crashes. So I decide to run large fft, and sure enough, it fails around 10 minutes every time. Do note I was able to complete 40+ hours just a week ago.

Hmm, I have been trying to lower vcore, so I pull it back up. Still fails after 50 minutes. ?! The only thing I've done was lap the heatsink, which lowered temps about 5 degrees. Hmm..

[Bail_w]

Well, i was able to complete the OCCT 1 hrs stress test and fail orthos in 30 mins. Usually, i do 8 hrs of orthos and it fail OCCT 30 mins test but i will run OCCT for 4 hrs and see what happen.

[ziddey]

&^@%^^&#^%&#$ Had to clear the cmos and it gave me bad ram beep. One of my d9gkx's will only boot with 2.1v now. Only used it for a week and a half too. Practically new. Only exposed it to 2.2v with fans over them, keeping ic's under 40'c at all times according to my thermistor.

oh no

[The Nemesis]

I'm starting to wonder if the whining sound near the dimms when stressed, the pickiness of this board with ram, the death of my best G. Skill kit, & sudden combustion near the dimm slots are all pointing towards something.

[MADMAX22]

well heres a few hours at 500fsb figured id throw this up here, memory cellshock and cooled with a mach ii gt, ignore the orthos clock it apparently doesnt adjust for multi changes, cpuz is accurate

[erwinz]

I'm starting to wonder if the whining sound near the dimms when stressed, the pickiness of this board with ram, the death of my best G. Skill kit, & sudden combustion near the dimm slots are all pointing towards something.

DFI should investigate this..

maybe the board is stressing the rams to much..

any DFI tech here???

[ziddey]

I did some testing and found something interesting about the squealing.

I used rmclock for the job. I put full load on cpu. Forced 6x with lowest voltage and got low pitched squealing. When I upped to 7x and lowest vcore, I got higher squealing. When I moved to 8x and max vcore, it went up more.

So it seems the squealing might not be entirely related to ram necessarily. I'm too lazy to get a tube and see where I can pinpoint the squeal from

edit: Since the dfi doesn't have a way to adjust vcore on the fly in windows natively like the gigabyte's do with easytune, but it does employ vid switching with c1e even with altered vcore, this can be helpful. 1.325 or whatever the native vid is will be the vcore set in bios. By dropping this, it will drop vcore. cpuz will show improper results, but enough to show that the vid is being manipulated. speedfan/smartguardian will indeed show that the vcore is changing. Definitely useful for on the fly vcore adjustment to find lowest needed vcore for stability, instead of constantly switching around with bios and then more testing.

I'm down to 1.47v stable at 3.6ghz and lowering at the moment. A lot better than the previous 1.5v I was using. Load temps have dropped about 3'c so far.

edit: 1.45v now

[nealh]

3420 8x380..stock + 175mv L47G431..not a great chip.. bios reads 1.5v...DMM idle is 1.5v, load is 1.48v
NB,SB,Vtt default...initially ran all defaults on ram..only change was cpu vcore and fsb(last night it ram P95 for about hr..this am..it would have an immed reboot....

so....

well I have issues with my ram

Ram vdd..default is 1.9v..DMM reads 1.83v
set 2.0 v and DMM reads 1.9v
set 2.1v and DMM reads 2.08/2.09 idle..load is 2.04/2.05

I am using Crucial Ballistix DDR1000..when running at 2.0v(dmm reads 1.9v) at 3400..380x8...4-4-4 I get immed reboot in P95.....set to 2.1v, still 380 1:1 4-4-4 reboots after a few min

380x8 1:1..5-5-5-15 @2.1v....P95 is running 45min

I am surprised this ram is not able to do DDR760 at 4-4-4 at 2.1v

[luihed]

hey i it my ram thats holding me back me thinks.

i have tried 4-4-4-15-4-25-3-11-12-3-8-6-6-5-6-4

and the ram still fails at 465.should i try cooling it or uping vtt to 1.3 ??

also this ram should do 1000 shouldent it ?

Try 4-4-4-12-4-40-A-13-11..... leave everything at stock volts.....

Im not sure if it was mentioned here before but I find giving too much voltage in this mobo is a no-no..... so start at stock voltage and only raise voltage as needed....