DFI Infinity 965-S Tests (535 FSB 1:1)

[oshox]

I have had trouble booting to any CAS latency other than 4, when I have Fast Adressing enabled. The system is perfectly stable at a 445 mHz FSB @ 1:1 on my e6420, and the only voltage I had to increase is the cpu core, which is at 1.48. This board is amazing, and I would recommend it to anyone. As soon as I can get it 4hrs stable and get a screen, I'll post it.

PS: I've scored 11,500 3DMarks in '06 along with my 8800GTS 640

[ziddey]

hmm interesting. i thought it was just me. i didn't really narrow it down or anything, but I found less stability with 500fsb cas5 than cas4

[luihed]

Im using micron D9GMH ( ballistix pc5300 512mb x4 ) and here are my settings to bench superpi1M.... hope it helps..

M3, M2, fast fast 4, auto auto 4-4-4-12-4-40-2-13-12-2-8-6-6-5-4-4... my voltages are vtt=1.275v vmch=1.45v vSB=1.7v & 1.1v vdimm 2.35v....



Same setting for 32M @ 520fsb but a lil tighter timings....



I think the trick to this board is not to overvolt anything.... too much volt ( even only one setting up ) makes it unstable or wont boot.... First board I actually took notes and I must have 15 pages full lol.... Tricky board but rewardingly fast....

hmm interesting. i thought it was just me. i didn't really narrow it down or anything, but I found less stability with 500fsb cas5 than cas4

I think theyre all like that.... I had to use 4-4-4 for my max screenie at 556fsb lol.... and I can only give 2.35v vdimm cos this mobo doesnt like anything higher ....lol....

[ziddey]

what would you guys say is max safe vsb? I've been using 1.6vsb with added stability, but think I may need 1.7v, although I never ventured to try that high

[luihed]

I personally think 1.7v is ok as long as its not getting hot..... for 24/7 I dont use high fsb so I leave it at stock....

[ziddey]

hmm I'm not that worried about the southbridge itself but vsb also fuels pcie from what I understand, and I'm worried about our good friend, Mr. Jay Micron, blowing up

[Rattle]

just ran orthos 9 hours at 400x8 with the dram timings you guys gave me
funny I couldnt pass 2 minutes before.

[The Nemesis]

Finally figured out what setting was stopping me from getting 3.8ghz stable. Silly me, the VTT voltage was set too low.

[nealh]

Nemesis: I have the same serial number mobos from newegg as you(actually 2 of them)

Mine runs a much hotter NB temps..idle at 40-42C...I placed a thin layer of AS5 on it...

anyone know what the safe voltages are on this mobo for:
NB
VTT

is cas5 really issue on this board....

[Bail_w]

NB: 1.4-1.6
VTT: 1.36 - 1.4x

[The Nemesis]

Nemesis: I have the same serial number mobos from newegg as you(actually 2 of them)

Mine runs a much hotter NB temps..idle at 40-42C...I placed a thin layer of AS5 on it...

anyone know what the safe voltages are on this mobo for:
NB
VTT

is cas5 really issue on this board....

My newegg board is the one that went up in smoke. This board is the one from eWiz. Max safe northbridge hasn't really been discussed since this most boards do 500FSB @ default 1.25v, not many people have used alot of nb volts unless running the 4:5 divider. Pay no attention to my temps, my board is on an open test bed and my northbridge is cooled by a Cooler Master Blue Ice Pro. Intel data sheets say the max cpu-vtt is 1.5v. This cpu needed 1.36v on RD600, 1.45v on P5B, & 1.5v on Evga to hit 3.7-3.8Ghz stable.

[nealh]

NB: 1.4-1.6
VTT: 1.36 - 1.4x

thanks

Nemesis..thanks as well

I sure hope the newegg batch are not bad..although I got mine probalby a week after yours and they had sold out so the boards I got where a different batch

I am really having a hard time getting just 380 1:1 fsb stable....(forget 425x8..nb 1.35-1.4/vtt 1.275 I get immed reboots)
I have left voltages except vcore at stock and set cas5 with 2.2v on my Ballistix DDR1000
one night it is stable for 5 hrs and next it reboot immed at same settings on Prime

I tried last night nb 1.35-1.4 and vtt 1.275 and still no stability

The cpu was given a short run on my badaxe2 underwater at 1.5v-1.525v at 3500 and prime95 was run for 30min....to get a quick sense of speed....I dropped ot 3400 to hopefully have a better chance of stability

I am going to try cas4 and 2.2v and see what happens...if this fails I will open the seconf board and see what happens

[Rattle]

cant get this board to do more than 400 stable unless I give it VTT and NB volts.

running orthos at 450x8 with 1.32 vcore, NB 1.5 and vtt 1.425 everything else stock volts.

at this point I am regretting the sale of my p5b-deluxe...

[nealh]

cant get this board to do more than 400 stable unless I give it VTT and NB volts.

running orthos at 450x8 with 1.32 vcore, NB 1.5 and vtt 1.425 everything else stock volts.

at this point I am regretting the sale of my p5b-deluxe...

better than me
e6600@ 3420...+175mv=1.5v
I cant run 380 1:1 with Ballistix DDR1000 at 5-5-5-15 or 4-4-4-12 @ 2.2v
with stock NB/vtt

trying to up the vtt and NB..Vtt 1.325/nb 1.35..going to try this now

[Rattle]

well my cpu is a good one but my ram and this mobo dont wanna play nice, i had better stability above 400 and above 3300mhz with my p5b-deluxe

[ziddey]

NB: 1.4-1.6
VTT: 1.36 - 1.4x

where do you get those numbers?

and as for vtt going to 1.5v, that's only on cpu side. on the nb side, vtt has a real low max/abs max

[nealh]

where do you get those numbers?

and as for vtt going to 1.5v, that's only on cpu side. on the nb side, vtt has a real low max/abs max

so is 1.325v on vtt safe?(the max in bios is 1.4+)..so are all the bios values safe?

[The Nemesis]

vtt voltage available increases as you increase the sb voltage. Raise sb to 1.6v and cpu vtt 1.5v becomes available. Raise southbridge to 1.7v and cpu vtt 1.575v becomes available and so on.

[memory]

Im using micron D9GMH ( ballistix pc5300 512mb x4 ) and here are my settings to bench superpi1M.... hope it helps..

M3, M2, fast fast 4, auto auto 4-4-4-12-4-40-2-13-12-2-8-6-6-5-4-4... my voltages are vtt=1.275v vmch=1.45v vSB=1.7v & 1.1v vdimm 2.35v....



Same setting for 32M @ 520fsb but a lil tighter timings....



I think the trick to this board is not to overvolt anything.... too much volt ( even only one setting up ) makes it unstable or wont boot.... First board I actually took notes and I must have 15 pages full lol.... Tricky board but rewardingly fast....



I think theyre all like that.... I had to use 4-4-4 for my max screenie at 556fsb lol.... and I can only give 2.35v vdimm cos this mobo doesnt like anything higher ....lol....

I am having issue with this mobo as well as long as I set my vdimm @ 2.4volt and my rig will AUTO shut down...

Do u have this issue as well....

[Rattle]

ok I have changed my mind, after playing with the mobo a couple days.

been playing with 500x7 needs high board volts though for some reason.

450x8 is doable as is 400x9, pretty much same as my p5b-d

[nealh]

I finally got some stability
8x425..

I need 1.325v vtt and 1.3v nb
I am concerned at the vtt...I thought over 1.25v was not good for the cpus from intel papers or was this only for 975 chipset and not 965 chipset??

[Reverendrun]

ok I have changed my mind, after playing with the mobo a couple days.

been playing with 500x7 needs high board volts though for some reason.

450x8 is doable as is 400x9, pretty much same as my p5b-d

Glad to see you are having some success--I've been following this thread for a while to see people's results and noticed you were having some problems.

I just placed an order for this board and it should be here next week. I had it down to the DFI and QuadGT, but since I currently have a DFI board and it's been great--I decided on the DFI.

When you say 'high board volts' what are you talking about? To the northbridge, southbridge, cpu, RAM, or all of them? I'm just doing a ton of research now so I can test all this out when I get my board.

[The Nemesis]

ok I have changed my mind, after playing with the mobo a couple days.

been playing with 500x7 needs high board volts though for some reason.

450x8 is doable as is 400x9, pretty much same as my p5b-d

Okay I'm getting a little concerned with your results. It took nothing, no changes on anything other than vcore & memory for me to get 500 X 7 orthos stable.



I know not all chipsets can perform the same but I thought the whole point of this board was 500 FSB on the 266 strap. I'm beginning to think the 266 strap @ high FSB's on this board is requiring something that every ram kit can't do. I find it hard to accept that every boards NB can differ this much that some are clock so much better than others. I did however notice that smart guardian reports less drop in voltages in windows on my first board than my second board (compared to what you set in bios).

[RealTelstar]

It would be therefore important to know which rams performs better with this board.

[Knight]

How many people are having to use high motherboard volts to get to 500mhz fsb?

Is Rattle the only one having issues?