Dividers GOOD! :droooooool:

HOLY CRAP!! :eek: :eek:

Ok, last night my system was locking up at 200fsb while listening to music fnar (for no apparent reason).

Well, just for the heck of it I decided to try running 5:6 and of course the board said no to that and gave me "Bad bios" symptoms... I powered on and off about 15-20 times using several different methods trying to get it to post, but nothing would work... :confused:

Well, I finally pulled the CMOS battery itself and powered on.. BAM, she's flying! :toast:

Now, here comes the interesting part...

My VDIMM has always shown itself to be about .05v to .1v low... dunno why, but it just was. Now? Its OVERVOLTING by a FULL 0.33v!!!! :smileysex :flame:

Here are my readings (monitoring programs dont show extra 0.33v for some reason?)...

2.6v = 2.92v
2.7v = 3.07v
2.8v = 3.13v
2.9v = 3.24v
3.0v = 3.34v
3.1v = 3.45v
3.2v = 3.53v
3.3v = 3.65v

Oh, and extra system info...

XP2000+ @ 2338.7 (14x166 (167.5 real)) @ 1.8v
2x256mb XMS3202 1.1 double sided
1x256mb KHX3000 double sided

I havent tried 200fsb yet incase this is a bios bug @ 166fsb, but damn.. :banana:

Oh, 429-12/18 Evilous LPB bios, too.

u sure its at 3.6v?

What mobo was that with? Sorry if I just missed it.

Did you leave the battery out, or take it out and put it back in?

Yes, i measured it with my DMM.

DFI Infinity

1: Remove battery
2: Post, go into bios.
3: shut down
4: Insert battery
5: Post, go into bios, set stuff.

Ok, i think I may have found out what I did to cause this.

When I set the board to use 5:6 I also increased the vdimm to 3.3 from 3.0 which I had been running.. when I did that and the board tried to post at 5:6 it may have done a "partial flash" so that 3.0 is now "3.3".

Does anyone know how I can backup my current bios and settings a .bin or .rom so they can be comparred to the stock Evilous 429-12/18 LPB Bios?

Looks like these boards may be bios vmodable :beer:

AWDFLASH will give you that option if you try to flash to a different bios.

1. Make a bootable dos disk
2. Stick awdflash.exe on it
3. Stick a bios file on the disk as well
4. Type "awdflash.exe bios.bin" hit enter
5. It will go thru its basic stuff and then ask if you want to save your current bios
6. Stick in another disk to save the bios to because the boot disk will be to full
7. After saving is complete just remove the disk and reboot the computer without confirming to continue the flash

Your working/modded bios will still be intact in the system, and now you have a copy of it to look over.

That should work man.

3.65V sounds impossible

maybe the 3.3V rail of ur psu overvolts strongly...

DFI does not use the 3.3v rail to power VDIMM

Quote:

---

Originally posted by .::$k4T4N::.
3.65V sounds impossible

maybe the 3.3V rail of ur psu overvolts strongly...

---

Not impossible, I have run a Vdimm over 3.5v before.

You are mostly correct, on most boards you need a very high 3.3v rail in order to achieve a high Vdimm because the Vdimm is dependent on that power rail.

I run my 3.3v rail at 3.7v 24/7 just in case I need the headroom.

the dfi's use the 5v rail for vdimm :)
Can't wait to see the outcome of this, probably give it a shot myself on thursday when i have off work.

hehehehehehe.......

maybe this will work with an NF7? :D

no, will never work on nf7 lol, impossible...

NF7s run off the 3.3v rail ... but who knows, something funky is happening...

Why would it be possible on the dfi and not the nf7? You could raise the 3.3v rail as HF has said, and do the same thing as you guys would like to see done on the DFI. Might still be impossible on the nf7, or both for that matter.

What do the motherboard readings give? The same as the DMM?

More importantly, can you OC the RAM anymore? :D

Motherboard still shows "correct" values.

I'm only getting 3.55v @ 3.3 now, too..

thats freakin weird......... COOL! lol

someone else try!

TRY IT WITH VCORE!!!!

http://www.wcroller.com/los/downloads/vdimmbios.zip

there is a link to the bios.

I DO NOT SUGGEST FLASHING TO THIS BIOS JUST YET, IT WOULD BE BETTER TO COMPARE WHAT HAS CHANGED IN THIS BIOS VS. THE ORIGINAL EVILOUS 4/29-1218 LANPARTY-B BIOS!!!

There are 2 bioses in that zip, a "high" and a "low" vdimm bios. The "high" one was done at 3.3vdimm (3.55 real) and the "low" one was done at 2.6vdimm (2.8v real) so they can also be comparred against each other.

Quote:

---

Originally posted by STEvil
http://www.wcroller.com/los/downloads/vdimmbios.zip

there is a link to the bios.

I DO NOT SUGGEST FLASHING TO THIS BIOS JUST YET, IT WOULD BE BETTER TO COMPARE WHAT HAS CHANGED IN THIS BIOS VS. THE ORIGINAL EVILOUS 4/29-1218 LANPARTY-B BIOS!!!

There are 2 bioses in that zip, a "high" and a "low" vdimm bios. The "high" one was done at 3.3vdimm (3.55 real) and the "low" one was done at 2.6vdimm (2.8v real) so they can also be comparred against each other.

---

who mod that bios.... top class mod... respect..... :)

ok i done checking it... the bios files didnt change

u trick the bios save block.... :)

try this... for those who want to overvolt their vdimm without volt mod....

1st set vdimm to max....

shutdown.....
pull of cmos baterry

turn back on (without cmos battery)
go to bios
lower the vdimm depends how many overvoltage u want(Overvoltage = Max - current setting)
save
shutdown

put back cmos battery...
u will get overvolt on vdimm u want

try it on Vcore.... who knows this will work to on other voltage...

Nice finding .... Congrats

p/s: measure voltage with DMM...

on nf7... the vdimm came from 3.3v so u wont get higher than this unless u have 3.3v higher than 3.3v

same trick...

1st set vdimm to the max.. 2.9v
save
shutdown

pull of cmos batery
power back on
set vdimm to 2.6v
save
shutdown

Put back cmos batery
power on...... now measure ur vdimm voltage with dmm(Digital multimeter)
should be 0.3V overvoltage on vdimm

i supposed it will work to... :)

can you confirm this on the NF7-S MoBo's? any have tried?!

one problem, your saying to boot with no battery and save and reboot with no battery... dont that clear the cmos? or does the power never get cut since the comp is on?

nice find Stevil!
already modded my NF7 so this isnt much help for me anymore.. but if it works its great way to mod Vdimm AND keep warranty! muahahaha :D
Stevil, did your Vtt follow?
________
Honda MT250

sweet mod, nice find :D

but where on the board do i measure the vdimm so i know it works?

Quote:

---

Originally posted by tictac
on nf7... the vdimm came from 3.3v so u wont get higher than this unless u have 3.3v higher than 3.3v

same trick...

1st set vdimm to the max.. 2.9v
save
shutdown

pull of cmos batery
power back on
set vdimm to 2.6v
save
shutdown

Put back cmos batery
power on...... now measure ur vdimm voltage with dmm(Digital multimeter)
should be 0.3V overvoltage on vdimm

i supposed it will work to... :)

---

Are you saying that on save and before it repost, to shut it down?? Confirm that for me.

I will try it....why the hell not.

backside of 1st mosfet under dimm1 or leg 1 of mosfet next to it (under dimm 2). Both readings will be within ~.02v of each other.

vtt/vbt supposedly tracks automatically on these boards.

I wonder if this could be related to some "dead" boards, and if we can modify the bios to do this without having to trick the board?

Well, it initially worked.

When set to 2.6v in bios, actual was 3.03v. Set to 2.9v and rebooted....before post I turned it off, Vdimm was reading 3.29v actual at that point.

Popped out battery and reposted, went into bios and it was automatically back to 2.6v.

Put battery back in and reposted....thats where the problem is.

On the first post after removing battery it shows cpu @ 1100mhz....although bios shows it set to stock 2500+ settings. So, after battery is back in and reposted, it detects theses settings and does a Save. That wipes out the 0.3v addition.

I've still got .2v of the +.3v.....

We're getting somewhere at least

Hell-Fire, on which MoBo did you try it?! NF7-S or DFI LP?!

Quote:

---

Originally posted by Hell-Fire
Well, it initially worked.

When set to 2.6v in bios, actual was 3.03v. Set to 2.9v and rebooted....before post I turned it off, Vdimm was reading 3.29v actual at that point.

Popped out battery and reposted, went into bios and it was automatically back to 2.6v.

Put battery back in and reposted....thats where the problem is.

On the first post after removing battery it shows cpu @ 1100mhz....although bios shows it set to stock 2500+ settings. So, after battery is back in and reposted, it detects theses settings and does a Save. That wipes out the 0.3v addition.

---

thanx for trying...

i think there is some difference on nf7 bios

coz when we save setting
it will boot 2 times
1st boot: Safe mod...
2nd boot: boot with our changes setting

maybe this is the reason why it cant be done on nf7 bios

any other try on DFI?

Quote:

---

Originally posted by naelq
Hell-Fire, on which MoBo did you try it?! NF7-S or DFI LP?!

---

NF7-S V2

Quote:

---

Originally posted by tictac
thanx for trying...

i think there is some difference on nf7 bios

coz when we save setting
it will boot 2 times
1st boot: Safe mod...
2nd boot: boot with our changes setting

maybe this is the reason why it cant be done on nf7 bios

any other try on DFI?

---

I am gonna try a few other things to see what happens.

Here is another odd thing. When using the battery out and in test, the board totally partially disregards the Vdimm mod I have in place. When set to 2.6v in bios, it is reading the Vdimm correctly at 3.03v....but when I swith to 2.9v, it reads 3.29v versus 3.45v which is what it reads with the battery in.

or try it difference way arround...

on the 1st save(save with 2.6v vdimm) let it to boot 1st.. then shut it down

power on back without cmos battery
set vdimm to 2.9v... save
let it post 1st

shutdown

put back cmos baterry... :)

Yeh, thats what I was thinking as well....or take out battery while it is booted, then repost and go into bios...set to 2.9v, save and repost, then put battery back in....reboot.

I'm confused as to what I did exactly I tried so many things, lol :D

Quote:

---

Originally posted by STEvil
I'm confused as to what I did exactly I tried so many things, lol :D

---

Yea but it was real fun! I felt like I was on speed again. Like the old days!! It was worth the rush :toast:

wut if u already have your board vdimm modded? would it bump it up even more?

Well heres what I did on a unmodded nf7,
I set it to 2.9 in bios and powered off

I removed the battery and Powered back on

Set the voltage in bios to 2.6, my settings were still saved in bios.

Powered off, then put the battery back in.

Powered back on and went directly into windows.

I dont have a multimeter right now, so Im just going by what I knew was stable at 2.9volts. Before max prime stable @2.9v was 217@2,2,3,11. Now with the bios set to 2.6volts, I am prime stable at 217@2,2,3,11.

So far so good for me. What Im wondering is, At what point will this reset itself back to 2.6v? If I change a setting or something or save bios again will it reset?

Quote:

---

Originally posted by FlyingHamster
wut if u already have your board vdimm modded? would it bump it up even more?

---

yes.. it will bump it up even higher on DFI board... :D

Quote:

---

Originally posted by Shroomalistic
Well heres what I did on a unmodded nf7,
I set it to 2.9 in bios and powered off

I removed the battery and Powered back on

Set the voltage in bios to 2.6, my settings were still saved in bios.

Powered off, then put the battery back in.

Powered back on and went directly into windows.

I dont have a multimeter right now, so Im just going by what I knew was stable at 2.9volts. Before max prime stable @2.9v was 217@2,2,3,11. Now with the bios set to 2.6volts, I am prime stable at 217@2,2,3,11.

So far so good for me. What Im wondering is, At what point will this reset itself back to 2.6v? If I change a setting or something or save bios again will it reset?

---

Wow, thats awesome ...

now how can we set at 2.9, remove battery and power on, then set to 2.6, put battery back in, then somehow set it up to 2.9v, so we get a real 3.2v ?

Nice work... :banana:

Hey, has anyone tried this with VDD ?

Well I cant get anyhigher clocks out of this mem, I think the chipset might need more voltage now. I get errors in prime at anything higher. Good new is, I try upping the fsb and voltage but it didnt help so i went back to 2.6volts at 217 and ran prime again, still stable:) So it seems to stick after the bios changes it self for me.

Quote:

---

Originally posted by Shroomalistic
Well I cant get anyhigher clocks out of this mem, I think the chipset might need more voltage now. I get errors in prime at anything higher. Good new is, I try upping the fsb and voltage but it didnt help so i went back to 2.6volts at 217 and ran prime again, still stable:) So it seems to stick after the bios changes it self for me.

---

Well... if the voltage still sticking on Bios memory

trick it for 2nd time

set to 2.9v again
save
shutdown

pull away cmos battery
set vdimm to 2.9v
save shutdown

put back cmos battery back
power on... straight to windows..

2.6v in bios = should give 0.6v overvoltage

"The bios is stupid enough.. dont worry we are clever then them... :D"

Heres something I dont get, I tried dropping my vdd now too 1.4volts and my board is still stable at 217. so on minimum vdd and vdimm it hits its max. Maybe that why this board isnt my overclocker.;)

has anybody tried the same trick on the NF7 Vdd?
________
volcano vaporizer

maybe this is why some boards get brick walled at 217 and vdd never seems to help and such?

Are you L12'ed? I am.

I'm currently testing out....

My highest stable(Prime 95 for over 12 hours) FSB with 2,2,2,6 timings was about 218~219MHz.

Now testing at 227MHz FSB :)

Gonna leave it over night, and hopefully come back in the morning to find Prime 95 still running :)

Wish me luck. Really should probe the voltages with my DMM. Ok, not sure what it read before, but with the BIOS set to 3.3v the back of MOSFET 1 is reading 3.34~3.35v.

for those doing this mod on an NF7, remember to cool the Vtt mosfet (underneath Dimm3 latch). If Vtt is tracking the increased VDimm this mosfet will get VERY hot and may burn out without cooling.
________
Honda C77

has anyone tried this on an A64? more vcore on the NF3 boards would be nice.

Nice find. Vcore would be best... vdimm is modded ;)

hehe, as long as voltages are stable. :thumbsup:

Quote:

---

Originally posted by STEvil
http://www.wcroller.com/los/downloads/vdimmbios.zip

there is a link to the bios.



There are 2 bioses in that zip, a "high" and a "low" vdimm bios. The "high" one was done at 3.3vdimm (3.55 real) and the "low" one was done at 2.6vdimm (2.8v real) so they can also be comparred against each other.

---

STEvil,
Infinity and LP? same bios?

hehe this news is spreading like wildfire on other forums. :D

Sachs - they are LPB bioses (and both are the same, I just did 2 backups at seperate voltages to see if it would pin down the cause), as I flashed my board to an LPB.. not that it turned me into a low ping bastard, but... :D

Its definately a neat find, but I wonder how many boards this has been the "death" of and they werent even dead maybe?

:(

Tried it the way Shroom said and no voltage raise :( I checked with a DMM

oh ya nf7-s with one of tic-tacks 22 bioses

This thread is getting major attention from all the web.

I am hoping we can figure out other methods for the NF7 series, and maybe other voltage mods for the DFI from the bios "trick"....Vdd and Vcore..??

Hell of a find STE.

vdd & vcore should be same trick... give it a go... :)

Dumb luck is more like it ;)

As to if it helps.. I can sure say hell yes!

Got my BH-6 doing 250 @ 2-2-2-2-11-9-12-eefe right now, although it ate my registry on boot.. lol.

3.53v for that, guess my BH-6 kinda clocks like crud :D

not working for me on my DFI Lanparty????

I tried it about a dozen times, nothing..........

Please help!

Does it show the "higher" voltage in the bios?

So I guess its a no go on the NF7 ??

sribble - You have to measure with a DMM to see the higher value.

thesaucier - It worked for Hell~fire, kind of......

UPDATE:

Swapped ram to slots 2+3 and now am doing 9x252 2-2-2-2-11-9-12-eefe in prime with no problems.

253.5+ errors within 30 seconds.

The guys with success how did you Exacly do it, like i said i tred a few times:

2.9 > power-down > remove battery > boot and change to 2.6 > pop back the battery> boot ...............

and nothing, checked with DMM and i just get 2.68 ish just the boards normal overvolting

Try 2.9 -> power down -> remove battery -> boot + change to 2.9 -> save+exit+power down -> +battery -> boot.

Other than that, maybe this could be bios related?

I'm not convinced that it works....

I set VDIMM to 3.3v, powered down, removed battery, powered back on, set VDIMM to 3v, powered down, battery back in, powered on, set VDIMM to 3.3v.

Now although I didn't know what my board was reading previously when set to 3.3v, it now reads 3.34~3.35v from the 1st MOSFET.

And my PC is no more stable at FSBs of 225~227MHz, still crashes Prime 95 after about 1 hour.

I'll perhaps give it another go later.

It works, its just a pain to make it work..

Tell me how to get it working then :)

Keep trying, basically.

i think ill give it a try tomorow night.. see what i can see.

Quote:

---

Originally posted by .::$k4T4N::.
3.65V sounds impossible

maybe the 3.3V rail of ur psu overvolts strongly...

---

Nice nick ;) :D :smileysex

Quote:

---

Originally posted by Evil_Spork
i think ill give it a try tomorow night.. see what i can see.

---

Why do I see quotes from GeForce in so many peoples sigs?? :p:


STE, any word on the other V options in the bios?? ....Vdd...Vcore....

so.....has anyone goten it to work on a NF7, and stick?

I just tried this mod 3-4 times, it dosent seem to work for me...

And i am measuring with a digital multimeter too...


I learn my board undervolts by .1 which explains why ive been getting semi-good FSB's (i run 245 stable in windows)



This is exactly what i did:

Boot, went to bios, set Vdimm to 3.3

SAVED BIOS AND SHUT DOWN

while comp was off, popped off CMOS battery.

With cmos battery out, booted, went into bios, changed vdimm to 3.0v,

SAVED BIOS AND SHUT DOWN

while comp was off, popped CMOS battery back in.

Boot, went to bios, set Vdimm to 3.3 again

SAVED BIos AND SHUT DOWN

Boot streight to windows....



measured from mosfet, and still getting hte .1 undervolt i always get.......





Anything i am doing wrong?



edit: Are you supposed to mess with the BIoS jumper at all? Leave it on on or on off? Or take it off completely?

Is there any "timed" trick to this? Do you need to wait a long time before popping in and out of battery or inbetween booting? ......etc............etc.


I ask the people that had sucess what bios they were using and if they are using the L12 mod or any other wiremod

Dont need to touch the bios jumper.

It could have to do with timing
It could have to do with certain BIOS'es

Helll~Fire - My VDD seems to have calmed down. It used to overvolt when I first got the board but now it is running what it should.

I'll check my vcore right now.

1.88v when set to 1.825v in bios measured from the choke... same overvolt as its always been.

argg, I tried this over, and over, and over, and over ... nothing.

Then I rebooted and nothing happened :D Screwed my BIOS chip ...

Luckily my dad has an NF7-S, just had to wade through all his HDs and messy cables and hotswap my BIOS chip in to flash it ;) Actually cracked the casing that holds the chip in the board on his NF7 lol, he won't notice ...

We need to start over, everyone who has succesfully found a way to do this, post your board, and bios version, and exact steps you took...

I also tried it and it did nothing. So, a question I have is this - when you say 'shut down' do you mean just power off, or PSU off alltogether? Anyone who got it working please respond.

f10+enter is what I did, then flipped the psu switch before post.

Bad news in a couple minutes, though.. gotta get pics off my camera and stuff.

ok, time for the bad news.

The reading we are getting off the mosfet is bad, I believe. *:stick:

As you will see in the attached picture below there is a point where you can measure your VDIMM next to the 3rd DIMM slot.. it appears to me that what I have been reading off the mosfet is somehow incorrect... somehow. :stick:

I also found another point just below the VDIMM mosfet that apparently is either VTT, VBT, or VREF.. dunno which of the 3 it is, but its the correct voltage for match the first measure point in the pic (both measure points not on the mosfets are spots for capacitors.. could putting capacitors there help with high fsb?).


Still, this leaves a question: How the hell are the mosfets showing an incorrect voltage?!?! :stick: :confused: :stick:

ok, please let's get organized a little! :)
now it seems that this "trick" is working like a charm on the DFI LP, but how abou the NF7-S? another question is, which is very important IMHO, how long does this "trick" lasts? i mean, should i redo it after i shut down my pc? or after a restart?!
answers would clear a lot of stuff for all of us, especially NF7-S users! ;)

Quote:

---

Originally posted by STEvil
ok, time for the bad news.

The reading we are getting off the mosfet is bad, I believe. *:stick:

As you will see in the attached picture below there is a point where you can measure your VDIMM next to the 3rd DIMM slot.. it appears to me that what I have been reading off the mosfet is somehow incorrect... somehow. :stick:

I also found another point just below the VDIMM mosfet that apparently is either VTT, VBT, or VREF.. dunno which of the 3 it is, but its the correct voltage for match the first measure point in the pic (both measure points not on the mosfets are spots for capacitors.. could putting capacitors there help with high fsb?).


Still, this leaves a question: How the hell are the mosfets showing an incorrect voltage?!?! :stick: :confused: :stick:

---

My voltage reading at the mosfet raises about 0.03V everytime I do the trick although it should be 0.3V.

I have 3.33V on the mosfets( 3.6V in BIOS). I have 3.5V where you get 2.88V.

So no workie for the NF7 series (1.2 & 2.0 ) yet .. ?

Glad to see some poeple having luck with this but tried about 12 times today. Measured with multimeter an does nothing for me. Tried a various voltages an checked each time an no change.

Quote:

---

Originally posted by nOx34
So no workie for the NF7 series (1.2 & 2.0 ) yet .. ?

---

No...not yet sir.

Well, I take that back...it sorta works.

I am still trying to figure a sure fire way to get it to work consistently.

Hell-Fire - Are you using an ATI 9500 or higher card for testing?

Mine are all back to normal now that I have the GF1 in..

Seems to me it's a freak incident.

:banana: :banana: :banana: :banana: i fried my bios chip twice tryin this, i give up

Yeah, probably a good idea to leave it alone until we've got it figured out.. ;)

tictac

can you program a bios to trick the save block everytime so we can get the dimm voltage mod everytime?

Quote:

---

Originally posted by STEvil
Hell-Fire - Are you using an ATI 9500 or higher card for testing?

Mine are all back to normal now that I have the GF1 in..

---

I actually tried a 9800Pro and GF2 GTS just outta morbid curiousity.

Me and STE went thru several different bios flashes between a NF7, AN35N, and DFI bios files.

The DFI file certainly would not work on the NF7....to bad.

compare the 3.3 reading with the 9800 vs. GF2, too.

with my 9700 the board read 2.96-3.2

with the gf1 i'm using now i get 3.2-3.22

bump.

Anyone else measured the mosfet vs. capacitor and found their readings to be out?

Any updates? Mainly on the NF7-S???

nadda.

Bump for lost thread.

Bumpage...STE what bios where u using?

Its mentioned about 3 times in the thread.


Evilous 4-29 with 12/18 romsip.

Hello

Registered on porpuse to share my ideas about this nice finding..

First of all, I didn´t manage to put 2.9V real w/2.6V in bios, using the method you said...I run a NF7 2.0 w/22 3dfire (great job tictac:toast: ).

Soooo, I just managed to get it in another way.

If you take the battery out, and the plug off the power suply cable for 10 secs, all the defenitions in BIOS are wiped out, and when booting without the battery, the NF7 wiil boot w/ 1.5V no Vcore and 2.8V on Vmem, although in bios its 2.6V choosen.

If you save it w/ other voltage, it will boot up with the exact voltage you´re selected, if you just go all the way into windows with no bios checking, it wiil show 2.8V on MBM(and on my DM), and when opening 8rdaVcore it will show 2.8V selected, so no way to rise it up above 2.9V.

Sory to ruin all the karma, but for what I´ve tested, it seems like in NF7 ins´t possible to Vmem using this method...

[]z

id rather have more vcore. anyone tried to see if vcore works?