GTX285 various mods.. which may work for GTX260/280/295 also

*** please mod at your own risk ***

before any further my cooling to vmodding recommendations
- stock cooling: step 1 + ocp mod
- better air/water/phase cooling: steps 1-3 + ocp mod
- multi phase/ln2/helium cooling: crotales vgpu/ocp + napalms steps 1-3

the following pics and mods @ evga ssc 285
http://i38.tinypic.com/2czexck.jpg

http://i35.tinypic.com/1zml668.jpg

10v 1500uf rubycon blackgate nx cap @ vacant C561-4
http://img135.imageshack.us/img135/6677/capmodvh5.jpg


crotale vgpu/ocp mods: http://www.xtremesystems.org/forums/...d.php?t=220607

before attempting any vmod please measure your current voltage/s

measure points
- red:vgpu
- green:vmem
- courtesy:CryptiK
http://i720.photobucket.com/albums/w...surepoints.jpg


step #1
- for better understanding please check dengyongs post #80 cryptiks post #112 and #118
- simple way to increase vgpu from 1.15v reference
- mod increments vgpu ~1.2-1.3
http://img511.imageshack.us/img511/998/49555888ru5.jpg

step #2
- for those of you that dont want to do step 1
- bridge 2 vacant from factory
http://i45.tinypic.com/nwwk03.png

- for those of you that dont want to do step 1
- 1/2/3/11 bridged from factory
http://i50.tinypic.com/2jg17cm.png

- for those of you that still want to do step 1
- for those of you not gaining much @ step 1
- for those of you that have already done step 1 and want to take it a bit further
- 1/2/3/4/11/12 bridged
http://i46.tinypic.com/24erssz.png


step #3
- one more vgpu mod :)
- mod increments vgpu ~1.35-1.4+
http://i45.tinypic.com/ml349t.jpg


http://i48.tinypic.com/651ooy.jpg


thanks goes to
- Cryptik for all the help/confirmations
- crotale for his vgpu/ocp mods
- dengyong, andressergio and everyone else contributing @ thread

:toast:

Interesting ... Thanks for sharing !!! :)

I wonder if having an eight pin power cable in place of the 2nd six pin would help any ?

Thank you for sharing. :up:

Outstanding, thank you for this.

Where did you get your SSC from? Newegg...Mwave...Tiger??

Quote:

---

Originally Posted by Orangeblast

Where did you get your SSC from? Newegg...Mwave...Tiger??

---

Would that make a difference ?

i got my ssc from newegg..

which gtx285 brands @ 4908 pcb ?

Quote:

---

Originally Posted by dengyong

Would that make a difference ?

---

It might...never know.

Quote:

---

Originally Posted by NapalmV5

i got my ssc from newegg..

which gtx285 brands @ 4908 pcb ?

---

Not sure, that might be mfg. specific. XFX and BFG have different PCB numbers? Can anyone check? It also might just be a batch number to a set of PCBs...anything is possible.

evga ssc @ 1.20v
my ssc @ 1.23v
vmod @ 1.26v

no oc gain

vgpu mod does not work on the ssc until ovp is removed/increased

so if anyone could/would share that would be greatly appreciated

Quote:

---

Originally Posted by NapalmV5

evga ssc @ 1.20v
my ssc @ 1.23v
vmod @ 1.26v

no oc gain

vgpu mod does not work on the ssc until ovp is removed/increased

so if anyone could/would share that would be greatly appreciated

---

I think everyone is looking for OVP for these 285/295 cards... Tough to find though. From what I gather the best way to increase Mhz is better cooling. with a 295 @ 1.26-1.3v (OVP/OCP kicks in around 1.3) on stock cooler you can get about 750-775 core. With watercooling people have been getting 800-830. I imagine once we can get our hands on some Tek5.0 slims and LN2 these babies we will see big gains in the core clocks

yeh 1.3v ovp :down:


heres the 285 vcore mod by Dural @ luxx forums:

http://www.forumdeluxx.de/forum/show...postcount=1340


22k fixed resistor gets 1.26v on my ssc

deleted

Don't ya cool these memory and pwn parts originally touched by the stock HS? :eek:

What did the last mod gain you in clocks?

wow, i'd be scared to kill my board with that much capacitance.
jesus.

this is fantastic. I can't wait to try this out on my 2 SSCs

wow... nice...

I think you would be much better using low value (~100nf) distributed capacitance in high load areas to catch spikes and high frequency noise than addling large ammounts of bulk capacitance to the outputs of the regulators, especially with the long lead lengths which it looks like you're using. Using long lead lengths will increase ESL and ESR of the capacitors and will likely negitively effect the response of the voltage regulators.

yeh i agree persivore.. this was just the first recap.. more recapping to come

the capping on the back of the card ^ at first i used 4 x 6.3V 470uf blackgates

but the 4 x 10V 1500uf blackgates/results @ 1st ^ post is what made me post the mod up

im looking into tantalum/film caps aswell

there are more components in the way of this recap which are deemed to be replaced

for an sli/trisli setup id recommend 4 x high quality 10V 330uf tantalum caps @ C561-4.. if you got the space for electrolytic caps go ahead

at the moment i dont recommend the recapping of the 8x100uf.. i need to recap some more.. when i do find the better the caps/capacitance ill let you guys know.. still experimental

the memory runs fine @ 1500 naked cooled just by the airflow

having said all that im playing ETQW @ 800/1750/1500

20mhz core +
50mhz shader +

the hr03 cooler itself @ idle is only about 1 C better than the stock cooler.. @ load probably 2-3 C

^nVIDIA skimping on memory vDD and vDDQ chokes. No surprise.
tsk-tsk...

But good to see they put some sense in the layout of GTX285's vGPU VRM area after the mess on GTX260/280s.

Nice work Napalm, thanks for sharing. Those 4 x 6V-10V 1500uF blackgate caps have a huge can size for 1500uf. I suppose that's one of the benefits of using high quality audio caps. The lead length is a worry though, if you can reduce that perhaps the results will be even better. I have some Sanyo WG's and Rubycon MBZ's that may be suitable for this mod.

It's also somewhat surprising that the HR-03 is only marginally better than the stock cooler.

sure as persivore stated shorter leads would benefit even more.. but space on the card is a beatch.. not only that but the ram on the mobo is in the way aswell

yeh its the nx 10v 1500uf bgs.. sounds as good as it performs on the 285

Quote:

---

Originally Posted by NapalmV5

how much gain ? ill find out later today

---

waiting like a little kid @ xmas. :P

lol ok ill get to it in a moment..

i was trying to get true120 copper on.. it fits! :D

though ill go back to hr03 after this test run

Dayum... I can hear that card moan because of the weight on its shoulders (and maybe because of pimped guts that got no chance to "heal" yet :D). :lol:

Interesting mod, but video cards are not 12v car amplifiers... it would seem the initial load on the FETs from charging the caps would eventually kill the card. It is one thing to reduce ripple, but another to overload the card with capacitance....

well last night after true120 copper install (stupid idea) :shakes: everything went to hell

at first the card didnt boot.. after adjusting the cooler i got it to boot.. the weight of the cooler pulled on the backplate after i removed it i noticed one of the small ceramic cap on the back of the gpu might have been touching the metal plate.. or at least it looked like it im not certain.. most im concern about.. as for the caps ive recapped several times to no available.. the card is in bad shape.. everything gained was lost even default clocks stability is gone lol kaos!

hope the ceramic cap @ the back of the gpu is fine.. recapping to tantalum caps is what i wouldve done in the end anyways.. the next mod was gonna be the inductors but im gonna leave it that for now

let you guys know in a couple of days how the recapping to tantalum caps goes

btw.. have been playing around with a zotac amp 285.. vcore modded @ same voltage as ssc.. its doing exactly the same: 750/1700/1490 only 10mhz less on memory just a slight v increase to match ssc's voltage/clocks is all it needs

Quote:

---

Originally Posted by NapalmV5

well last night after true120 copper install (stupid idea) :shakes: everything went to hell

at first the card didnt boot.. after adjusting the cooler i got it to boot.. the weight of the cooler pulled on the backplate after i removed it i noticed one of the small ceramic cap on the back of the gpu might have been touching the metal plate.. or at least it looked like it im not certain.. most im concern about.. as for the caps ive recapped several times to no available.. the card is in bad shape.. everything gained was lost even default clocks stability is gone lol kaos!

hope the ceramic cap @ the back of the gpu is fine.. recapping to tantalum caps is what i wouldve done in the end anyways.. the next mod was gonna be the inductors but im gonna leave it that for now

let you guys know in a couple of days how the recapping to tantalum caps goes

btw.. have been playing around with a zotac amp 285.. vcore modded @ same voltage as ssc.. its doing exactly the same: 750/1700/1490 only 10mhz less on memory just a slight v increase to match ssc's voltage/clocks is all it needs

---

Must be luck of the draw cause my vanilla zotac does over 1500mem stable with no mods. (yet)

^ nice.. what core/shader @ ?

yeh im not surprised.. its luck of the draw since all 285s use the same memory

Quote:

---

Originally Posted by NapalmV5

^ nice.. what core/shader @ ?

yeh im not surprised.. its luck of the draw since all 285s use the same memory

---

Using precision I run 753/1675/1507

I think the memory will do a bit more but I haven't tested.

i precision set @ those clocks and when i run ETQW the system reboots lol vcore modded!!

looks like i got the vanilla and you got the amp for a vice versa $$$ :rofl: not fair!!

Quote:

---

Originally Posted by NapalmV5

i precision set @ those clocks and when i run ETQW the system reboots lol vcore modded!!

looks like i got the vanilla and you got the amp for a vice versa $$$ :rofl: not fair!!

---

I'm getting a full cover block for it as soon as they're available, I have to turn the fan up to 90% to play crysis for more than 15min.

water should help get 50/50/50 + on top of mods


amp @ 1.26v i can game ETQW stable @ 800/1764/1485 @ precision.. no cap mods

though a few mem mhz less thats smgt my ssc is not capable of @ same voltage without the cap mods

deleted

update: ssc gpu is fine.. ssc is back!

i just tried 4 x 820uf 4V chemicon caps where 4 x 330uf tantalum caps normally occupy and now at least stock clocks are stable again

now i can breathe easier.. :) whephew

time to find out what really makes the 285 tick :D

He he he....

I think that poor SSC needs to be booked in for some trauma support :D

Napalm, have you tried RT softmod to overcome OVP?

I like your true120 copper on GPU, almost 2Kg is pretty extreme :up:

^ lol its finally out of intensive trauma unit and doing well now

^ unfortunately isl6327 + rt: no go


clocks back @ ~800/1750/1500 @ ETQW

no loss @ clocks/performance

as for gain.. instead of the bulky caps on the back.. these kemet caps perform just as good and keep temps down

the cornell fim caps are bulky themselves and the leads would be even longer so i didnt even bother.. the small film caps would go better

vmem
http://i48.tinypic.com/651ooy.jpg

10K : 2.25-2.29v
4.7K : 2.34-2.37v

Those caps look much better, although having as much value as the card in caps added to it was good for the extreme factor ;) If they give so much benefit (30MHz core) I wonder why for the sake of saving a few cents Nvidia didn't put caps on the vacant spaces in the first place.

Just to clarify, was it the addition of the caps on the 4 vacant spots on the back of the card that gave the initial 30MHz increase in core? Then the swapping of the solid caps for the 1000uf 16v blackgate fk's that gave the extra 20MHz?

Nice work on the vmem mod too :up:

thanks

yeh the caps on the back and front pushed the ssc further.. it wouldnt budge no matter what.. vcore only .03v higher didnt help anything.. neither the cooler

without the caps 800/1750/1500 no way in hell @ ETQW

Nice work Napalm! :up:

thanks


well guys i believe this is the ovp mofo:

http://img204.imageshack.us/img204/5889/ovpod3.jpg

pretty easy to find thanks to the isl6327 datasheet

the 100K resistor rings all the bells to me.. what do you guys think?

it would have to be removed and replaced with a ~50k or less or vr

can you guys confirm? if someone confirms ill replace the damn thing in a heartbeat :)

I know you know what you are doing, but I have to ask are you certain it is OVP triggering the shut down though? That is, if you raise vGPU during idle, does the card get a black screen and vGPU ~0v or reboot when ~1.30v+ is reached? Or is it only under load it results in black screen or shutdown? If its the second option it is most likely OCP being triggered, unless under load there is .

I'd need to see a close up pic of the area surrounding the ISL627 chip, but from the circuit diagram it appears OVP is controlled by one pin (44), and you may very well be correct with your identification. Can you locate pin 44 and follow where the trace from that pin leads?

To completely remove OVP it says to leave the pin open so removing the resistor completely would remove OVP. However if you wanted to increase OVP point you could add a 100K resistor in parallel giving 50K total, giving a 50% reduction in OVP signal, or you could use 100K VR in parallel and tune total resistance to allow and adjustable OVP point.

Doing the OCP mod is harder obviously, as you'd need to do something similar to what I did with the G80 OCP/vdroop mod which was solder SMC resistor across each capacitor in each phases R/C circuit controlling Isen for each phase (linked to Isen+ & - pins for each phase).

pin 44 doesnt lead to anything on the front so it leads to the back.. right to that resistor.. at least how i see it


even @ 10k/15k resistor the vcore still remains @ 1.26 and the card doesnt os boot.. i say it has to be ovp kicking in.. i dont think ocp would be active before the ovp at this voltage

thanks for the help.. ill try that ^

Ok sure, can you measure if there is continuity between pin 44 and one end of that 100K resistor? (I'm not sure if the pins are large enough to probe individually). If so, I'd go for it.

If it won't even load OS @ 1.26v then I'd agree its probably OVP triggering shutdown. It would IMO be safer to add a 500K VR to the 100K resistor in parallel so you'd only be getting around 83K total resistance reducing the OVP signal by ~17%, so if its not the correct resistor it wont be causing a very large alteration in circuit behaviour. If this at least lets you load OS at 1.26v then you know you're on the right track, so reduce it to 50K and try again.

i added a 100k resistor and now the cores finally getting 1.3v @ 10k but no os boot :(

added 47k 22k 10k (1.32vcore) no os boot

ill try 200k next

edit: @ 200k added to the 100k vcore @ 1.27 @ 10k

still no os boot

Do you mean you'll now remove the existing 100K resistor and put in its place a 200K resistor?

The minimum OVP point with a valid VID is VID + 150mv, so if EVGA have set VID to 1.20v OVP should not trigger until 1.35v minimum. However if they are adjusting vgpu above the set vid by bios or another method, while retaining the standard VID of 1.15v, that puts OVP at 1.30v like is happening.

Since that lowering that resistors resistance is having the effect of raising the vgpu the core receives, but even adding 10K to it decreasing OVP signal by ~91% didn't allow OS boot at only 1.32v, perhaps it's worth considering removing it entirely leaving the OVP pin open? This isn't 'safe' but shouldn't cause any negative effects as pin 44 is only used for OVP. If that doesn't allow boot at any voltage, something else must be limiting it.

That's if adding the 200K resistor doesn't work.

yeh i was thinking the same.. removal

im doing this on the amp 285

havent tried it on the ssc yet

It doesn't show on the circuit diagram what OVP pin connects to. The pins that are connected to ground through a resistor is pin 11 (I out), TCOMP, OFS, FS, SS. SS is pin 43, right next to OVP.

Sure pin 44 is connected to that 100K resistor?

The only real way to work this out is find continuity between pin 44 and a nearby component with the DMM.

i removed the 100k resistor.. 10k @ vcore: 1.27vcore still no os boot

1.26vcore os boot
1.27vcore no os boot

something else in the way..

it sure looks ovp pin goes to the back of the card to that resistor.. i could be wrong

You'll have to use DMM to find continuity between pin 44 and a nearby component.

EDIT - perhaps altering the VID resistor arrangement is a way around this issue? See the table in the data sheet.

^ indeed.. vid resistor arrangement for higher vcore should work


ive read more @ datasheet.. looks like ocp occurs @ 2v+

Quote:

---

If the voltage is higher than
2V, ISL6327 will go into the OCP mode

---

do you think vrrdy (pin 45) has also smtg to do with this?

Quote:

---

VR_RDY - VR_RDY indicates that the soft-start is completed
and the output voltage is within the regulated range around
VID setting. It is an open-drain logic output. When OCP or
OVP occurs, VR_RDY will be pulled to low. It will also be
pulled low if the output voltage is below the undervoltage
threshold.

---

check this out:

Quote:

---

Overvoltage Protection
Regardless of the VR being enabled or not, the ISL6327
overvoltage protection (OVP) circuit will be active after its
POR. The OVP thresholds are different under different
operation conditions. When VR is not enabled and before
the 2nd soft-start, the OVP threshold is 1.275V. Once the
controller detects a valid VID input, the OVP trip point will be
changed to the VID voltage plus 175mV.

---

VR_RDY appears to trigger the phase shutdown after OVP or OCP is triggered, so although it does have something to do with it, it's not the cause of the problem. The main issue here is the data sheet doesn't show how OVP is configured.

As you saw, before a valid VID is read its 1.275v and after its VID + 175mv. Getting around this is not straightforward. At this stage all I can think of is VID alteration unless I can figure out how to get around the OVP sensing the actual voltage.

However it seems that this VID + 175mv is set internally, as for example with the OCP point they explain which resistors control this, but there is no mention of how OVP is controlled with resistors or an R-C circuit.

vid resistor arrangement ?

http://img244.imageshack.us/img244/2594/vmodph8.jpg

ssc has a resistor at the spot where the amp does not

my ssc: 1.23v
my amp: 1.13v

looks like they have to be as there are none on the front side i can see corresponding to the VID pins.

However all those spaces are filled and I cant see any VID's that have all 1's. Unless the resistors differ in value so for ex a 30K is read as open and a 50K is read as closed.

What do those resistors measure? Are they all the same? Just wondering where the 30-50K on your pic comes from. They may be part of a circuit that adds to the resistance so the actual resistor may be less than that.

Also I have a non ssc model coming in a couple of days and i'd like to add a resistor in the space to raise VID to SSC level if i can confirm they are the VID resistors and confirm the value of the resistor to add there.

dont mind the 30-50k

thats what i used to increase vcore and vmem on the amp.. posted that @ news forum thread

i did measure but i forgot.. let me measure again

be right back

So what exact resistance did you use in that space on the zotac to raise vgpu & vmem? (I will do this on mine when it arrives).

I'm just thinking if we can see a high/low resistance pattern and match it to the open/closed pattern in the VID table we could figure out which resistor is which in the VID table.

22k produced 1.26vcore/2.17vmem


so there are 16 spaces.. on the ssc 1/2/3/11 are occupied where 1/3/11 are occupied on the amp the rest are vacant

on my not so great dmm all resistors are of the same value: 1ohm

gotta get a better dmm for such finesse measuring

I can't see an obvious pattern there. Nvidia have been known to put VID resistors in strange arrangements, like 1,4,2,3 left to right for example.

Are the VID pins on the ISL6327 chip large enough to get the probe of a DMM on? If you can find which VID pins have continuity with specific resistors in that row, that should allow resistor identification.

after more fiddling around i doubt any kind of vid resistor arrangement allows past the 1.27/1.3 limit.. they built it such.. no way around that if theres limit elsewhere

the 100k may be the vr_rdy why it allows 1.32vcore @ 100k added

all 3 right here are 10k which possibly have to do with the 1.3ovp
http://img293.imageshack.us/img293/7918/10kst7.jpg

which one.. ill look into later on tonight

tried 10k resistor at all 3.. still no os boot

as for finding continuity my dmm is not up to the task.. will get one more appropriately

only so much info to go by from the datasheet.. im gonna let the experts have at it

once vcore freely gets more voltage.. clocks will take off

in the meantime.. cryptik/if you guys have anymore ideas.. let me know


after reading some more @ page 18-19

Quote:

---

ISL6327 begins the
soft-start and ramps the output voltage to 1.1V first. After
remaining at 1.1V for some time, ISL6327 reads the VID
code at VID input pins. If the VID code is valid, ISL6327 will
regulate the output to the final VID setting

---

ill remove all mods and play around @ vid pins

well looks like vid pins are more effective than resistor @ vcore

after several combos.. 1.00-1.25vcore most im getting

no resistor required just by connecting the points

so what is the combo to 1.3 1.4 1.5vcore ?

anyone ? :)

The OVP and VR_RDY pins are only logic level outputs to indicate the status of the voltage regulator, and shouldn't have any effect on the OVP threshold. The OVP threshold is derived by comparing VDIFF (effectively the output voltage of the regulator) and the reference VID value from a DAC + 175mV.

You might be able to tweek the reference voltage from the DAC by pulling the DAC pin up to a higher voltage (depending on the output impedance of the DAC), although doing this might damage the voltage regulator. If that doesn't work, the only other way to overcome OVP is to change the VID values.

I thought about that option but deemed it too risky considering the expense of these cards.

That's why I thought the only safe option is to attempt to modify the VID value through addition/removal or resistors/bridges. However its the identification of the appropriate resistors/spaces that is the tricky part without being able to test continuity between the VID pins themselves and the resistors.

I did the VID mod to get 1.22 vgpu, 2.13/2.14 mem. I had 1.15/1.16 vgpu and 2.11v mem stock.

Did you figure out if bridging any of the other VID contacts raises the vgpu any more than just having 1/2/3/11 bridged?

Quote:

---

Originally Posted by persivore

The OVP and VR_RDY pins are only logic level outputs to indicate the status of the voltage regulator, and shouldn't have any effect on the OVP threshold. The OVP threshold is derived by comparing VDIFF (effectively the output voltage of the regulator) and the reference VID value from a DAC + 175mV.

You might be able to tweek the reference voltage from the DAC by pulling the DAC pin up to a higher voltage (depending on the output impedance of the DAC), although doing this might damage the voltage regulator. If that doesn't work, the only other way to overcome OVP is to change the VID values.

---

any idea to 1.3 1.4 1.5 heck even 1.6 ? :)

Quote:

---

Originally Posted by CryptiK

I did the VID mod to get 1.22 vgpu, 2.13/2.14 mem. I had 1.15/1.16 vgpu and 2.11v mem stock.

Did you figure out if bridging any of the other VID contacts raises the vgpu any more than just having 1/2/3/11 bridged?

---

bridge 4 and you may get 1.24-1.25v

after numerous combos on the amp i got to remove all 1/3/11 bridges.. i got as low as 1v and as high as 1.25v

havent had the time since then to try some more

its possible that the vid code validity goes hand in hand with the vid pins + vid bios.. im not sure since i didnt get to try

looks like nibitor maker finally released 4.8 but not the 4.9 going around that no one is sharing

dont know if 4.8 supports 285


still doesnt support 285

anyone out there willing to share 4.9 ?

Quote:

---

Originally Posted by CryptiK

I did the VID mod to get 1.22 vgpu, 2.13/2.14 mem. I had 1.15/1.16 vgpu and 2.11v mem stock.

Did you figure out if bridging any of the other VID contacts raises the vgpu any more than just having 1/2/3/11 bridged?

---

CryptiK
Did you use a resistor at vid point 2 or just join the contact pads to end up with 1/2/3/11 on a std card?
Has it helped at all ?

just bridge the points by solder/wire

if you use 22k resistor @ one of the points or by bridging the points with a 22k resistor then you get 1.26vcore/2.17vmem

Quote:

---

Originally Posted by bito

CryptiK
Did you use a resistor at vid point 2 or just join the contact pads to end up with 1/2/3/11 on a std card?
Has it helped at all ?

---

As Napalm said just bridge it with solder, the 'resistors' they use are 0 - 0.1 ohms anyway (basically a dead short between the 2 contacts). Max stable clocks did improve after the mod.

On my card (base model EVGA):

1/3/11 = 1.16 vgpu / 2.11 vmem
1/2/3/11 = 1.22 vgpu / 2.13-2.14 vmem

EDIT - bridged the 4th contacts (so now have: 1/2/3/4/11 bridged), now have 1.23 vgpu / 2.15 vmem under load, basically at SSC voltages now.

Raising voltage hasnt had a large effect on load temps, before mods on auto fan speed it was ~80*C in really heavy scenes of crysis, at 1.22v is was around 83*C, and now at 1.23v it's around 85*C (but today is warmer weather too). Fan doesn't go over 72%.

Great.....I have some silver conductive paint here, from the old Athlon days which i used for the same thing :D
Can be removed as well if need be. Thank you for sharing your efforts.

Quote:

---

Originally Posted by CryptiK

As Napalm said just bridge it with solder, the 'resistors' they use are 0 - 0.1 ohms anyway (basically a dead short between the 2 contacts). Max stable clocks did improve after the mod.

On my card (base model EVGA):

1/3/11 = 1.16 vgpu / 2.11 vmem
1/2/3/11 = 1.22 vgpu / 2.13-2.14 vmem

EDIT - bridged the 4th contacts (so now have: 1/2/3/4/11 bridged), now have 1.23 vgpu / 2.15 vmem under load, basically at SSC voltages now.

Raising voltage hasnt had a large effect on load temps, before mods on auto fan speed it was ~80*C in really heavy scenes of crysis, at 1.22v is was around 83*C, and now at 1.23v it's around 85*C (but today is warmer weather too). Fan doesn't go over 72%.

---

Thank you CryptiK and NapalmV5.

I bridged 2 and 4 as well today, for a nice little performance boost on my vanilla zotac. http://service.futuremark.com/compare?3dm06=9927577
Attachment 94365Attachment 94366

nice ^ :up:

your welcome :)

@NapalmV5, is there any chance for pencil vcore mod for GTX 285, as softmod is impossible for GTX 285 and GTX 260 55nm?
Thank's!

sorry nickless no pencil but all it takes is silver conductive paint to bridge a vid pin

http://www.xtremesystems.org/Forums/...2&postcount=56

if you bridge vid pin 2 and 4.. take a look at post #71 by cryptik

and results @ post #73 by dengyong

Quote:

---

Originally Posted by NapalmV5

sorry nickless no pencil but all it takes is silver conductive paint to bridge a vid pin

http://www.xtremesystems.org/Forums/...2&postcount=56

if you bridge vid pin 2 and 4.. take a look at post #71 by cryptik

and results @ post #73 by dengyong

---

Thank's mate, you are great. Last question, can I remove this silver conductive paint and how, when I decide to sell the card?

Yes silver conductive paint is easily removed with some isopropyl alcohol or acetone on a Q-tip. If you use acetone be careful what you get it on, it is quite a strong solvent, but I have not had an issue with it yet.

Napalm, would you mind measuring the shader voltage on your cards? I use one of the small caps a centimeter or so towards the rear of the card from the shader chip (on the back of card) - I'm pretty sure this is the shader voltage anyway.

For anyone confused about the vmod here's a pic.
All credit to CryptiK and NapalmV5.
Attachment 94504

I made it, I bridged 2 and 4 pin with silver conductive paint and now I have 1.22v Vcore and 2.12v Vmem. Thank's a lot to NapalmV5, dengyong and CryptiK :worship:.
Get some test now :hehe:

Quote:

---

Originally Posted by nickless

I made it, I bridged 2 and 4 pin with silver conductive paint and now I have 1.22v Vcore and 2.12v Vmem. Thank's a lot to NapalmV5, dengyong and CryptiK :worship:.
Get some test now :hehe:

---

NapalmV5 and CryptiK did all the work and deserve all the credit. :up:

hmm, i have an eVGA SSC - the first three from the left already have resistors on them. thats all though, 1,2 and 3 - the rest are blank.

would bridging set 4 help any you think?

actually, looking again - 1,2,3 and 11 are resistored already.

1/2/3/11 is normal for SSC version, this gives ~ 1.23vgpu/2.15vmem. If you bridge 4 you will get a little extra voltage. I currently have 1/2/3/4/11 bridged and I'm experimenting with other arrangements.

Quote:

---

Originally Posted by CryptiK

As Napalm said just bridge it with solder, the 'resistors' they use are 0 - 0.1 ohms anyway (basically a dead short between the 2 contacts). Max stable clocks did improve after the mod.

On my card (base model EVGA):

1/3/11 = 1.16 vgpu / 2.11 vmem
1/2/3/11 = 1.22 vgpu / 2.13-2.14 vmem

EDIT - bridged the 4th contacts (so now have: 1/2/3/4/11 bridged), now have 1.23 vgpu / 2.15 vmem under load, basically at SSC voltages now.

Raising voltage hasnt had a large effect on load temps, before mods on auto fan speed it was ~80*C in really heavy scenes of crysis, at 1.22v is was around 83*C, and now at 1.23v it's around 85*C (but today is warmer weather too). Fan doesn't go over 72%.

---

hmm. i just saw this. adding the 4th pin does infact give the extra .01 on vcore?

this is a great thread. i thank you guys for your willingness to share!

your efforts are appritiated :)

Yes for me bridging 4 gave me an extra 0.01v on both vgpu & vmem.

Thanks for this guide guys ,well done. I would like to know where to measure v.core and v.mem ? where are the points to check with DMM ? i just got an xfx GTX 285xxx and i would like to measure my stock voltage and try this mod after testing the card on stock volts.
Any help and guidance will be greatly appreciated.

chispy.

Quote:

---

Originally Posted by nickless

I made it, I bridged 2 and 4 pin with silver conductive paint and now I have 1.22v Vcore and 2.12v Vmem. Thank's a lot to NapalmV5, dengyong and CryptiK :worship:.
Get some test now :hehe:

---

I was going to reply to your PM but it looks like you have it sorted now.

Quote:

---

Originally Posted by chispy

Thanks for this guide guys ,well done. I would like to know where to measure v.core and v.mem ? where are the points to check with DMM ? i just got an xfx GTX 285xxx and i would like to measure my stock voltage and try this mod after testing the card on stock volts.
Any help and guidance will be greatly appreciated.

chispy.

---

Hey Chispy, here's the points I use. Red = vgpu measure point, green = vmem measure point.

http://i720.photobucket.com/albums/w...surepoints.jpg

^ dengyong/cryptik - thanks

nickless - looks like you got it figured out thanks to cryptik/dengyong

if you guys also bridge vid 10 you may get 1.24-25vcore

if you guys want ~1.26vcore which is the most doable right now/~2.17vmem.. just solder a 22k resistor from ground to one of the points @ vid 2

your welcome guys :)

Quote:

---

Originally Posted by dengyong

NapalmV5 and CryptiK did all the work and deserve all the credit. :up:

---

forget credit.. sharing is more rewarding :)

Quote:

---

Originally Posted by CryptiK

Napalm, would you mind measuring the shader voltage on your cards? I use one of the small caps a centimeter or so towards the rear of the card from the shader chip (on the back of card) - I'm pretty sure this is the shader voltage anyway.

---

is it the smaller cap above/right of C1285 ?

I measured 1.18v at both caps to the right of where 'C1277' is written on the PCB. I may be wrong though this might not be shader voltage.

Also I just bridged VID 10 (as well as having 1/2/3/4/11 bridged) and got no change in voltage. Then I removed VID 4 leaving 1/2/3/10/11 bridged and got 1.21v load and a BSOD at stock clocks. No idea what caused it but I'm removing VID 10 and hope stability returns.

If 2/4/ give 1.22-1.23v, can I try 2/4/10 for some extra vcore?

[off] Are commands for GTX 285 in nvflash the same like GTX280:
For Normal Flash: nvflash -5 -6 newbios.rom
For Forced Flash: nvflash -4 -5 -6 newbios.rom [off]

Quote:

---

Originally Posted by CryptiK

I measured 1.18v at both caps to the right of where 'C1277' is written on the PCB. I may be wrong though this might not be shader voltage.

Also I just bridged VID 10 (as well as having 1/2/3/4/11 bridged) and got no change in voltage. Then I removed VID 4 leaving 1/2/3/10/11 bridged and got 1.21v load and a BSOD at stock clocks. No idea what caused it but I'm removing VID 10 and hope stability returns.

---

sorry about that.. after so many combos..

i had:
1/2/3/4/5/11
1/2/3/4/5/10/11
1/2/3/4/5/9/10/11
many others

i dont remember exactly how i reached ~1.25vcore

probably vid 5 so: 1/2/3/4/5/11

im gonna confirm that in a bit..

Quote:

---

Originally Posted by nickless

If 2/4/ give 1.22-1.23v, can I try 2/4/10 for some extra vcore?

[off] Are commands for GTX 285 in nvflash the same like GTX280:
For Normal Flash: nvflash -5 -6 newbios.rom
For Forced Flash: nvflash -4 -5 -6 newbios.rom [off]

---

no just 1/2/3/4/11

let me see if i can get it ~1.24-25vcore


yes same commands

[QUOTE=CryptiK;3640847]I was going to reply to your PM but it looks like you have it sorted now.



Hey Chispy, here's the points I use. Red = vgpu measure point, green = vmem measure point.
QUOTE]


Thanks a lot Cryptik , appreciate your help bro. Testing today i will let you know how this mod goes on my card.


chispy.

Quote:

---

Originally Posted by NapalmV5

sorry about that.. after so many combos..

i had:
1/2/3/4/5/11
1/2/3/4/5/10/11
1/2/3/4/5/9/10/11
many others

i dont remember exactly how i reached ~1.25vcore

probably vid 5 so: 1/2/3/4/5/11

im gonna confirm that in a bit..




no just 1/2/3/4/11

let me see if i can get it ~1.24-25vcore


yes same commands

---

Don't apologise, I'd been bridging things experimenting myself, and was minutes away from bridging VID 10 of my own accord.

I tried 1/2/3/4/5/11 and it didn't cause any change in voltage for me. I'd like to get 1.25v, I think that would be a good balance between heat & speed.

Quote:

---

Originally Posted by chispy

Thanks a lot Cryptik , appreciate your help bro. Testing today i will let you know how this mod goes on my card.


chispy.

---

No problem mate, hope you get some good results with it.

i just redid 1/2/3/4/5/11 and @ startup its ~1.24vcore and @ load ~1.25-26vcore



next ill try 1/2/3/4/5/6/11

edit: startup @ 1.23vcore load @ 1.25-26vcore

http://www.xtremesystems.org/Forums/...2&postcount=77

Quote:

---

Originally Posted by mikeyakame

fgw,

Shall do that mate and get back to you.

I just finished reading and re-reading the Intersil ISL6327CRZ multiphase closed-loop feedback and compensation PWM Voltage Reg datasheet, the VR used by NV in the die shrink G200 reference cards, GTX 260 55nm and GTX 285 55nm.
Seems NV screwed you guys over with VR cost cutting this time around, mind you the IC itself only costs $5usd or so.

It's a VR designed with cost savings and affordability in mind so it lacks any kind of external low-pin count bus interface accessibility and relies on the pcb designer to set up all the necessary operational values in design and testing phase. There are 8 VID pins which when pulled up the base VID is read by which pins are 0 or 1 and converted to digital signal by DAC. '

Voltage adjustment is done through current offset via the OFS pin, which is polarity dependant for negative or positive offset, switch to Vss (gnd) for neg, switch to Vcc (vin) for pos. Offset is configured through external resistors connected between the DAC and REF pins, the impedance dynamically controls the current which in turn determines the Vout (voltage out).

Long story short, unless NV have some kind of low pin count programmable interface to adjust a pull up or pull down resistor connected between DAC digital VID signal pin and REF voltage reference pin there is absolutely NO CHANCE AT ALL to control Voltage Adjustments via software on either the GTX 260 55nm or GTX 285 55nm. The only way would be to change the resistor between DAC and REF pins, to give a greater current offset to drive Voltage out higher.

If EVGA is saying it is possible on their cards, short of physical changes to the NV reference cards to add this basic functionality I can't see how it is possible at all. I read over the datasheet a good 4 or 5 times just to make sure I completely understood the electrical design and don't see any way to do it.

---

lower resistance @ the 1kohm resistor dac/ref might lead to the back of the card lowers vcore.. higher resistance doesnt allow os boot @ 1.33vcore

suggestions?


cryptik.. what you think about the 1k resistor on the back.. its the 1st resistor at the bottom right of the vid pins.. the 26.6k resistor is the 2nd one

Interesting, what do you get with just 1/2/3/4/11 ?