Evga Voltage Tuner

[AgentGOD]

Out of respect for EVGA co. I am removing this link. I know it won't stop people from getting it, but it's the least we can do for our great friends at EVGA - runmc

[Cold Fussion]

My card's overclockability doesn't increase when i increase to voltage from 1.125 to 1.2, 702/1404/2600 is still the max i can even with that significant voltage bump :\

[AgentGOD]

My card's overclockability doesn't increase when i increase to voltage from 1.125 to 1.2, 702/1404/2600 is still the max i can even with that significant voltage bump :\

A moderator on EVGA said the shader clock goes higher when you have better cooling.

[mikeyakame]

A moderator on EVGA said the shader clock goes higher when you have better cooling.

That's exactly right. If I keep my core temp low to mid 60s I can run 1566 shader, otherwise the limitation is 1512.

Others with vmods and chilled water setups have run upto 1674mhz shader on their GTX280s. Temps were down in the mid to high 30s though.

[jaredpace]

I've been using Riva Tuner to adjust voltage on my GTX 260. Running clocks at 706/1501/1306, on 1.15V (up from 1.13V stock),in COD:WAW I'm getting a solid 60 fps with V-sync enabled, all eye candy enabled, run at 19x12 resolution. Other games behave similarly. I also force triple buffering with Riva Tuner. How much more could I get with a GTX 280 or 285?

Just being able to adjust voltage to stabilize clocks helps these 65 nm cards render at the same level of the higher cards. I really don't think that you'll see all that much of an improvement on the higher level cards as compared to the 260's. With this voltage tuner, EVGA just extended the life of the 65nm 260's.

The original 192 shader / full copper / 14 layer PCB / volterra VRM / 4 phase Pwr / 65nm gtx260's are obviously the most well built & robust and ATM are cheapest:

At the end of last year, GeForce GTX 260 became the first one of NVIDIA GTX200 series graphics cards to use 55nm processing technology. It featured P654 reference design, which cost less than earlier P651. The number of PCB layers reduced from 14 to 10, and it has abandoned the expensive Volterra chip to cut cost. Quite soon, the third-generation GTX260 design plan with codename of “P897/D10U-20″ will be also surfaced.
You might be interested in reading: The First Review of 55nm GeForce GTX260

The schedule drawing of P897 PCB design
According to the P897 design plan that NVIDIA sends its partners, it uses 4/6 phase NVVDD power solution ADP4100. It changes FBVDDQ power solution from 2-phase to single phase. And the MOSFET package has been changed from LFPAK to DPAK to save cost. Another noticeable change is the PCB layer decreases from 10 to 8. The length of PCB keeps unchanged, while the height of it is reduced by 1.5cm. In order to cut cost further, they will change DVI connector, crystal and probably change BIOS Rom from 1M to 512K.

But for the GT200 and NVIO2 chip, you might mistake GeForce GTX260 of P897 PCB design for GeForce 9800GTX+ (as they look so similar). Compared with P654 design, P897 GeForce GTX260 is expected to save cost of $10 to $15, which will undoubtedly improve its competitiveness. According to our source, this product will be available in the third week of this month.
We also manage to get some pictures of new GTX260 from a Chinese manufactuer Colorful. Coming from their iGame Series, this GeForce GTX260 card is designed based on P897 PCB design. They have just changed the TV-out connector to HDMI and added a set of overclocking jumper.

http://en.expreview.com/2009/02/07/g...-cut-cost.html

[Sailindawg]

The original 192 shader / full copper / 14 layer PCB / volterra VRM / 4 phase Pwr / 65nm gtx260's are obviously the most well built & robust and ATM are cheapest:

If you can still find one.

OOOPS....Get them while they're hot

[PizzaMan]

Tried this with the patch on my 9600GT. GVT loaded up with the slider bar all the way to the right. Didn't work......

[Leeghoofd]

A moderator on EVGA said the shader clock goes higher when you have better cooling.

Well I had my first GTX280 under water and it had the same limited clocks for the shaders as the stock fan at 100%... ram and GPU were able to go higher... more volts and better cooling help to make the shaders go higher then I would agree ( or a binned GPU :p )

[T_Flight]

Currently still bumping my clocks after installing my watercooling lastweek. I still haven't bumped voltage, but may be getting in that range now. I'm at 730/1566/1250 stock voltage on my EVGA GTX280 SSC. It has a DTek GFX2 waterblock and Unisink. GPU Temps are running 38-39C currently, and GPU-Z is showing VRM Temps ranging 54-63C with a GPU2 Folding client folding a 511 point WU.

Cooling is what does it on these cards. Before my temps idled higher than this thing runs at loaded and OC'd as high as it is now. It used to load in the 60's. An OC like I have now would've been impossible without watercooling. It just wouldn't do it. I tried. Especially on the shaders. Shader power rules all with these cards.

[jaredpace]

Would someone mind emailing me Agent GOD's Evga voltage tuner patch? They're pulling links left & right. jaredpaceatgmaildotcom

thank you!

[T_Flight]

Might wanna put some spaces in that e-mail addy or the bots are gonna have a field day with it.

[jaredpace]

Ah thanks. And thanks I've received it now. hehe - quick!

[Leeghoofd]

SSC are sort of binned GPU's and mostly run at slightly higher volts then vanilla's..; I had 3 vanilla cards ( 2 EVGA and one POV) and all capped out around the same shader speed for benching (480 region) GPU around 745 and rams were at 1250-1300 max,;

Very nice clocks T-Flight, praise yaself lucky mate hope the tuner brings out even more !!

[Sly Fox]

An OC like I have now would've been impossible without watercooling.

Naaaaaaaaaaaaah.


You'd just need the fan at 100%, no big deal as long as your PC's kept in a seperate noise dampened room.

[T_Flight]

I did have my Fan at 100%. Stock cooling will not do it. As soon as you put a load to them they go over the 50's and can get into the 60's. This thing loaded never gets above the low 40's. It makes a HUGE difference. I couldn't get above 675/1462 before no matter what I did. It just wouldn't do it. They like to run cool. The watercooling is what did it for me.

[Sly Fox]

I did have my Fan at 100%. Stock cooling will not do it. As soon as you put a load to them they go over the 50's and can get into the 60's. This thing loaded never gets above the low 40's. It makes a HUGE difference. I couldn't get above 675/1462 before no matter what I did. It just wouldn't do it. They like to run cool. The watercooling is what did it for me.

That surprises me, I would have figured that in the 50's and 60's stability wouldn't really be an issue yet. But as you pointed out, sounds like these cards like to be cool.

With all this voltage tuning talk and whatnot I want to get a waterblock myself.

Just hope that an MCR-320 could handle it + a Q6600. At the moment... I'm not that optimistic.

[T_Flight]

It should. Look in ym sig for my setup. I'm running a single loop. Those Kazes are running at 2000 via a controller. The HW Labs Rads are a little bit more efficient with the faster fans, but I still think you should be good, becasue I have an i7 and you have a Q6600. I'm running at 4.2 with HT also.

I do believe it's the cooling that's doing it. It seems like all the stock watercooled EVGA cards seem to do better, and the ones with custom watercooling do also. That's not saying they all will. You know YMMV on that when it comes to OC'ing, but I really believe it has something to do with the reduced temps.

Another thing I noticed also with the Unisink is it brought down VRM temps which might add to stability, but that's only a theory of mine. I have to do more research on that. It was said by Unwinder on the EVGA Forums in a thread that these VRM's can handle up to 150C. I wouldn't ever get near that, but it's nice to know there is some wiggle room in case something would go wrong. It might be that when they are cooler, they give better electrical properties though.


Aw, and guys, if you read this, please don't push those VRM's. They have a design spec of that, but what we don't know is the time:temp ratio. In other words how long will they handle that? It might be seconds or it might be minutes. I wouldn't push them and find out. hehe

[noobzed]

someone could pm me the patch link ? ty

[jaredpace]

someone could pm me the patch link ? ty

removed at the request of clo007

[clo007]

They removed the link for a reason. why don't you be respectful of that and stop linking more.

[Sly Fox]

It should. Look in ym sig for my setup. I'm running a single loop. Those Kazes are running at 2000 via a controller. The HW Labs Rads are a little bit more efficient with the faster fans, but I still think you should be good, becasue I have an i7 and you have a Q6600. I'm running at 4.2 with HT also.

Yea, I'd imagine that there's no way my Q6600 is going to produce more heat than an i7. Those things are beastly.

Another thing I noticed also with the Unisink is it brought down VRM temps which might add to stability, but that's only a theory of mine. I have to do more research on that. It was said by Unwinder on the EVGA Forums in a thread that these VRM's can handle up to 150C. I wouldn't ever get near that, but it's nice to know there is some wiggle room in case something would go wrong. It might be that when they are cooler, they give better electrical properties though.

Good points. I think I'm gonna get the mcw-60 + unisink as well. It just seems like the all around best setup in terms of price/performance/future compatability.

[NapalmV5]

@ AgentGOD/and your friend - awesome work guys !

i got both patches and thanks to the patches especially the ~2V.. ill be getting a gtx280 to play with after 2x gtx285

i hope you guys will spread similar patches after a voltage tuner is released for the isl6327

[T_Flight]

Yea, I'd imagine that there's no way my Q6600 is going to produce more heat than an i7. Those things are beastly.




Good points. I think I'm gonna get the mcw-60 + unisink as well. It just seems like the all around best setup in terms of price/performance/future compatability.

I also have a fan blowing directly at the face of the card more towards the right hand side where the power section is. It too is an Ultra Kaze 3000 running at 2000 rpms. It made a 4C difference in VRM temps. When I install everything in the case I'm gonna mount that diretcly to the bottom of the case with an air hole I'm gonna cut. Airflow does help with the Unisinks. It is much more efficient than the stock cooler though by far.

It's there in most of these cards. The higher binned GPU's do seem to do a little better also. I really am amazed at the clocks these things will do when they run cool.

[LiquidReactor]

I also have a fan blowing directly at the face of the card more towards the right hand side where the power section is. It too is an Ultra Kaze 3000 running at 2000 rpms. It made a 4C difference in VRM temps. When I install everything in the case I'm gonna mount that diretcly to the bottom of the case with an air hole I'm gonna cut. Airflow does help with the Unisinks. It is much more efficient than the stock cooler though by far.

It's there in most of these cards. The higher binned GPU's do seem to do a little better also. I really am amazed at the clocks these things will do when they run cool.

Yup GTX 280s love low temps. Mine can do 756/1566 as long as I can keep the temps under 60C, but with stock cooler at 100% fan speed that means I need ambient temps of 60F or lower.

[Roger_D25]

Indeed, lower temps made huge difference with GTX295! Just installed DD-GTX295 block and the load temps never go above 45C (and they used to hover upwards of 85C-90C when gaming, even with fan @ 100%). Increased voltage now allow significantly higer stable clocks than before (I realize this seems obvious, lower temps = more stable clocks, but it might help with people who are dissapointed that increased voltage isn't offering much headroom over stock clocks on stock cooling.