Folding@Home: THGC Needs You -Team 40051

Page 435 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
Just FYI - I've seen WUs range from using 70% of power limit to 100% of power limit, so the computational load varies wildly between cores and projects (settings) (so you could do 100 of one project and never have a failure and then fail every single one of another project). If you push the OC you are going to have to babysit the card.
 
I also had a wu that pushed my 980ti to 111% power, 1st and only one I've seen.

Each 10mhz step of my oc I leave for 3 days now.

Also to factor into the equation is there are such things as bad wu's, regardless of the oc.

Currently dealing with one atm core 18 and the power load is only 52%!!
 
Well so far 2 WU's stable (one from core 21 and 18).

Although guess i'll have to run it a bit longer to really find out.

But now it's a good time as any for me to really kick it into gear and run it 24/7 for a little bit till the 6th. (been only running it overnight till this point.)

Wont have enough time to really reach near my peak PPD but I should be somewhere around 500K to 600K if i could.

Sadly i must return to Massachusetts and leave my computer behind again so F@H will be cut off again for a while.
 
Its funny, the voltage slider on Afterburner only registers after you add +20mV.

Then after that any additions aren't registered on the core votage, I mean I can whack it up to +80mV and it stays on 1218mV.

Wonder if it is adding votage and not registering it or if it is limited, this is on the 980ti.

Also noticed the 980ti has lower default voltage than the 980, 980 is 1212mV, 980ti 1192mV.

Apparently 1.25v is safe, 1250mV.
 


Well... i can say my card has been stable after applying the 10mV....

So im going to guess that it's applying but not registering or that voltage is being used somewhere else.
 


well did a little bit of overvolting but didn't want to slow down all that much on my Folding progress but here is what i was able to get.

After applying the 20+ mV, it did read a higher reading.. It also strangly was able to boost to a high clock speed than I put in from 1510 to 1523 mhz without touching the clocks.

overvolt_zpsgdrgbe1k.png


Although after a little bit, when temps hit 67c, mV dropped down to 1192mV and speed hit back to 1510MHz
overvolt%202_zps5zmnn4n0.png
 


just tried it and that is a negative. just like your seeing, going beyond +20 does not register on either afterburner nor gpu-z
 


im not sure if i should laugh or cry for beanoslim.... :lol:

It feels like your just toying with him EXT.
 
Well, i've just finished my last WU before I had back to massachusetts tommorow.

Glad that i got back into folding for a little bit during this time (oh so short). Although im thoroughly impressed by just how many points I've generated during this time.

Months of folding from 2010 to 2013 (fell off folding after that) and I only got up to to 200k total points (to be fair, only was folding on a laptop that could do 1k PPD and i wasn't folding 24/7).

Then in theses 3 short weeks, I currently have 3.3M(+) total points (last WU hasn't been counted yet). About 3.1M of that was generated during this time. Also glad i did hit close to the 600K point mark yesterday on this graphic card. Meaning it is feasible for me to get (if not very close to) the 600K PPD mark. (there was some pausing and testing of clocks/voltages yesterday so thats my exuse of not hitting at or above 600k points. 😗 )

Although I wont be running it during my leave so I will be falling of the charts again. So everyone that seen me in there rear view mirror can feel at ease again. :lol:
 
I've been running at 1414 MHz successfully for 48 hours now, and have just bumped it up another 10 to 1424 MHz. My temps are staying around 69-71C, and I show no errors in my log. But I've been following warmon's efforts and I'm wondering at what point do I need to start considering increasing the voltage and at what increments? I suppose I would wait until I started getting errors. This card is already OC'd from the factory. How do I determine that it is pushed close to its limits? I don't wnat to burn it up, it's brand new and I won't be able to replace it for quite a while.



Thanks for your work warmon, we'll miss you here.

 
At those temps your card is probably throttling also.

Set fan to manual by clicking the auto tab and increase the fan to around 70%

If you sit there and watch the temp fall at some point around the 65C mark you should also see the core clock increase.

To unlock voltage you need to go into the settings and tick the box to enable voltage control, that will also show you your default mV.

+20mV is never going to do you any harm but if you have no errors don't worry yet.

Sounds like you need to rework the fan profile to keep that baby a bit cooler but find out where that point is 1st^
 
Wow! I did as you suggested, bumped my fan speed up to 70% and my temp instantly dropped to 50C and my core clock went up to 1437 MHz. But man is it loud! Maybe I'll play with it a little and see if I can find a sweet spot between temp ant clock speed.

I dropped down to 50% fan speed -- temp went up to 55C but clock speed remains at 1437. But I am getting a wee bit of coil whine at this speed.
 
In settings under the fan tab there is an option to enable user defined software automatic fan control.

That's the one you want to have a play with so when the card is cooler the fan drops.

At least when you keep it below the heat threshold you know exactly what clock speed it will run.

At what exact point does your card throttle, it must be higher than 50C, mines around 63C so I ramp my fans up around the 60C mark.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom