• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Nvidia GTX Titan

Titan will have the advantage of lower power usage and faster turn around times while GTX 690 has more raw power. Taking into account scaling inefficiencies of work units to larger GPUs I think PPD/W will probably be similar with more points coming out of a GTX 690 at higher power draw. F@H is FP32 so the GTX 690 isn't really missing anything that Titan has as far as folding is concerned.
 
I don't get the point of this card, if you want a powerful GPU why not get a Nvidia Tesla or a AMD FirePro
 
Spazturtle said:
I don't get the point of this card, if you want a powerful GPU why not get a Nvidia Tesla or a AMD FirePro
Click to expand...

Hit the Videocard section of the forum and read the 'GTX Titan first pics and specs' thread. In addition to the $$ difference, depending on what you're doing compute-wise it might be a perfect fit.
 
It basically IS the tesla, for less than 1/3rd the price!


Who's buying one! I bet it's faster than 690 overall too, ppd wise.
 
If only we could get a Kepler optimized core....
 
Spazturtle said:
I don't get the point of this card, if you want a powerful GPU why not get a Nvidia Tesla or a AMD FirePro
Click to expand...

This has more Cuda Cores than the Tesla K20, and is only $1k, not $3.5K. ;)
So technically, it is faster and cheaper.
 
I have 680 and 690 already, and I'll be getting a Titan as soon as the come out and ill post up come comparative stats.
 
I have a Titan coming in Tuesday. I'll run a couple of folding units through it for you guys before I put it on some BOINC.
 
DooKey said:
I have a Titan coming in Tuesday. I'll run a couple of folding units through it for you guys before I put it on some BOINC.
Click to expand...

Oh man am I jealous! :D That card is going to be a beast for CUDA projects like GPUGrid!
Way to go DooKey!
 
DooKey said:
I have a Titan coming in Tuesday. I'll run a couple of folding units through it for you guys before I put it on some BOINC.
Click to expand...

Thanks Dooks but Uhh - make sure you fold a few WU's BEFORE you go boincing. wouldn't want you to get sidetracked or anything ;)
 
Nathan_P said:
Its being worked on, as well as an AMD core that is worth folding on........
Click to expand...

I have TWO 7970's that I actually took out of my computer and put on the shelf for the time being as I do all my gaming on my laptop now and haven't been using them. If we get a mood F@H core for them I would love to fire them up! ...Imagine a good core w/ GPU QRB on a pair of 7970's... it could be pretty beastly!
 
extide said:
I have TWO 7970's that I actually took out of my computer and put on the shelf for the time being as I do all my gaming on my laptop now and haven't been using them. If we get a mood F@H core for them I would love to fire them up! ...Imagine a good core w/ GPU QRB on a pair of 7970's... it could be pretty beastly!
Click to expand...

Well, so far they have manged to get it down to 0.8% CPU usage on a core, that's a good start. PPD is still a way off from what i can tell - they want to get it fully functional before they start running "live" test units on - all current testing is done for the love, not the points. i.e zero points
 
Nathan_P said:
Well, so far they have manged to get it down to 0.8% CPU usage on a core, that's a good start. PPD is still a way off from what i can tell - they want to get it fully functional before they start running "live" test units on - all current testing is done for the love, not the points. i.e zero points
Click to expand...

Well, I will hold off for now as I converted my 3930k desktop system into a dedicated linux folding box as I dont really use it anymore. I will need to go back to windows, I assume, when I start up the 7970's again. It's not that it is a big deal, as I have linux installed on a different SSD than windows, but power is expensive! I will wait until a good GCN core is official.
 
One of the devs from Pande Group hinted that the new core 17 will come with Windows, Linux and OS X support or that Linux and OS X support will be a priority after Core 17.
 
Last edited:
Spazturtle said:
One of the devs from Pande Group hinted that the new core 17 will come with Windows, Linux and OS X support or that Linux and OS X support will be a priority after Core 17.
Click to expand...

Core 17 will be for windows at first release. 'soon' ;)
 
First FAH wu is a 7660 on my Titan and only getting a little over 34K PPD. The kepler core really needs some work.

Project 8043 is getting just over 31K PPD.
 
Last edited:
which is about 3K more than my 570s get. Until something changes, I'd stay away from these or any other 600 series card for folding.
 
musky said:
which is about 3K more than my 570s get. Until something changes, I'd stay away from these or any other 600 series card for folding.
Click to expand...

Agreed. Fermi owns gpu folding and until an optimized core comes out for Kepler you are spinning your wheels a bit.
 
Although Kepler is pretty impressive from a ppd per watt perspective.
 
DooKey said:
First FAH wu is a 7660 on my Titan and only getting a little over 34K PPD. The kepler core really needs some work.

Project 8043 is getting just over 31K PPD.
Click to expand...

Holy crap, you got TWO of them, damn!

So, is core 17 the new ATI core?
 
DooKey said:
Agreed. Fermi owns gpu folding and until an optimized core comes out for Kepler you are spinning your wheels a bit.
Click to expand...

That is simply not the case anymore since Stanford gimped Fermi production with core 2.25. My GTX 680 would get 4-5k more PPD than my GTX 580 on the same WUs. When I could still use core 2.22 on my 580 it would stomp all over the 680, so I guess their method of "optimizing Kepler" was just lower the production of Fermi to make Kepler look better.
 
ThunderDan said:
That is simply not the case anymore since Stanford gimped Fermi production with core 2.25. My GTX 680 would get 4-5k more PPD than my GTX 580 on the same WUs. When I could still use core 2.22 on my 580 it would stomp all over the 680, so I guess their method of "optimizing Kepler" was just lower the production of Fermi to make Kepler look better.
Click to expand...

This isn't completely correct. In order to get the Keplers working, PG had to remove some of the optimizations that worked on the Fermis in 2.22, hence the now lowered PPD on Fermi in comparison.
 
My "castrated" GTX570's endorse this message. Probably because a X70 (a 570) stomping on a X80 (680) was unacceptable.

Core 15 ver 2.25 is a load of crap
 
You must log in or register to reply here.
Back
Top