Boost FPS on a NVIDIA card !

I found this thread after not playing PA for a month or two because somehow the game was locked to 60 and 30 fps. So I searched for PA vsync settings and found this thread. And as I said, I disabled Vsync, restarted the game, reconnected and got 70 fps. Then I activated Thread optimizing and after a reconnect my FPS was around 90.

I did get a blue screen after a while though, but that is probably related to my recently set overclock.

On a 60 Hz monitor this is no myth, but a truth if you only talk about visuals. Input registration may still be improved, but this depends on the way a game has implemented it´s core mechanics!

On a 120+ Hz Monitor it´s of course a myth...

>overclocking an nvidia
they practically already run hotter than fusion reactors

I said CPU overclocking, not GPU And on 100% load my 770 only goes to 60 degrees celcius so I won't call that a fusion reactor..

Still hotter than the AMDs run, it's always been that way. I had two nvidias burn out within a week of each other once, in 2007 iirc. The only reason why I switched back to nvidia was because I ordered an R9 280x and it was DOA, and so was its replacement. I can understand the first one being DOA, it is, after all, the first wave of the new generation cards, but the replacement being DOA too really shows that GPU manufacturers really need up their quality standards.

The two nvidia cards I bought are still running, the 6 year old one in my parents' pc, next and the new one in mine, so no problems there.

I have a 32mb Radeon 7000 series that's well over a decade old and still works, so?

So I'd call both anecdotal evidence. AMD is considered to use more power for the performance, and more power means more heat. However, I think that this applies more to very high end cards rather than mid range because I've never seen a mid range card with a decent cooler that ran hot. (More anecdotes! Yay!)

I think maybe you're information is a bit outdated and/or based on skewed data, e.g. comparing notably different cooler styles or PCB design.

Edit: part of this is because they release on different cycles and AMD uses a larger manufacturing process.

Lets end the derailment here, both cards are great but one could have bad bad luck with one or another vendor.

So framerate improvements it is!!

If people still want more performance from the game and don't care too much for fancy effects and extra beautification i'll post a list of shaders that can be disabled or make the performance booster mod up to date with the current build.

*cough* 120/144 Hz monitors.

I liked a few posts in this pit of irrelevance.

Turning on threaded optimization is an interesting find, I'll certainly give it a go soon and, if I remember, I'll update this post with my findings.

Well actuallyscreens only support two framerates : 30 fps and 60 fps. (unless they are advertized as 100Hz, 120 Hz)

An example : Here whe have a 60 Hz screen displaying a 150 fps video :


So while the screen was drawing the GPU updated the image three times.

This means that although there can be a definite increase in responsiveness (a couple ms), there can be as much as 16ms difference between the different render chunks, same as with V-sync, except that you get tearing.

In the end, if you have a system that does not struggle running games, I would advise playing with V-sync (except if you can play with G-sync, that is)

If you want to improve your display fidelity, get a G-sync monitor because that **** is crazy (im saving for it )

Ok so I did some tests. I put my system in a stress situation : A locally hosted game with 16 planets and 10 AI I was spectating. And There good gain with Threaded optimization, but it was on by default here, so no free surprise boost for me.

WITH threaded optimization :


WITHOUT Threaded optimization :



The thing i did not test yet is the end game difference. Maybe its a tech that really shines when there is a gross number of units drawn on screen?

My tests were late game indeed, where bigger 100 to 200 sized unit groups were roaming around

NVIDIA's OpenGL threaded optimization feature allows offloading the CPU computational workload to a separate processor thread. This feature is designed to benefit CPU-heavy workloads but can potentially worsen the performance depending upon the game/application's particular OpenGL calls.

The default setting is AUTO fot multi core CPU and OFF for single core CPU.

Please note that these optimizations can currently only be enabled if the target application dynamically links against pthreads. If this isn't the case, the dynamic loader can be instructed to do so at runtime by setting the LD_PRELOAD environment variable to include the pthreads library.

it's also worth noting that if you have more than 60 fps trying to display on a 60hz monitor, you WILL get image 'tearing' effects.

I disagree, I can tell the difference (and so can many others) in CS:GO.

I play graphics all the way down to get max fps, even though I can run full settings at probably above 60.

His post is 1.5 years old, so he might have changed his opinion

And I'm more of a graphical lover than an FPS lover, but my 144hz screen allows me to see every frame generated anyway. ^^

I meant I turn down the graphics in CS:GO, in PA I leave it pretty high. Source gamws don't really lose too much graphical fidelity for the trade off of really high fps.