Leon86

Lower cpu overhead has been possible with opengl for years, but switching to opengl gets you a bunch of other problems apparantly, so everybody keeps using dx11.

on par means the same anyway, I've read that altis life is a quite heavy on the cpu, not sure how it'll run, on some servers you'll have higher fps than others.

He's already using a 3970x according to his sig. Anyway, clock it further than 4.2 :P Arma can't be maxed, and yeah, 100 v 100 fights can be problematic, like in any game, although it runs a lot better now than a few months ago.

the i5-4200u runs on 2.3 Ghz with 2 cores under load, it'll run, of course not on par with a recent desktop. Performance will depend heavily on settings and the mission. It'll run about on par with Z4layeta's system I think.

The dpi war with gaming mice is really pointless and crazy. Even the highest sens players will never benefit from anything above 2000 dpi. The highest dpi's are always faked, mouse outputting more counts than the sensor reads, or even worse, created in the driver. This is why high dpi always gets you jitter. If you're using a normal sensitivity it's best to set the dpi low, 400/800 range so the mouse has a clean output and the counts dont oversaturate the usb, otherwise you'll have jitter agian.

That depends on altis life, not sure how cpu heavy it is, might vary from server to server. see if you can find some info on performance on altis life specifically, and what cpu's they're using.

5600 dpi is ridiculous, well beyond what any mouse sensor will accurately register. I recommend finding out what the dpi of the mouse sensor actually is, then setting it to that or half and ajust game sensitivity until you get your preferred cm/360.

Shadowplay if you have recent nvidia gpu, otherwise I'd recommend msi afterburner.

Some video from the steam dev days, they talk about mantle in the end. nvidia advocates modern opengl to reduce driver overhead probably interesting, but I dont understand most of it :P

you don't need gcn/kepler to play arma, runs fine on my gtx 470, it should run similar on a 5870 or 6870, more than fine on a 6970. I've even seen someone on this forum claim he runs it on the 3770K onboard video, if that's tolerable it'll probably still be very playable on gtx280 and 4890. Anyway, supposedly they'll have a new round of devoper invites soon-ish with mantle, public sdk "later this year" is not very specific unfortunately.

just run with almost no ai and low viewdistance like other games, that'll make it optimised, or something.

Just max out everything but viewdistances. 200% sampling rate is pretty neat and loads up the gpu's nicely, if fps is low anyway, might as well enjoy good image quality.

It would help with mouse-screen latency yes.

but how? moving from 4 to 6 cores gets you loads of extra cpu time and does very little for performance :(

Meh, I get 60% with peaks to 80% cpu use on my quadcore in agia marina without ai, about 80 fps. With a big battle you can get it to go under 30 all the time, but if you then press escape, freezing the simulation you get something like 60 fps, even though the draw calls remain the same. While it would be cool to have some lowlevel api for massive viewdistance, but I'd rather have the simulation go faster first (although some people here have already lost hope this will ever happen).

I don't think so. Arma uses multiple cores just fine on an empty map, it's just a large part of the battle simulation that's singlethreaded.

With mantle it'd be the same, it's pointless to implement it until the simulation runs faster.

They probably have a fully multithreaded simulation, so every bit of cpu time saved on the directx overhead can be put to use in making the simulation go faster. Or they have a simpler simulation, meaning the rendering overhead is more of a bottleneck, or both, which is most likely.

Because they're being paid to do so.

Unlikely. the battle simulation is what's holding back performance, not the rendering overhead. On an empty map performance is fine. Or, make a big battle in the editor, get low fps, press escape, freezing the battle simulation and watch the fps go up dramatically. It'll only allow for higher viewdistance at roughly the same performance, not increase performance.

FX 4100 only has 4 cores (or 2, depending on the way you look at it). 50% is perfectly believable.

Offtopic, but you really dont have to reinstall for a new mobo. Changed a winxp system from pentium4 to core2duo, changed my own system from p5k to p5e to p55-gd65 all with the same windows install. Mantle is not the only way to reduce cpu overhead on rendering. and it's not just the 5000 cards, also the 6000 and all nvidia cards are not supported. Could also use dx11.1, would make more sense since win8 has a higher market share than amd gcn.

Could also make an opengl renderer with proprietary extensions. But why bother when it won't really help performance in a big battle and is not supported for a large part of the user base.

from http://forums.bistudio.com/showthread.php?147533-Low-CPU-utilization-amp-Low-FPS/page253&highlight=mantle

draw distance is mostly cpu.