white

It is worse, you are wrong. this is how it should behave: 2 Cores Enabled: GPU usage = 27% / CPU usage = 99.8% / FPS = 31 4 Cores Enabled: GPU usage = 76% / CPU usage = 91.1% / FPS = 47 6 Cores Enabled: GPU usage = 95% / CPU usage = 89.7% / FPS = 70 cpu used as needed, fps scaling accordingly to more cpu cores being avaiable to be used and gpu becoming the bottleneck eventually since cpus nowdays can pretty much easily handle any game when all cores are properly used. and games like bf3, cryssi 3 and farcry 3 being very gpu demanding behave pretty much like this. there are no "loops estressing the cpu there" or "bad multicore scaling" like bis devs like to keep repeating. thats a fairy tale created to sustain the false idea of multicores not scaling well when properly coded, and only scaling in stuff like 3d rendering. those are lies. http://forums.overclockers.co.uk/showthread.php?t=18459152 64bits alone? no, but with all the things i mentioned, yes. i understand syncing AI bottlenecks everything anyway if coded with 1 thread for each ai, but why is it coded like that? ive seen terrible behaviour on arma without any lag, like me shooting someone, the guy not acknowledging the hit, killing me and then acknowledging the hit and dying (cpu bottleneck on the server and everyone else perhaps?). or someone driving a vehicle and me seeing serious desyncing and not following the same path on my screen. ive never seen anything as bad as those situations on regular games. if syncing exists to get rid of that kind of thing, its doing a bad job.

acctually supressor makes the gun more accurate because it counterbalances the recoil aswell as diminishes the muzzle flash. afaik there are no disavantages. but like i said on the other topic, they do not supress sound as much as in this game does, they supress around 30db which still leaves guns pretty loud.

the thing is, in this topic someone posted a table with different supressors and weapons, in average they took down sound from 160db to 130db. 130 is still as loud as a jet engine, and sub sonic rounds apparently only lessen the noise by about 10db. even if it came down to 80db, it would still be a loud noise, and would be perceivable at some distance, the thing is, the sound becomes very different, its like banging wood togheter, thats the "clack" sound i was talking about. im sorry because i really wasnt very clear. still, veeery far from the silent "only trigger+air pressure" sound, to me that only exists in movies. http://en.wikipedia.org/wiki/Suppressor but i guess in 30 years they could make an eletronic silencer that reproduced back the exact same sound from the gun and countered the soudn waves neutralizing them completely. who knows. all in all, i just think that it should be a lot louder ingame.

because in reality theres a loud high pitched "clack" when firing even subsonic rounds at 100db minimum while in arma its like 10db and sounds like just the air pressure, like in movies.

high end pcs barely have enough fps to run this, the average household pc has no chance of having any reasonable playability. that pretty much ends the 32bit being necessary argument.

lots of people care so much about realism and how vehicles and weapons behave, i thought they would also care about a completely arcade and unrealistic feature such as this. everytime i argued about fps, theres always someone saying "shut up this is a simulator not cod, it doesnt need fps". but the same person clings to fictional weapon attachments and third person, which breaks that line of reasoning. personally i would prefer melee as an stealthy alternative.

not yet i believe, but i woul guess its somewhere between the 2500k and the 3930.

people just like to play it tomb raider style, deal with it. but i agree third person view sucks on immersion, noone ever comes near me without me being aware because of it. even with walls on the way. edit.: fps too low? its a simulator! take off third person? its not a simulator its a game!

just fine for a lot of people is 20 fps, and 20 fps is complete shit for anyone who isnt blind, even 30 is bad. but 30 minimum considering how bad it is right now would even be acceptable. but when anything goes on ingame the fps drops inside that range, or even lower. and you are wrong, frostbytes 2 bf3 for instant scales fine with 90%+ usage/ fps gain no matter how many cores you throw at it, but only 64 multiplayer maps need that much. also there are source, udk and cryengine to name a few.

it sees but doesnt use them properly, have you ever looked at armas threads on an api monitor? theres mainly 1 thread handling the game, sync and ai on the first core, and crap like clouds and physx spread on other cores that have to wait on the first core to begin with. i have a 6 core aswell, and ive posted screenshots of performance test with only 2 of them avaiable to the game, guess what, same fps, higher cpu usage 90%+ on both cores. the game main game/ai doesnt have enough threads to use more cores, thus, a 6 or 8 cores perform the same as a dual core because the first core is always bottlenecked with anything going on. (worse if the dual core is on a better platform and have higher frequency) http://forums.anandtech.com/showthread.php?t=2306247

but its giggly! well i guess its about phisics in general and descructible environments that go along with it. that could be fun.

+1 Should be 64bits +1 should have more async threads to make use of more than 2 cores +1 should have more stuff running as shaders on opencl/cuda 100% in gpus like the ocean/clouds i mentioned on another topic. i bet anyone would be able to get 50-100fps then.

im not sure how exactly it works, but ive seen a lot of people stating that when on low some of the settings are handled more by the cpu, and on high goes to the gpu. i havent been able to duplicate it nor found detailed information about how it works, but might be the case.

cool gun, i agree. (not the shiny one of course)

thatd be cool, maybe nvidia can work a miracle.

i phrased like that because most of us grew up with silencers in movies, and in movies they are indeed almost completely silent. sometimes they are shown with just the trigger sound, sometimes with a muffled air pressure sound and sometimes with a "ptew" laser gunlike noise and those dont exist but the supressor name is way more accurate given what they do, changing the sound from 160db to 130db is very far from silencing anything.

i guess since arma plays like tomb raider arma got all the improvements aswell huh. placebo/wishfull thinking imho.

i love female soldiers, and i believe all 3 images represent accuratelly how they will look on arma: https://www.google.com.br/search?q=israeli+female+soldiers&hl=en&newwindow=1&safe=off&rlz=1C1GTPM_enBR501BR501&source=lnms&tbm=isch&sa=X&ei=cJZEUca0NO2D0QG8-YEw&ved=0CAoQ_AUoAQ&biw=1340&bih=859

i believe this is relevant to this thread: Silencers - Why do they exist in ArmA 3? http://forums.bistudio.com/showthread.php?150368-Silencers-Why-do-they-exist-in-ArmA-3 - if bohemia fix the performance issues for launch, me and a ocuple of friends that work with 3d archviz are considering making an alcatraz prison island 1:1 would be pretty cool for quick/silent "takedown" missions, but since i dont believe silencers exist, it would need for melee to be implemented for the takedowns to be silent. reminds me of the "The Rock" movie.

samco just showed a table with how much noise supressors reduce. from 160db to 130db its still pretty damn loud. so yeah you would be able to know where it came from, jusr would be weirded out because it sounded different, like mega posted. 130db is like a jet engine.

thank you for your usefull post. exactly.

that does sound quiet but in a lot of other videos theres a high pitched sound much higher than that in rifles, from what ive seen at least. and well you can barely hear the non subsonic round so my guess is that the audio was poorly captured. yeah theres still a high pitched "clack" sound when firing. not like this which sounds just like the air pressure and yes it does sound like movies: arma 2 mp5sd -

they are still very loud compared to this holywoodian approach.

when things are light on the cpu side it lets the gpu do its job, so the gpu usage and fps goes up. when you increase visibility and the cpu cant handle shit, the gpu usage goes down and so does the fps. why the cpu usage is also low you ask? because the game main thread + sync code runs on 1 core, and since it doesnt properly handle multicores, most of your cpu is never used, ever. (unless you have just a dual core) someone posted an anandtech forum cpu benchmark comparison in which an intel dual core was neck to neck with an amd x8 8350. imho, the ocean should be a shader completely handled by the gpu, so should be the clouds. (yes it could be done) running on cuda/nvidia gpu: (could be opencl code and run on both nvidia and ati i suppose)

if the game is released with less than optimum current gen textures, then yes please.