General Category > Unofficial F/A-18 Acceleration Pack board

we should make our aircraft

<< < (2/5) > >>

fael097:
graphically speaking, should be as detailed as i can make it. i mean, as long as it doesnt kill the frame rate, so it should have a clean geometry, but i believe i can do something that looks nice, animates correctly, etc. but about flight dynamics, im not really sure, as i have no idea how to set this up.
maybe we can find someone who does, or in last case, use some sdk preset for it lol.

but creating an aircraft from scratch, and having feedback from users, im sure we can frequently improve it as needed.

i just didnt suggest the f-16 because aerosoft has a good looking one, but its actually a good choice. the f-22 might be a little easier to model, and just a little, as the f16 isnt that complex.

i'd just stay away from f18s (little tricky to model, as im making a highly detailed one, and another fsx one would be just pointless anyways) f14, or any carrier planes, at least for now. should keep it simple for a start, functions wise.

Razgriz:
You already have an F/A-18A made, why not translate it into FSX?

SpazSinbad:
Perhaps an F-35C would be useful? Dino Cattaneo is not likely to make that version from his F-35A version. He said this on his website. BTW he is updating his freeware Goshawk T-45C now.

fael097:

--- Quote from: Razgriz on January 21, 2011, 02:19:59 am ---You already have an F/A-18A made, why not translate it into FSX?

--- End quote ---

heh, funny question... the quick answer: something you model for rendering an animation, Computer Graphics (cg) movie, or a detailed still shot is very different than something you'd use in a game, due to, making it simple, the number of polygons.

ROUGHLY speaking, and i just cant be precise on any of the informations im gonna say, a game airplane model might have around 10.000 to 100.000 polygons. my f-18 already have around 2.500.000 polygons, without meshsmooth. an average/good computer would handle it, without textures, shading effects, nothing, just the plane model, less than 30fps, inside the 3d application's viewport. if i apply 3 levels of meshsmooth into it (standard to rendering) well, you do the math, for each smooth level, every polygon gets divided in 4. if im not wrong, that would result 160 million polygons. no game, no matter the computer, could handle it. even without all the smoothing, just the 2 million polygons plus high res uncompressed textures, plus the game graphics shading effects, im sure it wouldnt get any faster than about half a frame per second.

this is all rough information, my plane is far from done, i dont have landing gears, wells detailing, cockpit, weapons, etc. the tiny details are yet to come, so im pretty sure the model can, and will pass 10 million polygons unsmoothed.

the long answer: even the way you model those two is different. for example, when modeling a game model, you'll make everything as you gonna see in action, while when you model for rendering, you have to make everything thinking of the way the smooth modifier will make it look. that means lots of extra subdivisions , even on flat surfaces, where you'd need only one polygon for a game obj.

thats the difference between them. (thats only the polycount difference) and thats why games graphics differ from cg movies. a game can render a complex scene in a fraction of second, well, at least 30 renders (frames) per second, while rendering a single frame of a whole scene from a movie can take hours, even days (probably if you were going to render a frame from Avatar on a home gaming computer, it would take a week, thats why they have render farms with hundreds of computers working together), but in that case, many other variables count, like raytracing, global illumination, anti aliasing (that goes way beyond games 2x 4x 8x AA) etc.

to sum it up, when you make something for a cg animation, you want to make it look as closer to reality as possible, (trust me, in some cases, you cant tell if its real or cg) while when making something for a game, you have to deal with limitations, and make it not kill your frame rate.

hope that im clear enough, and that i didnt miss the point, though its kinda hard, since this is a very abrangent subject, and also im about to fall asleep in this chair :P

ps.: check this pic http://www.lucbegin.com/wp-content/uploads/2010/06/Luc_Begin_The_portrait-cgs2.jpg
this is CG, thats the level or reality cg artists want to achieve, and thats what i want for my hornet lol

cheers mate

fael097:

--- Quote from: SpazSinbad on January 21, 2011, 02:40:27 am ---Perhaps an F-35C would be useful? Dino Cattaneo is not likely to make that version from his F-35A version. He said this on his website. BTW he is updating his freeware Goshawk T-45C now.

--- End quote ---
edit.: ok, the C model doesnt hover, so skip the firs paragraph below :P i'd happily make the A version, if dino wasnt going to make it already.

indeed, its an interesting aircraft, but if its the hovering one, then i'd rather not mess with it. ive flew many hovering planes, and i didnt have good experiences with any of them, even payware ones from known companies. and im just saying that based on what someone willing to make the flight model for us would actually be able to make. for me, i'd model any jet, but we should keep our feet on the ground.

otherwise, idk how a carrier jet, would differ from a simple one, but for me, a simple one is more reliable to start with, since we wouldnt have all the carrier operation variables, we could focus on the navigation/landing features, and perhaps develope something more decent than aircrafts with lots of eye candy.

thats just what im thinking right now :P idk, perhaps someone has a different point of view... let me know

Navigation

[0] Message Index

[#] Next page

[*] Previous page