spent last night figuring out how to make these now behold


http://static3.filefront.com/images/personal/l/lordspidey/78836/otqtaebihx.jpg

and the multilayered gimp XCF file
http://files.filefront.com/Dustydantiaxcf/;9800734;/fileinfo.html

huh?

Ohhh shiny!

Yay, something else that's not renderable in real-time.

Don't go getting anybody's hopes up, spidey.

shaddup rogue

go play the second mission in halo one and look at the sun though a tree :P

done on a much smaller scale but it is possible to render in real time

even if it aint it could make for some damn awesome backgrounds

Listen to roguelazer, Spidey.

But i saw it in halo :(

Ugh.

bah whatever, still a nice pic to look at :).

It's not impossible to do volume light and shadows in real time, but it would make for a big performance hit, and potentially require pretty high-end hardware, depending on how it was implemented. Regardless, it's not something we're going to be spending time on in the near future.

Gameplay is the priority.

Shucks.

Genka go eat a devalon please.

I think spidey is just showing some awesome artistic skills and I hope to be able to learn a bit from him in the future. :)

bah its not art the computer does all the work :)

learn to take a compliment spidey he was being nice after all it could have been Lecter giving you crap or is that what you wanted

Elbryon

Beyond the Red Line looks a bit like this in the demo I have played. Did not run slow.

thats cause its the background in BTRL

Graphics in videogame development are based on a series of trade-offs. The engine design requirements of an MMO based on the idea of 6 degrees of freedom and dynamic "levels", which may be unexpectedly occupied by very large numbers of players, are quite different from that of a first-person-shooter that are largely single player. Usually, game engines use a lot of "cheats" that are specific to the engine requirements, to make things faster and better looking, and push the envelope on what's possible, within the very limited range of what they're trying to achieve. If Doom3 hadn't occurred largely within small enclosed spaces.. no one would have been able to run it. Conversely, if iD had had to deal with the prospect of a hundred people suddenly showing up in one area, with totally unknown (and perhaps complex) ship configurations, doing unknown/slow things like firing swarm missiles.. well, that would have made it rather difficult as well. The point being.. different types of game engines are "good" at different sorts of things. Everyone in the game industry tries to strike a balance of the best eye-candy, while keeping things playable.

The upshot here is, you can't really say "Well I saw something that looks vaguely similar to that effect, which was fast, in a totally different game that has vastly different engine requirements".. and have it be very meaningful. I'm not saying that putting volume shadows into our game is impossible.. not at all (as I posted earlier). Actually, we've looked at that sort of thing for years (and general stencil shadows, and soft shadows, and blah blah). But, to do it in a modern, fast way would require some overhead, and it might be pretty damned slow in sectors that have a lot of asteroids (or with a lot of moving players). And we're going to be adding a lot more sectors with a lot of asteroids. So, then it becomes "well, would you rather have largely empty sectors with neat-looking lighting affects, or a universe with a lot more Stuff". Having more asteroids/stuff is an important gameplay improvement.. especially with the addition of fogged sensors, radar occlusion, etc. Volume shadows add very little to gameplay, they're almost pure eye candy. I'm a graphic artist, I love eye candy, but if I had to choose between the two in our game, as it stands right now.. I'll choose gameplay.

Somewhere down the road, when we've solidified our Big Ass Space Battles (which people are currently complaining are slow), and reworked a lot of the sector contents, and are making more use of our existing gameplay/eye-candy features like fogging and radar extensions (which we've hardly used, storms are just the tip of the iceberg), we'll probably revisit this, and figure out what balance is right for us. But it's just not that important right now.

If you want to read more about this sort of thing, here's an older article that discusses a couple of methods:

http://www.gamedev.net/reference/articles/article1873.asp

Just in case anyone was wondering, I have a hand-modeled opacity-mapped example of three asteroids with volume shadows that I made as an engine test for our game about 10 years ago. This is really not a new topic for us, at all. I'm a total volume-shadow nut. But, I'm also a CEO, and have to make decisions that apply the very limited resources of our tiny company towards the best interests of the Game As A Whole (and Company As A Whole). Right now, this doesn't rate. Maybe down the line, but not now.

thank you for explaining everything. :)

Fair points all round .

Anyway Spidey, one day computers will be able to do in 3d real time the stuff you envision in 2d . Keep at it : nothing wrong with having aspirations .

Ecka

bah the GPUs in a year will have no problem rendering this stuff and in two year you will be able to buy an 8800 GTX for 50 bucks