Game developers are lazy!
Posted on Wednesday, February 4 2009 @ 03:00:27 Eastern
No they are not, you've all said it at one point or another...they took this content out because it was too hard. they make ps3 version crappy because they're too lazy to harness it's awesomeness....just NO.
it's easy to proclaim things like this when your naive to the reality that is game developement. and i'm saying this from the semi naive viewpoint of trying to get into game crafting.
my views on the programming side of things are limited to say the least i'm trying to get into the world art. level design etc.
but i'll run you through my own project, an asset pack for the leadwerks realtime 3d engine.
i'm approaching this idea as though i'm an art director, whether they follow this kind of workflow i wont know for quite some time. maybe i'm approaching this all wrong.
but first i needed to find a "keystone" a piece of art that would drive every other aspect of what i'm going to build. in case your not aware of this there is no software that makes rocks for you, if it does they look bad anyway. likewise tree's, walls textures etc. if you let a pc automate anything it makes it look fake.
and if you take a photo or a rock and glue it to a tissue box it wont look like a rock.
so why do i need this "keystone"? what good is it? like i said everything will be made in reference to this piece of art. the detail level in the normalmaps (more on these later if you don't know what they are) the colour key that will drive the colours available through the whole scene. what kind of stones are valid for the landscape in my case.
this one piece of art, in my case a stone wall, will be the pivotal piece of art to everything that follows. so once it was settled on i needed terrain textures.
i hand made the grass using the colours from the stonewalls, finding good grass textures on the net is a fools errand, they need to be topdown photo's they almost always have fisheye and no idiot has worked out to take the photo's with neutral lighting yet, so you have the choice of taking very large textures removing two thirds of the outside edges which are blurred anyway, then in most cases trying to strip the daylight from the image.
end result hand made is quicker and doesn't look any different.
once the stonewall, grass and dirt textures were made, i made the terrain, for those unaware, terrains come in a few variants i'm using a similar tech to the crysis/farcry terrain, one image the dark pixels are low the light pixels are high. i decided i'd play with the automatic tools this time.
i made some fractal patterns but didn't like the results, i ran the erosion algorithm's on those patterns..they looked more lumpy but not better, i added noise to the hieghtmap, lumpier but worse. i ran the smoothing tools less lumpy but now it was just rubbish.
i manually sculpted the terrain, i left the outter lumps alone, having full intent to later sculpt them into proper mountains, i spent six hours on one of them and have yet to do the rest.
at this point i decided vegatation was needed i ahd a flat grass texture, but now i needed to try and find yet more complete rubbish grass examples. hand painting flat grass is easy, vertical grass is harder to fake.
after hours of searching the internet for some decent examples, i found something, bt it was from a site that you had to pay a royalty to use their images. a one off fee and they would scan the image....i think the scanning fee was about 30 pounds and would happen the next time they were in their office.
so i decided i needed to hand make the grass texture again. photoshop would be dificult but not impossble. i wanted achievable consistant results though. and hand painting something that organic in photoshop wouldn't give me consistant results.
so i tried an experiment. i'd generate hair in xsi, render the "hair" onto a flat plane, then use that as my vertical texture, make my grass models off that.
it worked very well and the saved scene can be tweaked easily to make variations with little fuss.
i tried my hand at tree's but i got nowhere, i gave up for the time being and went on with shrubs.
ok that sucked too, how about ground cover, i made some ivy, a few variants later due to bad results with the sway shader, i have a result. just don't look too closely the sway shader wont respond to black vertex colours properly and still modulates the xyz positions when it shouldn't be. i should log that as a bug.
well at this point i have some life, flora but alive and moving gently in an imagined breeze, the next step is more props...specifically rocks.
i wont bore you with these details but lets just say YOMF_terrainprop_rockA was a few hours in xsi modelling the low polygon mesh, an hour to unwrap it's co-ordinates for the texture to map too. several hours of sculpting a 2.8 million polygon hi res rock, the texturing was easy because of z brush's tools, exported it all back into xsi, three separete attempts at rendering the normal maps, ambient occlusion maps and albedo maps
and i'm still working on the final texture...the resolution is too washed out.
anyway i know that's a lot of text and not too interesting
so that's where I'M upto in this project look at the dates on that forum and you'll see it's been a few weeks of my spare time. and so far all i have is seven models and three terrain textures. if i want a job in world art i need to get results BETTER than those in crysis and i need to do it faster. matching the success's of the past is of no use, games like crysis were getting results that look like crysis for a year or two prior. my art skills need to make art as good as the games of 2011 will look like.
even if that overshoots the mark it can only make me more employable.
and i'm well aware the current stuff falls short of the mark.
so in light of all that am I lazy? and if that's what i'm going through imaine what game devs like those who made tombraider underground went through? evan a game as simple as lego indiana jones was a visual feast. the x360 port of kung fu panda had incredible graphics. were those developers "lazy" no game devs work their guts out.
[ 0 Comments
] [ Post a Comment
To co-op, or not to co-op.
Posted on Monday, October 6 2008 @ 01:44:56 Eastern
that's my question, and i find myself not wanting to play games that aren't co-op..
don't get me wrong i enjoy blowing a few too many hours on oblivion or mass effect. but as i get closer to my ambition of scoring a job in game art. i find myself running out of time for gaming.
i tend to take time off my full time job, blow through the game as quickly as i can, and get back to my art. i'm becoming a better 3d artist , but i'm becoming a lousy gamer.
however the games i tend to remember in the fondest light are co-operative titles. games where me and some buddies from live can blast through some AI set to prohibitively tough standards and laugh at each others misfortunes. halo3 campaign is leagues better with 3 mates a bunch of active skulls on legendary. crackdown is hilarious when the two of you start doing cat and mouse games of who can write the other off first.
heres a hint drive the supercar wedge car at the monster truck suv car and get some enormous air when you hit the jump button at the right time.
i'm really looking forward to gears 2 and fable2, and i'm starting to realise it's more because i know i'll get online with my mates again and just have fun.
i'm looking forward to taking on the locusts with a mate. calling out positions, when the actions high and shooting the breeze when the characters need to walk to checkpoint 99807608.3
i'm also looking forward to getting into albion joining a mate and just roughing up some town for ***** and giggles.
co-operative gameplay is just so much more than plain singleplayer, and i'm honestly disappointed in any x360 title which doesn't feature it. the way things look now i think fallout 3 may be the last single player only game i play for quite some time. most likely until mass effect ships or some deus ex game rolls around.
[ 2 Comments
] [ Post a Comment