Naughty Alien

..is it possible to get whole project media/docs and form maybe another Bmax crew to do same project ? Have you guys think about that ?

..thanks Paul .. ..plenty of thigs are tweaked to get it visually same between two different rendering systems..also, new episode has entirely different way how game/game logic is executed (all crappy things from first one, are kicked out, and progression is now based on utilizing levels, characters and so on, to achieve progress), camera works entirely different and its totally dynamic, inventory is totally redone and there is certain 'economy' system involved in to a game, etc..i was hoping that new incarnation of LE will appear so i can use it to expand game on Intel and similar weird cards, but it didnt so i had to find a way .. it is just incredible how many people using those, so called, 'crappy cards' .. I never expected such amount of people to be in a 'row'..

..thanks pixel..new episode should be out within 3 months..its kind a tight time frame because im porting it at same time to another rendering system, to cover crappy Intel cards and what not, so during installation, user will be offered to install/run version matching underlying system..

..ah..got it..assign / insert did it..luv it ..well..insertion happening at the end only during creation process, and i do require direct, random access by indexes, so i think vector is way to go..

..reinventing wheel doesnt feel right to me..Boost lib seems to be consensus among many googled forums..anyone maybe used it ?

..what are my options regarding slicing vector container, apart from using for loop, performance wise ?

..numbers Josh exposed are quite realistic, and i have seen it from very different source, and its pretty much as he said..also, having in mind that Unity crew is running based on massive debts actually, makes things more less attractive for developer, especially consoles, so while whole Unity mantra looks attractive, I have some realistic doubts about going that route, even I do have license by the way..

..ahh..that was not a bug..actually that pathfind lib syncing character on to its proper position since frame rate is very low..otherwise it will go off the 'charts' ..calculated distance to travel/set sensor is smaller than distance character was moved so it was corrected..its kind a 'safety' feature, in case frame rate is low, what is case in provided video..its about..few FPS, i guess..

..because many people work on inputs given by very few..if everybody performing its own way of doing things, it will be chaos..execution itself, yes, all those listed people were involved, but at decision level, it has to be very few people..studio has 27 folks on board..some people listed in credits are from outher studios (sound, voice acting, etc), so its not all us..

..hey guys..thanks a lot for feedback..yeah, entire art is changed a bit, few new shaders are plugged in to get this visual style..characters looks different too, and i hope for good .. areas of interest will be now much more detailed without long transversing, there is physics fluid involved (jelly glue) and corrections over things done wrong in first episode.. @Pixel Yes, some pixel hunting is present in EP1, most of it was corrected in latest patch, but some are just bad design, simple as it is..also, im not sure about 'poor path execution', in terms what does it mean..path was always correct if im not mistaken and character automatically slow down when reach destination..eventually on few places there is 'zig-zag' present, but its because navigation surface is done with unnecessary triangles..other than that, it was okay, i think.. @Digital Hax We are small studio, but for art itself, im working with only 1 guy, with inputs from other folks, but you can sum it up as work of 2 persons only..

..this is some screenshots taken from one of the levels in Hoodwink, episode 2..cartoon art style is a bit different and color correction more exposed than previous one.. http://s7.postimage....w7tsb/SHOT0.jpg http://s14.postimage...320sx/SHOT1.jpg http://s9.postimage....d00a7/SHOT3.jpg

..mm..well..Im trying just to figure out, best possible structure, in data terms, for 'learning curve', and not particularly simulation of humans..I have some limited success, but somethings missing (when my NPC 'unlearn' something, cant get it back if certain situation at initial state was actually making NPC to learn)..i must admit, its really hard, but not impossible..I just need some more read about this, but I cant find any single article talking about this or similar things..

..recently, I started some 'experiments' with some of ideas i have regarding procedural AI..what i mean is basically, capability that NPC's could 'learn' about world surrounding them, as well as property of such world..NPC's should also be able to 'forget' under specific circumstances, and so on..I did some base skeleton of such structure and im rather happy with it..however, i need to make a further move and start 'determining' way of learning..so, I was wondering, do you guys know about any good read, related to such topic, where i could gain some ideas how to properly do such thing, like making NPC 'learning'..

..yes, you can use OGL while LE running..my HUD/GUI library does it..

..OpenAL is welcome, and its nice to have it, newton should be 'deleted with shift button' .. i would really want to have latest version of LE2, like that..

..is there chance to get LE2, modified, in such way that it contains renderer just and only (with corresponding entity system of course) ?? I mean, if you have some 'free time' ??

..hehe..yeah..i hope it will work fine..it will include translations to German, French, Simplified Chinese, Russian, Spanish, Portuguese and Italian languages, so i REALLYY hope, such hard work will be rewarded ..fingers crossed

..something like this, should work.. varying vec3 lightDir,normal; uniform sampler2D MyDifuseTexture,MyTextureOverDifuse; void main() { vec3 ct,cf,c; vec4 texel; float intensity,at,af,a; intensity = max(dot(lightDir,normalize(normal)),0.0); cf = intensity * (gl_FrontMaterial.diffuse).rgb + gl_FrontMaterial.ambient.rgb; af = gl_FrontMaterial.diffuse.a; texel = texture2D(MyDifuseTexture,gl_TexCoord[0].st); ct = texel.rgb; at = texel.a; c = cf * ct; a = af * at; float coef = smoothstep(1.0,0.2,intensity); c += coef * vec3(texture2D(MyTextureOverDifuse,gl_TexCoord[0].st)); gl_FragColor = vec4(c, a); }

..well..this is IOS game i mentioned few times, we did as a experiment, how whole thing should go, and is it worth doing .. here are some reviews http://www.egmnow.com/8bitenvy/split-review-escape-from-the-prison-of-your-boredom/#.T9628luqj3w.facebook http://www.gamingbus.com/2012/06/18/review-split/ ..as for Hoodwink, its out 28th this month on Origin..stay tuned

..actually, you should look for PolyVox library..it will voxelize mesh/cut off for you all unnecessary surfaces and pass it to renderer as a triangles..tested it and it works great with LE 2.x

Roland, capsule will not make any serious difference. With capsule, you may eventually end up that some parts of your character (feet) cutting inside terrain, so its not a solution..best way is along lines Scarlet Thread Studios suggested, but if that is too much, just simply make radius of controller to wrap character mesh very tight..its common thing and you may see that a lot of games have slight 'floating' characters. Typical example is Tomb Raider..unless some serious accuracy is a factor, either capsule or cylinder will do well, and instead, it has to be some custom made system..

..i understand you Rick..I assure you that provided physics is not good enough for any serious project..thats why i use Bullet..also, AI should guide/control/check avoidance and distance for NPC's and it shouldnt depend on physics collision, but one set with level.

solution is very simple..make controllers to collide with, just and only world geometry, not other controllers...then, create cylinder bounding space, with same size as controller, and position it on same position as controller is, and final, check cylinder bounding spaces intrusion, instead of controllers..similar technique im using for my system and it can run hundreds of characters with no slowdown at all..

@Roland Yes, of course..what I said is just a simplified way of process itself, but basically, there is a loading manager what will make sure that all assets are loaded before render/flip call is executed..it does work fine so far, even more elegant solution will be levels set in such way that initial camera cant see them, regardless of rendering/flip execution or not..