Naughty Alien

..of course button.entity is not null..in fact button.entity is just an instance, pointer on to original entity handler MyEntity, so its NOT nulled by simply freeing MyEntity...entity.valid is true or false based on state of original handler (false if nulled, true if not), while pointer (button.entity) still has value different than Null (pointer address)....this indicates that you haven't structured properly your classes, where none of instances (button.entity) should exist if original handler (MyEntity) is freed..

all you need to do in Bmax really is FreeEntity (MyEntity) MyEntity = Null GcCollect() ..after handler is nulled, thats it..some objects may require FreeEntity, such as meshes, but some DONT, such as sound sources, or sound files...either way, they have to be nulled after FreeEntity is performed (if FreeEntity, apply on to such entities), and thats all..im just not sure, how is possible to have entity handler available with its properities, after is nulled..

..well..equalizing with null should certanly do a job..after that you should call GCCollect(), not before.. MyEntity:TMesh=LoadMesh("abstract::MyMesh.gmf) FreeEntity(MyEntity) MyEntity=Null GCCollect() If MyEntity=Null DrawText("YAY, its freeeeeeeed !",0,0) End if ..what is the reason behind use of GCSetMode(2) ?? EDIT: Also, keep in mind that garbage collector DOESNT flush out things imidiatelly, once command is called...

..you should call garbage collector after FreeEntity, and given example is from Bmax, and it works fine..so im not sure what you using..other way is to have a flag what will determine is your mesh unloaded or not...that will require writing of some Loader/Unloader class, what is a must anyway..so withing that class should be member determining state of given entity, something like 'Loaded:Int' So, when you load your entity then Loaded flag will be set to true..soon as you use FreeEntity, it will be set to false...on that way you can always know whats going on with entity by accessing its state member from loader class.. however, after use of FreeEntity, always Null given handler..so, previous example should look like this MyEntity:TMesh=LoadMesh("abstract::MyMesh.gmf) FreeEntity(MyEntity) MyEntity=Null If MyEntity=Null DrawText("YAY, its freeeeeeeed !",0,0) End if

..sure.. MyEntity:TMesh=LoadMesh("abstract::MyMesh.gmf) FreeEntity(MyEntity) If MyEntity=Null DrawText("YAY, its freeeeeeeed !",0,0) End if

..as for 3D optic, set your camera far away and then use CameraZoom to get 'flatten' orthogonal look like screen projection..as for physics..you dont have much choice with Newton, but something along Josh suggestion...

..easiest way to create laser beam is to create quad on runtime and scale it up as it fits your design, then simply paint it by animated texture with proper blending mode, and you are set...texture size for such thing (laser) can be rather small and by having them scalled up together with corresponding quad will make them a bit blurry what is actually, more than welcome for such thing as laser beam..other way to do it is trough custom made shader where everything will be done in runtime, including beam itself (no texture anymore), but it will require some shader writing skills, so i guess first one is more than enough, considering that artists should be able to easy modify/tune eventual media for laser or beam of any kind, rather than making shader api to do such thing..

..im seconding what gamecreator said..few lines of example code with each parameter description is way to go..also, i hope that under 'programmers' expression, you primarly thinking on people whos using raw C++, C# and Bmax, rather than scripting language..having said that, i also hope that system will be fully available in fashion as current LE is, because to me personally, none of future editors has any value , because from my experience i have so far, most of tasks i needed to perform, and requests from artists, were way way beyond offered toolset capabilities (not just LE, but Phyre library as well) , so i would like to be able to build my own tools from scratch for certain targets .. i really hope for this things because im planing to use LE 3 fully on mobile devices + mac .. if not..then..ill have to say good bye..

..every engine i know, has a limitation related to polys, materials, textures...so do LE..however..I just have no idea how come you having such low FPS results, and i suspect, as a one of the users whos really pushing it to the edge ( i believe) , that you doing something real bad with your code/shaders..I dont know number of materials/textures you using, or size of level and number of polys you use but for instance, one of my heaviest levels has approx 290-340K polys, it has about 39 different materials and having about 28 characters moving around..all that with corresponding AI and stuff, running at 22-25 FPS at 7600GT card (6-7 years old card?) .. I wish i could help , but some access to your data is a must in order to provide proper input..

.is there any particular reason why you want to go recast route ?

..my library deal with such problem (dynamic obstacle avoiding) on such way that each generated navigation grid cell extracted as a path node, updating itself with bounding space and soon as it is triggered looking for alternate route..quite simple and fast..

..very nice work and touch up..wonderful..

..well..if you ask me, ill drop entirely 'Leadwerks' as a name of the product..Leadwerks is name of company, so let it be..and product itself can be named something entirely different and way more appealing , so options getting much wider than stick with 'Leadwerks + something' in name..

..i do not know what are you doing to get such low FPS on such machines...just to small comparison, my Hidden Dawn level with 350K polys, 12 characters on scene, and plenty of AI controlled NPC's , running 20-25 FPS on dinosaur nVidia 7600GT .. ill check carefully whats going on with code or drivers for given machine.. EDIT: It may be some issues with LE version you running, so just to update, im using version 2.3 (never update after that because, i was scared things whats working will be broken, so i dont know is that case with you)

..very interesting..

..hi guys..here is the main character in game, so you can see that whole idea about cartoon comic book style makes sense just by observing character anatomy..ill post images without shader processing tomorrow, but its looking uber ugly without shader because I have purposely degraded textures (plastered) and saved with quite low quality , what helped me to get, nice dirty details, after shader did color exposure.. right now im very tired (its 2:22 AM here, and im looking like a zombie after looking at monitor whole day until now..huh..), and im just finished this character fixing, im about to post, so i hope you will like character as well..

..thanks guys, your comments are really constructive and thats what i like about this forum sooooo much... @Pixel Perfect Marleys Ghost already pointed plan, Hidden dawn is completed, if publisher doesnt find something to add (fingers crossed)..as for Hoodwink, Episode 1 should be out by the year end, and new episode should be out each 4 months, but we will see..just look at DaDonik shader...quite interesting way of painting additional color on to texture bumps.. @Engineer Ken Yes, textures were 'photorealistic' , and processed a bit in Photoshop , and saved with lower settings (PNG for web) what did great deal of surface effect i wanted..

hi guys ..im glad you like it..ill have exterior level around next week, this time done, with some flying trains and what not .. in episode 1 there is one outdoor level only (but quite large actually), others are indoor..style i tried to follow is one seen in game 'XIII'...later ill post some characters (tiger man, main character, second chancer, etc ) ..

..as i mentioned before, im in dev cycle for my 2nd game (i use to post some menu interface related to that game http://www.leadwerks.com/werkspace/topic/3533-user-interface-shots-from-my-new-game/ )..its comic book style point and click adventure, and this is one of the levels from Episode 1 (game is episodic)..anyway, this is how its look like..I hope you like it..

..im personally not impressed with such things..games new days already looking quite 'convincing' , visual wise, so, let say today we have this unlimited detail tech available and ready to use..well, based on past experience, it appears that cheaper way to 'progress' for all lead game companies is simple update of rendering cycle, to get visuals 'realistic', while gameplay itself, is a total failure, or better say, AI didn't evolve much at all from PacMan pathfinding routine, and its quite true when you look at things in to bits..so, having said that, real procedural world feature that hasnt evolved or upgraded at all is AI or for that matter, some really new, good gameplay, and to make things worse, even existing system of delivered gameplay in huge majority of games is not used in to its maximum potential..I find it rafter sad when i realize that games some 7-12 years old, has more of story-gameplay than games today..HL2 is typical example or even older game, Suffering, whats so scary that you are really really kept all time under some sort of 'psycho pressure'...not to mention System Shock, also, quite old game, but yet, brilliantly delivered fear detail on to player.. new stuff..well just shine and polishing, but junk..i wish i could see something like 'Ulnimited AI detail' or something like that..because all this graphics push ups are so non productive for making real game that i wanna puke..

@Roland Well..as i said, some scene management tools provided are fine , just for sake to see/preview scene or level, with some basic setup what will allow artists to see how it should be, but to go completely by that concept, I strongly feel/believe, is wrong. Im dealing with artists on daily basis, and what i learn is, that NO TOOL out there is good enough...there is always something missing for those guys..I was dragged whole path of building fully functional editor for LE2, with everything requested, visual material/shader creation, bullet physics, AI, physics partticles, and guess what ? I end up writing few shaders for 3dsmax what will do all things they ever needed and then exported to LE 2.x .. Bloody Editor even does everything requested and more (testing AI, grouping, combat, etc) , nobody ever tuched..everything fall back to 3dsmax + small utility i made, for setting placement of objects and triggers and testing same..and thats all..

..to me its look like quite a lot people actually looking in to Unity, based on description of what are requests..to me, I believe, engine should be accessible to programmers as an engine, without anything but code..on other side, for artists, should be provided some tools for scene composition and scene setup, with some scripting ability, but to expect whole system to be literally drag and drop, sounds very bad from my perspective..

..yes..you have to put all skybox images in to row and proper order..after that during saving, you have to select ALL mipmaps levels, before save as cube environment dds, or your skybox will be just black..

..so how this suppose to work for code/applications ? I have to do what exactly, and how selling goes?? I mean, if I set price 20USD for my application, some percentage of it goes to AppStore or my price is topped up with that particular AppStore cost ?

..i would like to see LE+LISP..oh boy that would be a combo