Leadwers engine?
Archives Forums/General Discussion/Leadwers engine?
| ||
| What do you think about this 3d engine? I really don't know what do.. Change from b3d to bmax with this or not.. Give me 2 or 3 comment plz. Aza |
| ||
| As it will always be, its completely based on what you want to do. From what I've seen, you'll just get frustrated if you get Leadwerks. You have to decide for yourself or give us complete details on what your trying to achieve. |
| ||
| Ok I'll try.. We are 4 persons and we want create a big adventure 3d in a huge map (it's a city in real dimension) we found a way to do it in blitz3d but it's not really a fast way. And obviously we don't use shadow and expensive graphic system (there is a lot of 3d object in the screen), the ai it will more elaborate and that's another expensive resource. We use blitz3d because is fast to write, but we sacrifice many graphics detail just to concetrate us in gameplay.. I don't know if this can help you.. If you need more information just ask. Aza |
| ||
| TV3D. Sorry blitzmax/b3d never impressed me as much as tv3d did via 3d. It's windows only, but honestly is there even 10% of a market in mac/linux compared to windows for 3d? It is also so unbelievably easy to work with, and it'll work on any language you prefer. You can also use it for free until you distribute it commercially for a craptastic 150$. I suggest it because you have a team together and in my experience with tv3d you can get shit out world editors in a couple weeks by yourself. The community also provides many many tools and mods for generating grass, trees, particles, etc. Alas unlike b3d, TV3D is still updated frequently and the people in their forums are usually as generous as here. http://truevision3d.com Edit: By the way you can also use tv3d on blitzmax. I think some guy made a wrapper for it, but for an extra charge. Personally I would go with whatever other language you are comfortable with (c++,c#,vb6,delphi,etc) if you go with tv3d. Good luck. Hope you find what you're looking for. |
| ||
| I would find a faster way in Blitz3D to solve your problems. It may be basic but it is a fantastic environment for coming up with innovative solutions. Also I would design game with gameplay first and then consider graphics style that suits target audience. In terms of gameplay, if your team is not sitting round enjoying the prototypes of your game already then perhaps your graphics guys should go on holiday for a few weeks. In terms of graphics, the choice is either dx7 or sh3 current gen shaders. This years shaders have made all others obsolete so I think perhaps it is pointless targetting any graphics card that has shipped in last 2 years and it is the time to jump on a generation of shader cards that seems to finally have solution correct. You have to consider seriously how different in size and type the two markets are for your game. If you are doing casual mass market game dev or haven't finished a single project I would stick with blitz3d/dx7. |
| ||
| Rather strange opinion about shaders but well... |
| ||
| I think some guy made a wrapper for it, but for an extra charge. I made the wrapper for it, but there's no charge for it. There's a suggested donation if it's useful, and a few people have kindly done just that, but there's no obligation and no restrictions to using the wrapper at zero cost. If anyone is thinking about TV3D and BlitzMax, might want to hang on a few days as I'm currently updating the wrapper to work with the new version of TV which was released last night. |
| ||
| You can also use it for free until you distribute it commercially for a craptastic 150$. This is incorrect. Development requires a license, and one license is for one person on one computer. |
| ||
| It certainly looks very impressive from the screenshots/features. Personally its no good for me because its win only & im more interested in all 3 platforms. However if you are only interested in windows then its certainly worth trying it out. is there even 10% of a market in mac/linux compared to windows for 3d? I would say most definitely. a 10% of the market of the worlds hundreds of millions of computers is a fair bit. Considering the vast majority of windows users are average joe emailers and "employees". Also bearing in mind the kind of people that (stereotypically) go for mac tend to be creative, arty types and linux people tend to be more technically minded & programey types (although those stereotypes are starting to shift about a little these days). To both of these kinds of people "3D stuff" has a certain appeal (& almost everyone loves to dip into a game of some sort occasionally). Theres a lot of people who take this kind of computing seriously that are desperate to drop windows for an alternative, but cant always find a certain tool alternative or something because of less software choice. - creating a huge market. The windows market on the other hand is totally saturated. |
| ||
| This is incorrect. Development requires a license, and one license is for one person on one computer. No it isn't, it's completely correct. (You do realize he's talking about TV and not your engine, don't you?) TV is free for non-commercial development and it does cost $150 for a license. The license is not per-computer it's per-project, and you can generate multiple developer licenses if your team has multiple members using multiple computers. |
| ||
| Oh, I didn't realize he was talking about TV. |
| ||
| I think the Leadwerks Engine thing is quite exciting. Not much compared to Blitz3D, but it totally owns BMax. |
| ||
| I think Leadwerks engine can do much better looking graphics than Blitz3D, since it has dynamic shadows and dynamic lights. Also for making games it's much easier since it has built-in physics. I made some games with Blitz3D, but I got stuck when I tried to have more than 200 enemies, but Leadwerks can handle over 3000 meshes and over 300 million polygons. But Leadwerks is not for everyone, it focuses on good looking games, and if you want to make games which work on every PC or every computer today, you would have to look for some generic OpenGL engine which doesn't use GLSL shaders or GPU instancing. I'm not sure if even Ogre3D can do that. |
| ||
| Ogre3D is good. For example Neoaxis Engine Ogre3D render based is very good solution. www.neoaxisgroup.com Engine3D, Map Editor and Resource Editor all together. Indie License 99$ But ... Only for .net (C#) Good luck |
| ||
| Oh how I hate C#, it's horrible to code with with all the unnecessary typing and then it's even slower than BlitzMax with it's Managed Code thing, and last but not least it's only a Microsoft language. It should be forbidden to use the letter C in that language, and it should be called M$# or something. D is the only real potential succeedor of C++ if any. |
| ||
Yes Lumooja, sure ... |
| ||
| I cant see limit in poly count with use of B3D if used properly..I had 4 million polys scene and it was around 35 fps on my 7600GT..also regarding shadows, what would you like to do? Dynamic shadows over buildings? |
| ||
| Ogre doesn't have to use shaders, it's up to you what you want to use. There's even a DX7 renderer for ogre thats fallen on the wayside and is no longer supported. Still, on occasion you see it as an option. If you don't want to use cg or GL they are just plugins. Just remove: Plugin_CgProgramManager.dll cg.dll and you have no GL shader support. Remove RenderSystem_GL.dll and you have no GL rendering. RenderSystem_Direct3D9.dll is the plugin for DX9. RenderSystem_Direct3D7.dll for DX7 Leadwerks has a very strong looking product. I think I saw that he was using a screen space ambient occlusion shader to render a diffuse global illumination effect, mixed with some nice advanced texture shadows blur to make soft shadows and per pixel attenuated lights with decay. All very cool stuff :) that it is impossible to do in an engine that doesn't support shaders properly. |
| ||
| Leadwerks is a great engine for tech demos. But for actual production, its requirements are just too high, its too focused around "use the current tech" instead of use the basic technology and extend where current tech is present but still offer similar effects (similar = have basically a similar effect but not as nice) WITHOUT a rendermonster, especially with effects that have FFP (which thanks to Intel still is the majority as vertex transformations are done on CPU on anything before X3100 which better would do it as well due to 3 FPS on Crysis Minimum Settings on DX10), SM2 and SM3 counterparts |
| ||
| Hi thanks for many comments.. What i search is a good engine (i think to complete my game in 2 years), i am normally don't waste my money in non usefull programm, like texture or fx sound, for texture i can find it in the web for commercial use too (or i can acquire it with a photocam). Fx is the same.. The problem is the music, when my project it will be near to the end we choose if buy it and how many money use (depend from the result of our game!!) :) But a 3d engine is a bit far away from what i can do by myself.. Than doesn't matter if i spent 150 dollars for a 3d engine (it' like 100 euros, and it's nothing if you think i pay 2200 euros per year for car ensuranced..) I need to make a choose from blitz3d or blitzmax and a good 3d engine. I see tv3d and i think is good, in this moment i am a little bit confused.. But is better than leadwers? The price is the same than is not really different to me. Thinking that i need to choose the best one. If it's up to you (keep in your mind that i want a good result) what choose you'll make? Thanks in advice. Aza |
| ||
| If the Leadwerks engine can run on SM3.0 hardware and beyond, then I think I could use it for my project. As long as it's got; - Speed - Shadows, global lighting - Interesting water - Physics - CSG (should I bother? Here: http://www.opencsg.org/ ) |
| ||
| Hey Aza! We talked before . It seems that Leadwerks canceled developing on their old LW. Engine 1.0 Now he is through with the 2nd version, and he has only released screenshots yet, but no demos, but still you can prepurchase it. It sounds a little mystical. I recommend you to try out miniB3D in BlitzMax, or start programming your game in Blitz3D, as i am through with my Kimo Engine (former called Nexus Engine), and this engine uses all the old Blitz3D commands, so you don't have to change much of you code to transfer your game from BLitz3D to my Kimo Engine. Hope it can help you ;-) |
| ||
| Hi Qimmer, I remember :) and thanks. How many time you need to complete your engine? Approximately? Try to understand me i can't write all the game in a language and after that change language and 3d engine. If you think that will take no many time i can wait it, you trust of your word, but in other cases.. Thanks again for all your help Aza |
| ||
| I am pretty sure a beta is available for the mid-july to the start of august. And maybe the release will be available about autumn. But as i said, if you write a game in Blitz3D now, it is very easy to migrate to my Kimo Engine, as it uses all the same commands as Blitz3D. The only difference is a lot of new commands for controlling Post Processing, shaders, render2texture, and of course, the new Graphics3D10() command (Direct3D10 mode), and a lot more to come! ;-) |
| ||
| what is Kimo Engine?? |
| ||
| what is Kimo Engine?? I think it's something to do with killing Cancer. :) |
| ||
| Yeah i understand and i am sorry i searched for another 3d engine because i use blitz3d from only 1 year and many other problem try to create a 3d engine, directx 9, for blitz 3d (the owner too) and all this people failed.. I don't know how good you are in c++ and my question it was this if you can't accomplish your "job" i can't translate all my code in blitzmax (i never see a code of it but i understand that are a little bit different..) Ok i hope i can wait to see the beta (and i'll continue to use blitz3d). Do you know some feature of your engine? Shadows for example, bone animation, poly on screen (poly in total) etc..? Aza |
| ||
| take a look at S2 engine...its fully Shader 3 powered thing and VERY easy to use trough provided script...and its 120 euros |
| ||
| The S2 Engine has 3 pricing models: 100€ for scriptable only, 500€ for programmable and expandable using C++, 5000€ for source code and no royalties for each title you write. http://www.profenix.com/S2License.html |
| ||
| My engine (Kimo Engine) is an engine on the way i am developing (and won't stop until finished!), which used almost all the old Blitz3D commands, and has the following features: 100% Done: Geom. handling (AddVertex, AddTriangle, VertexColor, VertexBone etc.) Shadow mapping Hardware Skinning Render2Texture Post Processing (including shaders for: Depth-of-field, Dizzy, Sepia, ...) HLSL effects like Parralax Occlusion Mapping, Ocean etc. LOD Terrains with the ability to cut out quads (for holes in terrain) Basic physics using PhysX like cubes and spheres on a plane On the way: Soft shadows Full physics with ragdolls, cars, ropes etc. Screen Space Ambient Occlusion ... and more to come. ;-) What do you think about a Blitz-like shading language? ;-) Give a suggestion if i missed something? |
| ||
| I think a Basic shading language is a bad idea. I initially thought it would be cool, but there are a few problems. There are many things in GLSL that have no correlation to anything in BASIC. What is the BASIC equivalent to a varying? A uniform? There is none. Most shader code relies heavily on overloaded vector objects, which have no BASIC equivalent. You could support these in your pseudo-language, but it keeps looking less and less like BASIC and more like an unnecessary variation on GLSL. GLSL has a lot of niggles that a CPU language doesn't. Things like dynamic loops are not possible or not supported on <SM4.0 hardware. Or for example it is not allowed to get a value from an output variable like gl_FragColor. So the end results would be something like the following. GLSL: uniform float appTime;
void Main() {
vec4 pos = ModelViewMatrix * gl_Vertex;
gl_Position = pos;
#ifdef _GLSL_CG_TYPE
gl_ClipVertex = pos;
#endif
}BASIC SL: uniform appTime# Function Main() Local pos:TVec4 pos = ModelViewMatrix * gl_Vertex gl_Position = pos #ifdef _GLSL_CG_TYPE gl_ClipVertex = pos #endif EndFunction So instead of making your shader code cleaner and more compact, it just makes it nonstandard and confusing, being neither BASIC nor GLSL. Not really worth it, in my opinion. |
| ||
| ...what Josh said, makes sense..however, some nice visual tool for setting up your shaders and exporting it as material or whatever suits engine, would be nice... @Josh.. I have carefully read all your posts regarding your system..I would like to buy it, but, i saw that you are a bit dissapointed with Bmax and your system is written with it(far as I know)...does that mean that you will abandon it and redo things in some other language? And regarding your system, is this version going to be stable final one for SM3, inclusive physics and stuff?? Im sorry if my questions sounds suspicious, its just that me personaly would like to know more details since im interested, and im sure some other folks too.. |
| ||
| I totaly agree with Josh, All you need at the basic level are some simple templates for common material functions. These would offer basic diffuse, bump, specular, ambient, illumination etc. Some set up parameters number of lights, falloff, decay. These could be set up so artists and basic coders can get simple materials done easily. But there really should be support for industry standard shader languages so people can use their own tools and have plenty of reference materials and tools they can use to develop more sophisticated materials and FX. I realised a long time ago that most people on the BRL forums havent much of a clue about shaders at all. Understandable when B3D doesn't support them and thats all most people have used. But like Josh says Shader programming isn't a simple thing that would sits easily inside basic. |
| ||
| How many people would need to write their own shaders? |
| ||
| i saw that you are a bit dissapointed with Bmax and your system is written with it(far as I know)...does that mean that you will abandon it and redo things in some other language? And regarding your system, is this version going to be stable final one for SM3, inclusive physics and stuff?? Eventually I think moving to C++ would be a logical step, but the only way I would do that is if it could be a seamless transition with no effect on the users. Leadwerks Engine 2.0 is stable. The meshlayer API might chagne slightly so that I can add support for LOD and physics in meshlayers, but that is only 1-2 commands. I am not making any more changes to the command set or how anything is done on the end user's side. I am very happy with how it is now. |
| ||
| Thanks for all yours post. I 'll think about that and i'll take a decision in a few days. Thanks guys. Aza |
| ||
| Here is another possibility: Just stay with B3D. Then make sure you got a clean code with separated game and graphic logic. Most of the game logic should be outsourced for example to BriskVM-Scripts. And when the time is ready for a new 3D-Engine change to that language and engine you like. (BriskVM supports B3D, BMax, PureBasic and C++, I think). |
| ||
| How can you stay with Blitz3D, when it's impossible to make good looking games with it. Even if I would never sell a game, at least it must look good with dynamic lights and shadows and physics. |
| ||
| How can you stay with Blitz3D, when it's impossible to make good looking games with it. It might be impossible for *you* to make good looking games with it, but the first few pages of the gallery seem to indicate that plenty of other people are able to make good looking games with it. |
| ||
| I think we have quite different levels of quality standards. Maybe I should explain that a bit: I want to make games which look as real enough, that it's hard to tell if it's a photo or a in-game screenshot. Yes, I know it's possible to do with Raytracing and make a game like Dragon's Lair which just shows different pictures in Raytracing quality. But I think it's also nowadays possible to do that in real-time with some limitations. Long time ago there was a game called Crysis (almost a year ago) which showed us that it's possible even with non-GeForce 8800 cards, but if we go a step further and say that GeForce 8800 is minimum, it should be possible to refine the quality even more. I saw some months ago that there is a new card called GeForce 9800 GX2, but it's apparently not really +1000 better than a 8800. I think I will wait until the GeForce 10800 GTS comes out and see if it has native OpenGL 2.5 or even OpenGL 3.0 support. I'm still happy with my 2 years old GeForce 8800 GTS, not even Crysis could bring it down to it's knees with hacked very high DX9 settings (=same as DX10), so I'm still looking for bigger challenges. |
| ||
| when it's impossible to make good looking games with it Well of course (rolls eyes). From my point of view, Blitz3d is nice and simple and flexible to use. There's a number of powerful libs that extend its capabilities (ParticleCandy, SpriteCandy, shadow libs, pathfinding code etc..) and I have an extensive codebase to work from. It's also stable - which is not something to be overlooked. When your game finally hits the public, having a forum full of people angrily complaining about crashes/freezes is neither desirable nor fun. I suspect that the people who put down B3d for not having the latest shader fx don't take enduser stability into much account. I've had a few bugs pop up with TU but the worst of these (Vista crash), was actually down to a programming error - not Blitz. Having said that, Blitz is not very fast - both in running code and the graphics engine. Which is a quandary - if I go to Blitzmax I lose all of the excellent libs much of my codebase depends upon - and my codebase would need to be rewritten. If I stay with Blitz3d but use a new engine (via dll), I'd still have a lot of recoding to look at. So from my point of view, Blitz3d2.0 would be the best solution. The core speed of blitz may not be improved, but the graphics engine would likely be faster. Mark mentioned something may be available around Chrissy next - I certainly hope that would be the case. |
| ||
| I think we have quite different levels of quality standards. Nope, I don't think it's about our levels of standards, because I could pick a few games out of the gallery which I'd say look better than Crysis. I think the difference between us is that you think fancy technology makes things look pretty and I know that it's 95% good art and only 5% about technology. |
| ||
| Of course art matters also, but it's not nearly 95%. And you have to fit the music also into that 100%. Let's say I want to make realistic looking white cube in a white room. Not much arts needed I think, just a plain white texture. Well, maybe I need to specify what material it should be: paper, marble or metal. But then it's up to the engine to render the light and shadow to make it look real with all reflections and material properties. |
| ||
| I don't get it - Crysis was developed by a specialist company with a budget of millions and a large team who spent several years working fulltime to develop a cutting edge engine. Now Blitz/Blitzmax is great, but I think using Crysis as the benchmark for your bedroom coding efforts is a bit unrealistic. |
| ||
| Of course art matters also, but it's not nearly 95%. At this point, I'm guessing you don't have any shipped titles behind you. And you have to fit the music also into that 100%. I really don't think music is going to affect how pretty a game looks, but if you're sure. Let's say I want to make realistic looking white cube in a white room. Not much arts needed I think, just a plain white texture. But then it's up to the engine to render the light and shadow to make it look real with all reflections and material properties. If you think you're going to beat Crysis by feeding your engine white textured cubes and expecting it to do all the beautifying, I fear you're in for a major disappointment. |
| ||
| I am mostly staying out of this discussion, but I would like to point out that next-gen engines are easier to use than old technology, in terms of programming and content creation. No lightmap or vis compiling is required, so you can just load a mesh, create a light, and it just instantly works. Occlusion culling is performed on the GPU, so there is no need for portal systems or BSP trees. The materials system works the same on everything, so you don't have to worry about bumpmapping lightmapped surfaces or baked vertex colors...the engine doesn't know or care if it is rendering a wall, a car, or a person, it's all just triangles running through the same render and lighting path. Now that I have been working with this system for a while, I could never go back to a system of baked low-res textures that sort of look like blurry shadows. I feel like I have forgotten that lightmaps ever existed...they just seem so silly to use. It's so much easier when you can just have colored lighting with realtime soft shadows on everything. |
| ||
| why not just mod crysis engine, its perfect for your needs. If money is the issue, then license it. If your game is hot a publisher will pay the licensing fees. If its no good then you will have been wrong about yourself. |
| ||
| I really don't think music is going to affect how pretty a game looks, but if you're sure. Actually good music changes your whole mind about what you see and what you don't. It's an essential part, if not the most essential part of 3D graphics. I could just do a black screen with music playing, and people would start to imagine 3D graphics if they assume they can move around in the darkness. Have you ever read a book? You can sure vision the things what the book describes. |
| ||
| @Josh I see that your engine doesnt require Bmax to run(ver 2.0), regarding features list...is there any limitation regarding that? I mean, is it possible to completely finish game coding without touching Bmax or any other language appart from provided script, and if so, is there slowdown caused by using only script? Is there any editor or toolset you providing with engine, for shaders used ? |
| ||
| The debugging in script would honestly not be as good as BlitzMax or C++ or another language, but it allows access to the entire command set. There is no tool for writing shaders. 3D World Studio can be used with the engine, and the upcoming Sandbox editor is edits terrain and physics in real-time. |
| ||
| I want to make games which look as real enough, that it's hard to tell if it's a photo or a in-game screenshot. Yes, I know it's possible to do with Raytracing and make a game like Dragon's Lair which just shows different pictures in Raytracing quality. Dude, that's one really confused statement. Dragon's Lair was 2D disney-style animation. Plus, there's no such thing as 'raytracing quality' as a standard of image quality. Creating a game which has photo-realistic imagery is isn't as easy as people imagine it to be. Only the very best renderers (people) can manage something close to real in appearance, even though there are thousands and thousands of other highly trained people out there rendering images with the same tools, trying to achieve the same thing. Plus, 'photo-realistic' is a silly term. People use that term when they're referring to something that looks a lot like real-life, but there are vast differences between human eyes, camera ccd technology and the limitations of monitors. I don't know about you, but my photos usually need a lot of digital darkroom editing before they start to look 'real' to me, and I have a nice camera. Photos are not a good basis for real looking imagery. The point is, I agree completely with Gabriel. I'd bet more on better artwork and design than trying to reach reality with modern graphics tech. Unless you're a very, very skilled artist, all the tech in the world won't help you. |
| ||
| Using a shader based engine once you have your core shaders for handling lighting and shadows is more like using the native renderer of your 3D app. The problem is that shaders are not that artist friendly yet and very specialised for specific tasks. But once you have your core shaders established for your game creating good looking assets is much easier with shaders than fixed function. Instead of having to put lighting in your textures, you just use bump, specular and masks for the majority of surfaces. Plus you can bake normal maps from high res meshes, which can be quite tedious and time consuming but gets easier as you get more familiar with projection cages and where errors in the baking are likely to occur and how to avoid them. |
| ||
| Another issue is that the only kind of game where you can conceivable use a truly raytrace denvironment are limited to the likes of Dragon's Lair: something with an extremely scripted, linear storyline, with little deviations. It's very, very difficult to create a game like that that's actually fun and that people want to play for more than a couple of minutes. Gameplay that makes you come back for more is more important than having photo-realism. |
| ||
| @Josh: How the heck can you claim, that you perform occlusion culling on the GPU? The GPU can only handle geometry that is getting rendered. That means, that the whole world is being sent to the GPU for being culled out? That is theoratical impossible, or it is a VERY big performance loss. Maybe i am wrong, i may be. But can you explain this? :-) |
| ||
| hmm unreal engine still used light mapping though its the automatic kind. you don't have to unwrap the second uv channel yourself or anything so frustrating. proper dynamic lighting is still a huge performance hit? |
| ||
| I made a demo with 128 dynamic lights which cast of course also dynamic shadows, and my FPS was like 10. This was done in a very simple way using 16 frame buffers (8 hardware lights * 16 frame buffers = 128 lights), and when it's done right, using deferred lighting, then it can only be faster. |
| ||
| We are 4 persons and we want create a big adventure 3d in a huge map (it's a city in real dimension) If you're heading for such a professional project you should be aware of some license conditions for Leadwerks engine V2.0: You may not create any programmable, scripted, or GUI-driven game engine or game creation sytem. You may not create a 3D game or other application that is modifiable by script, programming, GUI interface or other means, unless the system requires the end user to also own a valid license for the SOFTWARE PRODUCT (aka Leadwerks engine) I think, today its a common technology for larger projects, to modify the world by a script and even allow users to do the same. Restrictions like the one above reduces at least my readiness to have a closer look at the SOFTWARE PRODUCT to zero. By the way, beyond the joy about the announcement of Blitz3D2 the question arises also here: How will be the terms of use? |
| ||
| Josh is a very good coder, the LW Engine is also good, the only problem is loyal users are left in dark in most cases AFAIK. He had some old products which are less mature and we have to be forced to upgrade with prety steep price which is like to buy an all new copy. AND, LW Engine IS very advanced, but it bind so tight with his own tools so you had to expect spend more money than you think, it's not a big issue but I think I should tell it. I think you have many chioces among these engines anyway, I don't want to put money on Leadwerks anyway, that's just personal feeling, don't mind. |
| ||
| "Quimmer" - regarding the occlusion culling on the GPU: http://http.developer.nvidia.com/GPUGems/gpugems_ch29.html |
| ||
| I read it. But as you have to render all the bounding boxes, then it may be very slow on many batches instead of CPU driven algebra-driven occlusion culling? |
| ||
| There are a lot of people (outside of Blitzland) doing GPU occlusion. He might have his own system - he'll keep that secret. Might be something that is not too bad to do on an 8800 GPU which is what he is developing with - plus he has his new ATI cards. |
| ||
| hmm unreal engine still used light mapping though its the automatic kind. Unreal Engine requires meshes to be uniquely textured to be lightmapped. I know because I saw some internal design guidelines before the engine was even released. So they are really just requiring the texture to be unwrapped on the first UV channel. How the heck can you claim, that you perform occlusion culling on the GPU? Quite easily. The engine performs occlusion on a per-pixel level. :) I think, today its a common technology for larger projects, to modify the world by a script and even allow users to do the same. Obviously we don't want someone adding a few commands and selling our engine as their own. Early on, someone wanted to do this and sell our engine plus a few commands as their own engine. A scripted product is possible but requires a special license. He had some old products which are less mature and we have to be forced to upgrade with prety steep price which is like to buy an all new copy. Leadwerks Engine 2 is a free upgrade for LE1 users. The only upgrade I have ever charged for was CShop -> 3D World Studio. |
| ||
| I'm still waiting to see what 3DWS6 will become. I remember cheap upgrade prices promised, but considering how much version 6 will be tailored for Leadwerks (an engine I'm not looking to use) I'm left wondering if I should be waiting around for it anyway. If I can make one suggestion, it'd be to make 3DWS6 a tool for people who don't have a desire to use Leadwerks, as much as those who do. I hope that makes sense. |
| ||
| If someone could suggest me the best engine for making photorealistic graphics for games, and which doesn't cost more than 1000€, I will use it. OpenGL and Newton is a must. At the moment I think it's LE2, but I could be wrong. |
| ||
| You are also going to need some very good artists... |
| ||
| as i said before S2 can do a job.. |
| ||
| Yes. Leadwerks is good at the moment,just the high minimum spec is a kicker- but it looks good. No matter what you think, if your engine can render "photo realistic" scenes, you're still going to need a good artist in order to actually achieve it. |
| ||
| Ditto on the "need an uber artist" comment. Think of Leadwerks as a really good set of tools. They might be good, but you'd need more than a good set of tools to make the Eiffel Tower. |
| ||
| @Naughty Alien: S2 seemed considerable, but it costed 5000€, so it didn't fulfill my requirements. There's also no demo, so I can't tell if it's technically also suitable. I just looked again at the S2 site, and now they have raised the prices a bit, but now all versions have NO ROYALTIES anymore, which means that I don't need to buy the 5730€ version anymore, but the 540€ version will also be suitable. |
| ||
| I do have S2 and it really does what said...everything is real time, every single object casting/receiving shadows, and so on...I dont understand why you want to go for 5000Eur price when you can do everything with version for 500Eur?? And system comming with all possible editors, from material for shaders up to geometry, scene and animation editors, as well as exporter plug ins for major modelling programs(Max,maya,..) |
| ||
| Like I said earlier, the 100€ and 500€ versions had per title royalty, so I would have needed to rebuy the whole engine for each mini-game I wrote. But they fixed that now, and just a demo is missing. Blitz3D -> Demo -> Bought RealmCrafter -> Demo -> Bought Torque -> Demo -> Bought C4 -> Demo -> Bought Leadwerks -> Demo -> Bought S2 -> No Demo -> Not Bought Maybe they will fix that too :) A picture is more than 1000 words. A video is more than 1000 pictures. A demo is more than 1000 videos. |
| ||
| Any Next Gen engine imho NEEDS CSG CAPABILITIES in order to feel complete. LE 2.0 has a terrain system that is updatable, but it isn't enough. If you want to blow a hole in a wall, or if you want to break things like trees (aka Crysis) or if you want to create "on-the-fly" props from the engine, you absolutely need CSG. This is the *LAST* feature that is missing from the Leadwerks engine that is forcing me not to spend 150 USD. Fast realtime CSG is imho the next step in all 3D engines, as it accompanies the physics capabilities (things blowing up into bits and pieces, for example) which will make more realistic visual experiences. http://carve-csg.com/ http://www.opencsg.org/ This paper talks about techniques that employ the GPU for use in fast CSG operations; http://www.gvu.gatech.edu/~jarek/papers/Blister.pdf |
| ||
| I need to test a demo before to spend 150$, sorry. I`m waiting for ....  |
| ||
| If you want to blow a hole in a wall, or if you want to break things like trees (aka Crysis) or if you want to create "on-the-fly" props from the engine, you absolutely need CSG. Crysis does not use CSG at all. All those breakable items are pre-made to break at specific points. |
| ||
| Crysis does not use CSG at all. All those breakable items are pre-made to break at specific points. You're going to tell me that if I can take a tree, and slice it like ham, that they predefined hundreds of break points in the geometry? Absolutely not believable scenario Josh. Maybe you're refering to crates or wood that breaks in engine slike Half-Life 2. |
| ||
| Yes. If you look carefully those trees break at regular intervals, about every 8 inches. Their artists modeled each one. |
| ||
| "halo" is right - I have ripped and analysed a lot of the media. |
| ||
| Well, there you go, a good excuse not to implement CSG. |
| ||
| Procedurally breakable objects are really up to the physics system to implement, and they haven't really been done in any game to date. Julio of Newton Game Dynamics has an early demonstration of his implementation here: http://youtube.com/watch?v=Q3wF0Klk5wU It isn't graphically impressive at this point, but it is a proof of concept. |
| ||
 |
| ||
| >>S2 -> No Demo -> Not Bought<< Well, thats up to you...far as i know few other blitzers own S2 license and they are more than happy with things they see..I can compile as video one scene where everything is covered by dynamic shadows, lighting, normalmaps, parallax, specular and stuff, if you want to see.. |
| ||
| Procedurally breakable objects are really up to the physics system to implement, and they haven't really been done in any game to date. Julio of Newton Game Dynamics has an early demonstration of his implementation here: http://youtube.com/watch?v=Q3wF0Klk5wU It isn't graphically impressive at this point, but it is a proof of concept. I think it's a good start for breakable objects. Yes this gives me the same impression as this demo I seen on pouët.net not so long ago: http://pouet.net/prod.php?which=50062 Ultimately, a full set of CSG operations would clear the floor for all cases and would give full circle on building models from within the engine, as well as destroy them. for my project, I absolutely need CSG to go on. I tried making my own models procedurally and they end up incompatible with a shadow system. And I absolutely need to build my models from inside the engine as it goes with the concept of my game plan. When I saw CSG implemented for BlitzMax ( http://www.blitzbasic.com/Community/posts.php?topic=74293 ), I was more than happy, only to find thow, that the speed taken for one operation is just too demanding for realtime gaming in Blitzmax. So, that is why I haven't gone Blitzmax really. But having this in a game engine would of course facilitate having a really fast working implementation and ease of use. I do not think it has ANYTHING to do with a physics engine, but more with model building, maintenance. It is just a convenience to have CSG and physics talk together imho. I know I won't be the poor soul coding a CSG engine, but I absolutely need this for my project. It saddens me that DarkBasic has it, which is a language I will never program in, or use DarkGDK or PureGDK even thoe I gave it some reflection not so long ago. I think game development would change a lot if we had CSG, a lot of creative souls would rise. |
| ||
| I am sorry to say that s2 is not compatible with blitzmax.. I asked to the owner (he is Italian like me) and he send me an e-mail and say it's not.. Aza |
| ||
| If S2 is a DLL, it will work with BlitzMax. |
| ||
| If S2 is a DLL, it will work with BlitzMax. No it won't, there are many DLL's which won't work with BlitzMax. |
| ||
| Ah yes, now I remember that too, I tried to link directly the Notes DLL into BlitzMax and it crashed, so I needed to make a wrapper for it in C. But that means S2 could be used in BlitzMax too, by writing a wrapper in C. Actually I saw today another cool engine, called Unigine, it looked damn powerful with volumetric fog and volumetric heat effects. The only downside I noticed that it had only 17 dynamic lights, while LE2 has unlimited number of dynamic lights (limited by your CPU performance only). But the prices were a bit nuts, 50000 USD for 2 years, and 75000 USD with source for 2 years. |
| ||
| Yeah, the Unigine license cost is ridiculous for an unproven engine. As nice as it looks, I'm not aware of a single completed commercial product which uses it. A handful of announced games no one has seen much of is not nearly enough to justify those prices. |
   |