mjohansson
Supporter
- Joined
- Feb 10, 2011
- Messages
- 409
The homebrewcompeteion, finnaly I just got the idea, am I the only one here who are planning to some time beome a commercial game develioper?
I am all alone, none of my friends have the coding skill or time or even the interest. I have plenty of time, my coding skills are quite disapointing really but I certanly have the dedication and will never give up or take no for an awnser. So are there other people here in a similiar situation? Some people here are seemingly coding geniuses, but work with software development? I suppose its well paid and fun? Otherwise why dont you make games? And I also assume theres a bunch of people everywere who wishes to make games but that cant code and either think they cant learn or dont want to? I have no education at all, I just had the luck of finding the right book that finally managed to teach me the lingo or what its called, that was my main problem, if people would have talked about ram memory instead of variables I would have learned to code a lot sooner...
Well what I need is coding geniuses in particual, but game designers overall, so step up if you can imagine trying to make a game together as a group or random internet users who wish to pursue the same goal! If we dont end up working together commercially atleast we can help eachother, I will gladly share my knowledge and tutor people with my limited experience and poor coding skills.
What I can do as of now is C with SDL and some basics of openGL with SDL. After looking at special effects for openGL I pretty much gave up on that, and making a 3D game with only single textured surfaces and no shadows or any effects is not what I dream of... So now I got the idea of making a software rasterizer, my own code that I could handle and I could do any kind of effects I wanted to cos of my own formatting, interacting with .bmp files for ex is insane! But making my own image format I could handle with ease and then I could always keep check on individual pixel color data and dont need to fetch values from system memory after it has been formatted for a SDL surface for ex. Todays cpus should be plenty fast to do a decent amount of polygons and effects anyway, its not the competing with high end graphics in their way thats interesting but creating a unique visual style wich is a hole lot easier with your own engine done with code you can read and understand. I understand the concepts for stencial shadows for ex. I could code that Im certain, but making a shader?? Id rather not even think about it... Working out how visual logic can work on computer hardware is definitly my thing, working out how other peoples high level langues work is NOT my thing!
So what do you need to be able to do to make a 3D game? Rotate a point around another one, thats basically it I think, if you can do that you can do almost anything!
My plan is to use a array of precalculated cordinates, enough positions for enough angles wished to be used, then trianglute the distance from center of rotation outwards with the vektor gained from array of rotation cordinates. And to draw polygons on a screen you need to transform the 3d data into a 2d image, I have an idea of how to do that but I havent tested it yet, anyway theory goes something like this: if FOV is 45 the each line into the scene infron t of camera grows with 2 units from last, if resolution of screen is 640 wide the first z value is 640 wide, 5 z value is 650 wide, thats 10 units more then 640, 10/640=shift value, move all points towards centrum shift value., all the points on z5 value that is. For simplicity camera is sticky and scene moves around it, you also need to cut of lines, if one polygon corner is in scene and one is outside you cant write to videomemry ouside screen area so you need to create a new point in the suatable location.
I have also tryed raytracing with my own simplistic way of doing it, but framerate isnt good, thats on 1 core 2,6ghz AMD Phenom. If any of you Pandora creators read this, then have you thaught about making a stationary console? What about making one specifically capable of raytracing? Raytracing is an amazing thing and I so long for the day that hardware excist to make good framerates! It would be a unique thing, something no one else is doing, everyone is doing their GPU powered garbage, make a multi cpu box with software renderer that can do raytracing at high res and high framerates! PLEASE do that!!
I dont know what else to say, hope I get some replys, any kind of comments whatsoever, I have the distinct feeling a lot of people will be sighing and rolling their eyes, but dreams come true if you make them, harder things have gotten done...
I am all alone, none of my friends have the coding skill or time or even the interest. I have plenty of time, my coding skills are quite disapointing really but I certanly have the dedication and will never give up or take no for an awnser. So are there other people here in a similiar situation? Some people here are seemingly coding geniuses, but work with software development? I suppose its well paid and fun? Otherwise why dont you make games? And I also assume theres a bunch of people everywere who wishes to make games but that cant code and either think they cant learn or dont want to? I have no education at all, I just had the luck of finding the right book that finally managed to teach me the lingo or what its called, that was my main problem, if people would have talked about ram memory instead of variables I would have learned to code a lot sooner...
Well what I need is coding geniuses in particual, but game designers overall, so step up if you can imagine trying to make a game together as a group or random internet users who wish to pursue the same goal! If we dont end up working together commercially atleast we can help eachother, I will gladly share my knowledge and tutor people with my limited experience and poor coding skills.
What I can do as of now is C with SDL and some basics of openGL with SDL. After looking at special effects for openGL I pretty much gave up on that, and making a 3D game with only single textured surfaces and no shadows or any effects is not what I dream of... So now I got the idea of making a software rasterizer, my own code that I could handle and I could do any kind of effects I wanted to cos of my own formatting, interacting with .bmp files for ex is insane! But making my own image format I could handle with ease and then I could always keep check on individual pixel color data and dont need to fetch values from system memory after it has been formatted for a SDL surface for ex. Todays cpus should be plenty fast to do a decent amount of polygons and effects anyway, its not the competing with high end graphics in their way thats interesting but creating a unique visual style wich is a hole lot easier with your own engine done with code you can read and understand. I understand the concepts for stencial shadows for ex. I could code that Im certain, but making a shader?? Id rather not even think about it... Working out how visual logic can work on computer hardware is definitly my thing, working out how other peoples high level langues work is NOT my thing!
So what do you need to be able to do to make a 3D game? Rotate a point around another one, thats basically it I think, if you can do that you can do almost anything!
My plan is to use a array of precalculated cordinates, enough positions for enough angles wished to be used, then trianglute the distance from center of rotation outwards with the vektor gained from array of rotation cordinates. And to draw polygons on a screen you need to transform the 3d data into a 2d image, I have an idea of how to do that but I havent tested it yet, anyway theory goes something like this: if FOV is 45 the each line into the scene infron t of camera grows with 2 units from last, if resolution of screen is 640 wide the first z value is 640 wide, 5 z value is 650 wide, thats 10 units more then 640, 10/640=shift value, move all points towards centrum shift value., all the points on z5 value that is. For simplicity camera is sticky and scene moves around it, you also need to cut of lines, if one polygon corner is in scene and one is outside you cant write to videomemry ouside screen area so you need to create a new point in the suatable location.
I have also tryed raytracing with my own simplistic way of doing it, but framerate isnt good, thats on 1 core 2,6ghz AMD Phenom. If any of you Pandora creators read this, then have you thaught about making a stationary console? What about making one specifically capable of raytracing? Raytracing is an amazing thing and I so long for the day that hardware excist to make good framerates! It would be a unique thing, something no one else is doing, everyone is doing their GPU powered garbage, make a multi cpu box with software renderer that can do raytracing at high res and high framerates! PLEASE do that!!
I dont know what else to say, hope I get some replys, any kind of comments whatsoever, I have the distinct feeling a lot of people will be sighing and rolling their eyes, but dreams come true if you make them, harder things have gotten done...