Jump to content

zumwalt

Members
  • Content Count

    54
  • Joined

  • Last visited

Community Reputation

5 Neutral

About zumwalt

  • Rank
    Advanced Member
  • Birthday 02/21/1972

Profile Information

  • Gender
    Male
  • Location
    Houston
  1. Can you please take a look at the forum email I sent you back on December 17 of 2013, thanks
  2. I thought BVH was bone files only with animations attached?
  3. Actually LE can already handle unlmited level size, my generic term for this is called cube loading, for lack of any other term, basically you setup an array 3x3x3 each segment is set to 1/9 kilometer in size, although this depends on the type of game your making but back to the design, so each 1/9'th only knows about its neighbors, so when you travel into one of the segments you replace the objects in the outer most segments with the needed objects, rinse and repeat, the depth of field keeps the player from seeing nothing beyond the horizon anyway and the camera has a limited viewing depth. Your skybox then creates the additional illusion of infinity. Memory is handled very nicely in this model because each segment has a maximum amount of memory it will use by the largest fillable data set in that segment, this is basically your assets. You can reduce load time by caching objects that are nearest to the adjacent segment. While the player is playing your current quadrant of segments, you then continue to cache in chunks. Being mindful that you don't want your game to eat up more than about 1 and 1/2 gig of ram since 32 bit machines are limited to max of 4 gig of ram period and about 2 gig of that is eaten by the OS if they are on Windows 7/8 so you don't want to blow up there machine. Physics is the challenge, not the rest of the stuff, because it is the physics that can make or break your game. Set all physics to be calculated from position of dynamic objects that are in motion, when they are not in motion anchor them and disable their physics. The player / actor has or should have a limited enactment of physics by nearest sets of polygons, and immediate polygons, but this takes practice with ray casting and determining polygons that the player model will possibly intersect with in an immediate radius. You can cheat with this by creating a sphere that is invisible, anything inside the sphere should have physics enabled, anything outside the sphere disabled, you attach such spheres to players and also to dynamic objects, when the sphere touches an object or contains an object, that object should have physics enabled only where the polygons are within the sphere. Long of the story in any given game engine, you only can have about 32,768 worth of objects with physics enabled before you barf, each object only containing 12 polygons, (so basically a single cube was my test), however if you remove the physics from those cubes, your good until you run out of ram or the GPU pukes.
  4. Is it possible to just load BVH files in and use them progamatically? If not can we add this as a very needed wish item?
  5. The problem with this is that when it converts the model, almost all of them are a single object once converted and no child objects so I loose the "head", "body", "Shield" etc. I will have to find a way this weekend to get 'leadweks' a copy of my models (or rather just one or two) so you can see what is going on.
  6. I tried the import, that is always ignoring the texture, I tried to drag into my project folder and nothing converts automatically, I tried with a single image file to the model and I tried with multiple image files to the model, in all cases, naked model, no textures. I am clearly doing something wrong, will start again this weekend and see what I am screwing up.
  7. how do i import and keep the textures and animations without loosing them??? all my fbx files are importing without textures and loosing animations
  8. I have something now that I can generate heightmaps from both coherent-noise and perlin noise then save to bmp files, everything from grid type terrain to planetary level terrain to apply to spheres, so now I want to take those images and apply them as the heightmap, lets just say 1sq mile size grid objects then save the generated terrain result including applying the colored version of texture to the generated terrain as a prefab. Needs smoothing levels for the terrain so it doesn't look so, how do we say, choppy? Then I can simply save all of these out to a folder for my array of terrain objects and load only what I need when I need it.
  9. They support every possible platform on the market today, not to mention they have server loadbalancing available and cloud deployment. As far as total comparison between products, you would have to look at all of the features side by side with ENet to answer that question, I personally have never used ENet. I can seamlessly shift from one server to another in the event of heavy server load without the player knowing it, basically one game world can in theory have unlimited players. Only limit at that point is the rendering engine. To get any deeper into the pro's and con's between Photon and ENet you literally would have to look at what is available in both along with licensing and packet size. Photon is truly streamlined with great compression. Check out their reference section: https://www.exitgames.com/References
  10. Simple really, using Photon networking engine (which anyone can download and have 100 CCU license for free), I am building up the code where in one line of code you connect up to the server and then you can send serialized player data to the server for broadcast, etc, same for creatures, only thing that doesn't get sent is the obvious things like maps, etc, onlything anyone will have to do is download Photon and using the win32 SDK or iPhone / Android SDK setup instructions, get the basic setup done for the project then drop in the header and cpp files and off they go. Make simple calls such as Player->Send(); and Enemy->Send(); would be almost enough, trying to make it generic enough for any type of game. Right now I have the client talking to the server and getting data back from the server which was enough proof of concept using the Native C++ libraries.
  11. Just my little pet project... Incomplete but is working for testing purposes obviously, I just needed to figure out how to get this to work in Leadwekrs and I am happy I did. http://www.exitgames.com/ Attached is my current header and code files, for now the client simply attachs to my server and send/receives test packets, all the other methods I have in the header aren't ready yet. PhotonLib.h PhotonLib.cpp
  12. Ok so two new things to learn off this, if you do the text draw before rendering the world, you get stacked text and looks like ****, if you do it after you render the world then it looks good. Now here is my code that works thanks to both of you. bool App::Loop() { //Close the window to end the program if (window->Closed()) return false; //Press escape to end freelook mode if (window->KeyHit(Key::Escape)) { if (!freelookmode) return false; freelookmode=false; window->ShowMouse(); } lib.update(); jText = lib.getStateString(); text = jText.UTF8Representation(); if (freelookmode) { //Keyboard movement float strafe = (window->KeyDown(Key:) - window->KeyDown(Key::A))*Time::GetSpeed() * 0.05; float move = (window->KeyDown(Key::W) - window->KeyDown(Key::S))*Time::GetSpeed() * 0.05; camera->Move(strafe,0,move); //Get the mouse movement float sx = context->GetWidth()/2; float sy = context->GetHeight()/2; Vec3 mouseposition = window->GetMousePosition(); float dx = mouseposition.x - sx; float dy = mouseposition.y - sy; //Adjust and set the camera rotation camerarotation.x += dy / 10.0; camerarotation.y += dx / 10.0; camera->SetRotation(camerarotation); //Move the mouse to the center of the screen window->SetMousePosition(sx,sy); } Time::Update(); world->Update(); world->Render(); // lets draw that text, do this after render world context->SetColor(0,0,0); //context->Clear(); context->SetBlendMode(Blend::Alpha); context->SetColor(120,120,0); context->DrawText(text,pos.x,pos.y); context->SetBlendMode(Blend::Solid); context->Sync(false); return true; } And I can see that the networking component is now working, just got to figure out the lag.
  13. I used the first tutorial which works for drawing the text to the screen, had to add in the context->clear for it to work, but now I am having a totally different problem, I looked at the documentation at URL http://www.leadwerks.com/werkspace/page/documentation/_/command-reference/font/fontload-r43 This has something called "Draw" in the code sample that is not defined so I have on clue what that is in C++ because when I try to use it, c++ has no clue what it is, so my code that I tried is: bool App::Loop() { //Close the window to end the program if (window->Closed()) return false; //Press escape to end freelook mode if (window->KeyHit(Key::Escape)) { if (!freelookmode) return false; freelookmode=false; window->ShowMouse(); } lib.update(); jText = lib.getStateString(); text = jText.UTF8Representation(); // lets draw that text context->Clear(); context->SetColor(120,120,0); context->SetBlendMode(Blend::Alpha); context->DrawText(text,pos.x,pos.y); context->SetBlendMode(Blend::Solid); context->SetColor(0,0,0); //context->Sync(); if (freelookmode) { //Keyboard movement float strafe = (window->KeyDown(Key:) - window->KeyDown(Key::A))*Time::GetSpeed() * 0.05; float move = (window->KeyDown(Key::W) - window->KeyDown(Key::S))*Time::GetSpeed() * 0.05; camera->Move(strafe,0,move); //Get the mouse movement float sx = context->GetWidth()/2; float sy = context->GetHeight()/2; Vec3 mouseposition = window->GetMousePosition(); float dx = mouseposition.x - sx; float dy = mouseposition.y - sy; //Adjust and set the camera rotation camerarotation.x += dy / 10.0; camerarotation.y += dx / 10.0; camera->SetRotation(camerarotation); //Move the mouse to the center of the screen window->SetMousePosition(sx,sy); } Time::Update(); world->Update(); world->Render(); context->Sync(false); return true; } The problem is that this code, the text does not show, if I remove the remark lines from //context->Sync(); then it draws the text but the scene is black, but if I leave it commented out the scene shows but the text doesn't show, so how do I get both the text to show (and yes the text changes per game loop) and have the scene show?
  14. They need to be bigger so that it doesn't cut off half the word in Windows 8.
×
×
  • Create New...