Pages: « Previous 1 2 3 4 5 6 Next »
Summary Period: 2003-10-06 to 2003-09-21 (Commits 6-55 of 255)
Okay, this bug needs to go away :-) Think I finally fixed it (I had calculated the clamped values last time, but I wasn't *using* them!)
2 lines of code changed in:
There is a bug in WirePCC where we were getting NaN for the closest points because of giving a value just outside ofthe range of -1:1 for the arccos... I fixed it at school, thought I committed the fix, then got home and saw that CVS says that I didn't commit it.... oh well... Here's the fix.
5 lines of code changed in:
Ahhh... Phyxing physics bugs... (This lame misspelling of words in the log has to stop... I never realized how lame it really looks.) Anyways, I fixed the weird bug where we'd sometimes (though rarely) jump from the middle of the tube to the beginning. The bug was in WirePCC where I was taking the arccos of something that was *close* to 1.0, but just out of bounds. So, WirePCC now clips those bounds to -1.0:1.0. Also, I fixed the 'going to far backwards' bug where you could crash the computer by turning around at the starting point and going backwards to the first tube. It was a good ol' dereferencing NULL pointer error, since we were looking for a tube before the first one....
9 lines of code changed in:
Project Builder complains about the proper capitalization of #include filenames...
1 lines of code changed in:
Some more robustness work on the fizzix. I've added back in the check to make sure that we never go out of bounds (and if we do, correct it immediately). This bounds check occurs much closer to the collision time than the last time we implemented it, so we shouldn't see jumping too much, even if the collision occurs in between frames. I'm still kinda worried about the distance check, since before we had issues colliding with "walls" in between tubes that don't exist. I think that may be fixed, though, with the simple check that we reevaluate which tube we're dealing with every time we do the update. A similar check might be important for all of the physics calculations later.
37 lines of code changed in:
I'll commit this change, which is only lower polygon counts and a bigger texture scale, so that my laptop and other slower computers can handle the game better. Someday later, we'll probably have the polygon count somewhat configurable from the game.
3 lines of code changed in:
Worked on collision robustness... Tweaked some params to make the player go even faster.
22 lines of code changed in:
I added back in the collision response. I still have to work out some issues with double collisions and missed collisions. I think I may actually be getting missed collisions on the borders, but I'm not sure... seems kinda odd.
25 lines of code changed in:
More physics work. I've put in everything but the response code. I think it works :-)
24 lines of code changed in:
Almost done on the collision prediction code. All of the collisions should be predicted properly, but I haven't reworked the response code, so its still turned off. Is there a remote possibility that a collision could be missed if you were to go *exactly* into the crack between two tubes? Probably. Maybe I could do some sort of overlap in the angles and take the minimum hit on the edges. But it would have to be a *very* small overlap because there could be other problems. Well, maybe its not a problem anyway, since it'd take a pretty dead-on hit to make it happen, if it even happens. Am I rambling? Its late.
68 lines of code changed in:
Forgot to take out the debugging "printf" of the frameTime... It was cluttering up the console, and I think its unnecessary now.
1 lines of code changed in:
Fixed the stupid "off by a factor of 10" timer bug... Now the times returned by all of the timers should be correct, and our units in the game should be scaled up close to their proper values.
38 lines of code changed in:
Forgot an #ifdef on the Windows QPC code, so it wouldn't compile on the Mac... stupid me.
3 lines of code changed in:
Nothing to see here.... Just making sure the source is up-to-date, and commenting out some tests I don't need anymore.
6 lines of code changed in:
Worked on the tube textures a little. The texture size is now consistent and changeable (see PHTube.h for the constants which define how big the textures are). Only problem left to tackle is that the textures coords don't start at the right spot from tube to tube, so there is a "jump" in the texture at the tube seams. I'll get around to it later :-)
20 lines of code changed in:
Eh, took out the velocity scaling, since it will mess with the physics calculations... But, some tweaking in PHInputManager.cpp fixed most of the velocity scaling issues, and I'm sure there's much more tweaking to be done later.
5 lines of code changed in:
Fixed weird bug where we would sometimes have frames where the camera was facing backwards.... Still don't know *exactly* what caused it, but the lookAt vector was sometimes facing backwards before the gluLookAt call was made. I fixed it by setting the lookAt vector *right* before the gluLookAt call.
26 lines of code changed in:
Nothing much changed here... just took out a loop which doesn't need to run repeatedly and replaced it with the inner calls, and changed the screen size back to 800x600.
6 lines of code changed in:
Fixed some more issues with the WirePCC code. These are minor issues, but may increase portability of the code.
4 lines of code changed in:
Started work on a better timer system. Now using the Windows high performance timer instead of the SDL timer which had accuracy issues. Plans for using the OmniTimer high performance timer framework for Mac OS X.
51 lines of code changed in:
Started work on a better timer system. Now using the Windows high performance timer instead of the SDL timer which had accuracy issues. Plans for using the OmniTimer high performance timer framework for Mac OS X.
Added mouse smoothing. See PHPlayer.cpp for the MOUSE_SMOOTH_SPEED constant, which changes how fast the mouse approaches its actual target.
Added a VELOCITY_SCALE constant in PHRigidBody.cpp which allows us to further tweak the way that our player is controlled. The max speed should be the VELOCITY_SCALE constant exactly, and the accelerator accelerates to it quickly, while constant deceleration causes us to slow down near the max speed.
Tweaked some of the player accelerations, velocity scales, etc.
27 lines of code changed in:
Added mouse smoothing. See PHPlayer.cpp for the MOUSE_SMOOTH_SPEED constant, which changes how fast the mouse approaches its actual target.
Added a VELOCITY_SCALE constant in PHRigidBody.cpp which allows us to further tweak the way that our player is controlled. The max speed should be the VELOCITY_SCALE constant exactly, and the accelerator accelerates to it quickly, while constant deceleration causes us to slow down near the max speed.
Tweaked some of the player accelerations, velocity scales, etc.
1 lines of code changed in:
Working on a more robust physics engine... Collision prediction is on its way. The new red dots are there to tell us where the collision predictor thinks a collision will occur (on the current tube only)
233 lines of code changed in:
Felt it might be necessary to get the arc information computed from the Wires, so I added those methods.
7 lines of code changed in:
Some PhysicsManager robustness work.....
49 lines of code changed in:
Removed some unnecessary code in the PhysicsManager.... It may become necessary in the future, but for now, its commented out.
2 lines of code changed in:
Added _USE_MATH_DEFINES to prefix.h, so that we can try and get this M_PI issue worked out once and for all....
1 lines of code changed in:
fixed sliding collision response; it was only sliding tangent to the wire.
there is still a problem with the player snapping back
2 lines of code changed in:
added a repeating texture... not mapped correctly yet, but good for demonstration.
changed light to a positional light located at the camera position.
added fog, same color as ClearColor.
39 lines of code changed in:
More fizzix work.
7 lines of code changed in:
Fixed stupid "Mat4" bug, which had nothing to do with the "Mat4" class.... It was actually a reference to a copy of a wire that didn't exist... go figure.
3 lines of code changed in:
I think I did something to collision response
11 lines of code changed in:
Changed the slowdown method to help with turns.
4 lines of code changed in:
PHTimer now counts down to a Game Over status :)
Inching closer towards that functional minimum!!
13 lines of code changed in:
Readded slowdown to player movement when not accelerating or decelerating....
1 lines of code changed in:
changed pitch and roll to the right directions and clamped pitch to -Pi/2, Pi/2
6 lines of code changed in:
More work on physics. Not quite where I want to be yet.
Fixed timer display.
125 lines of code changed in:
Working on closed-form solution to the physics problem...
13 lines of code changed in:
Working on closed-form solution to the physics problem...
71 lines of code changed in:
Ahh, here's a bug.... didn't initialize the players position, velocity, and acceleration to be zero.. MacOS does this automatically.... windows fills them with garbage.
3 lines of code changed in:
Weird sloppy bug with Physics Manager.... closest tube is sometimes undefined...
20 lines of code changed in:
Fixed a header issue with APTextureFont
1 lines of code changed in:
Added texture font library
371 lines of code changed in:
An update, though I'm not sure its a very good one.....
120 lines of code changed in:
Took out #warning, since windows does not have that capability...
3 lines of code changed in:
Added windows.h to the WirePCC code for Windows compilation.
18 lines of code changed in:
Added windows.h to the prefix.h for Windows compilation.
4 lines of code changed in:
Fixing things slowly...
2 lines of code changed in:
Stupid physics engine. Lame input handling.
180 lines of code changed in:
Building the structure of the game engine....
638 lines of code changed in:
Pages: « Previous 1 2 3 4 5 6 Next »
Generated by StatCVS 0.2.2