haa, finally I found a hole in your code, it’s the input system. The rest is GREAT !!
here, when the game is loading the title, pressing Esc would skip intro, right ?
But if I press Esc several times, the escape command would be executed several times too, thus I did these several times :
I clicked the Single/MultiPly button (they just weren’t visible), chose pilot and launched the ship several times until the framerate went down.
and the result were these :
This is excellent. Thank you. I’ve been spending so much time looking into other things I haven’t been able to do much testing.
Thanks. I’ll start working on this. Hopefully I’ll get this fixed in the next release.
I’ve been adding a lot of features to this program. It’s almost done ( I think)
I just have to add a bunch of missions- but that’s not too difficult. I have 10 currently.
I mostly need people to test the thing and help me find errors. The source and an installer is on my panda3dprojects page:
Here’s some recent screenshots:
Looking at your 2d stuff, I found something :
[size=150]1.[/size] you use screen resolution query, but implement 2d position with 4:3 ratio hardcoded in mind ?
(Environment.py : 2058)
self.console = OnscreenImage(image = 'models/shipicons.png', pos = (1.1, 0, 0.75), scale = .2)For widescreen, that puts the icons too far away from screen edge. As treeform said, he would kill you for that.
This should be fine :
aspRatio=base.camLens.getAspectRatio()
self.console = OnscreenImage(image = 'models/shipicons.png', pos = (.83*aspRatio, 0, 0.75), scale = .2)[size=150]2.[/size] I guess you already know that P3D always downscales all textures to power-of-2 by default, unless you use “textures-power-2 none” option.
For small texture, the artifact would be obvious, especially if the opaque part is thin.
radarsight.png (orig. 25 x 25) :
Texture radarsight (from models/radarsight.png)
2-d, 16 x 16 pixels, each 4 bytes, rgba
repeat x repeat, min default, mag default, aniso 1, border 0 0 0 1
1024 bytes in ram, compression off
target.png (orig. 150 x 150) :
Texture target (from models/target.png)
2-d, 128 x 128 pixels, each 4 bytes, rgba
repeat x repeat, min default, mag default, aniso 1, border 0 0 0 1
65536 bytes in ram, compression off
And for me, there is 1 annoying thing when launching the ship :
you move the mouse pointer to screen center right after the LAUNCH button get clicked, and then the scene get loaded, which takes awhile. During that loading time, user may move the mouse, which would yield ward change as soon as the scene finally loaded, since there is mouse position offset. In that case, user would lose the 1st place view focus, which perhaps you’ve set carefully, which could be important according to the mission, and that’s really really bad.
Shouldn’t the mouse centering code be put at the very end of Environment’s init, or even better in inSimulation’s end ?
New Version 0.95 is up at my P3DP.com site.
The first tour of duty is now complete: 11 missions.
There’s a briefing screen map that explains the missions before each level.
I’ve updated a lot of the models.
Here’s the link to the video on youtube. http://www.youtube.com/watch?v=1w8X0xze3B8
@ynjh_jo:
1-I’ve fixed the widescreen ratio issues I think.
2-I’ve fixed the textures to be powers of 2
3-I believe the mouse pointer centering code is already in the Environment script. I’ve noticed the same thing you have but I don’t know how to fix it exactly.
[3] you just need to be sure that movePointer() call is the last call after loading & preparing everything in the mission, to minimize time lapse to its 1st rendered frame.
I dont like joysticks really , and was hoping there was a mouse control mechanism here somewhere 
Game looks nice beyond that, and I wish you luck with it )
cheers
nl
Well, I got hit by hurricane Ike so i’ve been out of the picture for a little.
@Neighborlee: I believe there is mouse control. I prefer the joystick but when I’m working on it with a different computer I just use the mouse control. This all should be automatically controlled by the program. If there is a joystick connected to the computer at the moment of start-up the program will use that, if not it will use the mouse for control.
I am unable to run the game or install pygame or psycho to install (error message about python not being installed)
When running the game get this error message:
"Assertion Failed
Program c:\epoch\python\ppython.exe
File: panda/src/dxgsg8/dxGraphicsStateGuardian8.cxx
Line: 1709
Expression: _d3d_device !=Null"
is there a way to alter the config file to use software.
I am running Panda3d inside of a Virtual Machine using Ubuntu Linux 8.04
Thanks
Alex
You can use software rendering mode, but it will be terribly slow.
But why not install Panda3D on your ubuntu box? There are installers for ubuntu at the download page…
well the ubuntu box is the 64-bit edition and panda doesn’t seem to want to run in 64-bit mode. that and I am not experienced enough to program in Linux… yet
It doesn’t? There are also 64-bits ubuntu binaries on the download page. I use 64-bits ubuntu and it works great!
In your case:
panda3d.org/download/panda3d-1.5 … _amd64.deb
I’m such a dummy, I have panda installed on my laptop and forgot to install it on this machine… LOL
oh well it’s working now, thanks for the link
I know it’s been some time since I’ve put up an update. I apologize. I have been working on Epoch and the most recent version is up on the p3dp.com site now.
Tour 2 is now half-way complete. It will be a few months until I can devote more time to the game. Until then I need the communities help to find bugs and issues with game-play. I could especially use advice about levels being too hard, easy, boring, etc.
holy cow - it’s a kickass more than ever
keep it up!
i am going to test it ! the game nice !.
The Environment.py says it’s the network version. I’d be interested how a local/SP version would look like.
The early versions of the program where written exclusively for single player. I know the code reads:
But this is an old line.
During development I realized that local LAN play was also possible through Panda and incorporated a few functions within the program to pass and accept data. In the beginning I made a network version separate from the single player version because I knew the single player version was working and didn’t want to contaminate the working version with code that I wasn’t sure would actually work in the end. I don’t currently maintain two separate development paths.
At the beginning of the program you are asked to choose single player or multiplayer. The single player mode is recorded by main.py and passed to environment.py . Remember, environment.py only loads and plays the individual levels of the simulator environment. main.py on the other hand is the master file that tells environment what to load up: level, player ship, multiplayer status, etc.
During single player play, most of the multiplayer centered classes are simply never initiated. For example there is a “Nonplayer” class that represents another human player in the LAN but not the player’s craft on this computer.
The easiest way to figure out what is happening in the program is to just focus on the classes. Each class is an object that is instanced.
At the beginning of a level main.py will create an instance of the environment class. This will then load up a mission file. That mission file will load up several Ship3D objects (these are the craft in the game). Sometimes the mission will give initial orders to these ships, sometimes they are simply created and the default “Attack all enemies” behavior takes over. The mission file also loads up a “task” which checks on how things are going in the level, if the win conditions have been met, or if a new stage of the level has been reached.
I have tried to comment a lot to help people understand what I was thinking but I admit that commenting hasn’t been my primary focus lately. Let me know if something is unclear.
Man, this was an interesting project, and it works on 1.6.2
I am curious if you are still maintaining and if there’s any progress, as there are errors that pop now and then.
Yes, I’m still maintaining this project. As I wrote in a previous post I was out of the ability to work more on the project for a few months due my situation at work/school.
But I’m back to development. In fact I have been contacted by some people who want to further develop the program. Any errors you have would be great information and most appreciated. All my time is pretty limited, so what time I have I spend on development and not play testing the game.
The newest, unreleased version has some improved graphics, shaders, opening screen, but no new levels as yet.
Also, we are trying to see about implementing it on a web browser interface for easy multiplayer play similar to David’s tagger program- more to come.