Pandora Tincs - Multiplayer Shooter For Pandora (video)


Butterman

Gief Pandara
Joined
Jan 30, 2009
Messages
776
Location
London
Website
Visit site
TINCS is currently inactive until Pandoras start being shipped, the development thread is here:

http://www.gp32x.de/board/index.php?/topic/48729-human-condition-development-thread/
 
Last edited by a moderator:
Looks pretty sweet, Though with the MD2 model system, how many models can you get on screen before we see a framerate hit? And from what you said, I assume that this model is near the 800 mark?

Personally I would prefer higher detailed models, but if it is going to slow the game down when we have many players on a map then it will just have to stay with the current format.

Also, I'm glad that you are using BSP, hopefully we will see some ports of old CS 1.5/1.6 maps :D
 
Mattz said:
Looks pretty sweet, Though with the MD2 model system, how many models can you get on screen before we see a framerate hit? And from what you said, I assume that this model is near the 800 mark?

Personally I would prefer higher detailed models, but if it is going to slow the game down when we have many players on a map then it will just have to stay with the current format.

Also, I'm glad that you are using BSP, hopefully we will see some ports of old CS 1.5/1.6 maps :D
Yeah, that's why I said we will wait until much later on before we decide on exact model formats, so we can get a balance between performance and player count.

The model is about 800 polygons, I reckon we could handle up to 1000 polygons without any major artifacting (with MD2 this is), we could probably run ~80 of these models on screen at once. But like i said, I won't know until I get a Pandora.

I'm aiming for very high graphical quality, that's why I'm writing this from the ground up :D, ports of old CS maps should be possible, I'm going to write a straight BSP reader that will read basic info (spawn points) etc and textures as well as the map itself, then pick up textures from different files (so we don't load more than one of the same texture or anything silly). I would probably have to write a converter for CS maps as I would be using the Quake 3 BSP format. That said, TINCS IS NOT COUNTER STRIKE. :D :D

EDIT: Just tested out 400 models at once (higher poly ones) on my rig, didn't take any performance hit at all. So it's safe to say we will change model in the future, possibly when I get my hands on a Pandora, maybe to .fbx. Video is here : http://uk.youtube.com/watch?v=_n43xENJan4
 
Last edited by a moderator:
the Pandora is capable of OpenGL ES, not full OpenGL. any possibility you could set up an ES dev environment? there aren't too many differences but it'll certainly take some amount of time to port your game form GL to GL ES.
 
CC_machine said:
the Pandora is capable of OpenGL ES, not full OpenGL. any possibility you could set up an ES dev environment? there aren't too many differences but it'll certainly take some amount of time to port your game form GL to GL ES.
I've taken that into account and have had the OpenGL ES 2.0 header handy, making sure I'm not using anything that isn't in it, so far so good.

Explain what you mean by setting up an OpenGL ES environment? Would I be able to run the programs under Vista?
 
Last edited by a moderator:
I think there are emulators for OGLES or for the SGX or something.
Here's some install instructions.
http://pandorawiki.org/OpenGLES_On_the_Pandora#Tools

Also, I think the Pandora's OpenGL ES is the new version that doesn't have any fixed-functionality stuff built-in, so you might need to write a simple fixed transform vertex shader and a texturing pixel shader for it to work in ES.
 
Are all the models running the same frames of animation? Because if they are, thats no a very good representative of actual speed... They should all be on different frames of different animations.
 
zacaj said:
Are all the models running the same frames of animation? Because if they are, thats no a very good representative of actual speed... They should all be on different frames of different animations.
Yeah, I'm just sorting out instancing right now. I can put up another video with different frames, but I'm working on rendering MD3 models.

lulzfish said:
And can you explain in a few sentences why md2 loses animation quality as you add polygons? I would think those would be totally separate.
It's down to the way MD2 stores vertex data, it just isn't very accurate. I'm working on implementing MD3 models right now.
 
Last edited by a moderator:
Hey, I'm excited about the project and I think that you may have something here. Don't let the any negative comments get you down, I've notice people tend to be fairly un-assuring and assume the worst until proven otherwise (and there is some merit to it).

I'm one of the gentoo.pandora.devs and I would be happy to run some benchmark tests in our environment once we have the right tools installed. We want to incorporate as many of these community/homebrew projects as we can and I will be happy to make an .ebuild and compile it when it is possible. I currently have a beagleboard B7 on my desk (128MB ram variant) and it should be able to give you a rough idea of what to expect when the official pandora platform is released.

Respectfully,
Viridior
 
viridior said:
Hey, I'm excited about the project and I think that you may have something here. Don't let the any negative comments get you down, I've notice people tend to be fairly un-assuring and assume the worst until proven otherwise (and there is some merit to it).

I'm one of the gentoo.pandora.devs and I would be happy to run some benchmark tests in our environment once we have the right tools installed. We want to incorporate as many of these community/homebrew projects as we can and I will be happy to make an .ebuild and compile it when it is possible. I currently have a beagleboard B7 on my desk (128MB ram variant) and it should be able to give you a rough idea of what to expect when the official pandora platform is released.

Respectfully,
Viridior
Wow, yeah, that's a resource I will definitely call upon in the near future. Cheers.

I'm going to stick with MD2 for this week at least, MD3 is a pain. I've got collision and stuff in, I want to code a basic server this week, so hopefully, next weekend, I can get some people together to start playing a basic game of TINCS, (with no map and on windows) and start shooting each other up. That reminds me, I'm going to need some beta testers, ;) .
 
Last edited by a moderator:
Butterman said:
That reminds me, I'm going to need some beta testers, ;) .

Are you gonna make the beta public or pick a few people and give them the beta?
 
Last edited by a moderator:
Username said:
Butterman said:
That reminds me, I'm going to need some beta testers, ;) .

Are you gonna make the beta public or pick a few people and give them the beta?
By beta testers, I meant people who I can call on during development to help test X/Y feature online (pre-alpha testers? :huh:). There won't really be a beta, I will be releasing binaries all the time through development.

That leads me on to something else, I will be keeping source closed, down to the fact that this is a multiplayer game. That said, if anybody wants to request portions of code, to help with what they're doing, or steal if they want, I will be happy to deliver.
 
Last edited by a moderator:
Possible name: TINYCS ??

"TINYCS Is Not Your Counter Strike"


I think it would be appropriate considering the Pandora form factor to be calling this Tiny CS.
 
Looks nice, but collision detection in use seems ripe for exploiting, dont know if thats just to show its working or is actually going to be implemented as such
 
A setting that turns on or off the random event of poop whales falling from the sky. Think of it as a kind of Smash Ball feature, except deadly and hilarious.

And yes, I do mean a whale made out of feces. Same size as a blue whale.
 
Back
Top