Presidium- An SF Citybuilder

WIP games, tools & toy projects
#141

Morgan, have you changed which class is the one for launching the app? A few months ago you said it was DebugWorld, but I don’t see that class in the source.

Thanks,
Mark

#142

You can try out DebugBehaviour or DebugPathing instead (TBH, the functions are pretty loose at this stage.) Cripes, I haven’t posted in a while. I’ll letcha know what’s happening soon.

#143

Morgan, have you changed which class is the one for launching the app? A few months ago you said it was DebugWorld, but I don’t see that class in the source.

Thanks,
Mark

#144

You can try out DebugBehaviour or DebugPathing instead (TBH, the functions are pretty loose at this stage.) Cripes, I haven’t posted in a while. I’ll letcha know what’s happening soon.

#145

Got back last night from a weekend getaway and I was able to use both classes.

All I can say is wow! I will be tracking this game and will buy when ready.

Thanks,
Mark

#146

Got back last night from a weekend getaway and I was able to use both classes.

All I can say is wow! I will be tracking this game and will buy when ready.

Thanks,
Mark

#147

Thanks for trying it out. If you don’t mind my asking, were you running it under windows/mac/linux? I’ve had some noticeable slowdowns on other OS/hardware, so external reference points are good. :slight_smile:

I’ve expanded considerably on the content associated with most of the structures in the game over the last month or so, and the economy mechanics are essentially sorted at this point. There’s a good deal of extra content to introduce or debug, but it’s quantitative rather than qualitative- I know with some precision what needs to be done, and just have to follow well-defined steps to do it. (Rather than being, e.g, paralysed with indecision over the proper tile resolution for terrain-mesh sections.)

This should be a representative sample of the current game-state:

http://s15.postimg.org/6fld09qsp/new_settlement.png

There is one technical point I could use some advice on, though: I’ve been updating the fog-of-war display so as to try and blend/fade visually between old and new fog values every second or so. I’d prefer to do this purely using the GPU and without shaders (since I don’t want to dive into those just yet.) Is there any simple way for ‘classic’ GL to get and apply the weighted average of two textures, including their alpha… or have I missed something obvious?

#148

Thanks for trying it out. If you don’t mind my asking, were you running it under windows/mac/linux? I’ve had some noticeable slowdowns on other OS/hardware, so external reference points are good. :slight_smile:

I’ve expanded considerably on the content associated with most of the structures in the game over the last month or so, and the economy mechanics are essentially sorted at this point. There’s a good deal of extra content to introduce or debug, but it’s quantitative rather than qualitative- I know with some precision what needs to be done, and just have to follow well-defined steps to do it. (Rather than being, e.g, paralysed with indecision over the proper tile resolution for terrain-mesh sections.)

This should be a representative sample of the current game-state:

http://s15.postimg.org/6fld09qsp/new_settlement.png

There is one technical point I could use some advice on, though: I’ve been updating the fog-of-war display so as to try and blend/fade visually between old and new fog values every second or so. I’d prefer to do this purely using the GPU and without shaders (since I don’t want to dive into those just yet.) Is there any simple way for ‘classic’ GL to get and apply the weighted average of two textures, including their alpha… or have I missed something obvious?

#149

Blend them?

#150

I’ve tried various blending options, but either they don’t do what I need or I don’t understand them properly. (e.g, GL_MODULATE seems to multiply the two textures together, GL_ADD is too bright, and others give special treatment to the alpha channel, which I don’t need.)

I could just do it on the CPU, but it’s something I’d rather not every 25th of a second.

#151

Blend them?

#152

I’ve tried various blending options, but either they don’t do what I need or I don’t understand them properly. (e.g, GL_MODULATE seems to multiply the two textures together, GL_ADD is too bright, and others give special treatment to the alpha channel, which I don’t need.)

I could just do it on the CPU, but it’s something I’d rather not every 25th of a second.

#153

http://www.opengl.org/wiki/Texture_Combiners

#154

I appreciate the reference, though it seems to be going out of it’s way to say ‘don’t ever use this, use shaders instead’.

I have thought of a different approach that might work. I could basically store the old and new fog values as 2 layers of a 3-dimensional texture, and change the UV mapping of the fog-layer-geometry so as to interpolate between them over time. I could either cache the intermediate results (about 1 meg per frame) or generate them as needed (maybe 1% of CPU) or simply apply the fog as one giant quad (which assumes perfectly flat terrain.) I think the lattermost approach might be an acceptable hack for the moment (I have the capacity for proper 3d-terrain, but I’m not making any real use of it at the moment.)

#155

http://www.opengl.org/wiki/Texture_Combiners

#156

Wait, you’re not using shaders? Sorry, that just seems a little silly :stuck_out_tongue: They’re only beneficial to your game, why wouldn’t you use them?

#157

Oh, I fully intend to move to a shading-based engine in future, I just haven’t gotten around to learning GLSL yet. (Originally, I wanted the game to be able to run on older hardware, but at this stage anything shader-incompatible is probably on a scrapheap.)

#158

Ah, I see. Sorry for assuming! It’s looking really nice so far! Do you need resources for learning GLSL? I’m sure I could dig some up, schools very boring today :slight_smile:

#159

I appreciate the reference, though it seems to be going out of it’s way to say ‘don’t ever use this, use shaders instead’.

I have thought of a different approach that might work. I could basically store the old and new fog values as 2 layers of a 3-dimensional texture, and change the UV mapping of the fog-layer-geometry so as to interpolate between them over time. I could either cache the intermediate results (about 1 meg per frame) or generate them as needed (maybe 1% of CPU) or simply apply the fog as one giant quad (which assumes perfectly flat terrain.) I think the lattermost approach might be an acceptable hack for the moment (I have the capacity for proper 3d-terrain, but I’m not making any real use of it at the moment.)

#160

Wait, you’re not using shaders? Sorry, that just seems a little silly :stuck_out_tongue: They’re only beneficial to your game, why wouldn’t you use them?