Hi Gabor,

sorry for the late reply, but I was in Italy and didn't have a good internet 
connection.

> sounds nice. would be great to read the configuration from a variable in the 
> ~/.fluxus.scm configuration script, although i think that runs later then the 
> glut setup. what do you think?

Sounds good to me, but wouldn't it be complicated to initialize the interpreter 
before GLUT?

> would be nice to see how it looks. don't you want to make a branch in the 
> fluxus repo? so everyone can try your modifications and we could easily 
> import them to master. dave can give you access.

I'll be happy to do so if Dave can give me access, I haven't used GIT yet, but 
don't think it's complicated.
In the meantime I added some simple classes for post-processing and render 
targets, and it looks pretty good. 
It would be possible to do the same with FFGL but I'm doing a framebuffer copy 
to keep multisampling and I don't think shadows would work when using the 
PixelPrimitive, since it doesn't provide a stencil buffer. It would work by 
using GL_DEPTH24_STENCIL8_EXT but I don't know how widely it is supported.

> yes, i use gdb.

A quick question regarding gdb, I'm quite new to using it through the command 
line, and wasn't able to set breakpoints yet ( having trouble with namespaces 
etc ) could you give me a quick example?
I must say that it's nice to code without using an IDE, maybe i'll get used to 
it : )

Fluxus is great fun!
thanks
Daniel


On Oct 20, 2011, at 11:43 PM, gabor papp wrote:

> hi Daniel,
> 
>> multisampling in GLUT and the string solution works. It would be also
>> possible to read the string from a configuration file, that way users
>> could modify rendering settings easily.
> sounds nice. would be great to read the configuration from a variable in the 
> ~/.fluxus.scm configuration script, although i think that runs later then the 
> glut setup. what do you think?
> 
>> 2: I had a look @ the shadow volume code, and I modified it to render
>> the shadows by subtracting light from the scene as a postprocess.
> would be nice to see how it looks. don't you want to make a branch in the 
> fluxus repo? so everyone can try your modifications and we could easily 
> import them to master. dave can give you access.
> 
>> 3: I added a reload current script function that gets called with F1,
>> This way I can edit my scripts in an external editor and reload them
>> quickly.
> this is a bit risky if you press f1 by accident during a performance :). one 
> of my colleagues added a modification that reloads the file automatically if 
> it has been changed. these are both nice, although i think we should make the 
> editor good enough so an external editor won't be needed.
> 
>> If any of these are useful I will be happy to send in the code. I had
>> a hard time though with debugging... do you use GDB on OSX?
> yes, i use gdb.
> 
>> Another question regarding the installation of modules. I wanted to
>> install gabor's anttweakbar module but I didn't find any info on how
>> to install it. I installed the lib, and then tried putting tw.ss in
>> modules/scheme but when I open the demo file I get this error message
> you don't have to put tw.ss in modules/scheme. if you have the command line 
> version of fluxus, not the app, you can just run fluxus from the folder when 
> the example and tw.ss are.
> basically if you use (require "tw.ss") then racket looks for tw.ss in the 
> current directory. the other way is to use the form (require something/tw), 
> where you have to place the tw.ss file on the collects path in the 
> 'something' folder. in osx ~/Documents/Fluxus/collects is in the collects 
> path, so probably that's the easiest to use.
> probably you also need to change the require's in tw.ss from fluxus-017 to 
> fluxus-018.
> 
> best,
> gabor
> 

Daniel Berio
[email protected]



Reply via email to