prt vs own radiosity engine
Author
Hello, community
How would Precomputed Radiance Transfer(PRT) compare to a custom built radiosity engine? Is it a good idea to built my own radiosity engine or something else? What problems are you having with the prt engine? I heard that it can take hours to process models.
Any help on the prt engine and.or building my own radiosity engine would be nice,
ProgrammingNerd
With PRT, you get the benefit that you can rotate the lights and still get nice lighting. But its quality is not anywhere near what a full-blown radiosity solution would do. But with radiosity, you can't have dynamic (rotating) lights.
Both PRT and a full radiosity solution will take a while to compute. Obviously, the full radiosity solution will take much longer.
Implementing a radiosity solution is not for the faint of heart. I don't know anything about your background, but usually full-blown radiosity solution is beyond the scope of most graduate level graphics courses unless massive assumptions are made about the input data. It's necessary to know the math behind radiosity extremely well to be able to implement a full solution.
If you want to finish your project in anything less than 4 or 5 years, I would suggest using PRT or some other scaled-down version of a full radiosity solution instead.
neneboricua
Implementing a radiosity solution is not for the faint of heart. I don't know anything about your background, but usually full-blown radiosity solution is beyond the scope of most graduate level graphics courses unless massive assumptions are made about the input data. It's necessary to know the math behind radiosity extremely well to be able to implement a full solution.
If you want to finish your project in anything less than 4 or 5 years, I would suggest using PRT or some other scaled-down version of a full radiosity solution instead.
neneboricua
Author
Hello,
Sounds like I have a bit of a choice to make. Do you know of any good resources on implementing a full-blown radiosity. I think that with that, I could make a better decision that what you just gave me, no affence, of course[smile].
Thanks for the help,
ProgrammingNerd
Sounds like I have a bit of a choice to make. Do you know of any good resources on implementing a full-blown radiosity. I think that with that, I could make a better decision that what you just gave me, no affence, of course[smile].
Thanks for the help,
ProgrammingNerd
Also note that with PRT, its not just about being able to rotate lights. The real win comes in when you move an object through a space where every point in space can have a different lighting environment. As the object moves through space, these unique area lighting environments will impact how the object is lit. You can see a demo of this in the IrradianceVolume sample that shipped in the August SDK.
btw, its not too hard to mix light mapping with PRT and traditional shadowing techniques.
Jason
btw, its not too hard to mix light mapping with PRT and traditional shadowing techniques.
Jason
Author
Hello,
Still a tough choice, though. I downloaded this program on fluid studios website that computes the radiosity solution for an input file. I must say, it produces excellent results. Also, it is open source, so developing my own out of it would be a six month expidenture. Anyway, if you would check it out and tell me what you think, that would be great!
Thanks for your input,
ProgrammingNerd
Edited by Coder: Fixed your link
[Edited by - Coder on September 1, 2005 11:58:22 PM]
Still a tough choice, though. I downloaded this program on fluid studios website that computes the radiosity solution for an input file. I must say, it produces excellent results. Also, it is open source, so developing my own out of it would be a six month expidenture. Anyway, if you would check it out and tell me what you think, that would be great!
Thanks for your input,
ProgrammingNerd
Edited by Coder: Fixed your link
[Edited by - Coder on September 1, 2005 11:58:22 PM]
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
