Ray Tracer pt2
414
0
Advertisement
So, I've fixed the graphical artifacts in the code, it was as suspected something todo with the refraction code;

However, while it "fixes" it I'm not too happy about the fix.
The problem was of course that it was saturating the colours leading to white spots on the image; now the clamping was handled before the conversion from floating point to 8Bit RGB in the main loop, however if I moved it from there to the RayTrace() function everything seems to work fine.
This bothers me on three levels;
1) why does it work when its there but no when it's in the main loop? Conceptually it's the same difference.
2) The code I'm using as a guide clamps in the main loop and works fine. Sure my code isn't 100% the same, but layout and maths wise it is.
3) This removes HDR abilities for now [sad] A nice HDR-tone mapped output is a future 'want' for this. However I'll worry about that later.
I'll ponder that while pushing on wards, chances are things are going to get re-written soon anyways to support things such as textures et al

However, while it "fixes" it I'm not too happy about the fix.
The problem was of course that it was saturating the colours leading to white spots on the image; now the clamping was handled before the conversion from floating point to 8Bit RGB in the main loop, however if I moved it from there to the RayTrace() function everything seems to work fine.
This bothers me on three levels;
1) why does it work when its there but no when it's in the main loop? Conceptually it's the same difference.
2) The code I'm using as a guide clamps in the main loop and works fine. Sure my code isn't 100% the same, but layout and maths wise it is.
3) This removes HDR abilities for now [sad] A nice HDR-tone mapped output is a future 'want' for this. However I'll worry about that later.
I'll ponder that while pushing on wards, chances are things are going to get re-written soon anyways to support things such as textures et al
Advertisement
Advertisement
Advertisement
Discussion