|
|
Post by jamesmcm on Nov 22, 2019 20:02:37 GMT
Hi, My refraction results seem slightly strange:  But all the tests pass. Is it possible to get the environment set up for the example in the book:  So it could be compared more directly? |
|
|
|
Post by Jamis on Nov 22, 2019 23:21:04 GMT
Certainly! Here's the scene description, in YAML:
# ======================================================
# refraction-bend.yml
#
# This file describes the scene illustrated at the start
# of the "Transparency and Refraction" section, in
# chapter 11 ("Reflection and Refraction") of
# "The Ray Tracer Challenge"
#
# by Jamis Buck <jamis@jamisbuck.org>
# ======================================================
# ======================================================
# the camera
# ======================================================
- add: camera
width: 200
height: 200
field-of-view: 0.5
from: [-4.5, 0.85, -4]
to: [0, 0.85, 0]
up: [0, 1, 0]
# ======================================================
# light sources
# ======================================================
- add: light
at: [-4.9, 4.9, 1]
intensity: [1, 1, 1]
# ======================================================
# define some constants to avoid duplication
# ======================================================
- define: wall
value:
pattern:
type: checkers
colors:
- [ 0.00, 0.00, 0.00 ]
- [ 0.75, 0.75, 0.75 ]
transform:
- [ scale, 0.5, 0.5, 0.5 ]
specular: 0
# ======================================================
# describe the scene
# ======================================================
# floor
- add: plane
transform:
- [ rotate-y, 0.31415 ]
material:
pattern:
type: checkers
colors:
- [0.00, 0.00, 0.00]
- [0.75, 0.75, 0.75]
ambient: 0.5
diffuse: 0.4
specular: 0.8
reflective: 0.1
# ceiling
- add: plane
transform:
- [ translate, 0, 5, 0 ]
material:
pattern:
type: checkers
colors:
- [0.85, 0.85, 0.85]
- [1.00, 1.00, 1.00]
transform:
- [ scale, 0.2, 0.2, 0.2 ]
ambient: 0.5
specular: 0
# west wall
- add: plane
transform:
- [ rotate-y, 1.5708 ] # orient texture
- [ rotate-z, 1.5708 ] # rotate to vertical
- [ translate, -5, 0, 0 ]
material: wall
# east wall
- add: plane
transform:
- [ rotate-y, 1.5708 ] # orient texture
- [ rotate-z, 1.5708 ] # rotate to vertical
- [ translate, 5, 0, 0 ]
material: wall
# north wall
- add: plane
transform:
- [ rotate-x, 1.5708 ] # rotate to vertical
- [ translate, 0, 0, 5 ]
material: wall
# south wall
- add: plane
transform:
- [ rotate-x, 1.5708 ] # rotate to vertical
- [ translate, 0, 0, -5 ]
material: wall
# background ball
- add: sphere
transform:
- [ translate, 4, 1, 4 ]
material:
color: [ 0.8, 0.1, 0.3 ]
specular: 0
# background ball
- add: sphere
transform:
- [ scale, 0.4, 0.4, 0.4 ]
- [ translate, 4.6, 0.4, 2.9 ]
material:
color: [ 0.1, 0.8, 0.2 ]
shininess: 200
# background ball
- add: sphere
transform:
- [ scale, 0.6, 0.6, 0.6 ]
- [ translate, 2.6, 0.6, 4.4 ]
material:
color: [ 0.2, 0.1, 0.8 ]
shininess: 10
specular: 0.4
# glass ball
- add: sphere
transform:
- [ scale, 1, 1, 1 ]
- [ translate, 0.25, 1, 0 ]
material:
color: [ 0.8, 0.8, 0.9 ]
ambient: 0
diffuse: 0.2
specular: 0.9
shininess: 300
transparency: 0.8
refractive-index: 1.57
|
|
|
|
Post by jamesmcm on Nov 23, 2019 10:57:10 GMT
Thanks! Are all of the demo YAMLs available somewhere btw? Also the refraction works okay:  But I have some issues with the walls (adding the walls behind the camera made the spheres invisible, I'll try to debug that now) |
|
|
|
Post by jamesmcm on Nov 23, 2019 10:58:06 GMT
Also for anyone reading this, note that the order of transformations matters!
Scaling * Translation != Translation * Scaling (since the Scaling would affect the amount translated)
Use Scaling * Translation for the yaml example.
|
|
|
|
Post by jamesmcm on Nov 23, 2019 14:30:46 GMT
Working (although the specular reflection is in a slightly different place wrt. the demo for some reason):  In the end all of the issues were related to the order of transformation matrices, and the code was fine  |
|
|
|
Post by manxome on Nov 23, 2019 21:05:45 GMT
I think the light moved! I am trying to gain confidence in my RayTracer implementation so I hope this doesn't just seem like nit-picking. Jamesmcm, thank you for posting your results. Jamis, thanks for the awesome book! OK, I just rendered this image using Jamis's yaml and noticed that I had shadows and speculars that were slightly off compared to the book picture that jamesmcm posted above. Here are the comparison pics:    If the light didn't move somehow, I have some serious debugging to do... |
|
|
|
Post by accolon on Nov 23, 2019 22:16:18 GMT
Using the YAML file in this thread, my Python implementation produces an image that looks exactly like those posted by you, manxome, and jamesmcm. I suppose that the image from the book is based on a slightly different YAML.
|
|
|
|
Post by Jamis on Nov 24, 2019 3:07:05 GMT
Good catch! I must have changed the position of the light source after rendering the image for the book...I don't know why. I suspect maybe I didn't like how the specular highlight was positioned over the image of the green ball, but then I never got around to replacing the image itself in the book.
|
|
|
|
Post by Jamis on Nov 24, 2019 3:08:33 GMT
Oh, and to answer your question about other YAML files: if you look in the "Gallery" forum and find the threads created by me, most of them include YAML files for recreating the images there.
|
|
