The setup is all about the light placement for this experiment, following where the lights are placed in my room and also the height was measured from the balls to the lights (the pictures earlier in the artefact show the lighting and the balls on my pool table.
Using a high self illumination map for the lighting and leaving it standard white – as the lights are in my room.
Rendering time: 12 Seconds / Resolution: 320x240
Above is the first render that uses a Raytrace reflective map on the balls, you can see the two white specs on the top of the balls which are the reflections of the lights above. This is physically correct as described earlier in the artefact.
To set this up the Raytrace reflective map was setup low at 20, the initial 100 was so intense that the reflection was overwhelming and it looked like a mirror.
A closer look at one of the balls shows the two reflections of the lights above. Notice to follow the pattern of the real world halogen lights a falloff map has been placed on the imitating lights in 3DS Max. What this has done has made the reflections a rounded gradient, looking closely the white is quite intense and dependent on the slope of the ball the white will slowly lose its transparency on the outer edge of the reflection.
The next stage of the experiment is to add a specular highlight and glossiness to the balls and see what effect this has on the outcome of the reflection of the balls. I believe this will add to the realism and follow real world patterns. Such as the larger area that the of white reflective colour amounts to on the top of the ball(this can be seen in the earlier picture of the actual pool balls).
It just a matter of experimenting with the amount of specular highlights to make the specular amount on the material greater or smaller when the light shines on it.
The above image is a close-up render of one of the balls. Notice that the specular amount has overwhelmed the initial reflection of the lights above. However this does not matter as much once it is zoomed out to show all the balls (below).
Rendering time: 12 Seconds / Resolution: 320x240
The final test of this experiment is to see how the image looks like when the reflection map is turned off and the specular highlights are left on to create the white marks on the top of the balls
Although we lose the reflection of the balls it is not very noticeable from a distance what the changes are. The colour of the specular white areas are not bright white but are a very light colour of whatever the ball colour is. The amount the ‘artificial’ reflective lighting has also been reduced in size, but again not too visible unless close up.
Rendering time: 9 Seconds / Resolution: 320x240
Success! Just using specular highlight can mimic a reflective light map on 3DS Max. Rendering at the same resolution, just using specular highlighting is quicker. So for a larger scene this can be adapted to save time. The next stage is to test the scene created and ask viewers of their opinion to see which reflective lighting method is best – the actual reflection of the lights, the cheated specular method that mimics the reflection of lights or finally a combination of the two. |
Pie chart to show favourite zoomed out scene of the pool ball experiment
Looking at the results above I believe that the test was a success. This being that the results are so close to each other that there is not a definitive favourite scene to choose from. Thus my experiment of using specular highlights to mimic the actual reflection map in 3DS Max worked. The Raytraced reflective map I believe gives the best results as far as reflections from the light and also alongside the use of global illumination the reflection of other balls gives an all round positive realistic result.
With the difference of only 3 seconds between the rendering results I have no objections in using this method for future scenes.
- Only two people didn’t notice the missing reflections on the balls when the reflective map was taken off for the last scene.
- Half of the people that voted for the raytrace reflection method thought that the other two rendered results had too large of a white light reflection. But the majority said it could depend on the size of the light (as they were informed of the distance between the lights and the balls).
- Only 6 people noticed the transparent gradient of the light reflections in the rendered scenes.
