More physically accurate light modeling
Posted: Mon Aug 25, 2014 2:06 am
To accurately model an incandescent light bulb you would need to have a heated filament using a black body emissive material, and not an octane light inside some bulb.
If you make lamps for product design, you often end up taking close-up shots of them, but if the bulb is not physically modeled as described above it will look unrealistic as you will have both the bright filament and a floating light, the lighting won't be accurate. But if you do take the approach described above, it becomes difficult to know how much light is actually emitted as the size of the filament VS the power and efficiency values will have an impact on how much light is produced. You won't know if your 45watt bulb is actually emitting as much light as it should, or too much.
Another example where I did use an Octane light was one where I wanted three differently colored lights inside a tube which used a scattering material. If I simply use three lights and changed their color, I would have no idea how much light is emitted anymore. The only way to make sure I was still physically correct was to have all three lights identical and using their correct values (4500k, power of 20 each), and use a colored glass bulb to filter the lights. This way I knew my lights were all emitting as much light, and the resulting color was only due to the colored glass, not by messing with a non physically correct color value of the lights. This increased rendering time due to the presence of a bulb, but the results were correct.
So what I think would be really useful is at least to have some sort of light meter, this way it would be possible to adjust the black body power value of filaments, or the color values of the lights, until you manage to get the equivalent you would have in real life.
Or maybe there is a better solution, but this is what I thought of.
Anyone has faced this issue?
If you make lamps for product design, you often end up taking close-up shots of them, but if the bulb is not physically modeled as described above it will look unrealistic as you will have both the bright filament and a floating light, the lighting won't be accurate. But if you do take the approach described above, it becomes difficult to know how much light is actually emitted as the size of the filament VS the power and efficiency values will have an impact on how much light is produced. You won't know if your 45watt bulb is actually emitting as much light as it should, or too much.
Another example where I did use an Octane light was one where I wanted three differently colored lights inside a tube which used a scattering material. If I simply use three lights and changed their color, I would have no idea how much light is emitted anymore. The only way to make sure I was still physically correct was to have all three lights identical and using their correct values (4500k, power of 20 each), and use a colored glass bulb to filter the lights. This way I knew my lights were all emitting as much light, and the resulting color was only due to the colored glass, not by messing with a non physically correct color value of the lights. This increased rendering time due to the presence of a bulb, but the results were correct.
So what I think would be really useful is at least to have some sort of light meter, this way it would be possible to adjust the black body power value of filaments, or the color values of the lights, until you manage to get the equivalent you would have in real life.
Or maybe there is a better solution, but this is what I thought of.
Anyone has faced this issue?