Reflected light, relative brightness?

[billw]

I am writing a space based game, and in it I will render suns, planets, moons etc. I am currently working on the lighting model, and wondering if I should bother taking into account reflected light. For instance, I think the reflected light from the Moon to the dark side of the Earth has such low intensity relative to the light from the Sun to the lit side of the Earth that it is not worth simulating. i.e. it would never be noticeable without extreme exposure settings. But for other bodies, for instance the moons of gas giants, where the sun is much further and the ratio of sizes between the gg and its moons much larger, reflected light could be significant. Could anyone point to me either images, numerical data or analyses that demonstrate the ratios? Or just give me their best guesses?

Thanks alot!

P.S. IANAA (I am not an astronomer!)

P.P.S For those interested: A short vid of where I am up to at the moment: http://uk.youtube.com/watch?v=iLAc1nm6w4E. Just about to move from a single planet to a solar system (a short jump from doing one planet to doing many), and trying to work out what directions to light from, hence this post.

[Merlin66]

Definately an interesting poject, even for a non-stronomer!!

Looking at the video, I'd say the reflected "hot spot" is far to large for the planet being illustrated. Depending on the angular size of the illuminary a sphere only give a spot reflection of about 1/300 its diameter ( PM me for the detailed calcs)

The back reflected light from a planet to its moon (we call it Earthshine for our moon) I'd say should be modeled to give that feeling of depth and solidity. About 10% of the incident on the planet would probably be a good starting point.

Hope this helps!

( I design flight simulator aircraft !!)

[billw]

Thanks! Well I will be coming back here alot for advice as I want to go as realistic as (realistically) possible. But I do have to make some (alot of) compomises to get the thing to run at a decent speed. This is really the point of this question, would it be straying to far from reality to not model reflected light between different bodies (so as to safe video card clock cycles) or is reflected light visually apparent under certain circumstances? Still I think you are right, I only really have to render one body at a time (what is the general term for a sun/moon/planet/asteroid anyway?) so I can afford to model reflected light, and it will give a much nicer look.

With regard to the specular hot-spot size, thanks for the tip. I haven't yet found a picture of the Earth from space that shows the hot-spot from a distance, so I can see its relative size. When it comes to finalizing my rendering settings I will be looking for data on albedo, average reflectance, etc, but for now all settings are just guessed.