- #1

- 3

- 0

I'm building a solar system simulator, designing it to be as accurate as possible. Visualization is done through OpenGL (although I may be forced to use a software renderer for higher precision).

I have set up all the planets and their orbits, tilts, and made them update according to the best models I have available. The model can, for instance, show Earth's solstices hundreds of years in the future.

I'm trying to show the upcoming transit of Venus, and unfortunately my model predicts the transit about a day too early!

Explanations:

-The sun is currently modeled as not moving, yet the sun wobbles significantly (like up to a solar width). I believe this accounts for at least part of the difference. Where can I find data to model this?

-I am not accounting for the moon's effect on the Earth either. I can work out how to fix this myself.

-The orbits of the planets evolve over time. My model is based on the evolving equations given on this site, and the computation is much the same. The model uses J2000 as a basis, so perhaps errors accrued over the following 12 years? Is this model correct?

-Relativistic effects

All of these I know figure into the model, but I'm not sure which are most important (though I suspect the sun's wobble and Earth/moon most), nor if I'm omitting something significant. -I am not accounting for the moon's effect on the Earth either. I can work out how to fix this myself.

-The orbits of the planets evolve over time. My model is based on the evolving equations given on this site, and the computation is much the same. The model uses J2000 as a basis, so perhaps errors accrued over the following 12 years? Is this model correct?

-Relativistic effects

Finally, how

*was*this transit predicted anyway? There must be some simulation or calculation somewhere. How was it predicted? Ideally, I could base my work on that instead . . .

Thanks,

Ian