You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently the Sector View uses a weird mish-mash for its vector units. The coordinates stored are in parsecs, a sector is 256 cubic parsecs. Stars store their radius in solar radii. The Sector View uses the coordinates without modification, so a "unit" for anything that is given to the 3d engine is in terms of parsecs. This means that the radii for stars end up being really tiny (10^-5 and less), and then they end up not even being rendered by SceneKit (I think). So, currently I'm scaling the Stars up such that 1 solar radius == 1 parsec. The camera, though, positions itself at an approximation of an AU away from the star scaled to parsecs. This approximation is kinda weird, it tries to place the star such that if it were 1 Solar Radius it would have an angular diameter of ~0.5 degree (the same size as the sun from earth), trying to demonstrate the scale in a familiar way. This approach doesn't really work though, because we have to set an FOV of the camera, so that requires some framing/cropping of the sky, so to speak... at which point there's really no frame of reference for what 1.0 solar radius really looks like. So it sets its FOV up so that if you had 10 suns, they could fit in the viewport side by side. ¯\_(ツ)_/¯
I'm not really sure what to do about it yet though. The current implementation looks decent for now, but once I get the luminosity stuff working (#2), I'd like to come back to this and make it make more sense, somehow...
The text was updated successfully, but these errors were encountered:
I adjusted the FOV stuff to be 6 suns and that seems to work a bit better.
But ultimately I need to get the scale and density of the Sectors in general figured out. Currently a sector is 256 cubic parsecs, and on average that should be on the order of millions of stars per sector, which is obviously way too many. So a sector should be more like 10 cubic parsecs? But at that point the Sector View is totally gonna fall apart so my coordinate system needs to be adjusted. I was thinking about doing light years (just over 3ly per parsec), but @timbotron mentioned doing decaparsecs... I like that idea, I'll need to do a bit more pondering on that.
Currently the Sector View uses a weird mish-mash for its vector units. The coordinates stored are in parsecs, a sector is 256 cubic parsecs. Stars store their radius in solar radii. The Sector View uses the coordinates without modification, so a "unit" for anything that is given to the 3d engine is in terms of parsecs. This means that the radii for stars end up being really tiny (10^-5 and less), and then they end up not even being rendered by SceneKit (I think). So, currently I'm scaling the Stars up such that 1 solar radius == 1 parsec. The camera, though, positions itself at an approximation of an AU away from the star scaled to parsecs. This approximation is kinda weird, it tries to place the star such that if it were 1 Solar Radius it would have an angular diameter of ~0.5 degree (the same size as the sun from earth), trying to demonstrate the scale in a familiar way. This approach doesn't really work though, because we have to set an FOV of the camera, so that requires some framing/cropping of the sky, so to speak... at which point there's really no frame of reference for what 1.0 solar radius really looks like. So it sets its FOV up so that if you had 10 suns, they could fit in the viewport side by side.
¯\_(ツ)_/¯
I'm not really sure what to do about it yet though. The current implementation looks decent for now, but once I get the luminosity stuff working (#2), I'd like to come back to this and make it make more sense, somehow...
The text was updated successfully, but these errors were encountered: