Celestia Forums

I've just identified a bug in the coordinates of the celestial grid.

At 0h, there are the usual angular coordinates 10, 20, 30, ..., 80 and -10, -20, -30, ... , -80. This is fne. But at 12h, there are only 10 and 20 on one side, and all the negative values are on the other side of the equator (-10, -20, -30, ..., -80). Where are the 30, 40, ..., 80 values, at 12h ?

This needs to be corrected.

While not a bug, I think we should also add the odd values 1h, 3h, 5h, ... on the equator.

I may have found the bug of the missing coordinates. In the 'render.cpp' file, at line 136, there is the number of coordinates defined :



Since there are six missing coordinates, that number should be 44, not 38. I made a build to test it, and it works. All the missing coordinates are back.

I think the wrong number was left there by Chris, when he changed the grid resolution, some time ago.

Not bad, for a non-programmer, heh ?


While we are at it, is there a reason to not show the odd coordinates too, along the equator ?

Cham wrote:I may have found the bug of the missing coordinates. In the 'render.cpp' file, at line 136, there is the number of coordinates defined :

Code: Select all

static int nCoordLabels = 38;

Since there are six missing coordinates, that number should be 44, not 38. I made a build to test it, and it works. All the missing coordinates are back.

I think the wrong number was left there by Chris, when he changed the grid resolution, some time ago.

Not bad, for a non-programmer, heh ?


While we are at it, is there a reason to not show the odd coordinates too, along the equator ?

That's definitely the problem . . . Nice find!

Unless you beat me to it, I'll add the odd coordinates as well. For 1.5.1, I'll replace the current grid with something much better.

--Chris

I have doubled the grid's resolution along its circles (line 7319) :



There's no impact on the frame rate, and the grid is now looking much nicer. It was a bit crude (polygonal) with nSections = 60.