Thanks for the constructive criticism. So let's discuss the issues brought up:
1) This
is a very large amount of space, and to be honest, I was completely unaware of just how much space these stars were taking up. As you have mentioned, this is due to the fact that most of the companion stars are distant bright specks that have the same resolution as any other large star in the vicinity. What I need to do is discuss with Admiral Nelson is just how large the stars in the missions should appear. This way, I can significantly reduce the resolution of companion stars if their representative images are not going to be used to the fullest extent.
2) Just as mentioned above, the resolution of distant companions need to be reduced to increase efficiency.
3) This was the biggest problem associated with the older template that I used to make the stars. As of recently, I have been making new stars that actually take up most of the image. The star itself sits at the center, but the halo of the star is what actually takes up the rest of the image. I had to compromise, because a star with a small or absent halo does not appear luminous at all. Flux alone is difficult to represent as it is. So now everything is scaled according to luminosity, where the final star has a halo that barely touches the edge of the image.
Thanks for taking the time to compose that spreadsheet. I have looked through the stars, and I notice most of these have not received the updates that I previously mentioned. However, I do see some of the new stars on this list. So either I did not to a good job at scaling the halos, or you haven't come across the newer images. For instance, there is a newer version of the Albireo system that was posted in one of the last few pages in this thread, along with a couple of others.
Anyway, the problem that I wanted to avoid is the presence of hard edges. What one might consider pure black is usually an RGB of (4,4,4) or something along those lines. It was my intention to have an RGB of (0,0,0) as close to the edge as possible so the halo of the star blends in well with the background. If too much is cropped from the edges, you will most likely be able to notice the edges of the image in the background. So if you shrunk the canvas size while ensuring that the hard edges are not present, then I'll go back and redo these stars.
4) I'm not too sure I follow you here. I have not played FreeSpace at all to verify how well the stars appear in-game from one mission to another. However, I would at least hope that players will notice the difference between say Altair and Antares A, for instance.
Stars that have similar luminosity types and similar spectral classes will in most cases look exactly the same in these images. This is basically the case with the Sun and α Centauri A, both of which are class G2V stars. The only true difference that one might be able to infer from looking at the images is that α Centauri A has a slightly larger halo to give the impression that it is more luminous than the Sun. So I agree, in cases like this, it is just a manipulation of
one image that can be used to represent stars of similar luminosity and spectral classes. Having two separate images, which are most likely going to be scaled in-game to the point where the difference in halo size is inappreciable, is unnecessary.
If you were referring to stars of different spectral classes and luminosities, then not only is scaling the most likely culprit of this indifference, but the positioning of stars within nebulae diminishes detail, as well. When it comes to the primary star of a particular system, I would hope that the in-game image size is very close to the original so the details of the star can be appreciated. Looking at Altair and α Centauri A for example, the primary stars in-game are significantly smaller than the original 512 X 512. Not only that, but it appears as though α Centauri A is next to a nebula, and these images have a habit of washing out the detail of the stars. Up close with no obstructions, the differences should be clear:
Admiral Nelson, what do you think about changing the size of the primary stars at least 50% of their original size? In my opinion, anything less implies too much loss in detail, and of course a loss in efficiency. We can discuss anything you want, but I think we should begin with 1. the size of primary stars, 2. the size of companion stars based on their orbital distances, 3. the size of the lens flares, 4. the positioning of stars within other celestial bodies. Let me know if you have anything else in mind.
Again, thanks for the criticism. If there are any other concerns about the stars, feel free to address them.