[nbos] [AS] Star density/frequency"Andrew Seel" Mon May 28th, 2007
Hello everyone! I'm Andreus, new on the mailing list. I'm an avid user
of AstroSynthesis 2.0 and have used it for months now. I just had a
little gripe with the system I wanted to talk about, and possibly
discuss a solution for.
Currently, when generating new sector data, the only way to control
the density of your starfields is "Star Frequency", which, in my
experience, tends to be highly arbitrary. I like using clustered star
distribution, as this tends to create interesting sectors with a lot
of variance in cluster shape, which makes for fun storytelling.
The "Star Frequency" slider seems to be broken, or at best, fairly
random. Take for example the 500 ly sphere sector I just spend the
entire night generating - it has over 400,000 systems, whereas a
500-ly sphere area around Earth only has roughly 42,000. We're looking
at an order of magnitude more stars than one would expect, and I
generated that system on the lowest setting. The sector was way too
big to be anywhere near manageable - it was 3.15 gb (!!), and even
trying to look at it in "Show hospitable systems only" lagged quite
badly. This was far from ideal.
Testing of the "Star Frequency" slider on 20/20/20 cubes produces star
frequency and density which seems to be way out of whack with known
models of space. The one I just generated for a test has 4,095 systems
on "even" distribution at the highest setting - I imagine this would
be something like what would happen close to the galactic core, but
planets probably aren't habitable there, so we would really venture
that close. "Clustered" on highest frequency produces a cube with 156
systems, still a bit high, but more acceptable.
I thought of a solution to fix this - I'm not sure how much work it
would take, but I was thinking you could change the "Star Frequency"
slider into two different ones, "Star Frequency/Amount" and "Star
"Star Frequency/Amount" would dictate the general amount of stars one
would expect to find in the sector, and thus the general amount that
whatever generation algorithm the program uses would be allowed to
create. If possible, I'd like this to have some setting which is close
to the star distribution we'd get around Sol.
"Star Density" would dictate the general distance one would find
between stars (which, depending on your star distribution template,
would still vary widely). This would further modify the star frequency
by only allowing a certain area of space to contain a certain number
of stars. Again, I'd like this to have some setting that's close to
what we'd find around Sol.
Failing that, does anyone have any tips that I could use to make
500-ly spheres that don't have an unmanageably large number of stars?
Nbossoftware mailing list