Reducing the return on investment for colonization of Ice worlds and even gas giants would be a good idea.
On the topic of gas giants its fairly clear from observations of some stars that what makes up a gas giant needs some updating from 3rdR.
I'm not sure what you mean by this sentence. Are you referring to Hot Jupiters
http://en.wikipedia.org/wiki/Hot_Jupiter, i.e. Type G planets existing in what Starfire would call a star's rocky zone? I'm aware of Hot Jupiters, though haven't fully decided on whether or how to use them.
And/or are your referring to "Eccentric Jupiters"?
http://en.wikipedia.org/wiki/Eccentric_Jupiter EJ's are Type G planets having very eccentric orbits, and supposedly exist in 7% of all known systems with planetary systems, which makes them more common than Hot Jupiters. However, EJ's are a very problematical type of planet to deal with in Starfire because of their eccentric orbits. Starfire assumes circular orbits. Highly eccentric orbits would be amazingly difficult to represent in Starfire, if one was using orbital movement. Also, an EJ's eccentric orbit would disrupt other planetary orbits, which would require calculating the inner and outermost distances of the orbit, so that the other planetary orbits could be wiped out. All in all, EJ's are just a pain in the butt that I intend to ignore for the sake of sanity.
Back to HJ's... If they exist in fewer than 7% of known systems, then they're not terribly common. That's not really a big deal. The problem for me is that putting an HJ in the rocky zone ought to cause rather more orbital disruption than a Type G out in the Gas zone, since orbits in the rocky zone are rather closer together due to the nature of the Titius-Bode relationship that Starfire uses to produce orbits. This isn't as difficult a problem as the orbits of EJ's, but it might require wiping out something like 1 orbit sunward and 1 orbit outward.
A number of examples of so called 'Super Jovians' some in fairly proximity to the star have been noted along with things like Brown Dwarfs in larger numbers as we get better at detecting them orbiting larger stars. Brown Dwarfs along with Red Dwarfs should be fairly common in space.
There are quite a few little details in these two sentences...
Super-Jovians are basically defined as gas giants possessing a mass greater than Jupiter. Not sure that that fact by itself is particularly interesting, unless we were to start paying attention to different masses for Type G planets for some reason. There's really not much point to saying that a Type G is Mass 1, 2, or 3, unless there's a particular reason... such as higher mass G's being able to hold more moons than lower mass G's. As for Gas Giants being fairly close to their stars, i.e. Hot Jupiters, I've covered that above...
Brown Dwarfs are sort of a mid point between being a very massive gas giant and a very low mass star, that don't really have enough mass to sustain nuclear fusion. As a type of planet, I'm not sure how different they'd really be from a Type G gas giant, other than to call them a Mass 3 or 4 gas giant. Could a BD have a biosphere? Perhaps, we'd be starting to talk about distances measured in tactical hexes rather than LM's. Ultra does happen to have an oddity that represents this situation. But all in all, this is a bit too complex for a standard sysgen process.
As for the commonality of Red Dwarfs (and Brown Dwarfs), this is actually a different sort of question. In real life, yes, Red Dwarfs supposed represent nearly 80% of all stars. However, an underlying assumption of 2e/3e Starfire was that WP's were more likely to be attracted to higher mass stars than lower mass stars, hence the distribution on the star type table. And frankly, I don't think that the use of this assumption violates reality, simply because it is making an assumption about the nature of WP formation that known reality cannot speak to. It doesn't say that there are fewer Red Dwarfs than there should be. It says that WP's are more likely to make connections to stars in proportions that do not match the same proportions that describe the numbers of each spectral class.
However, if I were to change the existing star type table's distribution to the real life distribution, it would almost certainly cause a drop in habitable planets of about 50% or more... thus causing an increase in exploration luck. This is a very touchy subject. Some people might love this, but many others might hate it. Those for whom realism is important would probably enjoy such an adjustment. Solo players might not mind it, since they don't have to worry that much about exploration luck issues. More competitive players might not like it so much though. But then again, more competitive players can already get grumpy when random system generation expresses its randomness is less than favorable ways (i.e. they don't get enough T/ST planets), so an adjustment that would cause a drop in T/ST's by 50% or more would probably not be met with much happiness from them.
As for Brown Dwarfs as "stars" in their own right, well, they seem rather useless, and would really, really complicate system generation. It'd be tricky enough if they were acting as simply large Type G planets. But as system primaries? You'd end up having to do the sysgen process in tac hexes, not LM's. And I'm not at all sure that we could make any assumptions about planetary formation zones, etc. Frankly, BD's as system primaries seem like a headache I'd rather avoid.
BTW, note that just because I may come off in this post as sounding a bit negative about these ideas, don't think that I'm actually entirely negative. I've actually put a fair amount of time and effort reading up on a lot of this stuff over the past year or so, and am open to considering such ideas as long as they can be done in a fairly simple, easy to use manner, and have some value. For example, it's been my intention to include Red and Blue Supergiants as a system anomaly... that is,
properly sized supergiants... that have stellar radii that could be measured in system hexes. (IIRC, the largest known star has a stellar radius that equates to about 6 system hexes.) And since WP's are attracted to higher mass stars, supergiants, though very rare, would almost certainly have considerable numbers of WP's.
Well, that's enough for now...