Currently all the orbits are at 1/10th scale and all the planets are at 1/5th scale.
To quote my side of a direct conversation with Andrew from a few months ago, copying it here in the suggestions.
I would strongly recommend you reconsider the mismatched orbit and radius scale. One of the most important things for an authentic space game, in my view, is accurately presenting the scale of space. You're laying a dangerous precedent in unrealism with this inconsistency. One of the first problems that arises is that you can't simply consider your game world to be a scaled down version of a real system. There is no good rule for scaling orbital periods, transfer times, and delta-velocities with inconsistent scaling between celestial bodies and orbits (I mean there is one, I imagine, that shouldn't be too difficult to figure out, but it's late so I won't right now, and it won't be as pretty as sqrt[rescale_factor]*period).
Another absolutely massive consideration is that of the "Orbit Is Halfway To Anywhere" rule. While that isn't a hard rule, it does point out something important about the relationship between the delta-v a vehicle will need to get into orbit versus the delta-v it will need to get to its destination. And this is something that inconsistently changes from the real world depending upon what you're doing. If you're going to Luna, the rescale skews in the easier direction, since the orbit is closer. However, it becomes slightly harder to go interplanetary, because while the distance between the planets is smaller, the actual velocity of the planets is higher and thus so is the relative velocity. That's why going interplanetary around a red dwarf star is harder. It also just flat-out breaks when you try to do this consistently elsewhere in the system. A conjectural Io analogue would be torn apart by Tydos, being inside its roche limit. Phobos would be nearly inside of Cyleros--and yes, again, inside the roche limit.
Another problem is what happens when you try to imagine what the star is like.
If it's a 1/5th scale analogue of our sun, with the same temperature of 5800 Kelvins and 1/5th the radius, it will have a luminosity L = R^2 * T^4 = (1/25). To find the habitable zone (specifically the orbit where you get the same insolation as Earth) for a given luminosity, take the sqrt of the luminosity. 1/5th. So we see that the habitable zone is at 1/5th of an au. That means that Droo is being baked with 25 times the sunlight and a scorching temperature of 273 fahrenheit. Sure if we cool down the star then we can find a working habitable zone. But cooler stars have lower masses, which means the star is no longer a consistent 1/5th scale analogue of the real sun.
If you ever want to create solar heating or light dimming with distance, this will be relevant. Same with solar panel curves.
I get that the game isn't supposed to be hard-realistic. Concessions have to be made to balance the game. But I genuinely don't think that inconsistently playing around with the physical constants of nature is the right way to go about it. My suggestion to balance the game instead would be either to focus on the rocket hardware or finding a different rescale factor that is balanced correctly with both orbit and radii scaled the same. For instance, if you find rockets are underpowered in consistent 1/5th scale, try consistent 1/6th scale.
I think in the long term this should be an advanced decision that the user can change when they start a new game with a new system. But I think it is absolutely necessary to set a precedent for the physically consistent. The user will barely notice that they'll have to fly a little further or make their rockets a little longer. There's no career mode, there's no reason you can't simply build a larger rocket. But at a fundamental level it is disingenuous to build a physics simulation on false pretenses.
The main reason to be honest, is that it's just wrong. I could try to find more practical reasons why you should keep it consistent, but at the end of the day my main reason for disliking it is just that it's not correct, and it doesn't have to be incorrect. I hope that alone is a good enough reason to reconsider.
(I did receive, back then, a response to the effect of "thanks, we'll consider it" but since I haven't heard anything of it, I'm reposting it in the suggestions box.)
To add on, I honestly feel like the game ought to run at real scale, with a planetary radius of 6,400 or so kilometers and an orbital radius of 1 AU for Droo, and with rocket parts balanced against the real world. It'd help make anything you do in the game more real--like hey, this is a rocket design that would actually work if you built it IRL.