Android Save Issues, and Gravity
Hey Folks! Still working through the recent Android issues resulting from patch v1.2.5. Though I did get a little prototyping done, as well.
The save issue looks like it might be a result of the new file path we're using. It looks as if the path we chose works on some devices (such as our own testing devices), but not others (see the app reviews). Furthermore, there was a discrepancy between the save path and the load path, which probably made things worse. Again, some devices resolve both paths equally, others must not.
So one fix appears to be just making those paths equal on save vs. load. But Tiago thinks there might be a more robust way to ask Android what path it prefers so we don't alienate any devices or OS versions. He'll be looking into that tonight and tomorrow.
As for the periodic crashes some users report, this is harder to pin down. Tiago thinks it could be performance related, as the symptoms seem to say as much. But no way to know for sure until we can repro it in-house.
On the prototype front, I spent some more time working on gravity effects in the plotter. I think I've got them working now, after sorting out some issues with celestial object timing, and fixing some typos. What I have now is a reasonable real-time simulation of gravity that tends to break down the faster it runs, or the closer an object gets to things like the sun. Those are both cases where the simulation step size is not granular enough to be smooth, and weird things happen.
In order to help me debug that weirdness, I've started adding a debug vector line to show the direction gravity is pulling and relative strength. I'm hoping this will reveal any strange inflection points or jumps in the smoothness, so I can narrow down where I'm looking for bugs.
Taking a step back, the point here is to make gravity at least reliable enough that the game doesn't seem like fantasy space. Most of the time, players will be thrusting hard enough to make gravity wells negligible. However, if the player attempts to fly too close to Jupiter or the Sun, they may get stuck and pulled in. Or their route may have to be multi-stage to avoid passing too close to a grav well. Or maybe they're just out of gas, and hoping the pull will bring them across someone's path (or overcome momentum carrying them out of the system).
Basically, enough realism to support drama, but not enough to simulate Cassini's trajectory.