This is a known issue, it's in how we are calculating what is to be shown on the screen at a given time. The current method is simple but quite inefficient. There are a number of optimizations that can be made there, but we haven't had time to get to those what with everything else. For us, the important things to test have been what performance we can get when the known-inefficient part isn't running (aka panning), and then the interturn stuff and actual gameplay bits.
The slowness of the far zoom is still a bit of a mystery as to exactly where the performance woes are, but it's something we will be profiling for later. There are also some issues when a bunch of saucers drop in and many events kick off at once, among a few other things.
The main request I have is that anytime you have a situation that you think is egregious, or a good test case for us to look at, then please upload that to mantis with steps for us to repro what you are seeing performance-wise. We can always use more test savegames, those are always welcome, because it's not always the same thing in all of them.
With the international incidents work out of the way for Keith, he's pretty much on fulltime bugfixing duty at the moment, and then is going to be having performance and profiling work slotted into that increasingly. I think that 1.0 should be pretty much night and day from this, but definitely getting reports and saves of particularly egregious spots of trouble is a big part of that. Thanks!
