What would help me more is a 'performance test', of sorts, that does a quick check of CPU and recommends a ship cap/#players. Occasionally I host a game and one of the clients gets 10 FPS on normal ship caps. Admittedly, I should know better and just play co-op on low or ultra-low when dealing with two or more new faces. And the whole #of players scales things up as well.
So nothing I cannot do with proper screening and note-taking, but I'd be curious to see some kind of performance rating system (with clients being the focus, not the host).
Something like
A: 50,000 ships, 30 FPS or greater which seems around smooth play
B: 25,000 ships, 30 FPS or greater
C: 10,000 ships, 30 FPS or greater
or:
A: 5000 ship system battles
etc. I don't know which metric exactly would be most useful, but I'm thinking...
Every 500 ships in a system/5000 ships in the galaxy the player's computer can handle counts for 1 point.
That is, if the player can handle a game with 20,000 ships in the galaxy (2,000 in system, whatever is more important), then they get a score of 4. If the player can handle 50,000, then they get a score of 10. If they can handle 100,000, then a score of 20.
just kicking around the idea, but really that's because I somehow organized a newbie training game that ended up with over 400 AIP (without fallen spire or superweapons even), and that's more of my fault for letting us stick with a fairly defensive pair of AIs, and not reigning in some aggressive tendencies that showed up.
Honestly, the original plan was to ride closer to 200 AIP, and if we hadn't conquered over a quarter of the universe, we would have avoided most of the lag, and the 10,000+ patrols. Also, I really should ban Neinzul Enclave Starships in co-op, they're just a Bad Idea (TM). I love 'em, mind you, it's just they only hurt performance more when you already have enough ships that you can't fit everything into a full cap of assault transports - and that's before the drones spawn and after you've already left all your MK1s behind.
Still, at least my hosting machine could keep the frame rates up, but we still had the 'waiting for player' that wasn't helped by the fact that the other two players were streaming the game at the same time.