My latest playtesting has convinced me to change the way ai attacks happen in the upcoming GSB campaign. Previously, there was a complex system involving local threat levels, which changed over time based on the strength of your fleet combined with the number of enemy-controlled systems linked to your world by hyperspace warp tunnel thingies.
And that works fine, and its cool, and mostly staying.

But what was happening was that you would conquer a system, push forwards and conquer the next system, and there was still a threat ‘behind the lines’ to recently conquered worlds. That was fine too, but I also coded a little ‘unlikely but possible sneak attack’ system whereby any of your worlds could get attacked at any time. If you didn’t have a big fleet sat there, this would be unopposed, you would lose the system, and maybe now have a gap in your supply lines.

Frankly, in game terms, this is a pain in the exhaust-port. It’s frustrating and annoying to lose a system behind the lines, and it’s wasteful to keep a fleet in every system just in case. The good old ‘pushing-back the frontier’ system is better.

I’ll keep the gradually lowering threat level thing, but ditch the sneak attacks. Once you have conquered a world, and parked a big-ass fleet there for a few turns, you can mvoe on and not fear losing it. It is, after all, a big map to conquer.

On an unrelated note, can whoever codes the cursor stuff at ATI get their shit together please? Multiple monitor setups in windows 7 are basically chaos with an ATI card. Random cursor corruption when swapping monitors, and an invisible cursor if it goes into text carat mode and back again on the secondary montior… These are not new bugs, from what I read, so why are you tweaking drivers to get an extra 1 FPS on starcraft when you should be fixing basic windows functionality? Bah!

When I started designing the campaign, I put in code for repairs to ships, where you could repair individual pieces of damage. An attack on a ship might damage a module, doing 25% damage to it. You could repair just that specific piece of damage, if you so chose.

The idea was that rather than just clicking ‘repair this ship, 1,240 CR’ you could spend less, if funds were tight, and just repair critical modules. I wrote a ton of code to generate slightly silly technobabble descriptions of each piece of damage. All of this is in the campaign and working.

In practice, when testing the campaign I find myself invariably repairing the whole ship, or if funds are tight, just leaving it to the next turn when funds are available. The fancy repair interface is a bit overdone and over-engineered. However, I see no good reason to remove it entirely, it’s kinda fun, and allows micro-managing for those who want to, so it’s staying in.

Something that did change today is the cost of repairs halved. Previously it was the same as construction. If a module cost 400CR and took 25% damage, it cost 100CR to fix. Now it’s 50CR. Why the change?

Well in practice, repairs are reasonably rare. It takes time for ships to retreat, especially cruisers, and in battle, by the time you realise your fleet is screwed, the chances of all of them managing to turn 180 degrees and warp out are quite low. You often lose half the fleet during the retreat, unless you make an instant judgement at the first sign of the enenmy, and call an instant withdrawal. Combined with this, there is is the issue of repair yards. Not every system has them, so the damaged ships need to make a few warp jumps back to the nearest connected repair yard, before they get fixed up. By the time you do this, you might as well build new ships.

So….. Repairs will be cheaper than new build. I think it balances out better this way, and it makes sense in gameplay terms anyway. The real frustration should be logistics, and having retreated and thus given up territory, rather than resenting the cost of repairs. I’m aiming for a ‘company of heroes’ style mechanic, where its sensible to retreat and fight another day, if outgunned.