SEV: Taking me to the edge of sanity

[Raapys]

Yah, that's really the worst part of it: The AI isn't using the long processing time to do something sensible, like figuring out what he should do next. It's used for the actual movement of the ships, since the pathfinding mechanism apparantly isn't optimized at all.

[Romulus68]

Try changing the ship movement speed in the options.

[Captain Kwok]

To clarify, it's not the movement itself but the pathfinding - so it is sensible really. This is due to the line of sight in the game, so it's calculating the path for the ship each hex that it moves. Although I don't think it calculates again in squares that have sensor coverage - but if it does that may be a point of optimization.

[Slick]

There have got to be other optimizations for pathfinding. Numerous other games use pathfinding, oftentimes with more complex pathing/obstacles without a noticable slowdown. SEV, in fact, really has a simple pathing, even with the rules of sight. I'm thinking this is a negative consequence of going from square grids to hex grids. Although I really like the hex grids over squares, somehow, when I first saw the demo, I had a bad feeling that this change would cause a slowdown.

[geoschmo]

Arralen said:
But there's an easy solution to this:
Atm, the AI programming is still following the concepts of the '80 .. but instead of giving the AI only some seconds to "think" between the player hitting the "end turn" button and him eagerly awaiting the results of his orders, the AI(s) should do its(their) turn(s) while the player does his.

I'm not sure really this would make as much of a difference as you would think. I can't say for sure with SE5, but I know with SEIV and simultaneous turn processing games on PBW large games can take a lot of time to process even when there are no AI involved. I think that at least with SEIV the amount of time it takes for the AI to make up it's mind what to do is small compared to the time it takes for the game mechanics to carry out those orders. And a lot of decisions, like combat, can't be made until the game is actually processing because what the AI does is dependant on what the players around it do.

[Slick]

Come to think of it, it appears to me (without testing) that longer paths have much higher processing time. It appears that if you break up a long path into multiple steps using multiple "move to" orders, you can achieve the same thing but now the pathing routine just calculates up to the next "move to" point instead of the destination. Thus this should cut down processing time for pathing. I have been doing this and it seems to improve turn processing time. The bad thing is that this method kinda makes waypoints useless. More testing required to confirm.

[Captain Kwok]

It speeds things up for the player since only 1 movement order a time is taken - but the individual AIs wouldn't be doing this and that's a bigger source of slowness.

[frightlever]

Arralen said:
Fundamental flaw in all those games (as Fyron remarked, it was the same with Civ3, isn't that much better with Civ4 at times, and Dominions3 can be problematic as well with 'big' games):

The games getting more and more complicated - several orders of magnitude, actually, if you want to put down the mechanics into formulas for an AI to use.
And we all want "smart" AIs, don't we? AIs which at least somewhat understand whats going on and can play the game on roughly even ground without cheating too much (able to see through for of war etc)

But does it get the 'power' to do so? Hardly.
Yes, o.c., CPU 'power' has grown enormously.
But power = work * time, or something like that, I was told.
And obviously, there's lots more time needed to reach to 'power levels' we want to see.

But there's an easy solution to this:
Atm, the AI programming is still following the concepts of the '80 .. but instead of giving the AI only some seconds to "think" between the player hitting the "end turn" button and him eagerly awaiting the results of his orders, the AI(s) should do its(their) turn(s) while the player does his.

That shouldn't be too hard to implement, with all those games being TCP/IP-enabled MP games .. just plug the AI(s) in like any other player in a MP game, and you wouldn't even need big changes to the game mechanics to make it work.

And with all those multi-thread, multi-core and multi-CPU equipped PCs out there, I doubt anyone could complain about the AI slowing down his PC or endless "end turn" waits ... .

It's like you're pretending you discovered GalCiv or something. Trust me, we know how GC works but it is a world away from SE. Sheesh.

[Fyron]

Slick said:
I'm thinking this is a negative consequence of going from square grids to hex grids.

Hexes have 6 adjacent locations to analyze, squares have 8. If anything, hexes should reduce the pathfinding calculations by virtue of 25% fewer adjacent locations to consider. Of course, if you optimize it to ignore squares/hexes that are in the opposite direction of the target, you pretty much have 3 versus 3 locations to consider. Sight and larger systems, plus double the movement, seem like they would have more of an effect than hex vs. square.

[Theonlystd]

the sim times are depressingly long


Grab a book,mow your lawn, do the dishes your turn might be done when you return.