

Moderator: Moderators for English X Forum
Acceleration is a function of speed, so yes, speed definitely does matter. Going from 200m/s to 0 in three seconds is going to feel a lot worse than going from 2m/s to 0 in three seconds.
Yes. This is the key point. A series should build on what went before but every X game goes back to square one on basic features and yes, the logic is independent of the specific platform.ScandyNav wrote: ↑Wed, 12. Dec 18, 15:25It's their problems that they can't adapt code to new game and rewrite it from zero each time. And each time same behaviors like: formation follow, auto trade, pathfinding, auto exploration become broken, useless, dumbed down. Yes, code language may be different, but not algorithms! Why they forgot the algorithms of all that stuff?!Vandragorax wrote: ↑Wed, 12. Dec 18, 14:48 Please, stop comparing a game with TEN YEARS of additional development onto it, with X4 which has only just been released.
Algorithms are easy - there are books on it, they are well defined, already been used. They have them. What is hard is implementation of those algorithms. It has to be implemented to take into account current code base. It has to be implemented with satisfying performance (hardest part usually). Usually it has to be done in a way which will not make further changes near impossible or prohibitively time consuming...SteveMill wrote: ↑Thu, 13. Dec 18, 09:26Yes. This is the key point. A series should build on what went before but every X game goes back to square one on basic features and yes, the logic is independent of the specific platform.ScandyNav wrote: ↑Wed, 12. Dec 18, 15:25It's their problems that they can't adapt code to new game and rewrite it from zero each time. And each time same behaviors like: formation follow, auto trade, pathfinding, auto exploration become broken, useless, dumbed down. Yes, code language may be different, but not algorithms! Why they forgot the algorithms of all that stuff?!Vandragorax wrote: ↑Wed, 12. Dec 18, 14:48 Please, stop comparing a game with TEN YEARS of additional development onto it, with X4 which has only just been released.
At a very minimum things like turret tracking, autopilot and the like, which are always broken on release, should have the crap tested out of them before release.
Sure. But I’d bet real money none of that was done here.Arghan wrote: ↑Thu, 13. Dec 18, 11:37Algorithms are easy - there are books on it, they are well defined, already been used. They have them. What is hard is implementation of those algorithms. It has to be implemented to take into account current code base. It has to be implemented with satisfying performance (hardest part usually). Usually it has to be done in a way which will not make further changes near impossible or prohibitively time consuming...SteveMill wrote: ↑Thu, 13. Dec 18, 09:26Yes. This is the key point. A series should build on what went before but every X game goes back to square one on basic features and yes, the logic is independent of the specific platform.ScandyNav wrote: ↑Wed, 12. Dec 18, 15:25 It's their problems that they can't adapt code to new game and rewrite it from zero each time. And each time same behaviors like: formation follow, auto trade, pathfinding, auto exploration become broken, useless, dumbed down. Yes, code language may be different, but not algorithms! Why they forgot the algorithms of all that stuff?!
At a very minimum things like turret tracking, autopilot and the like, which are always broken on release, should have the crap tested out of them before release.
Then there is one more problem. Balancing them.
Lets take turrets. They cannot make them too good or too bad. Too good is imho easier to implement. If turrets would always hit, hit harder and ship be able to roll to bear most turrets on enemy (all relatively easy to implement), fighters would either be obsolete, or would require inhuman skills to pull attacks off (so close to no player could execute them). So they need to be nerfed enough to provide a challenge and not be too weak or prohibitively OP. So algorithms must be adjusted, taking into account assumed player skills, how player are expected to fly, etc... And because each game has different ships and flight physics, you cannot copy-past past solutions. You need to eyeball it for that particular game. It become another hard problem not only connected with purely algorithmic skills.