Fighters & Carriers

This is for directed discussions on immediate questions of game design. Only moderators can create new threads.
Message
Author
User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Fighters & Carriers

#61 Post by MatGB »

All of this is possibly true (and for what it's worth I just stopped a game where I had clear superiority against all AIs in part as they had no chance of even seeing my stealth strike "subs" until I'd wiped out all their fleets), however...

In naval combat today, fighters and carrier groups outperform battleships because, on earth, in a water based navy, fighters are better. In space, that needn't necessarily be so, the objective is to balance them so that there's a rock/paper/scissors effect where a player that overly relies on fighters will, for example, suffer against a player that's invested heavily in, for example, unshielded cheap stealth strike ships (I love these), but that player suffers against a Big Battleship player (because a stealth strike needs to destroy a lot in the first round), who is vulnerable to fighters.

Whether we can get that balance right is a challenge, but the current balance favours big battleships over stealth strike, shields are too good for the most part, etc. (I am certain that the tactics I used to wipe the AIs in my last game would fail against even a moderately experienced player)

I hope we can get it balanced in such a way that mixed fleets are going to be the ideal (we are, separately, working on a replacement for the existing fleet upkeep system that costs different hulls separately meaning small hulls are worth using late game) but that a strong concentration in once aspect can give surprise or against-the-odds victories.

"Torpedo boats" are a strange case for FO as a direct analogy because they're normally thought of as coastal short ranged light vessels that are cheap and throwaway, we've deliberately, up until now, kept system defences as being planetary meters not individual ships, we may want to address that moving forward.

Because we have structure meters, it's a lot harder for a single plane to sink a battleship—my grandfathers boat nearly sunk multiple times from single individual lucky hits, that can't happen with our system, etc.

Getting the balance right to make it fun is going to be a challenge. I'm looking forward to it frankly.
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

UncleFred
Krill Swarm
Posts: 14
Joined: Tue Mar 15, 2016 5:51 pm

Re: Fighters & Carriers

#62 Post by UncleFred »

It's not about sea battles vs space battles. It's about the ability of some number of small 1-3 man vessels to destroy a capital ship. But beyond that it's about a complete change in thinking about offense and defense in ship to ship combat.

Up until now combat in freeorion has been much much closer to wooden men of war than high technology combat, and that is way beyond okay. Introducing a completely different combat mechanism changes all of that pretty much forever.

So consider that one way to deal with this is to allow a targeting progression. Where I can select a fleet and tell it kill carriers first, then battle wagons, then whatever is left.

Consider how aircraft are deployed. Not willy nilly but the complement of one or more carriers, are directed at killing one or more specific ships.

Consider something else. The Missouri class battleships became anti-aircraft weapons platforms with over 100 anti-aircraft gun emplacements. Yes they retained their 16" guns and were used effectively to support invasions, but when air attacks happened they were flak platforms.

If you are going to do this you need to modify random targeting.

I've had a number of thoughts on this.

One possibility is that for every class of ship, you allow them to have a progressive targeting list of opposing ships by class. Another is that you allow one fleet to target another. As it stands now it appears that the AI doesn't merge ship types.

In the larger sense it really doesn't matter. Once you introduce a different combat mechanic, you'll be forced to allow more nuanced targeting.

My recommendation is that you don't introduce this kind of change until you also rework the combat system and a mechanism to allow target prioritization.

Morlic
AI Contributor
Posts: 296
Joined: Tue Feb 17, 2015 11:54 am

Re: Fighters & Carriers

#63 Post by Morlic »

I implemented some basic AI-adaptations for the fighter mechanics: https://github.com/freeorion/freeorion/pull/576

I'd expect the changes to be stable enough to test/balance new fighter parts with it.
When doing so, please provide feedback on weird behaviour of the AI or actual errors (you should probably open a new thread in Play-Testing for that or post on Github)
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Fighters & Carriers

#64 Post by MatGB »

I'm 3/4 through a game at the moment, and I need to finish off some work to commit, then this'll be my top priority in terms of balance and tech introduction.

Curious, what is unlocked at game start, and what parts need to have techs to unlock them so I can get the basics in place (I haven't looked at the branch since testing the basics as the AI was so obviously unable to cope with it).
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Fighters & Carriers

#65 Post by MatGB »

OK, first bug I've not seen, might be unrelated, I've compiled fresh a copy of Morlic's AI-fighters fork that's been rebased to master, due to weird issues elsewhere I had to PR it into my repo where it's currently sat in the Master branch. freeorion

I can't create a new save. Whenever I open the save dialogue it gives me a filename without the .sav suffix and won't let me save, even if I add the .sav manually. However, I can select a pre-existing save and overwrite that, which is what I ended up doing with my first testgame. Observing it here as I'm compiled expressly for this thread.

Anyway, to actual topic: there appears to be an occasional glitch with combat initiation, if a stealthed carrier with no weapons is in a system where the enemy cannot see anything to attack, then it doesn't always launch fighters (although the enemy does, then recovers them again). Given I'm testing using organic stealth ships because, well, I wanted to see how that works, it's proven to be a slight annoyance.

The damage display for a ship with mixed weapons (ie fighter bays and lasers) isn't at all clear and at times misleading.

We need a "split ships with no fighters left from the fleet" to go alongside the split damaged and split unfueled. Not replenishing during a combat and the new supply system makes fighters harder to manage, in a good way I think, but being able to quickly remove the ships that need to go get new fighters is important.

I'm playing around initially with the fighter hanger stats, then I'll look into adding various techs/prerequisites.

I'm thinking it would probably be best if we had a dedicated fighters fork for a bit so the AI team and the balance/scripting side can be worked on in tandem till we're happy with it, if no objections I'll set it up tomorrow (partially because there's something weird going on between my git install and Morlic's github fork and partially because it's what git's for anyway)
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Fighters & Carriers

#66 Post by Vezzra »

MatGB wrote:I'm thinking it would probably be best if we had a dedicated fighters fork for a bit so the AI team and the balance/scripting side can be worked on in tandem till we're happy with it, if no objections I'll set it up tomorrow
Sounds very reasonable, as far as I'm concerned, go ahead.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Fighters & Carriers

#67 Post by Geoff the Medio »

MatGB wrote:I can't create a new save. Whenever I open the save dialogue it gives me a filename without the .sav suffix and won't let me save, even if I add the .sav manually. However, I can select a pre-existing save and overwrite that, which is what I ended up doing with my first testgame. Observing it here as I'm compiled expressly for this thread.
Probably related to https://github.com/freeorion/freeorion/ ... 21d4a4c62e

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Fighters & Carriers

#68 Post by MatGB »

Vezzra wrote:
MatGB wrote:I'm thinking it would probably be best if we had a dedicated fighters fork for a bit so the AI team and the balance/scripting side can be worked on in tandem till we're happy with it, if no objections I'll set it up tomorrow
Sounds very reasonable, as far as I'm concerned, go ahead.
Fighters branch by MatGB · Pull Request #587 · freeorion/freeorion Done, branch needs rebasing from master with a few updates but there's aminor clash in an AI script I don't want to risk mixing up.

Have to say I'm having a lot of fun playing with this, the AI seems to be making some interesting designs and combat is an entirely different beast with all the fighters flying around, stealth strikes are far riskier when you can't take out the fighters being launched in the first round. Fun fun.

Lot of work to get this properly balanced and the AI updated for that balance I suspect, worth it though.
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

User avatar
Kassiopeija
Dyson Forest
Posts: 212
Joined: Wed Jul 16, 2014 6:14 pm
Location: Black Forest

Re: Fighters & Carriers

#69 Post by Kassiopeija »

I can live with the random targetting (even tough in the presence of fighters it seems highly idiotic to not simply sink their carrier...) but what is really mind-boggling is the "let's still shoot at stuff already destroyed" mechanism....

And esp. in the presence of expensive modules such as Spinal Anti-Matter Cannon. Perhaps that part could see a "do not target Fighters" line of code as well...?

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Fighters & Carriers

#70 Post by Geoff the Medio »

Kassiopeija wrote:...what is really mind-boggling is the "let's still shoot at stuff already destroyed" mechanism....
It's not destroyed until the end of the combat round. Everything fires simultaneously during a round.

User avatar
Kassiopeija
Dyson Forest
Posts: 212
Joined: Wed Jul 16, 2014 6:14 pm
Location: Black Forest

Re: Fighters & Carriers

#71 Post by Kassiopeija »

But a Fighter only needs to take a single hit, so it's easy to deduce beforeahead that you only need 1 weapon fireing at it in order to destroy it. UncleFred does have some point when stateing that the FO shipcombat commonly reflects historic wooden shipbattles. Such flaws shouldn't happen anymore once you got into a technological age...

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Fighters & Carriers

#72 Post by MatGB »

You only need one weapon to actually hit.

I like that, currently, all firing is resolved as if it's simultaneous. I would like for some kind of target priority/leader system to allow some control over what's targetted, but part of the challenge of balancing this is going to be getting the costs and effects worked out, I like the idea that a swarm of fighters in between ships absorbs more fire than it needs to, that appeals to my space opera watching brain.
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

emrys
Creative Contributor
Posts: 226
Joined: Fri Oct 24, 2003 3:44 pm

Re: Fighters & Carriers

#73 Post by emrys »

Thinking about the problem, the regret comes at one end of the spectrum from wasting a hit from a big gun on a trivial target, or at the other from firing small guns pointlessly at something they can't even scratch when there are available targets they would have killed.

Perhaps some kind of over/underkill assessment could be added?

i.e. you generally want to fire at a target that both applies as much of your potential damage to the target as possible without wasting much (avoid overkill), and ideally that has as a good chance of destroying the target (avoid underkill, dead targets don't shoot back). Perhaps targets for each shot could be selected ,rather than the pure "first out of a hat" system I gather is currently used, by picking a set of some number of potential targets at random, working out a rough figure for each option that assessed these factors, and firing at the one with the best value.

Tweaking the number of targets assessed, or perhaps the balance between under or overkill avoidance, could even be a feature - either between ship/weapon types (fighters don't get much choice, they pretty much have to hit the first thing they see because they can't afford to wait for a better option, big ships are more survivable so they can, and their big weapons have more individual effect, so they care more about picking good targets for them) or as a technological option (better battle computers/inter fleet co-ordination techs/gear = more targets considered = better average battle performance)

I'd guess with numbers of targets assessed before each shot ranging between about 1 (i.e. the current system) at the low end, 2-4 typical up to maybe 6-10 at the extreme upper end for really big guns on big ships with lots of targeting boost, Fighter/chaff ship swarms would still have a good screening effect , but key ships and guns would be more likely to "see through" the screen to find a "reasonable but not as good as they might have found" target (and hence players would be less likely to be highly annoyed by having *really* bad luck and all their big guns being randomly stupidly wasted)

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Fighters & Carriers

#74 Post by Vezzra »

Oh hi, another veteran from FO's mythic ages stopping by! Always nice to see that some of you guys still silently lurk in the background (or at least take a look at us now and then). :D

Regarding your post: We had a lot of discussions in the past about the (obvious) shortcomings of our current combat system. And as has been stated before: yes, it has quite serious issues, but the current system has always been intended to be a stop-gap solution, until we can come up with something better. So, the concerns you raise will be addressed, but I don't think it's reasonable to just tinker with the current system to make it a bit better, but still have only a stop-gap solution that will eventually thrown out. Implementing something like you outlined won't be trivial, which (if only as a enhanced temporary thing) would be a waste of time and efforts IMO.

Once we decide to go back in and finally do something about the combat system, it should be in an attempt to come up with an approach that has a decent chance of making it into v1.0. There have been ideas/concepts suggested which I think look quite promising, at least I think they point into the right direction.

You can find the thread on that proposal here: http://freeorion.org/forum/viewtopic.php?f=4&t=9732 (the interesting stuff starts several posts down).

But as far as a stop-gap solution is concerned, the current systems works well enough, despite all its annoyances and issues (IMO).

User avatar
Sloth
Content Scripter
Posts: 685
Joined: Sat Mar 17, 2007 12:28 am

Re: Fighters & Carriers

#75 Post by Sloth »

I'm really waiting to test the fighters feature, but i have been too lazy to merge the branch and resolve the conflicts.

I already have some ideas for using fighters to create some content (especially species traits).

Isn't that the mayor feature of the next release? Is it just the conflicts that prevents it from being merged?
All released under the GNU GPL 2.0 and Creative Commons Attribution-ShareAlike 3.0 licences.

Post Reply