Steamrollering blitzkrieg—we need more pyrrhic victories

For what's not in 'Top Priority Game Design'. Post your ideas, visions, suggestions for the game, rules, modifications, etc.
Post Reply
Message
Author
User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Steamrollering blitzkrieg—we need more pyrrhic victories

#1 Post by MatGB » Fri Mar 07, 2014 9:09 pm

I'm nearing the end of another game, 350 systems, cluster map with a single annoying intersection. None of that really matters.

The only reason I'm not still using the shios that survived my conquest of my first opposing AI 150 turns ago is because I scrapped them to make building nanobots with death rays cheaper.

I've moved into an AI that dominates two clusters, and I'm steamrollering them. I withdrew from the first big fight to let reinforcements come up as their biggest fleet was scary, but when the reinforcements arrived I sat on it.

I outmatched them, but not by a huge amount. I won without a casualty having taken out 20+ of their ships. Take the homeworld it was defending (the intersection was full of empty systems, annoying), and they all get repaired by the drydock. It has other fleets coming up (the AI had conquered a different, Eaxaw, AI which is building most of its ships), and they're good fleets. If I'd taken casualties in the first fight I'd be in trouble. As it is, I might as well stop producing new ships as the fleets I have in theatre will take the AI out.

I know the combat system was meant as a placeholder until the player controlled rendered combat system came in, but that's not going to happen anytime soon unless a new dev steps up, right?

So the game, in the meantime, needs a combat system that doesn't give the victor so many spoils the game ceases to be a challenge once you've won a single big fight per AI. That means the winner needs to take losses.

Amongst other things, even if we do get rendered combat at some point, I doubt you'll want every fight to be fully fought per turn, having an 'auto' option that gives results that makes sense is good long term, right?

Now, I can't code any of this, but I can work out some mechanics that should be easy to code into the existing system (tabletop game design being my background after all). Without losses in victory, the blitxkrieg becomes far too easy, other changes being discussed elsewhere are helping tone it down, but fleet combat favours the better force far too much ATM and that's a problem.

ETA
What I'd like to see is the relative strength of the forces proportionate to the strength of opposing fleets squared, which is roughly what happens in ranged armed combats anyway, so if two sets of fleets of equal design go against each other the larger fleet should take some losses but mostly come out unscathed, 10 ships vs 5 ships should see 2 or 3 of the larger fleet die (10 becomes 100, 5 becomes 25, ergo 2.5 casualties), which would also make throwing away smaller fleets costly but actually valid, they would do some damage, as things stand you need to just run if a larger fleet is bearing down on you, rearguard actions, etc are pointless.
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
Geoff the Medio
Programming, Design, Admin
Posts: 12249
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#2 Post by Geoff the Medio » Fri Mar 07, 2014 10:42 pm

MatGB wrote:What I'd like to see is the relative strength of the forces proportionate to the strength of opposing fleets squared...
I either don't understand, or this makes no sense. Square the relative strengths of fleets will make the disparities in strengths larger, not give the smaller fleet a relative advantage as you seem to want.

I also have no idea what it would mean to "square" the "strength of fleets" for combat resolution purposes...

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

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#3 Post by MatGB » Fri Mar 07, 2014 11:04 pm

It's a half remembered phrase from my days as a history student (Archer Jones, The Art Of War In The Western World, am trying to buy myself a copy ATM anyway). And yeah, it wasn't clear, but...

Currently, it's way more than force squared, if you outmatch the opposing side you win with almost no casualties. That needs to be reduced, but a larger force should be disproportionately stronger than a smaller force. If you put my 5 vs 10 in the current system, the 10 will take a small amount of damage (which damage control techs will fix and we're discussing elsewhere), but will almost certainly survive without losing a ship whereas the 5 will die.

Since I learned of the rule, I've checked it out in a lot of the games I've played and in actual action reports, and it's broadly true. 5 tabletop space marines vs 10 in similar circumstances will kill 2 or 3, 10 longbowmen in a real battle would take out 5 but lose 2 or 3, etc. That's about right.

Note, it's broadly, it's obviously impossible to ensure it's always the case with the massive variety of designs possible, so it shouldn't be a rule that it always applies, but we do have 3 combat metrics (damage, structure, shields), comparing them before a fight should give you a rought idea of parity, etc.

The smaller fleet should be disadvantaged. But nowhere near as disadvantaged as it currently is, winners should take some casualties. At a very basic level, having ships that're already damaged made priority targets would make a massive change, but I've no idea how hard to programme it it would be.
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
Ta'Lon
Space Squid
Posts: 66
Joined: Sat Feb 01, 2014 5:03 pm

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#4 Post by Ta'Lon » Sat Mar 08, 2014 1:14 am

While I do not fully grasp how combat is handled currently (I'm assuming that values are assessed against each other, without any fancy mechanics for maneuver and such), if I had to guess why this happens, I'd point to one shot kills. I.E. if you have a 36 attack strength and the defender only has 15 structure, defender dies on round 1, delivering whatever damage it can. This gets exascerbated with shields, as the first 4/7/whatever damage goes away immediately.

Also, if the 36 attack is spread out amongst multiple targets in the same turn (say kill 2 of the 15 structure ships, with 6 damage to a third), well then the one shot kill thing is even more powerful.

I've made the mistake a few times now of underestimating the value of shielding. It almost seems like the 7 defense is per weapon and not per full attack, based on the large attack forces I've sent against a single ship a few times. In one case, as I remember, said asteroid ship with the 7 shielding survived essentially unscratched against 2-3x it's own structure rating, and double it's own attack strength. I had 4 laser 1's on the attacking robotic hulls (5 of them) in this case, versus a single large asteroid.

If combat was handled with the assumption of more combat rounds per turn, with either lower attack values or significantly greater structure, so that moderately sized vessels could survive the first volley, this would give the disadvantaged side a chance to score more damage over multiple rounds, instead of just vaporizing in the first round.

I.E. in the current model, a 27 attack versus 15 structure means the 15 structure ship will die quickly, but 27 attack versus 30 structure should allow the defending vessel two return volleys before it dies intead of one. I'm good with a full resolution in one turn in most instances (I've seen some opposing ship stacks last a couple of rounds against each other in some cases), but the disadvantaged side should be able to do better against it's attackers.

Shielding I'm more reluctant r.e. recommending changes, as for the most part I think it works pretty well. That being said, if it is per weapon system and not per volley, well that sounds a little OTT to me.

Another solution would be to allow the defenders to fire first, with only the surviving attackers being able to then return fire, but this might give the defenders too much of an advantage, particularly should planetary batteries also come into play. Of course, if a kill is 2-3x as hard to get in one battle round, then that might not be as bad.

This could be nuanced by having one defender fire, then one attacker, then the next defender, next attacker, etc. alternating until all ships on each side have fired. Once one side's ships have all fired, then the remaining vessels on the other side all fire in succession until everyone has fired.

As for combat speed, which as I understand isn't doing much currently, well you could have a hit formula for each weapon system each round, with stealth and speed reducing the chance of a hit each round. So, say 100% base chance to hit, with a (combat speed/2) deduction to accuracy, with the difference between detection and stealth also applied in some fashion (as a multiplier to modified accuracy, not a direct deduction). So that, say a 100% stealth rating after the sensor deduction reduces accuracy by 50% or 75%. Negative Stealth ratings would make ships easier to hit...

Another thing you could do to make combat a little more nuanced would be to abstract ranging. I.E. combat starts at long range, which reduces weapon strength by say a factor of 4. Starting range at long is assumed to be 100. If both sides are closing on each other, then subtract each side's combat speed from 100. 70 is medium range (weapon strength divided by 2), 40 is close (full weapon strength). If one side is chasing the other, then the difference in speed is added/subtracted from the 100 figure. So a faster fleet blitzing past/running away from a slower fleet will only endure one volley at long range, otherwise the chasing fleet will slowly close the range. This could be handled solely via formulas, without a complex combat model.

You could also go with a 50% reduction to attack strength at long, and a 25% reduction at medium. If you are really bored, you could reduce base accuracy by 25% at long, and increase it by 25% at short.

If multiple stacks are involved, then the speed of each stack would be used for comparison. So a fast attack stack may quickly close the range in one round, while the slower heavies take more than one round to close to short range. Or, if they have to 'cross' a system, they need to close from 100 to 0 range, then open the range back up to 100 (in the other direction) to get past the defenders. Or maybe just close to 50 then increase to 100 again as they 'flank past' the slower defenders.

With the above assumptions (more structure, ranging), I'd suggest something like 5-8 combat rounds in a turn, with longer combats taking multiple turns to resolve (similar to what I'm seeing now).

Also, with the suggested doubling (or whatever) of structure values, I'd recommend that the current 100% repair between combats be curtailed a bit, to say 50% repair between bouts, with the slower rates being adjusted in some manner (not necessarily reduced, perhaps increased based on how effective you want them to be). This would make damage a little more meaningful. Ships completely shrugging off damage between bouts just exascerbates the steamroller situation.
Conquering the galaxy, one planet at a time...

Any artwork that I submit for use in the graphics forum is submitted under the CC-BY-SA 3.0 License.

User avatar
Dilvish
AI Lead, Programmer
Posts: 4689
Joined: Sat Sep 22, 2012 6:25 pm

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#5 Post by Dilvish » Sat Mar 08, 2014 1:35 am

MatGB wrote:The smaller fleet should be disadvantaged. But nowhere near as disadvantaged as it currently is, winners should take some casualties. At a very basic level, having ships that're already damaged made priority targets would make a massive change, but I've no idea how hard to programme it it would be.
I agree this is a significant detraction to the current system, and think the necessary programming wouldn't be prohibitive, if we could just first agree that we want to do *something*, and then home in on what.

The last discussion I recall in this general area was focused a bit more specifically on combat 'streakiness'. I don't think anything significant resulted from that discussion, and the streakiness issue really is more of an issue for closely matched fleets; the potential solutions discussed there seem unlikely to me to significantly improve the problem you're discussing here. While I was writing this post Ta'Lon also posted; from my quick read of his post it seems those suggestions would mostly help regarding streakiness, but not help so much with the issue of how easy it is for a stronger force to win combat without any casualties at all.

I think the only thing that could really improve this is some type of more intelligent targeting (right now it's about as unintelligently random as possible). I have a vague recollection of some type of targeting intelligence having been discussed before, but couldn't find a thread.

One option is to leave things as they are.

Another option is to simply add a fixed degree of more involved targeting to the standard combat auto-resolution. One possibility for this could be making sure a ship selects its target from the available targets that it is actually able to damage (i.e. penetrate shield DR). Another possibility could be making ships of a fleet more likely (not necessarily guaranteed) to attack the same target.

A more ambitious option, which I would prefer, is to have this new targeting influenced by a new 'targeting intelligence' ship meter (or fleet meter), and possibly having multiple tiers/approaches gated by the presence of different targeting computers. Either of these would allow for research to have significance in this aspect of combat.

Another yet more ambitious possibility that just came to mind as I'm writing is to allow combat targeting to be controlled by a python script and extra process, just like we've started doing for universe creation. That would would free up the combat engine and FO executable in general from needing any targeting details; different targeting options would be part of Content and could be gated by tech. A new class of 'targeting computer' ship part could control which targeting algorithm was used for a given ship, or perhaps that could simply be controlled at an Empire level. The default would be the current random selection.
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
eleazar
Design & Graphics Lead Emeritus
Posts: 3858
Joined: Sat Sep 23, 2006 7:09 pm
Location: USA — midwest

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#6 Post by eleazar » Sat Mar 08, 2014 3:45 am

I agree that (based on previous play-throughs) victories tend too often to be with minimal or no losses.

Maybe it could be fixed simply by allowing multiple ships to fire "at once". IIRC currently a ship is randomly chosen to go first, it shoots, and if it does enough damage it's target is removed from battle before it gets a chance to do anything. What if multiple ships on both sides got a chance to fire, before the weapons hit and did damage/destruction? That would reduce the current strong snowballing effect, and tend to exact higher casualties on both sides.

User avatar
Ta'Lon
Space Squid
Posts: 66
Joined: Sat Feb 01, 2014 5:03 pm

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#7 Post by Ta'Lon » Sat Mar 08, 2014 4:08 am

OK, so I looked over the current combat system code.

If I'm understanding it correctly, every weapon on a ship attacks separately, against a random target, in turn. Also, the next ship that gets to attack is determined randomly, not in any particular order. One ship or planet attacks each round, with the total number of rounds being equal to 3x the total number valid attackers. Planets have an attack value equal to their defense meter value, which is why planets can be so scary - they 'mass' their entire attack strength against one target, whereas ships are spreading their weapons fire randomly.

There is currently NO variable in this combat method that 'counts' whether or not a ship has fired already, versus the other ships, so a given ship can be picked multiple times in succession as the next attacker, if the random picker dictates so.

This also means that in multi-ship battles, a ship will choose it's targets randomly for each weapon it carries; i.e. it does not concentrate fire on one target until it is destroyed, although it's random choice of targets may favor one target over another. It also does not look at defense strength in this random assignment.

This explains the wildly different results I've seen in battles, and would seem to confirm what I suggested above, i.e. that each weapon attack is compared against shield strength separately, not all weapons cumulatively.

Probably a good thing that shields aren't allowed to stack in this case.


So, if I was going to suggest any changes for now, that would be to insert a couple of algorithms into the current combat code that do the following:

1) After a ship selects a target, it fires it's first weapon. If the target survives the attack, it fires another weapon at the same target, continuing to select the same target until it runs out of weapons or it's target is destroyed. A new target is selected randomly only after the currently selected target is destroyed. You could also introduce a percentage chance that the next weapon system would select a different target instead of the same target, to make things a little less predictable; i.e. say a 75% chance of selecting the same target if it is still alive instead of automatically.

2) If the target of a ship has a shield strength that exceeds the weapon strength (i.e. won't cause damage), a new target is randomly selected. This process is repeated a limited number of times (5 sounds good), with the ship firing on the fifth selected target regardless, if the first four randomly selected targets have higher shield strengths than the current weapon strength (to prevent an endless loop situation). Yes this could mean that the same target is selected and rejected multiple times, up to the fifth target pick. This reflects the chaotic nature of combat, i.e. more suitable targets just weren't near that particular attacking ship at that point in time.


With the first suggested change, I do think that combat results should be a little more predictable, i.e. ships will concentrate their weapons fire instead of spreading it out, hence increasing the chance of them leaving significant amounts of damage behind on their selected targets before they are wiped out.

The second change might be a little more tricky to implement, as a 'shield strength' check will need to be introduced.


You could also just 'lump' a ship's attack strength into one single attack variable, instead of resolving each weapon separately. This 'masses' fire, but does not allow attacks against multiple weak targets in a given round, if you have a larger number of weapon systems. It also reduces the effectiveness of shields, which might call for an increase in shield strength values, which would make attacks by ships with only one or two weapons even less effective/completely useless. The game already penalizes a player enough for building multiple small ships already, no sense making small ships even less effective.

If Shield Strength were reduced by say 10%-20% every time a Shield was compromised (attack strength equal to or greater than shield strength), and regenerated at a rate of 20% per combat round, then a mass of smaller ships attacking a larger vessel could degrade the shields sufficiently to score more significant damage as the combat progresses. This would require some creative coding, however...

It might also be nice to break the combat rounds into 3 phases, with each eligible attacker being allowed only one attack in a given phase. This ensures that each attacker get a chance to attack once, before any attacker is allowed to attack a second time. This would help smooth out combat results significantly, although the random factor is kind of fun/keeps people on their toes. A 'counter/ship has fired this phase' variable would have to be introduced to do this, however.



The more involved stuff (introducing a chance to hit, with variables based on speed, range, and level of detection/stealth) would be a bit harder to implement, but I think that the changes suggested above shouldn't be too hard to code. I'm not a coding guy, though...

As for my previous post, increasing the combat rounds to 6 x number of eligilble attackers, and doubling hull structure strength should give smaller ships more opportunities to hit bigger ships before they are destroyed. This will be more effective if 100% repair between combats/each turn is reduced significantly, to a percentage each turn instead of the full amount. That way empires with weaker fleets are given the opportunity to capitalize on any damage they may have caused in a previous combat, rather than having to face 'fresh' fleets each round.

The reason steamrollering in the later game works so well is because you know that if you win, you'll have fully repaired ships at the ready to jump into the next system the next turn, and the weaker defenders simply cannot mass defenses quickly enough to stop you. However, if the damage is accumulating, and takes several turns to fully repair, then the attacker may need to pause for a turn or two while damage is repaired, instead of just charging at full steam to the enemy homeworld without a care in the world. Thus giving the defenders time to pull in some additional fleets for defense, plus build some last minute defenders, or perhaps mount a counterattack against your damaged fleet.
Last edited by Ta'Lon on Sat Mar 08, 2014 4:26 am, edited 3 times in total.
Conquering the galaxy, one planet at a time...

Any artwork that I submit for use in the graphics forum is submitted under the CC-BY-SA 3.0 License.

yandonman
Creative Contributor
Posts: 699
Joined: Thu Aug 30, 2012 12:32 am

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#8 Post by yandonman » Sat Mar 08, 2014 4:20 am

Destroying staryards during combat would slow blitzkrieg
Code released under GPL 2.0. Content released under GPL 2.0 and Creative Commons Attribution-ShareAlike 3.0.

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

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#9 Post by MatGB » Sat Mar 08, 2014 4:21 am

It's now past 4am so a brief reply to Ta'Lon, yes, shields are per weapon, we've had discussions elsewhere, pretty much ever since Geoff committed the code for them, that they can be a bit overpowering. they're also ridiculously expensive—I always do better and win faster when I don't use shields compared to when I do, it's very very rarely that I notice their lack, and that's normally if I've let the Experimentors develope their monster fleets. My personal opinion is that they should be reduced in effectiveness and vastly reduced in cost, but it's a much lower priority for me as their excess costs makes them not game breaking currently compared to other issues.

Some sort of UgoIgo mechanic would possibly be a good idea, possibly using something like the D'Hondt counting method for who gets to go next, that would at least reduce the streakiness and make it less unbalancingly bad, I think—and yes, I have had quite powerful ships destroyed by lesser ships sometimes, but only ever in small groups, generally if you've got at least 5 ships on your side and you outclass the opponent you win overwhelmingly, sometimes withdraw for repair is needed, that's it.

Dilvish, like the idea of targetting computers &c, but I've no idea how much of a rework it'd be, compared to more simple adjustments to the current system. I'd like to see something implemented fairly quickly to make the game more, well, interesting in the late game, ideally newbuild ships should be to replace older models, not simply reinforce so it's overwhelming.
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
Geoff the Medio
Programming, Design, Admin
Posts: 12249
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#10 Post by Geoff the Medio » Sat Mar 08, 2014 8:50 am

eleazar wrote:IIRC currently a ship is randomly chosen to go first, it shoots, and if it does enough damage it's target is removed from battle before it gets a chance to do anything. What if multiple ships on both sides got a chance to fire, before the weapons hit and did damage/destruction? That would reduce the current strong snowballing effect, and tend to exact higher casualties on both sides.
I've considered changing the mechanics so that there are only 3 "rounds" of combat, instead of the current 3*Number_of_ships_in_combat, and every ships fires during a round, before any are damaged or destroyed. This would have the "advantage" that a lot of shots would be wasted when a larger group attacks a smaller group. Currently, a ship stops being targeted when it is destroyed, so shots are instead used on other still-functional ships. If attacking simultaneously, a number of shots allocated to each target will be a random distribution which doesn't depend on how many shots are actually needed to destroy the target, so some targets will get more shots fired at them than are needed and the extras will be wasted. This is more likely to happen when a larger group attacks a smaller group, as the average number of shots affecting each target in the smaller group is probably higher.

A system of that style would be more suited to incorporating factors like range or ship speed, as well. These were discussed a while ago, but I didn't find the proposed systems satisfying.

Re: repair speed at drydocks: There are objections to non-instant repair, which takes several turns to incrementally do the repair, as this makes tracking which ships are repaired "quite annoying".

I'm a bit wary, but not entirely opposed, to having a few ship parts that can be added to a design to affect its targetting behaviour. Options might include all weapons on one target vs. all weapons to different targets vs. default random selection of targets for each shot. More advanced parts that avoid targetting ships with too much shielding, or target the ship with the least structure remaining might be OK. The latter might not be of much use if combat rounds have every ship fire its weapons once, as all ships with that part would be firing on the same target during each round, probably wasting a lot of shots.

I would not want to incorporate complicated python scripts for targetting, though.

User avatar
BraveSirKevin
Space Squid
Posts: 68
Joined: Wed Aug 14, 2013 1:42 pm
Location: Johannesburg, South Africa

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#11 Post by BraveSirKevin » Sun Mar 09, 2014 3:19 am

Geoff the Medio wrote:
eleazar wrote:I'm a bit wary, but not entirely opposed, to having a few ship parts that can be added to a design to affect its targetting behaviour. Options might include all weapons on one target vs. all weapons to different targets vs. default random selection of targets for each shot. More advanced parts that avoid targetting ships with too much shielding, or target the ship with the least structure remaining might be OK. The latter might not be of much use if combat rounds have every ship fire its weapons once, as all ships with that part would be firing on the same target during each round, probably wasting a lot of shots.
Would it not be better to handle this with Combat Stances? We've already got Aggressive and Stealthy stances for the fleets, but perhaps we could some more that dictate which targets they focus on, so for example:

• Point Defense: The fleet focuses on destroying the weakest enemy ships (ie. the fighters and troop ships) first. Ships in the fleet will always choose the target with lowest base structure provided it can penetrate its shields.
• Hunter-Killer: The fleet focuses on destroying the stealthiest ships first. Ships in the fleet will always choose the target with the highest stealth provided it can see it and can penetrate its shields.
• Swarm: The fleet focuses on hitting the biggest target in concert. Ships in the fleet will all choose a single target with the highest base structure and will only fire at that one this round. It will pick a new target in the subsequent round only if the original target has been destroyed.
• Skirmish/Harass: The fleet does no damage, but reduces the shields of the enemy force. Skirmishers always fire first, picking their targets randomly and the gross damage taken by the target is applied as a penalty to its shields until the end of the round.
• Fire-at-will: The fleet fires randomly as per usual.

This is just a suggestion, of course, but i think doing it this way would give a more tactical feeling to the current combat mechanic, and change up the composition of fleets and ship designs in a good way.

To eliminate some of the micromanagement involved here, it might be an idea to set the stance of a newly built ship in the design window so that whenever a new ship of that variety is produced it automatically adopts the stance you intend. (ie. a new hunter-killer would pop out with the hunter-killer stance already specified, a swarmer would automatically be set to swarm, a scout would automatically be set to hide, etc.).

User avatar
OllyG
Space Kraken
Posts: 151
Joined: Wed Jul 20, 2011 12:03 pm

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#12 Post by OllyG » Tue Mar 11, 2014 8:39 am

Ta'Lon wrote:There is currently NO variable in this combat method that 'counts' whether or not a ship has fired already, versus the other ships, so a given ship can be picked multiple times in succession as the next attacker, if the random picker dictates so.

...

1) After a ship selects a target, it fires it's first weapon. If the target survives the attack, it fires another weapon at the same target, continuing to select the same target until it runs out of weapons or it's target is destroyed. A new target is selected randomly only after the currently selected target is destroyed.
Using these two ideas would change battles involving large fleets quite a lot, but leave battles between two small fleets largely unchanged. Seems good.

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

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#13 Post by Kassiopeija » Sat Jul 19, 2014 10:51 pm

MatGB wrote:10 ships vs 5 ships should see 2 or 3 of the larger fleet die
I think the reason why this isn't happening lies within this mechanic:
Ta'Lon wrote: Also, the next ship that gets to attack is determined randomly, not in any particular order. [...]

There is currently NO variable in this combat method that 'counts' whether or not a ship has fired already, versus the other ships, so a given ship can be picked multiple times in succession as the next attacker, if the random picker dictates so.
So basically a fleet that has double the amount of ships as the enemy fleet has also its chance to be randomly picked increased by the same factor. So even if the smaller fleet has way better offenses (resulting in both fleets having the same attack value) the larger fleet would generally still do 2 times the amount of damage.
However, if the ship design of both fleets are equally then the larger fleet had a base attack value of 200% of the smaller fleet, which would result in them doing 400% the amount of damage as the smaller fleet. --> which is why you win with your bigger fleet with apparently no lost ship!
And this is even further exacerbated because the larger fleet will faster destroy one enemy ship, which will additionally increase their chance to be picked (or the enemy ships to be not picked)

The same mechanic also promotes to design ships cheap, so you can build more in lesser time. Another drawback to shields (which are very expensive) and as is seen here, severely tune down your offense!

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

Re: Steamrollering blitzkrieg—we need more pyrrhic victories

#14 Post by MatGB » Sun Jul 20, 2014 9:26 pm

Problem is now, basically, solved in Trunk, the new combat mechanic of everyone shoots then ships are checked to see if they're dead or not means you take damage and can cause hits, and bigger fleets can end up massively overkilling a single ship in the smaller fleet, thus increasing chances of smaller fleet surviving into round 2 or 3.

Needs a bit more testing but it's far better.
Mat Bowles

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

Post Reply