'Fixing' Bigger = Better (open discussion)

Please post here for questions and discussion about scenario, art and sound modding and the game editor for Distant Worlds.

Moderators: Icemania, elliotg

Ikke
Posts: 13
Joined: Tue Aug 05, 2014 2:01 pm

'Fixing' Bigger = Better (open discussion)

Post by Ikke »

Greetings,

for the past few years I've been playing distant worlds on and off, it's one of those games I like to pick up when I want to emerge myself in some far away universe and shut my door for half a week like a proper junkie. So lately, I've been thinking about modding some issues that have bugged me with this game over the past years, on which I'd appreciate some feedback. I'm thinking about doing some sort of rebalancing mod. Rebalancing the weapons, defences, etc is easy enough - but some issues aren't that straightforward.

To start with, the given that bigger ships are better ships. I'll give my analysis in a bit, but first I'd like to explain why I dont like it. I see in distant worlds a potential for a more interesting game with multiple ship types for multiple purposes, rather than one beat all design. The game is set up as such too - with different ship types - escorts, frigates, etc - but the balance does not reflect this. Thus, by understanding the game balance, we are forced to arrive at the conclusion that the winning strategy is that bigger = better. We can choose to ignore this for romantic reasons, but that's not proper game design.

Why is bigger better?

-Tactically superior.

Lets imagine a very simple game and a very simple situation. Just hit points, attack damage and units - all other stats being equal. One side has one ship with 20 hit points and 2 attack damage. The other side has two ships with 10 hit points and 1 attack damage each. Now this would appear to be an even matchup, and for a while it is. Each side deals 2 damage to the other side with each volley. However, after 5 volleys and 10 damage dealt, the side with two ships loses a ship, and from there on out, the battle is no longer equal. Though both sides now have one ship and 10 hit points, the side which had two ships now only has 1 damage, as opposed to the two damage from the side with the bigger ship. The side with the bigger ship will destroy the second ship in another 5 volleys, while that ship itself will only suffer 5 more hit points of damage.

Long story short: Assuming the side with more ships can coordinate perfectly and get his forces to engage the an equal force of bigger ships at the same time, then still, doubling the ship size makes them 25% more effective, give or take a few percent.

-Economically superior.

Each ship has a certain amount of components that must be on it, but only one is needed. Command module, hyperdrive, targeting, etc. Lets call these components the fixed costs to having a ship (usually somewhere around 150 maintenance). If a fleet was made of small escorts of roughly 600 maintenance, these costs would be around 25%. If a fleet was made of capital ships of 3000 maintenance, this would amount to but 5% of the costs.
Add to that the fact that hyperdrives cost the same to operate at any ship size, and you can see that a fleet of 5 escorts at 600 maintenance guzzle up 5 times as much fuel on interstellar flight as your single 3000 capital ship.

Long story short: the bigger, the more bang for your buck.

Conclusion: The combination of tactical and economical benefits make bigger far to good.


What counterbalance could there be?

-Strategical inflexibility (aka rework the AI).

Bigger could be an over-concentration of force. Say I have one very big ship and 3 mining stations to defend, 3 smaller ships could attack these mining station simultaneously, and the one bigger ship can only defend one, losing two of its targets. The one target that is defended should have a threatening smaller ship close by, so that the defending ship can not leave it its defence without losing it.
If the AI could be reworked to pursue a strategy of constant threat against as many targets as possible, and defend appropriately, this would at least create a soft cap in ship size: you need an adequate amount of ships to cover all your assets, and threaten a fair amount of your opponents assets. Sadly, the AI prefers to lump up small ships into fleets of various sizes, and send them all on one target. This defeats sole advantage of having more smaller ships - the ability to be at more places than your enemy.

However: We have to scrap this option on the account that the AI is not moddable at this point. (note: by pursuing such a strategy yourself however, you can defeat almost any AI under almost any odds, allthough it requires excessive micromanagement.)

-Lowering the fixed costs or scaling them in proportion to size.

The economic part of the imbalance could be addressed by lowering the fixed costs, and let things like hyperdrive cost scale with size.
Taking this one step further, a slight exponential increase in operating costs in relation to size could counterbalance the innate tactical advantage. That is to say, if a ship becomes roughly 25% stronger by doubling its size, so should its operating costs increase by 25%.

However: While we can reduce the fixed costs somewhat by making the material requirement for said components minimal, that's as far as we're going to get. Hyper drives will still favour bigger ships. There is no way to add an "*ship size" modifier to any component, let alone work an exponential equation in there. We can get a little closer to economic equality, but nothing too influential.

-Using the overkill mechanic.

Remember our example from earlier, with 1 ship of 20 hit points and 2 ships of 10 hit points. Now lets say the first ship shoots a volley of 20 damage, and the two ships a volley of 10 damage each. After one volley from both sides, 1 ship from the side with two ships remains. This is due to the overkill of 10 points from the shot of the first ship.

However: This solution makes battle very sharp, very brief affairs, quite going against the notion of a game with multiple exchanges and gradually wearing down the opponents shields. It would go very much against the grain to implement such a solution. All though, as far as I can see, it's the only one that could be implemented.

-More random effect, making redundancy more important.

Say when you hit a reactor, it blows up the ship. Say shields reduce the damage let through by a percentage equal to how full the shields are. Say armour has a chance to block the remaining damage all-or-nothing, depending on how much armour is left. After a few shots there is a small chance any shot will penetrate both, hit the reactor and blow up the ship. This makes one ship a rather vulnerable affair - if you'd had more ships, you might face the same chance, but your redundancy means you wouldn't lose it all at once.

However: This cannot be implemented without access to the game mechanics, which we lack.

In conclusion:

All though creating bigger ships should carry its rewards, it is a research investment after all, at present it's too dominant an advantage in my experience.

I'd highly appreciate any and all feedback in addressing this, for I'm rather stuck at this point in my rebalancing redesign.
User avatar
robske
Posts: 59
Joined: Fri Dec 10, 2010 7:56 am
Location: The Netherlands
Contact:

RE: 'Fixing' Bigger = Better (open discussion)

Post by robske »

Fully agreed. Would like some lovely mechanics which support more inteligent ship design than just bruteforce (or torpedo force) the life out of those alien bastards [&o]
Scifi 4X game geek...

The universe has its mysteries which can only be solved through the curiousity of intelligent life...

My lesson about life:
Keep the past in mind, act in the present, look to the future.
User avatar
DeadlyShoe
Posts: 217
Joined: Sat Jun 01, 2013 10:15 pm

RE: 'Fixing' Bigger = Better (open discussion)

Post by DeadlyShoe »

i think bigger is intended to be better in order to keep ship counts down in the late game

otherwise the # of ships could get truely out of hand
Ikke
Posts: 13
Joined: Tue Aug 05, 2014 2:01 pm

RE: 'Fixing' Bigger = Better (open discussion)

Post by Ikke »

That might be true, but it's a tad inconsistent that the AI continues to build primarily frigates, destroyers and other small junk. It only builds a handful of bigger ships.
Also the ships designs are categorised by size, more or less stagnant at a certain size. Rather than say, role, and increasing in size to optimise the investment. (note: doing the latter as human player once again gives you quite a bit of advantage)

Though it might have been for other titles, from which the balancing conventions were likely taken, the game design is not consistent with that hypothesis.
User avatar
Icemania
Posts: 1847
Joined: Wed Jun 05, 2013 9:14 am
Location: Australia

RE: 'Fixing' Bigger = Better (open discussion)

Post by Icemania »

For the AI Improvement Mod I've ensured that the AI builds a lot of large ship as the game progresses, rather than the Vanilla junk. Any other strategy puts the AI at a disadvantage. But I very much agree some changes would be great to open up diversity in ship size strategies.

Nanaki
Posts: 306
Joined: Wed Jun 04, 2014 3:06 pm

RE: 'Fixing' Bigger = Better (open discussion)

Post by Nanaki »

There does need to be more diversity... An easy way to do it would be to have a price multiplier as mass increases, which makes super large ships prohibitively expensive to build and maintain.
I ate the batter of the bulge at Hans' Haus of Luftwaffles
User avatar
lurchi
Posts: 319
Joined: Tue Jun 10, 2014 4:25 pm
Location: LV-223

RE: 'Fixing' Bigger = Better (open discussion)

Post by lurchi »

I'd sure love to see a correlation between hyper drive and ship size, the drive size right now could be the minimum. Different upper limits for the various drives would open up interesting possibilities like faster smaller ships.

That said such limits should exist for thrusters as well enabling small gunboats to zip around huge dreadnoughts. The AI has to be updated so it can handle such tactics, though.

Way too much hassle for a patch, but this should be considered for DW2.
User avatar
lordmoore
Posts: 10
Joined: Sat Jun 23, 2012 7:43 pm

RE: 'Fixing' Bigger = Better (open discussion)

Post by lordmoore »

The only change I'd like to see is an increase in the amount of energy it takes ti initiate the drive based on ship size.

That way larger ships take more energy to jump and hopefully take a bit longer as well.
User avatar
feelotraveller
Posts: 1040
Joined: Mon Sep 12, 2011 10:08 am

RE: 'Fixing' Bigger = Better (open discussion)

Post by feelotraveller »

Have you tried permanently firing Strength in Numbers for all races? It's a bit clumsy but that would at least give a taste of gameplay where smaller ships have cheaper maintenance...
AKicebear
Posts: 205
Joined: Sat Jul 26, 2014 2:11 pm

RE: 'Fixing' Bigger = Better (open discussion)

Post by AKicebear »

Not sure of the correct solution, but just chiming in to say I agree on the need and would love to see (someone else) figure out a good way to mod this!
User avatar
feelotraveller
Posts: 1040
Joined: Mon Sep 12, 2011 10:08 am

RE: 'Fixing' Bigger = Better (open discussion)

Post by feelotraveller »

I should add that I'm not convinced by the 'bigger=better' argument. There is some truth to it... if we are talking battle fleets where destroying other ships/bases is the aim. On the other hand although you get more bang for your buck, like with resources, sheer quantity is often not as important as distribution.

Some counter examples

- if my purpose is to raid a colony (lets assume baseless) then I want to have 2 smaller ships vs. one double sized ship because it is almost a sure thing that one of the ships will succeed in dropping troops for the raid.

- if I am defending a new colony which has a base just starting to build I am likely to want more smaller ships, a 1500 size capital will not stop 4 size 300 frigates from destroying the base, even if it might destroy a frigate or two.
Ikke
Posts: 13
Joined: Tue Aug 05, 2014 2:01 pm

RE: 'Fixing' Bigger = Better (open discussion)

Post by Ikke »

@ Icemania I tried your mod, I loved it for the challenge, and hated it for pretty much this reason. ;) Not at all your fault of course.

@ Nanaki yes, if we had just a little bit more access, we could implement tons of solutions. Not just economic, but for example, we could modify the hit % with size, i.e. base hit % = ((weapon speed * target ship size) / (target ship speed * target turn rate)) * balancing constant. Something along these lines could be in the design already of course, but I fail to detect a big impact there. At any rate: should we have such access, we can chip away at the advantage with multiple factors, adding a great deal of gameplay choices in the process - at the moment however, I'd settle for just one way.

@ feelotraveller 1st post: That's an idea, do you know where the cut-off point is for what is a small ship according to that event? It would at least allow two viable ship sizes...
2nd post: good points, however I fail to see how a difference in ship size would alter the outcome. Not sure it always works either, I see AI ships ignore the distraction of other ships thrown at them often enough. Distracting the AI in such a fashion, by giving it something else to shoot at than it's objective is a tactic working by virtue of poor AI, not game balance at any rate.
Ikke
Posts: 13
Joined: Tue Aug 05, 2014 2:01 pm

RE: 'Fixing' Bigger = Better (open discussion)

Post by Ikke »

Having thought about it some more...

How about removing shields and armour completely - or greatly reducing them? This way, components would fall off pretty equally regardless of ship size.
User avatar
lurchi
Posts: 319
Joined: Tue Jun 10, 2014 4:25 pm
Location: LV-223

RE: 'Fixing' Bigger = Better (open discussion)

Post by lurchi »

ORIGINAL: Ikke

Having thought about it some more...

How about removing shields and armour completely - or greatly reducing them?

Who would want to do this? [:D]
martyran
Posts: 22
Joined: Fri Aug 30, 2002 3:37 pm
Location: Australia

RE: 'Fixing' Bigger = Better (open discussion)

Post by martyran »

Like this idea
I make a self rule when playing that the size of the ship if I customise it must be no more than the original size (I like a challenge )
how about build times reduce the output of the construction yards ?
make the really big ships years to make ?

User avatar
feelotraveller
Posts: 1040
Joined: Mon Sep 12, 2011 10:08 am

RE: 'Fixing' Bigger = Better (open discussion)

Post by feelotraveller »

No I am not sure about what the cutoff is. Would need some testing by someone interested in taking this approach... [;)]

The other tweak (which won't change the 'bigger' equation though) is to go through the policy files, and probably design files as well, and adjust the AI to build more bigger ships.

If you have specific criticisms/suggestions for AI behaviour improvements which may be easy(ish) to implement then the developer may adopt them. Many things have got into the game this way. But I would not hold your breath for an AI overhaul - demands of the game engine for processor time and developer coding time are both limited quantities, particularly at this point. (Still they could be useful suggestions for whatever comes next...)

Having had a look at your war strategy article I'm not sure the we really differ much about the bigger equation. I'm assuming your frigates and escorts (at least) are not max size? And that is all the point I wanted to make.
User avatar
Tanaka
Posts: 4474
Joined: Tue Apr 08, 2003 3:42 am
Location: USA

RE: 'Fixing' Bigger = Better (open discussion)

Post by Tanaka »

Great thread I completely agree with the OP!

Is there a way to make it so that smaller ships are faster and harder to hit but weaker with less firepower while bigger ships are larger slower targets and easier to hit but have more firepower? Isn't this how it should be?

I have this picture of large slow battleships firing broadsides at each other and launching ships while those fighters and pickets (frigates, escorts, destroyers, etc) zoom around attacking each other and taking pot shots at the big boys. Pretty much like star wars!

Other than diplomacy has to be one of the biggest shortfalls of the game that ship size means nothing and there are not advantages and disadvantages to size! Its just bigger is better all the way!

Image
AKicebear
Posts: 205
Joined: Sat Jul 26, 2014 2:11 pm

RE: 'Fixing' Bigger = Better (open discussion)

Post by AKicebear »

I like the ideas about about making building time scale with size, as well as hit chance penalties with greater sizes. Perhaps these are already in place but just need to be stricter/exponential?
User avatar
Blackstork
Posts: 802
Joined: Mon Jul 07, 2014 3:45 am

RE: 'Fixing' Bigger = Better (open discussion)

Post by Blackstork »

i think its better to ask devs about tracking features

smaller ship - harder to track wit weapons, more targeting needed/ more effect of countermeasure effect (example : 0.8 enemy targeting effect module applied / 1.2 countermeasure effect applied multiplier)
bigger starship - the opposite (example : 1.2 x on enemy targeting / 0.8 multiplier on applied countermeasure)
All classes should have gradation while destroyers will be 1/1

then the svasion pereference races, smaller ship races and countermeasure users will build towards hi evasion fleets , while on opposite side there also all will be better - bigger ships which easier to track, lesser cm effects, which will need to build on targeting systems which will raise their own value with such changes
Alien Immersion (WIP) : tm.asp?m=3651979
Beyond Extended Universe (WIP): tm.asp?m=3657646
Ikke
Posts: 13
Joined: Tue Aug 05, 2014 2:01 pm

RE: 'Fixing' Bigger = Better (open discussion)

Post by Ikke »

@feelotraveller: It is true I use smaller ships, minimal force to get the job done is economical force too. The AI uses that very poorly though. Not sure about requesting anything with regard to the AI - I can think of plenty of things there to be sure - but I am no expert in the field.

All though, my chess engine has no qualms about refining its calculations till it eats up all available computer power, it would be nice if the Distant worlds AI would do the same. I installed battle chess the other day, a chess game from 1989(?) which uses a simple min-max algorithm - and it manages to eat up all 2014 CPU power. Which was rather funny cause it's difficulty can be adjusted by the time given to the chess engine. It was hard, but I beat the easiest setting :p. Back to DW or RTS games in general, the entire game uses only like an eight of my computer (16g RAM, 8 core 2.6g CPU).

I never quite understood why and RTS AI cant be made to chop up a continues game into finite blocks to calculate ahead with - making those blocks smaller and more refined depending on available power. These blocks could be thrown in directly as gameplay elements; location wise: sphere of influence, empire, sectors, systems, planets - time wise: grand strategy, present war(s), campaigns (fronts), operations, battles, units, these blocks could be put in a hierarchy and an interface where a player can adjust/create them and the assets allocated to them. Assets could be chopped up in intel, logistics, etc. Larger blocks can run on very small resources, needing very infrequent updates - being the long term things. But maybe that should be left a university - cause I am rambling on with far too little knowledge.

I gather the DW team has no AI experts either - best thing would be to throw the code open for that part. Spring engine (open software rts) has very good AI for example, with automated micro which is pretty decent. Just people jumping in and refining the AI cause they want a better game - would help DW no less I think.

@ Blackstork I fully agree that some tracking mechanics would be nice. I'd like the option to relating it to actual ship stats such as speed and turn rate though. A very small ship sitting still is still a sure kill.
Post Reply

Return to “Design and Modding”