Another Look At Weapons Balance

Matrix Games Forums

Forums Register Login Photo Gallery Member List Search Calendars FAQ

My Profile Inbox Address Book My Subscription My Forums Log Out

Another Look At Weapons Balance

View related threads: (in this forum | in all forums)

Logged in as: Guest

Users viewing this topic: none

Printable Version

All Forums >> [New Releases from Matrix Games] >> Distant Worlds 1 Series >> Another Look At Weapons Balance

Page: [1]

Message

<< Older Topic Newer Topic >>

Another Look At Weapons Balance - 10/18/2014 3:26:52 AM

llanite_slith

Posts: 45
Joined: 10/23/2010
Status: offline

So I've played DW since the first edition, but I haven't really delved into modding until this release (mostly because babies and toddlers just suck up every minute of the day when I'm not at work� but it's getting better...)

I've read with great interest the various discussions of various weapons balancing issues, the thread on armor, and other discussions of the game's mechanics and I really appreciate the analysis from the DW community on this board - it's encouraged me to offer the following approach to visualizing and mapping weapons balance. My own goal was to understand the current system well enough to intelligently mod it. As people have noted, some weapon systems (like rail guns) have very limited utility, and there are only a few weapons (torpedoes, titan beams) that seem effective by the late game.

Methods:
To visualize all this, I've taken the trouble of assembling all the weapon systems, their improvements, and their respective tech levels into a single document. I then built a test case where I measured the damage of each system based on damage, rate of fire, and a test target case (100 shields, 50 armor, 100 hull points) and then calculated comparative damage models for all weapons and improvements at various ranges. To visually compare the results, I then normalized all the values and graphed the normalized results.

To understand what you're seeing, there are 2 normalized indices I use - a DE (Damage Efficiency Index) that is a normalized measure of the ability to kill the target case above at a particular range. 0 is the Mean of the entire population of weapons. Negative numbers fall below the Mean and are more efficient at killing the Test Target at that range.

I also use a second number - CE (Construction Efficiency) which measures how efficient a weapon system is at killing the Target Case at a particular range in terms of size and energy consumption. Again, 0 is the Mean of the population and negative numbers are more efficient in terms of space and energy.

I've accounted for all the weapon system's special abilities (bypassing armor, to-hit penalties, range fall off, etc.) in the model. The only thing I didn't do was wrap reactive armor effects into the equation. So, for example, a torpedo hitting the target case at range 200 will have to do 250 points of damage while accounting for the damage fall off at range 200. The DE is effectively how fast the torpedo will kill the target compared to all the other weapon systems. The CE is a measure of the energy it took to deliver that damage with the size of the weapon system modifying the answer. The same test for the graviton beam would only have to deliver 100 damage points since it bypasses shields and armor (but is penalized for its lower rate of fire and higher energy requirements).

So, here are all the weapon systems by Tech Level (X-Axis) mapped against their respective DE values (Y axis) at range 200:

[image]1.gif[/image]

Attachment (1)

< Message edited by llanite -- 10/18/2014 4:32:00 AM >

Post #: 1

RE: Another Look At Weapons Balance - 10/18/2014 3:34:20 AM

llanite_slith

Posts: 45
Joined: 10/23/2010
Status: offline

You can clearly see why people prefer certain weapon systems by the end game.

Here you can see the full nature of the development of all the weapon systems at range 99 (no distance penalty). X-Axis is CE, Y-Axis is DE, and as a convention I've reversed the scales - the upper right quadrant of the graph is the most efficient in terms of both damage and construction/energy. Systems above the DE Mean (horizontal black line on the graph) are more effective at damaging targets while systems to the right of the CE Mean (vertical black line) are more efficient at doing that damage in terms of size/energy:

Generally each weapon systems starts in the bottom left and progresses to the upper right with each tech level (labels made this graph unreadable).

Attachment (1)

< Message edited by llanite -- 10/18/2014 4:41:13 AM >

(in reply to llanite_slith)

Post #: 2

RE: Another Look At Weapons Balance - 10/18/2014 3:39:11 AM

llanite_slith

Posts: 45
Joined: 10/23/2010
Status: offline

Here's the same graph at Range 400:

Attachment (1)

< Message edited by llanite -- 10/18/2014 4:43:09 AM >

(in reply to llanite_slith)

Post #: 3

RE: Another Look At Weapons Balance - 10/18/2014 3:43:38 AM

llanite_slith

Posts: 45
Joined: 10/23/2010
Status: offline

And at Range 600:

Observations and Conclusions about the DW Weapon System:
The second graph explains a lot of things: first, you can see that rail guns (Orange) have a very different progression from all other systems. They start out very CE efficient and at the top in terms of DE. But they have a very flat progression and by mid-game fall out of favor - as their tech levels increase, they become more efficient in terms of size/energy, but never become better at causing damage to a target. Grav weapons are the most efficient at causing damage, but by late game can't compete against torpedoes and beams. Missiles are a bit hopeless - even at long range, by the time you develop Assault Missiles, Torpedoes already do a better job.

From my own perspective, this also illustrates why every game plays out the same by the late game. At the beginning of the game there are very real choices to make about different weapon systems because they are wildly divergent in terms of damage vs. construction trade-offs. But by mid-game, and certainly in the late game, all these different technologies converge toward a single point and there's only a couple of systems that make sense.

I hope this makes sense to people - and for the modders out there in particular, if you're designing a total conversion or a rebalancing of the weapon systems, it would be very interesting to see what might come of a set of weapon systems that stay divergent through the end game, with no clear winners in the tech race. This could give each race a lot of uniqueness and make the in-game ship combat more exciting as you have to redevelop your fleets to address different technological threats.

If anyone is interested I can also provide the underlying excel file I used for this - it combines elements from the Component.txt file, the Projects.txt file, and adds meta-data like the Tech level and Weapon Damage Type (phase, grav, etc) to each weapon system so that all of that data is available in one place.

Best Regards
Llanite

Attachment (1)

< Message edited by llanite -- 10/18/2014 4:45:00 AM >

(in reply to llanite_slith)

Post #: 4

RE: Another Look At Weapons Balance - 10/18/2014 11:03:02 PM

rxnnxs

Posts: 60
Joined: 6/1/2013
Status: offline

thank you for this work you didi a really time consuming and useful job!
i wish that those findings will influence the mods and the original game as well.
this work needs more attention and i hope it will get some more while i am typing :-)

(in reply to llanite_slith)

Post #: 5

RE: Another Look At Weapons Balance - 10/19/2014 2:26:31 PM

llanite_slith

Posts: 45
Joined: 10/23/2010
Status: offline

Thanks rxnnxs! I'll admit I did this mostly for myself as I dug in to what it would take to mod DW. I'll have some posts later this week about using all this as a tool to quickly design weapon systems for a total conversion. Hopefully people will see the advantages of an analytical approach with a real example.

(in reply to rxnnxs)

Post #: 6

RE: Another Look At Weapons Balance - 10/19/2014 9:40:10 PM

hewwo

Posts: 276
Joined: 4/22/2010
Status: offline

cool work :) I wonder how the weapon balance in the AI mod looks like in a graph like this.

(in reply to llanite_slith)

Post #: 7

RE: Another Look At Weapons Balance - 10/20/2014 3:28:40 AM

CyclopsSlayer

Posts: 583
Joined: 2/11/2012
Status: offline

I agree 100% with these conclusions.
For a long time now it has been apparent to me that the only two efficient systems, and the only two I develop, are Energy Torpedoes and Fighters. The best mix I could find of Stand-Off and local superiority combat ability.
Alongside the Warp blockers to tackle fleeing targets.

Believe me, an attacker trying to strike one of my forward Bases/ Mega Carriers will seldom actually get into gun range when faced with 40 Hanger loads of Fighters especially since Fighters once activated stay active in local space and engage FAR FAR beyond standard weapon engagement ranges.

(in reply to hewwo)

Post #: 8

RE: Another Look At Weapons Balance - 10/23/2014 2:09:01 PM

Icemania

Posts: 1847
Joined: 6/5/2013
From: Australia
Status: offline

Great to see some more analysis!

How did you account for special abilities (bypassing shields, armour bonuses/penalties etc)?

How did you account for different sizes and power requirements given the impact this has on ship design?

How did you account for ship tactics?

How did you account for armour and repair bots?

Which all leads to the key question ... have you conducted any testing in-game?

I started with the theory as well but to be honest found it easier to optimise ship designs and then focus on Battle Arena style testing at various technology levels and with different tactics.

Do you have any specific proposals to change the weapons and if so how they been tested in-game?

In those proposals what philosophy is being applied? e.g. some weapons strong early game but weak late game etc?

< Message edited by Icemania -- 10/23/2014 3:09:33 PM >

(in reply to CyclopsSlayer)

Post #: 9

RE: Another Look At Weapons Balance - 10/23/2014 5:20:18 PM

llanite_slith

Posts: 45
Joined: 10/23/2010
Status: offline

How did you account for special abilities (bypassing shields, armour bonuses/penalties etc)?
>I set up a dummy target with shields, armor, and a internal components and then calculated the number of shots to kill the dummy for each weapon including damage fall-off from range, rate of fire, raw damage, and special damage (bypass shields, less effective against armor, etc.) as part of the equation. The DE number is effectively a measure of how efficient a single weapon is at killing the dummy target for a particular range.

How did you account for different sizes and power requirements given the impact this has on ship design?
>That's the CE number - basically how efficient the weapon system is as part of a ship. Small or highly energy efficient weapons have negative CEs (missiles being the classic example). Graphing DE vs. CE shows you both what's best at causing damage vs. what's the most efficient ship system.

>An example at range 400 - the Titan Beam and the Shaktur Firestorm II (Firestorm with the first improvement) have roughly the same DE, but the Firestorm is considerably more efficient at causing that damage, making it a better overall weapon system at range 400.

How did you account for ship tactics?
>I don't- think of this as the weapons sitting on a lab bench and firing at a dummy target. Basically I'm setting all *other* conditions except for weapons performance as equal and comparing the results.

How did you account for armour and repair bots?
>Armor is accounted for in the DE since the dummy target is armored. Repair isn't accounted for- I'm just looking at raw ability of a weapon system to damage a target.

Which all leads to the key question ... have you conducted any testing in-game?
>This is next (see below)

I started with the theory as well but to be honest found it easier to optimise ship designs and then focus on Battle Arena style testing at various technology levels and with different tactics.
>I totally agree that this is the only way to really understand what's going on. DW has a pretty complex combat model (as opposed to, say, Total War's rock-paper-scissors approach) and the only way to see what's going on is to physically observe it. I'll be using your Battle Arena-style testing when I get to that stage of the mod's development.

Do you have any specific proposals to change the weapons and if so how they been tested in-game?
>Honestly, I like the system the way it is - especially with the AI Improvement Mod - that's the only way I play the game. I still find the game fun - I did this work to understand how the existing system works so that I could have starting place for my own mod. The main recommendation I have is that the game needs the kind of diversity that is present in the early game to also be present in the late game - that's something I'm hoping for in my own mod work.

In those proposals what philosophy is being applied? e.g. some weapons strong early game but weak late game etc?
>My general philosphy is that different weapon systems will have very divergent tracks on the graph above, with no single system being the clear 'winner', and less convergence in the late game. I'd like to guide the players to pick a fleet strategy and explore it's consequences. I'd also like a mod where there's a lot of uncertainty in what the player encounters. You see a bit of that in the native game, and your AI mod/ the Extended Universe mod explore this even more. By programming in a set of very divergent (but complimentary) systems into the AI's fleet design, I'd love for a player to meet each alien race with some trepidation, wondering if their chosen strategy can meet a new threat. I'll start posting what all this looks like to the Design and Mod folder soon - I should be ready for Battle Arena style testing by next week.

>Thanks very much for your attention to this. It means a lot coming from someone who's mod I've spent so much time playing for my own enjoyment. It's your work and the work of others on this board that have encouraged me to try this project at all.

Best regards
Llanite

(in reply to Icemania)

Post #: 10

RE: Another Look At Weapons Balance - 10/23/2014 5:39:09 PM

llanite_slith

Posts: 45
Joined: 10/23/2010
Status: offline

Here's an introduction to my mod work:

http://www.matrixgames.com/forums/tm.asp?m=3722655

(in reply to llanite_slith)

Post #: 11

RE: Another Look At Weapons Balance - 10/24/2014 10:42:48 AM

Icemania

Posts: 1847
Joined: 6/5/2013
From: Australia
Status: offline

Very cool.

Please note the impact of tactics is significant.

You may wish to consider the option of applying your analysis to Lurchi's Research Reloaded Mod once in Beta since we'd like to create a Mega Mod combined with Blackstork's Beyond Mod.

(in reply to llanite_slith)

Post #: 12

RE: Another Look At Weapons Balance - 10/24/2014 2:12:22 PM

llanite_slith

Posts: 45
Joined: 10/23/2010
Status: offline

Icemania - I agree, especially now that we can design ships with specific tactical dispositions. That's exactly what I'm hoping to explore in my mod.

I'd be happy to apply the same analysis to Lurchi's mod and compare results.

(in reply to Icemania)

Post #: 13