Important Message

You are browsing the archived Lancers Reactor forums. You cannot register or login.
The content may be outdated and links may not be functional.


To get the latest in Freelancer news, mods, modding and downloads, go to
The-Starport

Adding the "missing" missile levels with FL-consis

The general place to discuss MOD''ing Freelancer!

Post Sat Apr 23, 2005 9:58 pm

Adding the "missing" missile levels with FL-consis

Sparked in part by Argh's posts on missile weapons and in part by some wacky weapon hit point levels I saw in a Freelancer mod, I decided to do some research into the "missing" missile weapons in Freelancer. I'm sure someone has done something along these lines before, but I was personally surprised to find that there was a clearly self-consistent* pattern to the missile weapon levels and damages.
By self-consistent I don't mean logical or balanced or anything else, of course.

In any case, here's the background. Standard Freelancer "jumps" some missile levels - there are level 1,3,5,7, and 9 standard missiles in both guided and unguided flavors; there are level 1,4,7, and 10 EMP missiles; and there are 2 torpedoes with no real "level" associated with them.

So what would standard missiles in levels 2, 4, 6, 8, and 10 be like? Or level 2,5,6,8, and 9 EMP missiles? Or "some other torpedoes" if there were any?

The clue I used was the launcher hit points. I started with torpedoes since I was interested in them. There is a standard pattern to the Freelancer weapon hit points, where the personal hit points are level-based:

Level 1: 400 hp
Level 2: 548 hp
Level 3: 750 hp
Level 4: 1026 hp
Level 5: 1405 hp
Level 6: 1923 hp
Level 7: 2632 hp
Level 6: 4215 hp
Level 8: 4215 hp
Level 9: 6752 hp
Level 10: 6752 hp

If you look at the Starkiller and the Sunslayer, they both fit into this pattern: the Starkiller would theoretically be a level 5 (the launcher has 1405 hit points), and the Sunslayer a level 9 or 10 (launcher has 6752 hit points). This doesn't tell us precisely where the Sunslayer fits in, although since

Next we look at the unguided (missile01_markxx) and guided (missile03_markxx) missile damages, and try to compare them to torpedoes. the missiles are self-consistent - the guided version causes twice the damage of an unguided at the levels for "real" weapons. The Starkiller torpedo as a theoretical level 5 causes twice the damage of an unguided missile (2424 as compared to 1212), but the Sunslayer doesn't match the Cannonball's direct damage, so if there IS a pattern, the only place the Sunslayer could fit is level 10.

Going over to the shieldkiller or EMP missiles, we have only 2 points of comparison to the standard missiles: the level 1 and level 7. In both cases we see that the EMP missile causes 4/3 the shield damage of a like-level unguided missile's hull damage.
Level 1, 489 * 4/3 = 652
Level 7, 2091 * 4/3 = 2788
The EMP missiles also have a consistent hull damage - 1/40th the shield damage.

This suggests that we can then work out the damages for theoretical level 3,5, and 9 EMP missiles by using the same formula. We can also work out the damage for theoretical non-EMP missiles at levels 4 and 10 using the game's EMP missile data and multiplying by 3/4 (the inverse of 4/3) to get the unguided missile damage.

At level 10, the Paralyzer EMP missile causes 7824 hp shield damage. So in theory, a level 10 unguided missile causes
7824*3/4 = 5868 hull damage
And just for kicks, if we try to use THIS missile to get a comparable level 10 torpedo damage:
5868*2 = 11736
which is exactly what the Sunslayer is.

A HYPOTHETICAL DAMAGES TABLE

We can now use this to develop a table of damages; unfortunately, this does leave us without values for missiles at levels 2, 6, and 8. Whatever formula (if any) was used behind this originally, I can't reconstruct it now. It isn't a simple growth formula of any kind - the growth ratio between levels 9 and 10 is much less than 3-4 or 5-7. I finally punted and used GraphPad to do some regressions, then just picked one analysis I liked and plugged in numbers here. For the analysis I used the guided missile numbers augmented with my "calculated" versions for level 4 and 10 so all I was guessing from the regession was level 2-6-8 data. Still, those numbers are just armwaving and shouldn't be considered "consistent" because we don't know how they calculated them.

Level,Guided,Unguided,Torpedo,Shield,ShieldkillerHullDmg
1,244.5,489,978,652,16.3
2,303,606,1212,808,20.2
3,373.5,747,1494,996,24.9
4,474,948,1896,1264,31.6
5,606,1212,2424,1616,40.4
6,727.5,1455,2910,1940,48.5
7,1045.5,2091,4182,2788,69.7
8,1719,3438,6876,4584,114.6
9,2445,4890,9780,6520,163
10,2934,5868,11736,7824,195.6

BUILDING NEW MISSILE SYSTEMS
Knowing the actual damage cause by a missile is only 1 part of an actual missile system. Argh's posts about missiles are I think now a must-read for anyone developing missiles, particularly because the damage effects are complex. Understanding this complexity can help significantly with designing missile systems; there are interesting roles that have NOT been well-exploited, such as subsystem attack missiles.

For the standard missiles, however, there is a pattern to the data used for each one that can be replicated which I hope to go into later. The one critical item we're still missing for "standard" missiles at this point is the toughness data: everything else is either identical for all missiles in a class or varies in a way we can now see.

Post Sun Apr 24, 2005 7:35 am

For the most part, DA built things using straight ramps for damage.

Let's take a look at the Liberty laser weapons (which were, I am almost certain, the main area that they used as their models, since they're so crucial to early SP balance):

li_gun01_mark01:

hull_damage = 16.3
power_consumption = 5.1

energy/damage ratio: 3.2

li_gun01_mark05:

hull_damage = 40.4
power_consumption = 12.6

energy/damage ratio: 3.2

In the process of rationalizing the game design, I raised that ratio by quite a lot, because it made more sense to go by units of 5 as a starting point, and I figured that I could make things work out on the Shield/Armor end of things a bit better.

The problem that I saw when I began re-balancing things, though, was that the curves DA used aren't right when it comes to weapons that fire less frequently (i.e., have a lower hit-rate), have a lower effective range, a lower muzzle_velocity... or all three. Here's a typical example:

li_gun02_mark01

hull_damage = 97.8
power_consumption = 20.4

energy/damage ratio: 4.8

Hmm... a whole 1.6 (or 50%) increase in the ratio. Sounds great, eh? But when you compare that to my current system of balance... I gave a lot more bonuses to weapons with all of the key disadvantages... and damage was raised consistantly across the board. Players won't notice a great deal of change, however, because shield strengths have also been given a different ramp.

If you compare my laser and my plasma weapon, therefore... the difference in ratios looks like this:

li_gun01_mark01(the Justice): 5.0

li_gun02_mark01(the Lavablade): 7.25

And, I should note... this is after I removed shot-speed from the equation. If I'd left shot-speed in the equation, then the energy/damage ratio would be higher than DA's original numbers. So, my Lavablade may look like a worse deal- but the increase in shot-speed (plus the fact that it's "compatible" with all other weapons) more than makes up for it

People may or may not find what I've built to be "more" balanced, but I think that the approach I used was very similar to DA's. I just came to different conclusions than they did, and solved the problems in a different way. I could've just as easily left shot-speeds in, and assigned yet another bonus, but in the end, I decided that consistant weapon performance outweighed "variety". I hope that players will agree with me.

At any rate... my objective here, more-or-less, was to recreate the underpinnings behind DA's system, while removing all of the floats... and making it possible to show other designers how things were done.

My ship-balancing system has, thus far, proven to be a better way to do things than DA's, at least in my opinion- and people will now be able to put game-balanced ships into the design with some simple formulas, instead of just guessing as to what "feels right". Hopefully, that will help future modders avoid creating so many ridiculous uber-ships...

Post Tue Apr 26, 2005 2:14 am

After checking some standard weapons the way you did, randomly picking distinct weapon families that have identical level jumps - e.g. the level 8/9 Salamancas, Krakens, and Wyrms - it is absolutely clear that you're right about their ramps. Weapon efficiencies are identical across level within a specific family, and the ratio of increase in power with level across all families are identical for specific level jumps.

I don't know specifics of how you did the comparisons of power for their different systems, but I could see that the ratios gave us enough information that we could take any given weapon and "scale" the power and damage down to a level 1 to do very clear comparisons. Was that how you worked it out?

I'm not in the mood to tinker with it myself right now. Other things, and besides, my working data set is "contaminated". The Asgard 3 Beta has partially integrated some of your ideas about rationalizing weapon stats - not a complete rework, we just modified the refire rates so that 3.03 -> 3, 5.88 -> 6, and 8.33 -> 8. It's not as far as you're going, but even that minor change makes an amazing difference to how easy it is to mentally compare loadouts.

Post Tue Apr 26, 2005 9:07 am

Actually, since I was going back to the bedrock, I rebuilt every single weapon... and quite frankly, I discarded most of the values that DA had been using, because when I finally got done doing game-balance analysis, I wasn't happy with the way that they'd done things.

DA made a couple of distinct errors:

1. Different muzzle_velocities makes for aiming problems that the game engine doesn't handle correctly. FL's engine always uses the highest-velocity gun- so if you mix velocities, there's a substantial penalty involved, because you're going to miss a lot more with that low-velocity gun. Moreover, low-velocity guns will also miss more because they take a lot longer to reach their targets. So I removed that factor entirely. Every gun in my current balance has the same muzzle_velocity... what differs is the refire_rate and lifetime.

2. DA boosted range in several ways, including buffs on velocity. This was amazingly short-sighted, considering that it made the problems described above even worse- some guns are notably useless, because they're confusing to players. Players can and do learn to compensate for velocity-mixed guns, but when you have 4 different velocities (600, 650, 700, 750)... things are a wee bit messy. I took that factor out of the game design.

3. DA's game-balance also included energy/damage ratios that were noticably ramped to include the effects on the three Shield types, as well. They really did think that Shield=vs=Shield should be an important part of the game. But when all was said and done, they destroyed the integrity of that work when they made the hitpoint values of the three Shield types inconsistant... it just didn't matter any more. So I rebuilt that, and saw what it did to balance... and for the most part, making Shield values "neutral" has practically no ill effect.

In short... the system they used can be dissected, if you just start removing things. But the more I dissected their game balance, the less I liked it. What I've ended up building feels very much like what they did... but it's much more solid. For example... if I keep to my penalties/bonuses, it's going to be easy to adjust the weapons to fit the Faction-specific bonuses for most of them.

Overall, I found that the system they used, while "consistant", was incoherent, and didn't accurately reflect the actual usefulness of the weapons very well. Just like their balance for ships, it appears (to me, at least) that a lot of the balance done was done with an eye on SP, not MP... which was, to say the least, rather short-sighted, considering that MP play turned out to be a pretty big draw... but I have a feeling that MP was considered to be distinctly secondary at the time the game was designed.

Return to Freelancer General Editing Forum