We need to take a look at WMD balance across levels

[Song]

Right now we tend to look at WMD one weapon at a time. This works, but it tends to result in power creep, and causes problems in the long term.

I've done some work looking at average WMD per second for all non-launcher weapons in 1.6.3, and there's some interesting stuff there. First off, here's my assumptions:

1. All shots in a burst hit the target
1a. The game isn't lying about how particle hit detection works. (It's wizardry we don't quite understand yet)
2. No round-off or damage resistance is encountered, nor any bonus applied.
3. Every hit does exactly the average damage
4. Passthrough weapons (just the katana in this graph) don't.
5. The wiki values for what % of damage is done by each WMD category are accurate
6. WMD0 weapons do 1 point of WMD on each hit
7. When WMD weapons have less effective WMD than their rate of fire (thus, less than 1 WMD in average damage) this can be safely rounded up to 1.
8. Fragmentation weapons use their damage as calculated by the game's stats system. This is not especially accurate and I may redo the graph with revised totals to fix this. As it is, fragmenting weapons (Urak Howitzer, Kiloton Cannon, etc) are likely to do quite different damage to what's shown here.




Here's the graph I got at the end of it. Trend line is for all damage types.


(I would put it in spoilers, but those don't work)


A few things strike me immediately:

-Thermo is all over the freaking place, to the point that it's messing with the trend line
-Laser WMD/s trends downwards over levels, because rate of fire decreases and no WMD is applied.
-Against a WMD target, ignoring resistances (again, I don't know if armor resistance applies to WMD, but I suspect it does) the QAC actually does better than a Lamplighter. If you can hit, which is the main issue with the QAC.
-Particle weapons only really include one decent WMD weapon, the Nandao bolt cannon.
-The Mark VII howitzer is basically an APA, one level earlier. Theoretically at least: in practice damage resistance will make it less effective. Still, it's a joint-first place for those two weapons on the WMD/s front.
-Worth noting: the penitent cannon theoretically out-damages the Chimeric Graviton cannon against capital ships. It's 2WMD/s compared with 1.5. Wow...

Also immediately obvious: the Mark I howitzer sucks. It's out-performed by the supposedly worse AK505.

In the wider picture though, we need to work out what's actually wanted for WMD. Right now, it's a necessity. Internal compartments are really, really tough, and you basically want as high a WMD-rated weapon as you can get. Making the weapons balanced against internal HP that trends upwards by level would mean making a much simpler graph with a far clearer (and more linear) trend. However, this would reduce the value of howitzers and make a lot of weapons very samey. I don't really have a firm suggestion (although I'm tending towards thinking that WMD should be a "bleedthrough" effect: things take normal damage from everything, but WMD weapons do their WMD damage through armor to the compartment underneath what they hit....so they kill things faster but without crippling the other weapons), but it's clear that WMD-average and WMD/s needs to become a more common measurement of weapon balance, both in the community and in the vanilla game.

Edit: Oh, and if anyone wants my spreadsheet, or wants me to post the raw numbers then I can obviously do that.
Edit2: Tweaked my assumptions to be more accurate, fixed a problem with the AK505 (WMD was set to 1 in my sheet, not 5). Thanks to NMS for pointing this out.

[NMS]

Overall, that looks similar to what I'd expect. I'll have to get the rest of my data entered so I can compare my results.

There are a few weapons that seem to be missing:
- The AK505 (level 4 kinetic) does 42.73 compartment DPS.
- If you assume that each particle from a particleCloud weapon does at least 1 compartment damage, the PK25 and Ion-9 both do 187.5 compartment DPS. (I'm confident this is not true.)

[Song]

Overall, that looks similar to what I'd expect. I'll have to get the rest of my data entered so I can compare my results.

There are a few weapons that seem to be missing:
- The AK505 (level 4 kinetic) does 42.73 compartment DPS.
- If you assume that each particle from a particleCloud weapon does at least 1 compartment damage, the PK25 and Ion-9 both do 187.5 compartment DPS. (I'm confident this is not true.)

Yeah. I'm referring more to the "damaging" particles. I'm still unsure as to how the game actually handles particle hit detection.....it's a strange beast. But since the raw numbers are coming out of the game there's a reasonable chance they're accurate anyway (need to edit that assumption though).


The AK505 was there, but a typo in my formula sheet meant it was getting reworked for WMD1, not WMD5. This may have happened elsewhere (there's no way I've found to export WMD ratings, so it has to be done by hand, which is prone to error), so point out any other outliers that are missing.

Edit: OP is updated, refresh your cache if you're still seeing the old one.
Edit: Also it's impressive that the AK505 is the best of all the WMD weapons up until level six, and still beats all-bar-two weapons out of level seven.

[Song]

Digdug noted in IRC that the trend for this may actually be exponential. And that's likely to be correct...because the game currently has low WMD at low levels (despite the huge numbers of multihulled stations there for no good reason), and higher WMD as you go up through the levels. This results in a multiplier that (probably. I haven't actually tested this hypothesis) explains the exponential growth in damage.


I'd suggest that this is altered in some way to make the balance more linear (as most weapons are at least theoretically balanced to a linear equation). THis would mean granting WMD values based on the type of weapon as a uniform thing (possibly using a table to work out what value is appropriate for modders/creators similar to the weapon balance tables?), rather than the current system of WMD being loosely based on the function of thw weapon and its level, which leads to the current issues. This would also reduce the problem of weapons lacking WMD, as more weapons would gain low levels of WMD. As a downside, many wrecks would likely die for this change and a lot of tuning would be required.

[Atarlost]

My inclination is to set WMD solely by the role of a weapon with level not factoring in at all.

[Song]

My inclination is to set WMD solely by the role of a weapon with level not factoring in at all.

I think this has to be the case. And it needs to use the full spectrum. It's an easy trap to fall into (heck, looking at SM&M++....I've done it as well) to make a series for howitzers and just buff the WMD a little bit as you go up, just to make them definitely better. But doing so creates this problem of an out of control climb.


...of course, not everything in Transcendence does actually scale linearly. But the scaling on WMD seems.....too heavily weighted to those few outliers at the top end.


(Particularly since a lot of streamers have noted how damned long it takes to kill early-game stations with low level weapons. Which is entirely down to multihull getting in the way.)

[JohnBWatson]

My inclination is to set WMD solely by the role of a weapon with level not factoring in at all.

I'd agree with this. The rule I think would fit best would be for WMD to be highest for the weapons that are least effective against gunships, either due to unwieldiness or pricey ammunition. This both makes weapon slots more useful(due to more diverse weapon roles) and give people a reason to use some of the more neglected weapons. It also creates a choice between using safe weapons like howitzers, the Lancer, or the Flenser, but potentially having to fight some reinforcements, and using mags/missiles, but having to get in range of a station's defenses or waste valuable ammo.

[Song]

My inclination is to set WMD solely by the role of a weapon with level not factoring in at all.

I'd agree with this. The rule I think would fit best would be for WMD to be highest for the weapons that are least effective against gunships, either due to unwieldiness or pricey ammunition. This both makes weapon slots more useful(due to more diverse weapon roles) and give people a reason to use some of the more neglected weapons. It also creates a choice between using safe weapons like howitzers, the Lancer, or the Flenser, but potentially having to fight some reinforcements, and using mags/missiles, but having to get in range of a station's defenses or waste valuable ammo.

This does run into the problem of gunships generally being quite weak against howitzers anyway (due to their low HP total and the fact that slow-firing weapons aren't actually that slow in this game)....but I think this ease VS hitting power relationship is a good one to develop. Particularly if we also look at the velocity of projectiles....I know that kinetics are often neglected because of their slow slugs (also: excessive kinetic resistance in the early game), but it might be worthwhile to eventually look at making some other weapons have this same drawback. But that's a different matter that isn't really an immediate concern.

Having WMD tied to the nature of the weapon also improves the value of rare or hard-to-craft heavy weapons like the Dual Mark III that take a low rate of fire weapon and improve on it in some way...they get a bonus because they're essentially the same as a heavy howitzer, but perform better against gunships due to modifications (or vice versa if we ever get a crafted weapon that does the opposite).

Technically speaking, if you ignore damage resistance (which is the big thing my calculations do not look at), the AK505 is competitive out to at least level 7. But it's a very hard gun to use, which makes it more or less balanced unless you're a very good player who can accurately land hits on the same spot consistently. So there's definitely credit in this.

[Atarlost]

One caveat to making WMD not scale, low WMD is not relevant and should not be used.

Code: Select all

* WMD0 = 0%
* WMD1 = 4%
* WMD2 = 10%
* WMD3 = 20%
* WMD4 = 34%
* WMD5 = 52%
* WMD6 = 74%
* WMD7 = 100%

The roundoff is so terrible that WMD should actually decline with level.

Light WMD that practically every single weapon in the game needs under the current internal structure system would be the setting such that the weapon does 2 damage about half the time. On a laser cannon that's WMD5. On a Bolide it's WMD4. On a turbolaser it's WMD3.

Medium WMD for offsetting a slow projectile or poor damage type or on a fast, projectile that's supposed to base bust on the side (eg. the flenser) is probably around double that.

Howitzers should probably be WMD5. Missiles should probably be WMD6. Really dedicated base busters like the 3k siege MAG should be WMD7.

WMD0 should be reserved exclusively for safety weapons like ore scanner pulse thingies and weapons that substitute something else like particles, repeating, very high RoF, passthrough, radiation, or possibly device disruption for WMD. Actually, I'm not sure WMD has any effect at all on particles weapons. Possibly they may as well be WMD7 so they display as close to possible to their real damage.

The heavy hitters don't decline with level because >50% damage is always relevant, but the light base busters do because you need an awful lot of damage before WMD1 is any different from WMD0.

[JohnBWatson]

This does run into the problem of gunships generally being quite weak against howitzers anyway (due to their low HP total and the fact that slow-firing weapons aren't actually that slow in this game)

That's definitely an issue. I've had the idea of further reducing their fire rate in exchange for a light radius effect and higher damage. This would keep DPS the same, and make them more useful against capital ships with lots of armor segments, but reduce their effectiveness as a dogfighting weapon.

If the kiting/AI range exploit isn't fixed, this could also be taken further, nerfing howitzer range to weaken the exploit in exchange for even higher damage.

....but I think this ease VS hitting power relationship is a good one to develop. Particularly if we also look at the velocity of projectiles....I know that kinetics are often neglected because of their slow slugs (also: excessive kinetic resistance in the early game), but it might be worthwhile to eventually look at making some other weapons have this same drawback. But that's a different matter that isn't really an immediate concern.

I'd agree with this. A lot of kinetic weapons are not competitive with the basic laser cannon simply because their mildly higher DPS just doesn't compare with the drop in accuracy and ease of use. They definitely need some more love, but passing them over is just second nature to a lot of players at this point.

[Atarlost]

A lot of kinetic weapons are not competitive with the basic laser cannon simply because their mildly higher DPS just doesn't compare with the drop in accuracy and ease of use. They definitely need some more love, but passing them over is just second nature to a lot of players at this point.

This is not a universally held view. I usually pass over level 1 or 2 weapons of any type, but the one exception is kinetic. Partisans and Arbalests are both entirely usable at level 3. At level 4 there are three out standing kinetics and no out standing lasers. At level 5 the dual flenser is more flexible than the X-ray laser.

You are obsessed with projectile speed, but you are not the whole of the player base.

[Song]

A lot of kinetic weapons are not competitive with the basic laser cannon simply because their mildly higher DPS just doesn't compare with the drop in accuracy and ease of use. They definitely need some more love, but passing them over is just second nature to a lot of players at this point.

This is not a universally held view. I usually pass over level 1 or 2 weapons of any type, but the one exception is kinetic. Partisans and Arbalests are both entirely usable at level 3. At level 4 there are three out standing kinetics and no out standing lasers. At level 5 the dual flenser is more flexible than the X-ray laser.

You are obsessed with projectile speed, but you are not the whole of the player base.

Moderator warning: Let's avoid the personal attacks here.

You're largely missing the point that's being made: that projectile speed is a balancing tool, and factors into the "ease of use" that a weapon has. Watson strayed a bit far from it, but that's still the underlying thing here. Opinion is generally divided in the community about kinetic shot speed, with plenty of people on both sides. Personally I consider it a massive hinderance to the lighter guns, but generally fair to the heavier ones. Except the flenser, which is ridiculous....thus why I used it as an example. Very few views are universal, and the fact that people are not unanimous about something is not an argument against it. There is good discussion to be had, but this is not the way to have it.


More important, since this isn't the point of the thread (WMD is), we can move on and leave this alone and keep things nice and civil.

[Atarlost]

Some people disregard the experiences of others. Possibly because Transcendence is a single player game with no skilled LPs. Sometimes weapons are bad and everyone agrees they're bad. Like the GDC. Sometimes, though, vocal people are just lousy at using them. This can lead to buffs that make them overpowered for people who can use them. If you buff "difficult" weapons until even people who can't use them feel they need to try you've made them overpowered for those who can use them.

Projectile weapons already have increased damage over beam weapons to offset their lower shot speed. They don't need more except the WMD to effect internal structure that everything needs.

[gunship256]

Ease of use in various situations may be enough to justify a weapon's low or zero WMD rating. Weapons have some useful characteristics that are not all WMD-related. For example, this is what I'm typically looking for in a complete set of weapons:

1. Compartment damage at long or short range (missiles; WMD weapons; rapid-fire weapons; weapons with a particle, frag, or area effect)

2. Long-range base busting (missiles, howitzers, lancer cannon)

3. Shield against incoming missiles (autodefense turrets; weapons with a particle, frag or area effect; dual/alternating configurations; fast-firing weapons with some shot inaccuracy [don't exist?])

4. Omni or tracking capability for pesky gunships and for running away

Missiles or howitzers can take care of #1 and 2, which both involve the ability to damage compartments. They're not great at #3 and 4, either because they use ammo or because they can't track.

This is where kinetic weapons are really useful, even when they don't have any WMD. SmartCannon rounds can track. The dual flenser can block missiles, and the MorningStar blocks missles AND does compartment damage. The Ballista works for #1 but not #2 or 3, so even though it has WMD, it's actually an inferior choice to the MorningStar for the purposes of the list above.

Lasers don't have WMD or special configurations, but their fast fire rate, fast shot speed, and relatively long range make them okay at early-game base-busting (#2), something that the level 1-3 kinetics aren't great at.

Using this classification, there are a few weapons that stick out as being pretty bad. In particular, the late-game ion blasters can't do any of the four things above, so I almost never install them. The slam cannon is pretty bad too: long range, a slow fire rate, and low WMD makes it a poor choice for #1-4.

I usually use three or four slots for weapons because I can't fit everything I need onto one axis. It's hard to curve-fit in a five-dimensional space with such a small number of data points, so it might not be possible to make the entire universe verifiably balanced. I agree, though, that energy weapons in mid to late game need some kind of buff because of their lack of WMD.

[Atarlost]

Ease of use in various situations may be enough to justify a weapon's low or zero WMD rating. Weapons have some useful characteristics that are not all WMD-related. For example, this is what I'm typically looking for in a complete set of weapons:

1. Compartment damage at long or short range (missiles; WMD weapons; rapid-fire weapons; weapons with a particle, frag, or area effect)

2. Long-range base busting (missiles, howitzers, lancer cannon)

3. Shield against incoming missiles (autodefense turrets; weapons with a particle, frag or area effect; dual/alternating configurations; fast-firing weapons with some shot inaccuracy [don't exist?])

4. Omni or tracking capability for pesky gunships and for running away

Missiles or howitzers can take care of #1 and 2, which both involve the ability to damage compartments. They're not great at #3 and 4, either because they use ammo or because they can't track.

This is where kinetic weapons are really useful, even when they don't have any WMD. SmartCannon rounds can track. The dual flenser can block missiles, and the MorningStar blocks missles AND does compartment damage. The Ballista works for #1 but not #2 or 3, so even though it has WMD, it's actually an inferior choice to the MorningStar for the purposes of the list above.

Lasers don't have WMD or special configurations, but their fast fire rate, fast shot speed, and relatively long range make them okay at early-game base-busting (#2), something that the level 1-3 kinetics aren't great at.

Using this classification, there are a few weapons that stick out as being pretty bad. In particular, the late-game ion blasters can't do any of the four things above, so I almost never install them. The slam cannon is pretty bad too: long range, a slow fire rate, and low WMD makes it a poor choice for #1-4.

I usually use three or four slots for weapons because I can't fit everything I need onto one axis. It's hard to curve-fit in a five-dimensional space with such a small number of data points, so it might not be possible to make the entire universe verifiably balanced. I agree, though, that energy weapons in mid to late game need some kind of buff because of their lack of WMD.

That's a pretty good classification of weapon roles.

The light recoilless cannon, trident, and arbalest are as good at killing small multihull stations through number of shots as their laser peers. The recoilless, heavy recoilless, and partisan are no worse than the Rasiermesser lasers. It's not a damage type division but a fire rate division. The only division along damage type lines is shot speed.

I'm of the opinion that all of them need WMD. The internal compartment mechanics are completely broken for weapons with no WMD. Alternately internal compartments can be removed or changed completely.

Roles 3 and 4 might not absolutely need WMD, but I suspect the freighter needs to fill roles 1 and 4 in the same weapon.