Weapon balance stats

Post ideas & suggestions you have pertaining to the game here.
george moromisato
Developer
Developer
Posts: 2979
Joined: Thu Jul 24, 2003 9:53 pm
Contact:

Thu Jan 21, 2016 6:57 am

There's been some great discussion on weapon balance (http://forums.kronosaur.com/viewtopic.php?f=3&t=7419) and I thought I'd add some additional data. I wrote some quick code to benchmark each weapon as follows:

First we define the "benchmark time" of a weapon to be the number of ticks it takes (on average) for the weapon to pierce through a given armor segment. Next, we calculate the following values:

1. The "best armor" is the armor type that the weapon had the easiest time against (the shortest benchmark time).
2. The "worst armor" is the armor type that the weapon had the hardest time against (the longest benchmark time).
3. The average benchmark time of a weapon across all armor types of the same level as the weapon.

I calculated the values above for all weapons (not launchers/missiles) and ended up with this:

https://drive.google.com/file/d/0B3qNnR ... sp=sharing

Let's go through an example:

The first row is for a level 1 laser cannon. The average benchmark time is 39 ticks. This is the average benchmark time for all level 1 armors. That is, a laser cannon takes 39 ticks to fully pierce an average level 1 armor.

The best armor for the laser cannon is ultra-light titanium armor, which it can penetrate in 10 ticks (on average). The worst armor is light ceramic, which takes 100 ticks for the laser cannon to penetrate (this makes sense, since ceramic armor resists laser).

This data is very rough, so use with caution. For example, it doesn't take into account special damage effects (EMP, blinding, etc.) which would obviously affect the overall power of the weapon. Nevertheless, you can see some interesting results. For example, the TeV 9 blaster takes 108 ticks to pierce the average level 6 armor. In contrast, the mark III howitzer can do it in 82 ticks (almost 25% faster).

I still think your own intuition, honed by playing the game, is the best guide for weapon balance, but hopefully this table will add something to the discussion.

User avatar
Song
Fleet Admiral
Fleet Admiral
Posts: 2768
Joined: Mon Aug 17, 2009 4:27 am

Thu Jan 21, 2016 7:41 am

Cheers George. This should help with the numerical side of things.



......I now have an idea for something that might give us another piece of the balance picture, but I'll need to think about it (and maybe bother AP to see if it's possible).
Mischievous local moderator. She/Her pronouns.

User avatar
JohnBWatson
Fleet Officer
Fleet Officer
Posts: 1452
Joined: Tue Aug 19, 2014 10:17 pm

Thu Jan 21, 2016 10:27 am

Fantastic data. I observe a few things:

1. Looks like Dwarg Chiton armor and light Sung armor could use a level downgrade. Light Sung might work better with a buff. The repair ability of Sung armor is highly situational even for heavier variants, and is not sufficient to make such a low HP armor viable for the level that it is.

2. The difference in effective damage between kinetic and laser weapons is generally not sufficient to offset the increased difficulty of using the former effectively.

3. Dual recoilless could use a level decrease. Smartcannon, given its tracking abilities, could use a level increase.

4. OTL could use a nerf to bring its relative power in line with the omni laser.

5. Sabel 90 could use a buff. It's a slow firing, hard to use weapon. Right now, it's all but useless without the damage to make that pay off.

6. Dual Flenser/Flenser could use a significant range nerf. It's already an excellent DPS and WMD weapon that can destroy almost any projectile it encounters, it doesn't need the high range on top of that.

7. Slicer Cannon is far too weak, even without its inaccuracy. It's a unique weapon found at the end of a major questline with minimal use at range. It should hit a lot harder than it does.

8. Gusoku armor could use a level decrease. I've tested it, and it's definitely out of its range of usefulness.

9. The Kiloton may be long ranged, but it also does very spread out damage. It could use a buff to put it in line with the thermo cannon and make it a threat again.

10. Ion Flame cannon could use a buff, given that right now it's in line with the much more accurate yet still lackluster Ion Blaster.

11. Given that the Fusionfire/Hecates/Mark VII are roughly in line with their levels in terms of DPS, they could do to be given less range. At the moment, they're just plain better than the other fixed angle weapons of their level.

12. Omnidirectional ion blaster could use a buff to bring it in line with the other omnis.

13. Chimeric graviton cannon is a lot weaker than it ought to be for its level, but kills the player very easily due to showing up when the player's likely using level 9-10 equipment. Could do to be given less damage and higher WMD, and brought down a level.

Also, you may want to treat weapons like the Ares micronuke cannon as passthrough for computing things like this, given that they effectively are.

PM
Fleet Admiral
Fleet Admiral
Posts: 2400
Joined: Wed Sep 01, 2010 12:54 am

Thu Jan 21, 2016 2:05 pm

Smartcannon, as an ammo weapon, is fine; and if raised to level 4, it becomes much rarer in the New Beyond. Actually, level 4 devices are generally too rare for when they are useful (i.e., pre-Charon), and generally obsolete by St. K's where they are rarer than level 5 items. Keep in mind AK15 is the ammoless, non-tracking version of Smartcannon; using ammo is the price for tracking.

Omni turbolaser is a bit strong, but not overpowered. RK15 is same level and has even more DPS, at least until we use weapon enhancer device (+75% laser > +50% kinetic) and get swarmed by common Vikings (who resist kinetic).

Graviton is tier 6 damage, something we probably would not expect to see until level 16 (mid-to-late part 2). Chimeric graviton cannon should be a bit weak, and it is. If its DPS should be buffed, its damage type needs to be downgraded.
Download and Play in 1.9 alpha...
Drake Technologies (Alpha): More hardware for combat in parts 1 and 2!

Other playable mods from 1.8 and 1.7, waiting to be updated...
Godmode v3 (WIP): Dev/cheat tool compatible with D&O parts 1 or 2.
Star Castle Arcade: Relive classic arcade gaming in a new Transcendence adventure!
Playership Drones (WIP): Acquire more wingmen from Ship Broker. Unfinished, but updated to 1.8 beta.
Drake Technologies (Alpha) for 1.8: If you do not have 1.9 alpha 1...

User avatar
AssumedPseudonym
Fleet Officer
Fleet Officer
Posts: 1099
Joined: Thu Aug 29, 2013 5:18 am
Location: On the other side of the screen.

Thu Jan 21, 2016 4:37 pm

 Actually, looking at this reminded me of something I’ve been wondering about for a while and never gotten around to asking: How exactly is the 'damage property which tends to be my go-to balance metric, for better or for worse in typGetProperty/itmGetProperty/objGetItemProperty calculated, anyway?
Image

Mod prefixes: 0xA010 (registered) and 0xDCC8 (miscellaneous)

My mods on Xelerus: Click here!

Of all the things I’ve lost in life, I miss my mind the least. (I’m having a lot more fun without it!)

User avatar
pixelfck
Militia Captain
Militia Captain
Posts: 571
Joined: Tue Aug 11, 2009 8:47 pm
Location: Travelling around in Europe

Thu Jan 21, 2016 5:07 pm

While this is fantastic data, one very important variable is not included in these numbers, namely: can the weapon actually hit the armor.

High speed, many (long) projectiles score more hits than slow projectiles. Just like the omni-weapons score more hits on average.

~Pixelfck
Image
Download the Black Market Expansion from Xelerus.de today!
My other mods at xelerus.de

george moromisato
Developer
Developer
Posts: 2979
Joined: Thu Jul 24, 2003 9:53 pm
Contact:

Thu Jan 21, 2016 6:08 pm

pixelfck wrote:While this is fantastic data, one very important variable is not included in these numbers, namely: can the weapon actually hit the armor.

High speed, many (long) projectiles score more hits than slow projectiles. Just like the omni-weapons score more hits on average.

~Pixelfck
This is a really important point. The ideal benchmark value will vary depending on the characteristics of the weapon. As you said, omni weapons should have a higher benchmark value than non-omnis.

Think of this data as a variant on damage-per-second which takes armor resistance into account.

george moromisato
Developer
Developer
Posts: 2979
Joined: Thu Jul 24, 2003 9:53 pm
Contact:

Thu Jan 21, 2016 6:22 pm

AssumedPseudonym wrote: Actually, looking at this reminded me of something I’ve been wondering about for a while and never gotten around to asking: How exactly is the 'damage property which tends to be my go-to balance metric, for better or for worse in typGetProperty/itmGetProperty/objGetItemProperty calculated, anyway?
Good question--it's not that simple.

Semantically, the value should be interpreted as, "how many HP of damage is the weapon expected to do to a single target over a 180-tick period of time."

However, there are a lot of adjustments required to calculate:

1. We assume all shots of a multi-shot weapon hit (dual, repeating, etc.).
2. For shockwave and fragmentation weapons, we assume that only 1/8th of the fragments hit the target.
3. For radius damage, we assume the target is far enough away to take half damage.
4. For capacitor weapons, we adjust damage to account for the fact that some of the time the weapon can't fire.

User avatar
Song
Fleet Admiral
Fleet Admiral
Posts: 2768
Joined: Mon Aug 17, 2009 4:27 am

Thu Jan 21, 2016 7:23 pm

george moromisato wrote:
pixelfck wrote:While this is fantastic data, one very important variable is not included in these numbers, namely: can the weapon actually hit the armor.

High speed, many (long) projectiles score more hits than slow projectiles. Just like the omni-weapons score more hits on average.

~Pixelfck
This is a really important point. The ideal benchmark value will vary depending on the characteristics of the weapon. As you said, omni weapons should have a higher benchmark value than non-omnis.

Think of this data as a variant on damage-per-second which takes armor resistance into account.
Might be interesting to, at some future point, see about what the average damage resistance is per system. That'd probably take a rather hefty utility mod though.

I also support more qualitative balancing than the numerical work....there's all sorts of things that impact how good a weapon is, and this data is just a (useful) part of that.
Mischievous local moderator. She/Her pronouns.

User avatar
JohnBWatson
Fleet Officer
Fleet Officer
Posts: 1452
Joined: Tue Aug 19, 2014 10:17 pm

Thu Jan 21, 2016 8:39 pm

PM wrote:Smartcannon, as an ammo weapon, is fine; and if raised to level 4, it becomes much rarer in the New Beyond. Actually, level 4 devices are generally too rare for when they are useful (i.e., pre-Charon), and generally obsolete by St. K's where they are rarer than level 5 items. Keep in mind AK15 is the ammoless, non-tracking version of Smartcannon; using ammo is the price for tracking.
Given the ubiquity, weight, and price of Smartcannon ammo, as well as the vast increase in real DPS provided by being able to fire without a target in the crosshairs, I'd say it might be worth the level change anyways.
Omni turbolaser is a bit strong, but not overpowered. RK15 is same level and has even more DPS, at least until we use weapon enhancer device (+75% laser > +50% kinetic) and get swarmed by common Vikings (who resist kinetic).
Eh, it's much more accurate than the '15, and isn't anywhere near as heavily resisted. Given its much lower poweruse, it's overall a much better weapon than the omni particle cannon, which seems a bit excessive. From extensive playtesting, I'd say its damage could use a significant nerf.
Graviton is tier 6 damage, something we probably would not expect to see until level 16 (mid-to-late part 2). Chimeric graviton cannon should be a bit weak, and it is. If its DPS should be buffed, its damage type needs to be downgraded.
My reasoning is that it should be less deadly to the player(and other gunships, by extension) than it is now, but should still be useful to those who acquire it, given the difficulty of doing so. Giving it significantly lower raw damage, but high enough WMD to hit stations and capships, seems like it would work out well for that goal.

pixelfck wrote:While this is fantastic data, one very important variable is not included in these numbers, namely: can the weapon actually hit the armor.

High speed, many (long) projectiles score more hits than slow projectiles. Just like the omni-weapons score more hits on average.

~Pixelfck
My set of observations tries to apply that to the data presented. I think, overall, long ranged weapons that can strike their targets with impunity need a bit of a nerf, and weapons with slower, fixed angle shots need to be made more powerful in order to give them some degree of appeal.

User avatar
Atarlost
Fleet Admiral
Fleet Admiral
Posts: 2391
Joined: Tue Aug 26, 2008 12:02 am

Thu Jan 21, 2016 9:39 pm

JohnBWatson wrote:Fantastic data. I observe a few things:

2. The difference in effective damage between kinetic and laser weapons is generally not sufficient to offset the increased difficulty of using the former effectively.

4. OTL could use a nerf to bring its relative power in line with the omni laser.

6. Dual Flenser/Flenser could use a significant range nerf. It's already an excellent DPS and WMD weapon that can destroy almost any projectile it encounters, it doesn't need the high range on top of that.

11. Given that the Fusionfire/Hecates/Mark VII are roughly in line with their levels in terms of DPS, they could do to be given less range. At the moment, they're just plain better than the other fixed angle weapons of their level.
4) It's the omni laser that's out of band. The OTL is closer to the RK-15 and is constructed in relation to its base version the same way as most other omnis. The omni laser is constructed from a level 1 weapon with the customary fire rate penalty and should be level 2.

6) The flenser is a key utility for the Freyr and an important "not die to the primary Charon Fortress" option, particularly for the EI500 since it can't really risk using a slot on a missile launcher that can run out of ammo.

11) Range is the defining characteristic of howitzers. Nerfing that would make them completely pointless.
Literally is the new Figuratively

NMS
Militia Captain
Militia Captain
Posts: 536
Joined: Tue Mar 05, 2013 8:26 am

Thu Jan 21, 2016 10:47 pm

Looks like it's not taking overheating into account. The Advanced Tritium cannon should take 2 or 4 times as long, depending on whether counterUpdateRate is in ticks or game seconds. And it's assuming only a fraction of tracking fragments hit, when it would be more accurate (or more in line with the assumptions for other weapons, at least) to assume all of them hit.

It would also be useful to see how weapons perform against a basket of defenses that includes appropriate shields and compartment hp for each level.

george moromisato
Developer
Developer
Posts: 2979
Joined: Thu Jul 24, 2003 9:53 pm
Contact:

Fri Jan 22, 2016 12:02 am

BenchmarkChart.jpg
BenchmarkChart.jpg (168.41 KiB) Viewed 1908 times
Attached is a graph version of the table with selected weapons labeled. Note that most of the weapons cluster around 100 ticks to destroy an armor segment. Omni weapons cluster around 250 ticks, but some (like omni ion blaster) are clearly underpowered. Ammo weapons seems to cluster around 50 ticks.

Edit: I should also say that the data and chart come from 1.7 Alpha, which tweaks a couple of weapons. In particular the Lamplighter has been bumped up to 7d12 and the Qianlong to 6d12. Even at those levels, though, they might need more of a bump.

PM
Fleet Admiral
Fleet Admiral
Posts: 2400
Joined: Wed Sep 01, 2010 12:54 am

Fri Jan 22, 2016 1:32 am

JohnBWatson wrote:Given the ubiquity, weight, and price of Smartcannon ammo, as well as the vast increase in real DPS provided by being able to fire without a target in the crosshairs, I'd say it might be worth the level change anyways
I forgot to mention that level 4 Smartcannon would also be inferior to RK15 since the latter is ammoless. SmartCannon rounds are still ammo, which has all of the issues to go with it, even if the ammo is about as cheap and common as NAMI missiles. Level 4 Smartcannon would also mean I probably will not find one unless I find a Rasiermesser station before St. K's or anytime after St.K's where I want level 5+ weapons.
JohnBWatson wrote:Eh, it's much more accurate than the '15, and isn't anywhere near as heavily resisted. Given its much lower poweruse, it's overall a much better weapon than the omni particle cannon, which seems a bit excessive. From extensive playtesting, I'd say its damage could use a significant nerf.
If anything, Omni particle cannon (and most particle weapons for that matter) is underpowered because too many midgame enemies resist particle type and some reflect it. Only because midgame often features lots of slightly upgraded early-game enemies (mostly Outlaws, Miners, and Marauders) that particle is useful to some extent.
Download and Play in 1.9 alpha...
Drake Technologies (Alpha): More hardware for combat in parts 1 and 2!

Other playable mods from 1.8 and 1.7, waiting to be updated...
Godmode v3 (WIP): Dev/cheat tool compatible with D&O parts 1 or 2.
Star Castle Arcade: Relive classic arcade gaming in a new Transcendence adventure!
Playership Drones (WIP): Acquire more wingmen from Ship Broker. Unfinished, but updated to 1.8 beta.
Drake Technologies (Alpha) for 1.8: If you do not have 1.9 alpha 1...

User avatar
KaoS
Miner
Miner
Posts: 34
Joined: Sun Aug 19, 2012 3:25 pm
Location: 404
Contact:

Fri Jan 22, 2016 2:26 am

I have to ask, is weapon damage calculated before or during use? It has always been my understanding that a weapon with 1d6 will have a damage output of between 1 and 6, depending on what the RNG comes up with at the time of use. Is that true, or is it calculated once at load time and fixed forever after?

Post Reply