Study of Railgun Chance to Bypass Shields

Distant Worlds is a vast, pausable real-time, 4X space strategy game which models a "living galaxy" with incredible options for replayability and customizability. Experience the full depth and detail of large turn-based strategy games, but with the simplicity and ease of real-time, and on the scale of a massively-multiplayer online game. Now greatly enhanced with the new Universe release, which includes all four previous releases as well as the new Universe expansion!

Moderators: elliotg, Icemania

Post Reply
scbfromnc
Posts: 18
Joined: Sun Jul 22, 2007 9:54 pm

Study of Railgun Chance to Bypass Shields

Post by scbfromnc »

While testing some ship weapons using the game editor, I began to notice that ships equipped with rail guns were destroying other ships rather quickly. I remembered an advantage of rail guns is that they partially bypass shields and wondered about the frequency.

This quick study looked at the percentage chance for a shot to bypass shields of an enemy ship. I used the technique described by Icemania in his Weapon Balancing thread and designed custom ships as shown on the attached spreadsheet. The designs were added to the designTemplates/Customization folder. The target ship and my rail gun ship were identical except:
(1) my ship was equipped with a single rail gun instead of a missile
(2) my ship was equipped with a Trace Scanner so I could monitor the target ship
(3) my ship was equipped with 3 proton thrusters so it would overtake the missile ship
The ships were not equipped with warp engines so could not warp away.

I allowed the encounter to proceed until the target ship blew up. This occurred each time with about 80% or more of the target shields remaining. I counted the rail gun strikes against the missile ship (note that I didn't count shots that missed the target ship) and recorded each hit number that resulted in component damage. The final hit number represents the final strike where the missile ship blew up. I repeated this 20 times and the results are shown on the attached spreadsheet. I performed this study for Long Range Guns, Rail Guns, Heavy Rail Guns and Massive Rail Guns.

While shields are still intact, component damage only occurs when the shields are bypassed. So dividing the total number of components damaged by the number of strikes to achieve that provides the percentage chance to bypass shields. I'm not a statistician, though I play one on tv, so am not absolutely certain of the validity of this.

Results - percent chance that a shot will bypass shields:
LR Guns - 7.1%
Rail Guns - 8.5%
Heavy Rail Guns - 19.4%
Massive Rail Guns - 45.8%

Interestingly, the number of component damages resulting in the ship destruction was about the same for all rail guns -- five (7 for massive rail guns but that's still within about 1 std dev of the other guns).

Weapon stats on rail guns are not particularly impressive and many experienced players have said that they do not fair well in late game. As a relatively new player I don't have that experience to draw on, but shield bypass of rail guns could be effective is you've got a fast enough ship to get within range. I would appreciate any comments you have with late game rail guns.

I also would like to understand why these ships are blowing up so quickly. I thought armor would absorb the first hits of a shield bypass and protect internal components. These missile ships were equipped with 10 armor components (100 armor protection). So what's happening here and which component damage results in ship destruction -- the reactor?
Attachments
RailGuns.zip
(19.46 KiB) Downloaded 6 times
Aeson
Posts: 786
Joined: Fri Aug 30, 2013 7:36 pm

RE: Study of Railgun Chance to Bypass Shields

Post by Aeson »

I also would like to understand why these ships are blowing up so quickly. I thought armor would absorb the first hits of a shield bypass and protect internal components. These missile ships were equipped with 10 armor components (100 armor protection). So what's happening here and which component damage results in ship destruction -- the reactor?
If you want to understand why the target vessel dies so quickly, I would recommend swapping the sides - have the enemy ship be the railgun ship, and have your own vessel be the missile ship. Set both to point blank and never retreat in the designs, and they'll engage in a way that approximates the best-case scenario for a railgun vessel. Now you can watch how quickly the ship's armor breaks and how often a shot gets through the shields. As far as I know, there is no special damage bonus provided for hitting certain components, although destroying certain components has greater effect than destroying others - for example, destroying a reactor reduces the combat effectiveness of a ship by making it less able to power its guns, engines, and shields; destroying a command center, if I'm not mistaken, will disable a ship entirely regardless of what else has been lost unless there is another command center on the ship.

As for why railguns drop out of favor as the game progresses, I would suggest taking a look at the Guide to Armour thread (tm.asp?m=3600917) and then taking a look at armor through the various stages of the game. To make a long story short, the chance a weapon has to break armor is dependent on its ability to overcome the reactive rating of the armor. The reactive rating is essentially an additive debuff to the per-shot damage, with the magnitude of the debuff modified by the weapon's anti-armor effectiveness; afterwards, the chance to break the armor is dependent upon the weapon's damage at the time of impact.

If I've done my math right, the percentage per-shot chance for various types of railguns to break an armor plate are given below:

Code: Select all

        LRG.   RG.1   RG.2   RG.3   RG.4   HRG1   HRG2   HRG3   MRG1   MRG2
 ---------------------------------------------------------------------------
 S.A.   10.0   20.0   30.0   30.0   40.0   60.0   70.0   90.0   100.   100.
 E.A.   1.25   1.50   1.75   1.75   2.00   11.1   16.7   27.8   44.4   61.1
 R.A.   0.71   0.86   1.00   1.00   1.14   1.43   1.57   1.86   10.0   20.0
 UDA.   0.50   0.60   0.70   0.70   0.80   1.00   1.10   1.30   1.60   1.90
 

LRG stands for Long-Range Gun, RG for Railgun, HRG for Heavy Railgun, MRG for Massive Railgun, S.A. for Standard Armor, E.A. for Enhanced Armor, R.A. for Reactive Armor, and UDA for Ultra-Dense Armor, while the numbers after the abbreviation represent the stage of the weapon (so an RG.1 is the first unlocked railgun, whereas an RG.4 is the final upgrade of the railgun). The numbers given are the percentage chance for a hit by the weapon to penetrate and destroy the armor plate, calculated to the best of my ability given my understanding of the formulas in the Guide to Armour thread. As you can see, even the upgraded Massive Railgun is absolutely pitiful against end-game armor, and even the Long-Range Gun has a decent chance to break standard armor. Armor upgrades cripple railgun effectiveness as the game progresses, and this is one of the reasons why it is recommended that you stay away from these; blasters at extreme range may have worse chances of penetrating armor, but up close they'll perform far better. All railguns are additionally hampered by their lackluster range, which is the least of any other weapon in the game, and by the late-game this becomes fairly crippling - even phasers and blasters, the other two short-range weapons, have ~50% greater range throughout the game, and while this is reasonably acceptable early on (the 190 range of a Maxos Blaster I isn't a considerable advantage over the 120 range of a Railgun I), it's much less so in the late game when that translates to a difference of ~200 range units. You need a relatively greater advantage to close a ~200 range gap than you need to close a ~70 range gap, particularly since the DPS has gone up.

On top of this, railguns actually lose damage density as you go up the tech tree; a Massive Railgun offers only 0.32 (0.38 if upgraded) DPS per size unit dedicated to Massive Railguns, compared to a Railgun's 0.86 DPS per size unit as introduced (though the Massive Railgun has two significant advantages over the basic Railgun, namely greatly improved - but still rather short - range and significantly enhanced anti-armor performance), the Maxos Blaster I's 0.56 DPS per size unit at 150 range, or the Titan Beam II's 0.71 DPS per size unit at 450 range. So in addition to having a range shortcoming that gets progressively worse as the game advances and suffering severely against armor, they also fall behind on firepower density. Railguns are fairly exceptional weapons in the early stages of the game. They're not so hot once decent armor comes into play, and they really start losing out when weapon power and ranges start heading towards end-game levels. Beyond that, while at long range blasters will likely perform similarly poorly against armor as railguns do, mid-game and later blasters generally have higher rates of fire than equivalent-tech railguns, and in a situation where both weapons are dependent on a low-chance critical hit to break the armor, the rate of fire is a significant factor in which weapon is better.

(It's a bit harder to make a chart like this for blasters and torpedoes, as for these weapons the chance to break the armor is dependent on the range to the target. It can be done, but I'm not going to put up a chart like the one I gave for railguns as it's too much work to type up, and I don't feel like uploading a spreadsheet currently. If you would like to compare railgun anti-armor performance to the anti-armor performance of other weapons, I'd suggest reading through the Guide to Armour and working up a spreadsheet. Another thing that you can do is to compute the average number of shots required to obtain a certain chance of breaking the armor; the formula for this is

Code: Select all

     y = (ln(1 - x))/(ln(1 - n))
where y is the number of shots, x is the desired chance to break the armor, and n is the per-shot chance to break the armor; ln(a) is whatever your preferred logarithm happens to be.)
scbfromnc
Posts: 18
Joined: Sun Jul 22, 2007 9:54 pm

RE: Study of Railgun Chance to Bypass Shields

Post by scbfromnc »

Thanks Aeson. I took your suggestion and reversed the roles. The AI railgun destroyed components on my ship one by one -- and in some cases, two at a time. Armor was destroyed first and, only then, other components appeared to be destroyed at random. The ship was destroyed when the last component was destroyed. So this is what I expected to happen in my first study, OP. I have read the Guide to Armor thread, several times, and re-read again which confirmed my understanding per quote by Erik:

"Armor protects other components from damage. The only instance where other components should be damaged before armor is in the case of a hit from a gravitic weapon, which bypasses armor. Gravitic weapons damage components randomly once they hit and they can damage armor as well as normal components. No other weapon types completely bypass armor. By design, all other weapon types must first destroy armor before any damage occurs on other components."

So I still don't know what's going on in my first study when my rail gun ship blows up the AI missile ship after, on average, 5 components are damaged. Unfortunately you can't see inside an AI ship to look at its components. However the size of the AI missile ship was exactly what I expected, had the correct shield value, firepower and speed. So must be some disadvantage programmed into AI ships (this was on Normal Difficulty setting)?

And now I wonder if frequency of shield penetration is different with swapped sides. Will try that as soon as I get a chance.
Aeson
Posts: 786
Joined: Fri Aug 30, 2013 7:36 pm

RE: Study of Railgun Chance to Bypass Shields

Post by Aeson »

Another possibility is that the trace scanner may not be entirely accurate in its assessment of the damage dealt to your opponent, and in all honesty I would not be terribly surprised if it were the case that its accuracy leaves something to be desired. I would be more surprised to learn that it was a handicap. Also, if you want to take a look inside a computer-controlled ship, you can always pause the game, open up the editor, and change its allegiance, as long as it hasn't blown up yet. It's a test battle, so it's not like this is cheating the way it could be were you to do it during a fight against a computer opponent who wasn't just given a spawned-in ship, and you can always change the allegiance back if you want to do so.

The suggestion regarding reading the Guide to Armour thread was mostly in relation to understanding why railguns fall out of favor later in the game, not in regards to why the ships die as quickly as they seem to you to do in your tests.
scbfromnc
Posts: 18
Joined: Sun Jul 22, 2007 9:54 pm

RE: Study of Railgun Chance to Bypass Shields

Post by scbfromnc »

Thanks Aeson, that was handy info -- didn't know I could use the editor to make a ship switch teams. So I tried out your suggestion and took a look inside the ai ship as the battle was going on. The Trace Scanner is accurate -- good to know that. And the damage is done to ship armor. After about 5 armor components are damaged, on average, the ai ship is destroyed (even though the ship had 10 armor components as well as 23 other components). This average of 5 damaged components till destruction was consistent across three different races (on normal difficulty). At Extreme difficulty, the average actually dropped to about 3.5 components damaged upon destruction. Somewhat surprising that it would go down. But it was very apparent that the player ship accuracy went down at high difficulty level. Didn't try to quantitate this, but I'd guess about 75% of my shots missed at Extreme vs. about 25% or less at Normal.

Also want to mention that when I reversed roles with an ai ship having the rail gun and my ship was equipped with just a missile, something interesting happened. The rail gun damaged my armor components one by one after a random number of shots. But what I didn't expect was that, once all armor was gone, the other components were damaged on each subsequent shot. There was no more randomness to when components were damaged -- only to the order they were damaged. Since at least one shield was still intact and powered (at least one fuel cell and reactor for at least some period of time) throughout most of this, I didn't expect the shield penetration to be 100% for this last phase.

I think I'm done with this testing. The more questions I try to answer, the more I generate.
Post Reply

Return to “Distant Worlds 1 Series”