Print

Please login or register.

1 Hour 1 Day 1 Week 1 Month Forever

Login with username, password and session length

[Thiebs]

If SoB is calculated after the percentage reduction, the damage is improved by a few points, depending on SoB level and distance, but not a ton. It would be a much bigger impact on the knockback, though, since it would make hitting 14 damage much more common.

[Sylph]

I guess it shouldn't be impossible to test, given the fixed 10-hp of formers... Hmm...

If dropoff applies to SoB, then at 7 squares away (49% damage dropoff), a SoBx2 shotgun (18 average damage) should deal an average of 18*0.51=9.18dmg
If SoB applies after damage dropoff, then the same SoBx2 shotgun will deal an average of 16*0.51+2=10.16dmg
So all we have to do is shoot at former humans and sergeants from 7 tiles away, and see if they normally die, or normally survive.

I just ran 100 tests (savescumming and macroing was useful). 55 times out of 100, a shotgun with SoBx2 killed a former human at 7 tiles away. This suggests (although in no way confirms) that SoB damage doesn't suffer from shotgun damage dropoff, which has implications when it comes to ranged shotgun damage and armour reduction.
It also backs up my anecdotal evidence - the fact that SoB seems a surprisingly useful trait for shotgun use.

I'd encourage anyone else who is curious to join me in testing! I can't be bothered to shoot any more former humans at 7 tiles away with SoBx2 - my eyes have gone funny. :D


ps. If this is true, SoBx2 shotguns average 14 damage (double knockback) from 4 tiles away.
If not, then SoBx2 shotguns would average 14 damage from 3 tiles away.
Without any levels in SoB, a shotgun only averages 14 damage from 2 tiles away. (Bear in mind '1 tile away' is melee range.)
This difference, as Theibs suggested, is extremely important for cornershooting shotgun traps.

[Thiebs]

Very interesting. I'll try it out, too.

[Sylph]

Well if you do, just remember that a distance of 7 tiles actually means there are 6 empty tiles in between you and the former, and it has to be in a straight horizontal or vertical line!

Also worth mentioning that a FAR better test is 9 tiles away with SoGx5. That will almost never (<5%) kill a former if SoG suffers from dropoff, but should kill them over 60% of the time if SoG does not drop off.

[Thiebs]

That would be quite a bit harder to set up, but it also serves to point out how much of a difference it would make if it does drop off.

[thelaptop]

Urgh.  I hate doing this, but I have to.  I dived into the source code on the repo.

Shotgun damage is initialised as a dice roll object with number of dice, number of sides per dice, shotgun damage additions and bonus from SoB.  So, a shotgun with 8d3 and the trait of 2xSoB gives a dice roll object of 8d3+2 per enemy met in cone.

To determine the damage that an enemy gets from being hit by a shot of this shotgun, the dice roll object is rolled, discounted based on the drop-off and distance, and then rounded off.  If the resultant damage is less than 1, it gets set to 1.

The reason why it seems that SoB doesn't suffer from the drop-off is that more often than not, the stochastic part of the damage roll dominates the damage formula.  Also, the distance computation is wonky -- lots of rounding here and there and I have no idea how to decipher it properly.

tl;dr SoB won't hurt shotgunners, but the gain in DPS isn't as dramatic as regular rapid fire weapons.

Side note: Also I'm sure Sylph was referring to SoB and not SoG.

[LuckyDee]

While you're at it: is dropoff caluculated as (1-percentage)^distance or 1-percentage*distance?

[thelaptop]

Well, here's the whole line:

Code: [Select]

dmg := Round(Damage.Roll * (1.0-Reduce*Max(1,Distance(source,tc))))
You tell me.  =P

Aaand I'm no longer reading the source.

[LuckyDee]

Thanks. So my calculation should have been 18*(1-0.07*8) = 8 instead of 10 damage.

Although someone is probably going to tell me I did the math wrong, as usual ><

See? Told you so.

[Thiebs]

Something seems fishy here. Obviously the source code don't lie, but there seems to be something else afoot. Maybe something to do with the range is calculated and how targeting gets wonky, like was said. But I'm probably gonna do some more tests like Sylph did, anyways.

[LuckyDee]

Please bear in mind that:

1) This is the average damage. Testing 100 times might not be enough to reflect this properly.

2)

the stochastic part of the damage roll dominates the damage formula.

whatever that may mean real English. My statistics classes are too long ago (and too not in English) to grasp it. To myself, I'm translating this as "the theoretical average damage and actual average damage may not or do not correspond".

[Juice]

Ad 1) Probably true. There are statistical tests that could determine if it suffer from dmg dropoff or not, although I'd have to relearn my statistics to do it properly (and also increase the testing sample to maybe 10k :-D)

Ad 2) I'd say it means the random part of the calculation dominates the total damage sum.

[Thiebs]

Okay. Figured out why things seemed off. Sylph's test was at 7 spaces (which I didn't notice, I was thinking 8), meaning that you would see 10+ damage just over half the time, which matches the test. Which means that SoB 2 increases your odds of one-shotting a former by about 16% at max range. For knockback, I'm assuming armor DOES apply first, since the odds of getting at least 7 damage are quite good, (a roll of 15+, or about 74%) even without SoB. Which would mean that against an opponent with 2 armor (4 damage reduction vs. shrapnel), you would need 24 damage to score a knockback at max range, which is <1% with SoB at 0, 1, or 2, but jumps to 2.4% at SoB 3, and ~14% at SoB 5. Which is still nowhere near enough to take 5 traits. ;-P At point-blank range with 2 armor, the odds of a knockback are >99% to begin with, so no gain there. But a knockback 2 spaces from point-blank is 26% without; SoB 1 is 41.5%, 3 is 74%, and 5 is 93.5% and 6.5% of getting a triple. That's a little more interesting, though still pretty minor most of the time. I think the lesson to be gathered here is that SoB (of course) isn't meant to be used for shotguns, but if you're building rapid-fire, you can count on knockback with even a basic shotty in the prepared slot.

No promises on my math.

Oh, and if I had realized how in-depth this was going to be, I might not have put it on the quick questions thread. ;)

[Captain Mac]

Hey there, guise.

So. Angel of Darkness. The wiki tells me this: "Your vision radius is only 6 and anything outside of this radius is black, except for what maps reveal as well as Intuition. Scouts and maps do not know the location of stairs. Experience gain is doubled." I started it just today, on HNTR just to test the water.

Monsters are respawning. Should they be respawning? I don't think they should be respawning. Can anyone shed some light (haaaaaaaa.) on this for me?

[thelaptop]

Yes, and I'm merging this with Quick Questions.