Issue with the Crit "Clamping" introduced in 1.2

Post Reply
awalwang
Posts: 1
Joined: Mon Feb 27, 2017 3:44 am

Issue with the Crit "Clamping" introduced in 1.2

Post by awalwang »

On 1.0 and 1.1, I'd maybe have a 145% critical chance total after all bonuses on say, a ranger with a shotgun. Say the enemy is a Naga Serpens that reduces crit chance by 20% due to dodge.

Before 1.2 that meant a 125% critical chance, a guaranteed critical hit.

After 1.2, I noticed that my critical chance was being reduced to 100% first, and then the dodge critical chance penalty applied after that, giving me 80% critical chance in the end.

Is this just a bug on my end (perhaps I should just do a clean install of the mod)?

I'll probably try to reinstall the mod when I get off work and provide you with any other information then, should you need any.

Edit: Got some screenies of what I'm talking about. Set up a theoretical situation with all possible bonuses for my Assault.

https://i.imgur.com/Tur45pU.jpg

https://i.imgur.com/RORZYGx.jpg

Not just a UI bug, as shown here:
https://i.imgur.com/d8mA7LL.jpg

I should have 135% crit; 170% before the penalty is applied, 135% after.

Right now it's clamping my crit bonus to 100%, and then deducting the crit penalty from dodge.

Long War 2 is the only mod I have active at this time.
sectoidfodder
Posts: 21
Joined: Sat Jan 28, 2017 6:00 am

Re: Issue with the Crit "Clamping" introduced in 1.2

Post by sectoidfodder »

I'm going to argue against this being a bug. You can't be any more successful than all of the time in your crit promotions, so as long as the target has any dodge it makes sense that 100% crit chance isn't possible. Crit already reasonably counters dodge in 1.2, just not to the extent that OP expected. The result of those two mechanics countering each other should be the standard "hit" result, and OP still has 100% chance between hit and crit, though what he seems to want is to allow excessive crit to overwhelm dodge to the point where it doesn't exist.

The argument would make sense if promote/demote were one combined roll based on (crit - dodge), but it isn't.
TerribleName
Posts: 10
Joined: Sat Feb 04, 2017 6:23 am

Re: Issue with the Crit "Clamping" introduced in 1.2

Post by TerribleName »

Does anyone know if the roll is clamp(crit-dodge,0,100) or clamp(crit,0,100)-clamp(dodge,0,100)?

If its the former thats fine, everything is great. More crit overcomes enemy dodge and things work how the player expects. If its the latter thats not cool at all and it needs to be communicated to the player that crit over 100 is worthless and dodge trumps everything. Real work needs to be put into the shot wings and tool tips.
sectoidfodder
Posts: 21
Joined: Sat Jan 28, 2017 6:00 am

Re: Issue with the Crit "Clamping" introduced in 1.2

Post by sectoidfodder »

TerribleName wrote:Does anyone know if the roll is clamp(crit-dodge,0,100) or clamp(crit,0,100)-clamp(dodge,0,100)?

If its the former thats fine, everything is great. More crit overcomes enemy dodge and things work how the player expects. If its the latter thats not cool at all and it needs to be communicated to the player that crit over 100 is worthless and dodge trumps everything. Real work needs to be put into the shot wings and tool tips.
Neither. My whole point was that they're two separate rolls:

Start with 100% hit

Roll 100+% crit clamped between 0 and 100

You now have 100% crit

Roll 35% dodge clamped between 0 and 100

You now have 65% crit and 35% normal hit

The opposite example:

Start with 100% hit

Roll 35% crit clamped between 0 and 100

You now have 35% crit and 65% normal hit

Roll 100+% dodge clamped between 0 and 100

You now have 35% normal hit and 65% graze
Sporadix
Posts: 29
Joined: Wed Feb 01, 2017 9:27 pm

Re: Issue with the Crit "Clamping" introduced in 1.2

Post by Sporadix »

Two separate rolls, but crit/dodge over 100% is still worthless for the sake of these calculations it seems.

I think this is a pretty glaring problem, because the calculations are presented to the player in the UI the same way that hit chances are presented: with additive and subtractive modifiers, and a net result. But base hit chance DOES have value over 100 because it can overcome defensive modifiers plus the graze band.

You can see how this could be very confusing to players.
TerribleName
Posts: 10
Joined: Sat Feb 04, 2017 6:23 am

Re: Issue with the Crit "Clamping" introduced in 1.2

Post by TerribleName »

sectoidfodder wrote:
Neither. My whole point was that they're two separate rolls:

Start with 100% hit

Roll 100+% crit clamped between 0 and 100

You now have 100% crit

Roll 35% dodge clamped between 0 and 100

You now have 65% crit and 35% normal hit
So what youre saying is that its clamp(crit,0,100)-clamp(dodge,0,100), exactly the case I said was not great because it misrepresented information to the player and makes stacking crit worthless.

Roll 100+% crit clamped between 0 and 100 = clamp(crit,0,100)

Roll 35% dodge clamped between 0 and 100 = clamp(dodge,0,100)

You now have 65% crit and 35% normal hit = clamp(crit,0,100)-clamp(dodge,0,100) for max crit chance.

So exactly the thing that I said previously but you disagreed with and then went on to show was actually the case.
Sporadix
Posts: 29
Joined: Wed Feb 01, 2017 9:27 pm

Re: Issue with the Crit "Clamping" introduced in 1.2

Post by Sporadix »

So what youre saying is that its clamp(crit,0,100)-clamp(dodge,0,100), exactly the case I said was not great because it misrepresented information to the player and makes stacking crit worthless.

Roll 100+% crit clamped between 0 and 100 = clamp(crit,0,100)

Roll 35% dodge clamped between 0 and 100 = clamp(dodge,0,100)

You now have 65% crit and 35% normal hit = clamp(crit,0,100)-clamp(dodge,0,100) for max crit chance.

So exactly the thing that I said previously but you disagreed with and then went on to show was actually the case.
Technically it's not clamp(crit,0,100) - clamp(dodge,0,100).

It's actually something like clamp(crit,0,100)*(1-(clamp(dodge,0,100)) as far as determining your chance to crit because you're effectively making two independent rolls that apply simultaneously. They're joint probabilities, the values are not combined additively.

Still, crit/dodge over 100% has 0 value and it's not communicated to the player in any way, so we're totally in agreement there.
nmkaplan
Posts: 207
Joined: Thu Jan 26, 2017 5:14 pm

Re: Issue with the Crit "Clamping" introduced in 1.2

Post by nmkaplan »

Right, it's only subtractive when crit chance = 100%.
Jadiel
Posts: 214
Joined: Tue Jan 24, 2017 9:28 am

Re: Issue with the Crit "Clamping" introduced in 1.2

Post by Jadiel »

So the patch notes say that negative dodge is transformed into extra crit chance. Is negative crit turned into additional dodge? Seems like it probably shouldn't be... It would be nice to have an indication of how it's meant to work, so we can see if it's bugged or WAD.

Likewise, is dodge/crit over 100 intended to be worthless? Or should excess dodge remove crit, and vice/versa?
bountygiver
Posts: 74
Joined: Mon Feb 27, 2017 3:33 am

Re: Issue with the Crit "Clamping" introduced in 1.2

Post by bountygiver »

Negative crit is clamped, otherwise resilience will produce extra dodge.

Still, I agree excess crit should start eating away dodge.
Post Reply