Resist system

[Jungle_Boy]

I was thinking about trying to come up with a more universal, or at least easier to understand, resistance system. Do people think this is a worthwhile thing to look at? If so can someone point me to where I can find the max attack for each element and the current reductions due to resistance?

[buzzkill]

I take it that you're already familiar with FA/O style cumulative resists.

[fizzix]

I was thinking about trying to come up with a more universal, or at least easier to understand, resistance system. Do people think this is a worthwhile thing to look at? If so can someone point me to where I can find the max attack for each element and the current reductions due to resistance?

max attack for elements can be found in list-mon-spells.h

for the resistances you have to piece them together from: constants.h and list-gf-types.h

[dhegler]

Why don't the resistances be some % resistance from 10% to 50%, and then immunity, and make them cumulative? That way you can have 4 items that resist fire, etc and it may make some things more relevant if you make the resists cumulative...

[Jungle_Boy]

I take it that you're already familiar with FA/O style cumulative resists.

I am familiar with the fact that they have cumulative resists and if I remember right they are all percent based and items can have varying levels of resist foo.

What I am thinking of is a simpler system that would not involve changing items at all. At least for the basic four something like this, first resist works as now, second resist blocks half remaining, temp resist blocks 1/6 remaining. This works out to the same as double resisting currently, there is just an extra level of resist in the middle. this might work for the basic four but I have no numbers for the others so not sure how well it would do there. I'm not even sure it is possible to come up with a simple system that works for all elements without changing resists on items but I thought I might look at it just to see. Plus I love playing with numbers.

max attack for elements can be found in list-mon-spells.h

for the resistances you have to piece them together from: constants.h and list-gf-types.h

Crap, I was hoping it would be in a nice edit file. guess I'll have to download the source.

[Timo Pietilä]

Why don't the resistances be some % resistance from 10% to 50%, and then immunity, and make them cumulative? That way you can have 4 items that resist fire, etc and it may make some things more relevant if you make the resists cumulative...

There is a problem with that approach that you easily either get too many resists, or single one isn't enough and you don't find more. I prefer not to hunt more than one resist for each of the ones I need.

Resistances as damage reduction are pretty much irrelevant except for basic four and poison. We could as well remove them completely for rest of the elements and replace them with side-effect prevention and leave basic four & poison as they are.

[Jungle_Boy]

Here's a little something I whipped up this evening. Should be pretty easy to code and test but I don't have a compiler.

Current resistance system:

Element	Max	Divisor	    Res Adj	Dbl Res Adj	Vuln Adj		Res damage		Dbl Res damage	Vuln damage
Acid	1600	3	(dam+2)/3	(dam+8)/9	(dam*4)/3		534		178.667	         2133.333
Elec	1600	3	(dam+2)/3	(dam+8)/9	(dam*4)/3		534		178.667	         2133.333
Fire	1600	3	(dam+2)/3	(dam+8)/9	(dam*4)/3		534		178.667	         2133.333
Cold	1600	3	(dam+2)/3	(dam+8)/9	(dam*4)/3		534		178.667	         2133.333
Poison	800	3	(dam+2)/3	(dam+8)/9			        267.333		89.778	
							                                   Min res damage	Average Max res damage	
Nether	550	6	(((dam) * 6) / (damcalc(1, 6, dam_aspect) + 6))				471.429	347.368	275	
Light	400	6	(((dam) * 4) / (damcalc(1, 6, dam_aspect) + 6))				228.571	168.421	133.333	
Dark	400	6	(((dam) * 4) / (damcalc(1, 6, dam_aspect) + 6))				228.571	168.421	133.333	
Sound	500	6	(((dam) * 5) / (damcalc(1, 6, dam_aspect) + 6))				357.143	263.158	208.333	
Chaos	500	6	(((dam) * 6) / (damcalc(1, 6, dam_aspect) + 6))				428.571	315.789	250	
Disench	500	6	(((dam) * 6) / (damcalc(1, 6, dam_aspect) + 6))				428.571	315.789	250	
Nexus	400	6	(((dam) * 6) / (damcalc(1, 6, dam_aspect) + 6))				342.857	252.632	200	
Shards	500	6	(((dam) * 6) / (damcalc(1, 6, dam_aspect) + 6))				428.571	315.789	250	
Time	150	3	No resist		
Inertia	200	6	No resist
Gravity	200	3	No resist			
Plasma	150	6	No resist				
Force	200	6	No resist				
Mana	1600		No resist

Proposed New System:

Element	Max		Max	Average	Min	Dbl Max	Average	Dbl Min	Vuln Max	Average	Vuln Min			
Acid	1600		533.333	426.667	320	320	256	192	2133.333	2026.667	1920			
Elec	1600		533.333	426.667	320	320	256	192	2133.333	2026.667	1920		Single Res Adj	((dam*(damcalc(1,5, dam_aspect)+5))/30)
Fire	1600		533.333	426.667	320	320	256	192	2133.333	2026.667	1920		Double Res Adj	((dam*(damcalc(1,5, dam_aspect)+5))/50)
Cold	1600		533.333	426.667	320	320	256	192	2133.333	2026.667	1920		Vuln Adj	        ((dam*(damcalc(1,3, dam_aspect)+17))/15)
Poison	1000		333.333	266.667	200	200	160	120						
														
Nether	550		412.5	343.75	275									
Light	400		300	250	200									
Dark	400		300	250	200									
Sound	500		375	312.5	250		Res Adj							
Chaos	500		375	312.5	250		((dam*(damcalc(1,5, dam_aspect)+7))/16)							
Disench	500		375	312.5	250									
Nexus	400		300	250	200									
Shards	500		375	312.5	250									
Time	150													
Inertia	200													
Gravity	200													
Plasma	150													
Force	200													
Mana	1600

Basic 5 One resist blocks 67% to 80%
Double resist blocks 80% to 88%
Poison max damage changed from 800 to 1000

Result: Single resist is on average about 7% better than previous, while double resist is about 5% worse. Poison more dangerous

Others One resist blocks 25% to 50%
Result: Light, Dark and Sound are slightly more dangerous than previously, others have a slightly lower max but same min and same average damage

Vulnerabilities increase damage taken by 20% - 33% vs the previous 33%

Using the damcalc function in the numerator allows us to get an average in the middle of the range vs the previous skewing toward the low end of damage taken

Overall high resists and the basic four are more valuable while temporary resists are slightly less helpful but still a good thing to have. This is a boost to warriors and low level characters who cannot easily get the resist spell but also helps characters who's temporary resist has worn off unnoticed or ends in the middle of a fight

[Magnate]

Hmmm. I wonder if I can just clarify the basic proposition here. It seems to have two parts:

1. Resists should be cumulative in some way a la O/FA. (Currently only one perm plus one temp are cumulative.)

2. Resists should be variable in their effect. (Currently some are and some aren't.)

The first point was in the OP but not the reworked table. I'm quite open to the idea of moving to cumulative resists, but why wouldn't we just adopt the O/FA system? It's tried and tested; I'm not sure what its deficiencies are but I'd want to know that if there were any we were proposing something that improved upon them.

On the second point, I really don't like variable resists. IMO they are something that adds "realism" at the expense of gameplay (having to memorise the worst-case of the variable range). If there are good reasons for having variable resists, can we at least make them all vary by the same amount, and make the amount smaller, like 5-10%?

On the whole it's not clear how this improves on the current system.

[buzzkill]

What I am thinking of is a simpler system that would not involve changing items at all. At least for the basic four something like this, first resist works as now, second resist blocks half remaining, temp resist blocks 1/6 remaining.

I second Magnate's (FA/O) point.

In your above example, it would be an item with a 66% resist (numbers that big are uncommon/non-existent IIRC), a second item with a 50% resist (common), plus a temp resist (not sure how those work in FA/O).

As you progress through FA/O you end wearing many identical resists, so the individual resist numbers are a bit smaller, single resists are generally a bit weaker. 50% resist for base elements is common and 10% to 20% for high resists, with 20% being the magic number to avoid side effects.

Not trying to harsh your buzz, but it looks very similar to me.

[Timo Pietilä]

I second Magnate's (FA/O) point.

In your above example, it would be an item with a 66% resist (numbers that big are uncommon/non-existent IIRC), a second item with a 50% resist (common), plus a temp resist (not sure how those work in FA/O).

As you progress through FA/O you end wearing many identical resists, so the individual resist numbers are a bit smaller, single resists are generally a bit weaker. 50% resist for base elements is common and 10% to 20% for high resists, with 20% being the magic number to avoid side effects.

Not trying to harsh your buzz, but it looks very similar to me.

In vanilla with single resist being only 50% you are toast for most games. This kind of change would require complete overhaul of entire item base.

No thank you, I rather stick with current system than play several iterations of unbalanced games for few years to get them right.

[Magnate]

In vanilla with single resist being only 50% you are toast for most games. This kind of change would require complete overhaul of entire item base.

No thank you, I rather stick with current system than play several iterations of unbalanced games for few years to get them right.

Whereas I would be happy to iterate over several dev versions in order to balance a new system, if I was convinced of the advantages of moving to it.

;-)

[buzzkill]

In vanilla with single resist being only 50% you are toast for most games. This kind of change would require complete overhaul of entire item base.

While I personally favor the FA/O system, I have no opinion on whether V should adopt it or not, and wasn't suggesting that it be ported to V.

However, only in the early game, would you be likely to only have a single resist (or none at all). By the time resists start to really matter, there will be stacking. It would be hard to avoid.

Ripped from the FA ladder, a CL42 character.

 Acid:+.....+......  66% Confu:...+...+.....  86%
 Elec:+.....+.+....  72% Sound:....+........  45%
 Fire:+..+..+.+....  93% Shard:.......+.....  55%
 Cold:+.....+......  67% Nexus:.............   0%
 Pois:.............   0% Nethr:.............   0%
 Lite:.............   0% Chaos:......+......  44%
 Dark:........+....  39% Disen:.......+.....  43%

[EpicMan]

And while we're at it, we should remove the variable damage resist from the high elements. Right now they are useless as damage reducers and high elements have low maximums to compensate. They work very differently from the base 5 resists and that is not really explained to new players.

Why not make the high resists also block damage the same (down to 1/3) as the base 5? That way they work like the player expects them to. Otherwise we might as well remove the resistances and replace them with status effect protection. That situation would be mostly the same as today except that it is much more apparent to players what the "high resistances" really are.

[Nomad]

I would definitely be in favour of removing the random element from the higher resists. With different damage caps for the elements plus multiple different random resist ranges with bizarre, hard to conceptualise limits (Quick! No stopping to calculate it properly, what's 5/7 to 5/12 of 500 HP as a rough range?) the most information I am ever capable of retaining about how higher resists work is "resist = less damage than no resist".

Base resists, I know I can just divide the max damage by three and get a ballpark estimate of how low my HP can dip before I'm in danger of instadeath. And really, that actually encourages me to play a more risky game, sticking around to melee things because I think I know my margin for error instead of fleeing the scene early because I've got no clear idea of how much damage I might take in a turn.

[Timo Pietilä]

While I personally favor the FA/O system, I have no opinion on whether V should adopt it or not, and wasn't suggesting that it be ported to V.

However, only in the early game, would you be likely to only have a single resist (or none at all). By the time resists start to really matter, there will be stacking. It would be hard to avoid.

Ripped from the FA ladder, a CL42 character.

 Acid:+.....+......  66% Confu:...+...+.....  86%
 Elec:+.....+.+....  72% Sound:....+........  45%
 Fire:+..+..+.+....  93% Shard:.......+.....  55%
 Cold:+.....+......  67% Nexus:.............   0%
 Pois:.............   0% Nethr:.............   0%
 Lite:.............   0% Chaos:......+......  44%
 Dark:........+....  39% Disen:.......+.....  43%

Looks like defender weapon. A bad choice in vanilla. Cold 50% meet Tarrasque = toast.

It's really really easy in vanilla to go with only one item with basic four resist for majority of the game, because you simply can't get better gear with multiple. Have combination of Caspanion and Thorin and you are already in trouble finding all basic four covered.

I really don't see any reason to change this too to more complex version (which every single suggestion has been this far).