Topic: Resistance Roll Table Issue

[Dreven1]

Hello all!   

Well, I was writing a .NET program to automatically look up the RR's for me and I noticed something.  The numbers are not linier.
I was expecting the numbers to line up for certain level of attacker casting on certain levels of target.
According to the chart… if a 1st level Attacker casts a spell on a 1st level Target… the base RR on the chart is 50%.
If I look at the 10th level vs. 10th level its 50% and 15th level vs. 15th its 50% and so on…
HOWEVER, if you look at the resistances before or after that number they are drastically different.
If a 5th level Attacker casts a spell on a 1st level Target, the Target has a base resist = 70
If a 10th level Attacker casts a spell on a 5th level Target (still just 5 levels difference) the Target has a base resist = 65

If a 1st level Attacker casts a spell on a 5th level Target, the Target has a base resist = 30 HOWEVER,
A 5th level Attacker casts a spell on a 10th level Target, the Target has a base resist = 35

WHY?     
Shouldn’t this number be the same resist with the Target being a certain number of levels away from the Attacker?

So if an Attacker is 5 levels lower from Target shouldn’t the resistance roll be 70 no matter what the levels of both are? This seems to make more sense? 

Please help me with this! I am writing this VB.NET program and I need some logic behind the numbers on the RR table 5.10

Thanks!
Drev!

[Rasyr-Mjolnir]

It is impossible, at this remove (30 years after it was created), to know the reasoning behind why the RR table was designed in the way that it was designed.

But, if you look closely, the RR table essentially follows a pattern of diminishing returns.

Here is a general overview of what the pattern of diminishing returns looks like for the RR table.

• The table starts off with level 1 versus level 1 with a Target Number of 50.
• If attacker is higher level, then increase the TN by 5 points for every level of difference for the first 5 levels (i.e. to 5th level), then adjust it by 3 for the next 5 levels of difference (to 10th level), and then by 2 for the next 5 levels (to 15th level) beyond that, and then by 1 for all levels after that.
• If defender is of higher level, then reduce TN by 5 for each of the first five levels of difference (i.e. to 5th level), by 3 for each of the next 5 levels (to 10th level), by 2 for each of the next five levels (to 15th level), and by 1 point for every level beyond that.

Note: The adjustments are made first to the attack level, and then to the defender level.


The attached excel spreadsheet has the formulas plotted into it, and as you can see the formulas are simple, and it matches the RR Table exactly.

[Dreven1]

Great reply Rasyr! thank you for that explaination...

I was wondering though... does it not make more sense to have a more linear table?
Does the diminishing returns idea really make that much sense?

Do you think it would make more sense to start AT the level of the Attacker vs Target level at the 50/50 split and then work the table as if it was at the 1st vs 1st level?

so you keep the same formula but just subtract the target level from the attacker level?

so that a level 7 vs a level 1 would start +5 to 50 for each level below 7 up to 5, then 3 for the next 5, etc... like you stated below...

that way the level 1 would have a base of 81?

Make sense?

Drev

[Rasyr-Mjolnir]

I was wondering though... does it not make more sense to have a more linear table?
Does the diminishing returns idea really make that much sense?

The RR table has been untouched for approximately 30 years of Rolemaster. And it has seemed to work just fine for that entire time. 

(Note: please remember that this was WAY before my time when it got made).

[markc]

 I am just going to guess but IMO they did it that way on purpose.
MDC

[Marc R]

The result of the tables being as they are is that results are absolute, rather than relative. . .and that shifts in power level occur faster at lower levels and slower at higher levels.

i.e. compare RR to OB

5 levels of difference at 1st level will be a way larger gap than 5 levels of difference at 20th level.

The RR table keeps magic more in synch with the logic of skill and Combat proportions between people of differing levels, though the synchronization isn't actually 1:1 it comes close in overall ratios.

[Greyaxe]

It is impossible, at this remove (30 years after it was created), to know the reasoning behind why the RR table was designed in the way that it was designed.

But, if you look closely, the RR table essentially follows a pattern of diminishing returns.

Here is a general overview of what the pattern of diminishing returns looks like for the RR table.

• The table starts off with level 1 versus level 1 with a Target Number of 50.
• If attacker is higher level, then increase the TN by 5 points for every level of difference for the first 5 levels (i.e. to 5th level), then adjust it by 3 for the next 5 levels of difference (to 10th level), and then by 2 for the next 5 levels (to 15th level) beyond that, and then by 1 for all levels after that.
• If defender is of higher level, then reduce TN by 5 for each of the first five levels of difference (i.e. to 5th level), by 3 for each of the next 5 levels (to 10th level), by 2 for each of the next five levels (to 15th level), and by 1 point for every level beyond that.

Note: The adjustments are made first to the attack level, and then to the defender level.


The attached excel spreadsheet has the formulas plotted into it, and as you can see the formulas are simple, and it matches the RR Table exactly.

Interesting, the spreadsheet goes to level 20, instaed of level 50.  Presumabley because in a d100 system results less than zero and greater than 100 are not calculated?  or should the rule of thumb be if the oponent is greater than 20 levels you realy have no chance?

[Rasyr-Mjolnir]

You are reading too much into the spreadsheet.

The spreadsheet only goes to level 20 because that is where I decided to stop typing, and because that showed the first five levels of where the table would be using +/-1 per level.

I threw it together in about 5 minutes, right before posting it.   

And you ALWAYS have a chance... The defender could always fumble his RR (not open-end low, just plain fumble).

[Winterknight]

In EA#13, the Opposed Action system presented there is based on a simplified single-column version of the RR table.  In short, you can use that resolution table for ANY level of comparison, simply subtract the attacker's level from the target level to find the level difference on the table, and that's the roll required.

You can use that table for RR's without using the rest of the options presented there, and it should work.

[Marc R]

Rasyr's table goes 5 levels past the one in the book, which only goes to 15th in either direction. . .doing the table out to 100th both ways would need a really big page in the book, like a fold out map.

Racial and stat mods can cause major shifts, even before you toss magic and other effects in. . .like a dwarf with a high con, even at 1st level, has a decent chance vs poisons way over their level.

You could always open end high or low too. . .if you need a 150 to resist, even assuming no modifiers, you could still roll a 99 then a 51.

Only problem with a relative table, as I said above, is that the way everything else in the game works, a 5th level character vs a 10th level character is a much larger variation than a 10th level character vs a 15th level character, despite the fact both sets are 5 levels apart. . .unless you used that opposed resolution table for everything (including combat) it would toss magic out of synch with every other power ratio in play.

[Dreven1]

Good points all, I appreciate the discussion! This was most confusing to me when  I noticed it because I am trying to write the logic in an .exe file to do the RR resolution lookup for me.

The simple formulas presented will provide me what I am looking for!

that way, the program could calculate quickly the resolution for a 250th level character casting a spell on a 21st level target with one click!

By the way, does ICE (Rasyr, Lord Miller or anyone else) have any issues with me posting my .exe Utility program on these forums (as long as I don’t try to sell it?) as an free aid to GM's out there?

Please let me know! I love writing .NET code!

Dreven (Jason)

[Marc R]

Hmm, at 1-5th here's actually only 4 steps of 5, then 5 steps of 3, and 5 steps of 2, before the infinity of steps by 1. . .

AFAIK there's no such thing as a 0 level spell, but there are 0 level characters.

Table really should have started with 50/50 being the 0 vs 0 shouldn't it? (Though, you could trim the 0 level attack column off as pointless, the 0 level target row should be valid.)

[Greyaxe]

You are reading too much into the spreadsheet.

The spreadsheet only goes to level 20 because that is where I decided to stop typing, and because that showed the first five levels of where the table would be using +/-1 per level.

I threw it together in about 5 minutes, right before posting it.   

And you ALWAYS have a chance... The defender could always fumble his RR (not open-end low, just plain fumble).

I was being facetious. More directly how would one reflect number in the negative?  Above 100 is obvious but below 100, simple rule if you fumble you fail or would you actually have to get a RR result below zero before you failed. Further your table is now at 1, suppose you have 50 levles of difference between the attacker and defender, if the rule is 5,4,3,2,1. do you start to use decimal points to calcualte additional RR requirments or should one simpley say +1 per 5 levels of difference?  Is there a math formula one could use?

[Marc R]

It's an open ended roll. . .if you needed a -30 to fail, you'd need to roll under a -30 open ended to fail.

At least, that was what I always got it as.

Much like if you have a high enough OB, and a low fumble range weapon (fumble of 4 or less), it's possible to roll a 05, then open end low, get a -30 result. . .add your 150 OB to it, get a 120 and a 15C, roll a 97 and kill the target. . . .with a -30 attack roll.

At least, that was always how we played it. (Never did like the only open end high attack rolls, as with a high enough OB you reached a point where you hit or fumbled, never missed.)

[Dreven1]

I agree, if the logic was truly diminishing returns, wouldn’t you start moving the table at 21+ 0.5, 0.25., 0.12, 0.06, etc?   

Another comment about the diminishing returns for the Fighter (melee) vs. the Caster...

The BAR and the EAR's already have diminishing returns in them based on level (for skill and bonus) returns...  the casters are getting hit twice.

This would seem to lend more towards true linear table vs. diminishing returns for resistances.

Does this sound feasible?
Attack level vs Target level : Each level difference up to 4 away (a +5 for lower level or -5 for higher level), from 5 to 9 levels away (a +3 lower or -3 higher), from 10 to 14 levels difference (+2 lower, -2 higher) and over 15 levels difference (+1 lower or -1 higher)

so that a 7th casting on a 3rd is the SAME as a 5th casting on a 1st HOWEVER the diminishing returns kick in if a 10th is casting on a 1st or a 30th casting on a 5th.

Also a 10th casting on a 1st is the same as a 30th casting on a 20th.

I think I have the logic down and I think this actually makes more sense! 

Drev (Jason)

[Marc R]

Actually if 0/0 were 50, then 1 vs 0 (the first feasible target) is 55. . .so it's still "The first five levels of variation are 5" it's just the RR table doesn't happen to have a 0 level target row. . .I doublt someone arbitrarily selected the logic of 4/5/5/infinity, likely just the zero row was dropped as RRs for pre-adult 0 levels would come up rarely and was left off at some point, leaving the logic 5/5/5/infinity. Just my opinion, but it makes more sense than an arbitrary starting point of 4 steps.

If you like it, a floating point logic of that sort could be used, but recall that it's all resistance logic, and sometimes, there's no BAR involved. . .so it applies to the relative variation between a toxin and the target, or any other instance requiring an RR. . .it's not just used for casting.

Also BAR isn't a contest, like RR or OB vs DB are, the target doesn't affect the BAR, other than by wearing metal armor or taking cover. . .and those are not biased by level. The point where level variation comes into the contest is the RR, not the BAR.

[Rasyr-Mjolnir]

By the way, does ICE (Rasyr, Lord Miller or anyone else) have any issues with me posting my .exe Utility program on these forums (as long as I don’t try to sell it?) as an free aid to GM's out there?

Read -- http://www.ironcrown.com/index.php?page=cinfo/apps -- that gives ICE's policies regarding the sharing of applications you may create.

[Greyaxe]

I am going to start calculating the difference between the attacker vs defender and user the diference on the chart, rather than trying to calcluate levels vs levels.  I feel the chart works better if you user the difference between the caster vs target.

[Greyaxe]

I am going to start calculating the difference between the attacker vs defender and user the diference on the chart, rather than trying to calcluate levels vs levels.  I feel the chart works better if you user the difference between the caster vs target.

I do the same for poison, treat everyone as first level vs the level of the poison, which is always relatively low anyway.

[Marc R]

Shrug, it'll "even out" the effect of level variation on RR effects, unlike instances like OB vs DB or Perception vs Stalk which tend to get closer together as levels rise. It makes higher level effects and casters more relatively potent.

Removing that diminishing returns effect will make RR effects scarier at higher levels. . .since using the RR table as is, you keep getting more and more resistant in general as you crest the 5 level increments, then above 15th level it takes a lot of level variation to make a difference. . .going relative, a 20th level target is far weaker than a 25th level caster, rather than only slightly weaker. This will toss RR effects, especially magic, out of whack with contested skills and combat resolution.

i.e. a 1st level target should fear a 5th level caster/poison, but a 25th level target should be less fearfull of a 30th level caster/poison. . .much like a 1st level fighter should fear a 5th level one, but a 25th level not so much fear of a 30th. . .or how a 5th level thief can generally expect to be able to sneak past a 1st level guard, but a 30th level thief is not nearly so confident about sneaking past a 25th level guard.

YMMV, but in my experience, magic, poison and disease are strong enough already without giving them a boost.