Exploding Dice with Critical Strike

I've got a bit of an odd situation and I'm not entirely sure if it's even possible to set up the dice roll to work for this... but... Is there a way to set up a roll with the 3.5 templates using both the new 1d20cs>@{critrange} and 1d20! but only have the critical strike check look at the last die rolled? We're playing with a rule where a natural 20 on an attack roll is not an automatic hit but rather allows you to roll another d20 and add that result to your roll repeating every time you roll a 20. However an attack is only considered a critical if the final d20 is a critical threat for you. For Example: my character crits on a 12+ so if I were to roll a 20 for her attack and still not hit an enemy's AC I would then roll another d20 and add that to my attack roll if that result is enough to hit and the second d20 is >12 then I scored a critical otherwise it's just normal damage and the critical damage section should remain hidden. Edit: My DM has mentor status, so if this would require a custom template to be made then I'm sure I can get him to add it if someone could post the code for it.

If 1d20!cs>@{critrange} doesn't do what you're looking for (I can't test it at work), then you'll probably have to use an API script to either sniff the results of a 1d20! or implement the exploding part internally like // assume bonus = total bonus to hit, critRange = @{critrange} var total = bonus; for (var roll = randomInteger(20); roll == 20; roll = randomInteger(20)){ total += roll; } var wasCrit = (roll >= critRange); Incidentally, you may want to note that this mechanic means that weapons that only crit on a 20 can't ever crit (everything which would crit would instead explode a level deeper; the final die cannot ever be a 20). Not saying that this is good or bad, just that it may not be obvious.

Hm... that could be a problem... i guess we could just roll 1d20! In the crit confirm section instead... since the houserules also auto confirm crits... the only problem with that though is that it will roll the exploding d20 on all crits instead of just nat 20s... The ruling doesnt actually prrevent crits on 20s however. We are only suposed to re roll until we hit or stop rolling 20s whichever comes first... so i guess its probably better to just roll the exploding d20 seperate and figure out which roll was the last roll needed to hit if it hits and if that roll was a crit Maybe 1d20!>@{critrange} would work... an option that outputs both the number of dice in the crit range and the total modifier would be best though... and if we could change the behavior of the crit confirm section to only display on a nat 20 that would be even better

1d20!>{critrange}cs=20 might work for you. That should explode if higher than the critrange and highlight/trigger the crit criteria for the natural 20. Though you don't really need to add the cs=20 if the critical success on 20 is the only value you are looking for since that's the 'normal' crit value. You can also stack critical success values; I use d100!>95cs=66cs=100 for some rolls in a different game I play. That only shows the green box (and triggers the crit stuff in my roll template) if the d100 roll is a natural 66 or 100.

Diana, what your suggesting does similar to what I'm looking for but in the wrong order... I was looking for a way to have only nat 20s to explode, but highlight results where the final die is a crit... however as manveti pointed out this would cause issues if the crit range is 20 since the final roll on an exploding d20 would never be a 20... I did some testing though and.... 1d20!cs>@{weapon1critmin} will explode 20s and highlight all rolls that rolled >{crit range}... so far it's the closest to what I'm looking for... I had an idea to try rolling [[ 1d20!dl1]] [[d20cs>@{weapon1critmin}]] as the crit confirm and roll the attack as normal that way if a nat20 was rolled on the attack and it didn't hit the result of the crit confirm section could be added and if the 2nd number was a crit then the attack would still be a crit... but sadly it doesn't work as intended... the exploding d20 will drop the entire result if the roll is under 20... I had a few other ideas using math but they give the same result as above... any roll under 20 gets dropped entirely even if it was the only roll... I need the exploding die to only drop exploded results under 20 for this option to work... if the first roll of the exploding d20 is under 20 it should be kept and output.

Now that I'm home where my browser actually renders the colors, I can see that the 1d20!cs>@{critrange} highlights individual rolls in the output of /r and in the tooltip for inline rolls, but the box highlighting could be misleading for inline rolls. It might be better to use an API script (or just do a normal 1d20cs>@{critrange} roll and handle the exploding manually; unless there are other houserules which fundamentally alter the nature of 3.5 gameplay, I can't imagine these rules are going to come into play very often, given what you said about only rerolling until you get enough to hit).

considering that we will likely never roll more than 2 extra d20s... I think I'll just have the confirm crit roll [[d20cs>{critrange}]] [[d20cs>{critrange}]] and use that as a check for attacks that missed on a nat 20... would be so much simpler if I could set it to explode a limited number of times and compare the final roll to the crit range... then an exploding d20 that explodes only 2 times would still output a crit if the 3rd die was a 20 since it wouldn't be able to explode to infinity... a syntax like 1d20!#2cs>{critrange} would be nice... EDIT: Made a suggestion on the suggestions board. Limited Exploding Dice

Follow-up edit... I figure out a way to limit exploding dice >.< however it only really works for rolls that involve 1 initial die roll... 1d20!kh3 will output a result for a die that only exploded 2 times regardless of how many times it actually rolls 20... now if I could just hide the dropped rolls from the tooltip when using inline rolls and only highlight based off the lowest kept die... atleast I got something that will work for what I was trying to do though...