Replaced almost all calculations with references to abilities via %{selected|AttackCalc} and similar, making the macro vastly smaller, and the same thing happens when I add in any 5th "arrow". They work individually, they work in any combination of 4, but nothing I do will make all 5 work. It isn't a huge deal in the big scheme of things, as I can just make a second macro for one of them, but I'd sort of like to figure out what is causing the issue just for scientific purposes, and anyone else that might encounter the same thing. It cannot be a length of macro thing anymore, since it is significantly smaller, and to eliminate a few thoughts, replacing Arrow 5's formula with any amount of text, and even adding an Arrow 6 with more text, works just fine. Which leaves things like number of calculations, number of variable calls, number of attribute references, etc. To sort of test that, I replaced my long AttackCalc ability call (which is this: d20+@{attk-ranged}+@{repeating_weapon_2_enhance}+(?{Bard Song?|0}*@{BardSongMod})+(?{Haste?|0}*1)+(?{Point Blank?|0}*1)-(?{Rapid Shot?|0}*2)+(?{Favored Enemy?|0}*2)) with a simple 1d20, reducing the number of calculations/calls/whatever by 50 to 70 (2 for each Arrow, 5 arrows, 5 to 7 calls each), depending upon which are causing the issue. Arrow 5 then worked perfectly. So there is definitely a limit to *something* related to variable calls, ability calls, calculations, etc. I surmise at this point it probably isn't a Power Card issue though, and more likely a Roll20 macro limitation, but I can't find any documentation on such a limit, and obviously I'm not completely sure what is causing the issue. If I narrow it down more I will post further just for edification, or perhaps someone knows, but as I said I'm guessing this probably isn't a Power Card issue. Sorry for the derail, and thanks for the efforts to help. EDIT: After posting a thought occurred to me, so I went ahead and changed the AttackCalc to be simply d20+@{attk-ranged}+@{repeating_weapon_2_enhance}, so I still had the same number of @ calls, but still 50 fewer of the variable calls. It worked! So right now I am leaning towards there being a limit to the number of times either a single variable can be referenced, or the total number of variables.