Posted by The Arcane(VIP) on December 8, 2000 at 06:45:56:

In Reply to: reply posted by Graham on December 8, 2000 at 01:07:47:

...if my memory (based on experience) serves me correctly, failure rates are not a linear function. 75% in a spell doesn't mean you'll succeed 75% of the time. If it does, then either CF's random number generator is broken, CF's random number generator is sentient and enjoys toying with me, or I should buy lottery tickets more often. I can't think of how many times I've failed to cast detect invis upwards of ten times in a row when I had it at 75%.

The odds of that happening are something along the lines of .0015%, or 15 times out of every million.

At 99%, however, the failure rate seemed closer to 1%. And at 100% of course failure rate is 0.

Someone who knows the code can feel free to correct me if I'm wrong here.

Follow Ups:

• I've wondered about this now and then as well Nepenthe(IMM) 13:07:07 12/8/2000 (4)
  • Hmm.... The Arcane(VIP) 16:01:20 12/8/2000 (0)
  • Re: I've wondered about this now and then as well Death_Claw the Personal witness, Experienced in using Random Number Generators. 13:39:33 12/8/2000 (2)
    • On spells, saves etc, and the unholy RNG Drinks with Evil. 15:12:13 12/8/2000 (0)
    • You either misread what I said or you didn't read it Nepenthe(IMM) 15:10:38 12/8/2000 (0)
• With 100% in a spell you can still fail, but its rare n/t Stoneybalogna 10:55:48 12/8/2000 (1)
  • The check for % must be 0-100, not 1-100? House wins! (n/t) Drinks with Evil. 17:01:23 12/8/2000 (0)

Post a Followup