006 Shadow Hunter [60] v1.1.1

[Passarelli]

We ran close to 500 shots, with a 60% chance of vamp aura, and hex having a 50% chance of blocking, it should be a 30% chance of hitting with vamp aura. We had 5 or 6 29+ streams of shots not proc at all. The odds of this happening 1 time is
.3 ^ 29 = 0.00000000000000068630377364883
or 0.000000000000068630377364883%. Considering this happened multiple times it is a fairly good sign of something wonky going on. I know non quantum CPUs don't really have randomization, but that is still pretty low odds, and the fact we did it multiple times means it likely wasn't a false positive.

My math skills for probabilities is a little rough, haven't used the math in years, but I'm pretty sure I got it right XD

[What]

I'm not certain you know how percentages work.

If you flip a coin, you have a 50% chance of heads or tails. That does not mean if you flip it twice you are guaranteed to get 1 head and 1 tails. Each outcome has a 50% chance of happening, it happens in a vacuum, you can't add up all the chances and assume it should equal the percentage. It could equal out to the percentage, but it may very well be higher or lower than the assumed percentage.

[maynard]

I'm not certain you know how percentages work.

it's actually not 50/50, it's closer to 51/49...

[acolyte_to_jippity]

it's actually not 50/50, it's closer to 51/49...

depends on the coin. Quarters, yes (since they're two different metals combined, one half weights SLIGHTLY more which skews probability by a fraction.) dimes, no.

[maynard]

it's not just about weight, stains can also effect the outcome like dried coffee, more finger grease on one side than the other.... there's really a lot of variables...

a coin is closer to 51% chance to re-land on the same face that it started out on.... dimes are not excluded.

fun fact, the U.S. Lincoln penny (the one with the memorial on the back) will land tails side up roughly 80% of the time.... provided it's a newer one that's not all dirty from use.

[Passarelli]

I'm not certain you know how percentages work.

If you flip a coin, you have a 50% chance of heads or tails. That does not mean if you flip it twice you are guaranteed to get 1 head and 1 tails. Each outcome has a 50% chance of happening, it happens in a vacuum, you can't add up all the chances and assume it should equal the percentage. It could equal out to the percentage, but it may very well be higher or lower than the assumed percentage.

What, I'm guessing you have never done much in probabilities.

The probabilities of two sequential events having the desired outcome is the product of those events (you can also look at this as the odds of predicting said events prior to the first flip). Since these events have the same probabilities, the forumla was

T(P) = P^n

T(P) output is the total probabitity of P
P is the probability of one event
N is the number of times this event repeated.

So for your Coin flips, if you flip the coin 10 times and it came up heads each time, it would be:
T(P) = .5^10 = 0.0009765625 Which is .0977% or 1 out of 1024

If you did this 29 times it would be

.5^29 = 0.00000000186264514923095703125 or 0.000000186% or 1 out of 536870912.


EDIT: Clarified language

[Passarelli]

While I was at work, I figured out my original number was wrong, as I was doing the probability of vamparic procing through hex, then used it in the formula to show how rare of an occurrence it should be to not hit 29 times in a row.

It should be .7 instead of .3, so it becomes:

.7^29 = 0.00003219905755813179726837607 or
.00322% Much better odds of it happening, but still enormously small. As I said, the fact that this happened several times during our testing shows something strange is going on.

[CYBER]

Zero? Any chance you can send me a pastebin of the shadow Hunter code, perhaps commenting the section specific to HEX?

I'd like to take a look at how the hex works because Wolf and I noticed some internal cool down and fun facts about hex,

and I want to confirm that inconsistency we are seeing in a case without troubling you to do it

If you don't wanna post it publicly to preserve the code, can you send it by pm like last time you did? When we trying to fix the shadow void's changerace lace glitch?

[kionay]

what i wouldn't give to be able to try to debug the code for these problems

gah i hate not knowing >_<

i guess we're all forced to play the flawless races from now on /s

[ZERO]

From scratch reprogramming work started