Quake Tower is really inconsistent

[WindStrike]

Even with a chance of 40%, I've seen the Quake's shockwave not activate 10 times in a row, at which point too many creeps get past when none should at all. It's not just a one-time occurance either. A tower's damage should not come from random chance (ex, old drowning tower).

How about the Quake tower gets a shockwave off on every third attack? It'd be like going back to 33%, except that it's guaranteed 1 out of every 3. A way of making it work is that it uses energy of 0/3, 1/3 (first attack), 2/3 (second attack), 3/3 (third attack - autocast goes off, activating the shockwave and knocking it down to 0/3). Or if there's an easier and more passive way of doing that...

This way, there wouldn't be any "rush 20 ray towers in the back of the defense because the quake's shockwave didn't activate" or something like that. Granted, it does become somewhat more consistent for every extra quake tower you get, but it can still be quite annoying.

[holepercent]

Should be a pseudo-random system. Within every set of X attacks, it would trigger once randomly within that set of X attacks.

[WindStrike]

Eh, I suppose. Guess it's good to have one tower still be random chance, such that it stays unique in relation to other towers.

[Sykomyke]

I think it's fine as is. What you are modestly referring to is the "Gambler's Fallacy" in which you believe that with every non special attack that executes, your chance for a "Quake" attack gets higher. This is untrue. You had a string of bad luck mate.

[maverck]

im not sure if the pseudo random thing exists in starcraft. so... yea. something to check i guess.

[Ranakastrasz]

I think it's fine as is. What you are modestly referring to is the "Gambler's Fallacy" in which you believe that with every non special attack that executes, your chance for a "Quake" attack gets higher. This is untrue. You had a string of bad luck mate.

http://seanmonstar.com/post/708989796/a-le...andom-generator

Yup, thats why we want a less random random number generator. We are not asking a casino to modify the laws of probability, but rather how the dice are thrown (or rather requesting a change to a shuffle-bag, where removing a marble means that the rest of the marbles have a higher chance of getting picked) .I at least am not expecting that to be the case. I know full well that this is using a *real* random number generator, which is well known to be a bad idea in video-games. (or at least should be)

So a shuffle bag of 5 marbles, with 2 being *proc* marbles, would mean that the maximum *bad luck* streak would be 6 misses, but that would be far more rare than it is now.