A question regarding farms

[Absol]

Is the 18/6/6 setup more efficient compared to 24/6? I've checked the old thread here (http://elementscommunity.org/forum/index.php/topic,1091.msg11607#msg11607), and it says 18/6/6 is the best setup.
But here (http://elementscommunity.org/forum/index.php/topic,1091.msg317219#msg317219) it says 24/6 has almost 100% winrate. Which one is correct?

Also, how about 18/4/4/4? My guess is that it's not effective enough, but i just want to make sure. Thank you.

(before anyone asks, this is for a farm via PVP)

[dragonsdemesne]

From what I can remember, supposedly 18/6/6 was best, using 18 of any pillar except quantum pillar.  (not pends)

[kev]

I trust DarkWeaver's math.  24/6 is the best farm.  The best electrum farm is 24 pends / 6 rare critters.  It's better than 30 relics.

[dragonsdemesne]

I've never done the math on it, but I remember back in the t50, every farm was 18x pillar 6x rare 6x another rare, which is why I mentioned it.  It does seem like 24x pillar 6x rare would be better, but without knowing the math behind pillar spinning, it's impossible to say for sure.  I'd suspect 24x pillar/6x rare is better, but as I mentioned, absolutely everyone did 18/6/6 split for t50 farms.

[Sevs]

I've never done the math on it, but I remember back in the t50, every farm was 18x pillar 6x rare 6x another rare, which is why I mentioned it.  It does seem like 24x pillar 6x rare would be better, but without knowing the math behind pillar spinning, it's impossible to say for sure.  I'd suspect 24x pillar/6x rare is better, but as I mentioned, absolutely everyone did 18/6/6 split for t50 farms.

because people liked the better variety

[willng3]

I definitely seem to recall winning rares from 24/6 farms more often.  But as Rasta said in the second thread, there's a lot of RNG involved to make that observation untrustworthy.

Personally I'd put my trust in the more recent thread since the information's probably less outdated.  And Dark Weaver's a boss too, I'd trust his math more than my own certainty of 2+2 equaling 4.