I think that developing a Coinshift X11 pool is not a good idea.
1st: I don't like the idea of multipool
2nd and most important: Creating a multipool for an algorithm full of FPGA?

We do mine 100% LTC if our hashrate high enough and the other coins drop below 100%. However, even if we're mining another coin that is ~102% vs LTC, due to luck and spacing between blocks we may fluctuate between 97% or 106%, for example.

The more significant drop we saw today was due to Cryptsy's API being down ALL NIGHT. With their API down we have trouble estimating which coins are the most profitable to mine and we ended up not spending our hashrate as efficiently as we would have otherwise.

Honestly I spend about 20% of my time dealing with/working around Cryptsy's problems. Their wallets go down, stop accepting deposits, APIs go in and out and return stale data, etc. Last night I got pretty pissed and called them out publicly on their shit:
https://twitter.com/CoinShift/with_replies

But I doubt that'll help much -- and it's not like we can switch exchanges easily, the other ones just don't have enough volume.

Thanks sprite, but there are hundreds of people who were paid by them so i am not too awfully worried about getting "dragged in". Although i will have to find another affiliate website now to bank on. sorry to hear about your friend, i never heard of anybody getting a phone call from them, as a matter of fact i have heard a ton of complaints that they didnt have a phone number to call so i dont know how he managed that, but none the less sad to hear.

i have made $212 from there affiliate program and been paid every time. I dont care who does what, if they are going to pay me, i will wear a clown suit and walk on my hands. Free BTC is Free BTC. its a ton easier than wasting money on miners.