I don't know what the number is, I just know it is much higher than 114. I figured adding 30% to his number is a decent estimate.
The next difficulty change is estimated to be around a 50% increase. The difficulty changes every 2016 blocks, which works out to be every ~2 weeks. Considering this, along with the current ASIC development, and the history shown on the chart, I would think the break even point would be more like a year or more.
To anyone thinking this sounds too good to be true: well, it is.
It's extremely unpredictable. Even with the best ASIC out today, there's no guarantee that someone won't come out with a more power efficient ASIC next week, and with enough volume, all current ASICs could become unprofitable if the difficulty was increased enough.
Alternatively, if the value of the bitcoin dropped to say, $10, even at the current difficulty the existing ASICs would likely consume more $ in electricity than they produce in bitcoins.
In summary, high risk = high potential reward. but don't fool yourself, it's very high risk.