Technically, pure alluminum melts at 1220. The alloy, operational piston cooling jets, and airflow volume make it very difficult to nail the exact temp and duration required to actually watch the piston flow out of the exhaust valve. Of course, the conversion is not instantanious, so the engine usually self destructs well before that intriguing act could occur. Any temp above 1200 obviously puts you in the range where possibility for catastrophic failure exists. If you want to stretch beyond that, your betting that all of your piston cooling jets are correctly pointed, working, and doing well. How much they buy you, is anyones guess. One thing I have noticed, is that since I run an oil TEMP gauge in addition to presure, that long climbs with high EGT's definitely heat up the oil faster than the water. I would take this as a sign that the cooling jets are working pretty well. It also speaks to the efficiency of the oil cooler, by watching the water temps follow the oil temps pretty closely. I suppose if one were to somehow construct an external oil cooler that had a really high efficiency, so that you were actually spraying the piston with COLD oil (something under 100 degrees) then you might be able to extend it even further. The problem with experimentation, is that someone eventually pays the ulitimate sacrifice. And not many are willing to be that person.
Not only that, but you'd have to do it many times to establish a true base line.