I believe his root argument is that mining with these, or any other available ASIC today, is not profitable, and feels the need to preach this like the fate of our species relies on that knowledge.
...
The first batch was 2.1BTC - no chance in hell of getting that back PPS mining - they will probably get back 1/3 of that.
So yes clearly a BTC loss - clear for all but the mathematically challenged.
Bicknellski and his cronies overpaid for the chips (they were priced and called 'dev' chips for a reason that was clearly beyond their understanding) and once they realised their mistake, they decided to sell their hardware they were mining on, at a price to convert that loss into a profit, by providing a high loss to their buyers. Seems they have succeeded in converting that loss to a profit for themselves.
It's is simple to estimate which ASIC anyone buys will lose BTC.
A lot do lose.
I take it that is too complex for you to estimate?
FYI - just to add some more amusing information on the subject
Bicknellski sent one to ckolivas - even though ckolivas asked him to send it to me since ckolivas is rather short on time to do the driver at the moment ... though I had a requirement before I'd consider writing the driver, and that was that their next batch price wasn't a rip off like the first batch. Anyway, Bicknellski wouldn't send it to me. No big deal it is only 2.5GH/s - and wont make much BTC at all for the time and effort to support it - but it probably also means a driver delay for those who paid him all that BTC No big deal - one less BTC hardware item to add to my growing list of hardware that I support for the community even when they don't make a profit for me It will be done by ckolivas soon enough and he'll support it - so again no big deal. The only side effect will be that if anyone asks me for support of them I'll refer them to ckolivas - yet again no big deal