just wait

Dunno if the L3+ is a safe bet, either - the recent drop in price may very well presage the release of a more powerful ASIC miner that renders the existing generation all but obsolete. Again, I don't think any ASIC miner is a good deal because of the manufacturer's themselves.

And the $1900US for "1 + 4 free" Giant-N only applies if you buy them direct from Baikal, and I seem to recall reading that you actually have to buy 2 (ie - spend $3800US) to get that deal. But that requirement may not have existed or has already been relaxed; dunno, I don't follow the pricing shenanigans of these jokers too closely since I am, for a 3rd time, not interested in adding to their already fat-n-happy bottom line.

This is a classic error, thinking that there is some natural floor in the earnings you will make per day, or that what you can earn now will persist indefinitely when neither is the case.

Consider what would happen if even 2-3 coins stay on the original CryptoNight algo - every ASIC miner is going to mine the shit out of them and what do ASIC miners tend to do with their coins? They sell most, if not all, of them as fast as they can in a desperate race to pay off their "investment" ASAP. The inevitable consequences of this groupthink action are:

1. difficulty shoots through the proverbial roof, so earnings per day suck even with the massive hashrate advantage of an ASIC over a GPU; much fewer coins than expected are mined.

2. the price of the coin in USD or BTC plummets because everyone is dumping their mined earnings on the (usually tiny/shitty) exchanges as fast as they can.

For example, when network hashrate on ITNS spikes to 35 MH/s, as it has done several times over the past couple of days, 20kH/s brings in about 800 coins per day with a current value of $3.60US (56 Sats). Ignoring the cost of energy - because only 60W, right? - that will take 530 days to pay off if you bought the Giant-N for $1900US (or "only" 127 days at the current listed price on asicminermarket of $550US).

Earnings look a lot better for Karbo at this moment - around $18US per day - but that's because its total network hashrate is only 7MH/s; I don't follow this coin, but I suspect if a relatively unknown coin like ITNS has seen sustained spikes to 35MH/s then a much better known coin like KRB will see even more dramatic pumps in network hashrate.

I should note that despite my usual arguments against ASICs I am not anti-ASIC, per se; rather, I am anti-foreign-ASIC-manufacturer because all of them so far have engaged in predatory/deceptive behavior like mining with their ASICs for several months, selling them as "new", refusing to honor any sort of warranty for the 1/3rd of them that arrive DOA or break within a few days of use, and are all but untouchable in a court of law for these bad behaviors.

This miner it's not inefficient noisy heater. Even if it will earns only 2usd / day it will still be profitable, because the electricity consumption is nothing at 60W 

Once (if) ETN algo forks, which i heard they might next week block 250000 or something.. that is where most of the old cryptonight algo has is at so the profitability for the rest of the algo coins will plummet