Well, just some math.
576 blocks / day
4,65 DRK / block
-------------------
~2680 DRK / day
Currently 37.5% goes to miners, which is expected to increase, which will be 57.5% by this time next year.
If i understood correctly, ASICS are not expected to be out before that. This means only 42.5% will be distributed between miners.
--------------
~1138 DRK / day
In a year from now DRK supply will be reduced by 7%
---------------------------------------
~1058 DRK / day
To make it easier, let's use 1000 DRK / day for further calculations (feel free to adjust the end results by 5% due to this)
The three ASIC models in the article range from 900 to 7500 USD. Let's use the average of 2500 USD (assuming more units are sold from the cheaper version, which may not be true at all).
" For their business model to work, they reportedly have to sell 4,000 units before they produce one. "
-------------------------
Let's assume that 4000 units will be sold for 2500$ each. That's 10 million USD, basically the current market cap of DRK. It's easy to see that with today's prices it would be impossible to make a positive ROI. Why?
If 4000 units will be mining, the 1000 DRK / day generated will be distributed evenly, resulting in 0.25 DRK / day for each ASIC. With today's prices, that's an astonishing 0.25 * 1.6 USD = 0.4 USD
ROI in this case would be 6250 days, assuming no more people buying ASICs (impossible), DRK supply not decreasing (impossible due to the 7% yearly decrease), and the mining share of all coins generated not decreasing (unlikely based on the
current plan).
Nonetheless, let's calculate with this 0.25 DRK / day.
Should DRK to BTC ratio raise to 0.07 (3 times ATH), the result is 4 USD / day.
If at the same time BTC to USD raises tenfold as well to 2500, the yield is 40 USD / day. Which makes it roughly two months to get the initial investment back. Minus electricity.
The same 2500 USD would net ~1500 DRK currently, and within a year it would generate an additional ~180 (0.5 * 365). With the prices above one DRK would be priced at 160 USD. 1680 * 160 USD = 268800 USD
Conclusion:
-------------
There is no financial incentive currently for miners to switch for ASIC, assuming the minimum 4000 asic units needed, and the 2500$ unit price. Should any of these change by more than 1 orders on magnitude, it might be viable with today's prices. But even then, buying the coins directly and setting up a masternode would be a WAY better investment.
I could go on for pages with the analysis, but I hope this was enough.
The only reason why ASICs would be worth buying is if someone wants to attack the network. But seeing that creating such an ASIC costs millions, this is hardly an issue currently. So people, there is basically no reason to be afraid of ASICs for now.
If anyone finds a flaw in my logic above, please elaborate.