This seems like a very reasonable deal? Is there some hidden catch? Also, can anyone explain to me the $500 difference between buying first batch and second? It it simply because they think the difficulty will rise substantially following second batch? Just wondering. Thanks. Anyone have suggestions on hosting options?
I think the price difference can be explained in terms of economic value. When profit margins were skinny as they were in Nov->Feb, the value of the increased efficiency of an L3 over an A4 was much higher than it is now so the price was higher.
Inversely, with profits as high as they are today, reliability is in my opinion much more important than electricity efficiency.
Using 250Mh/s for comparison and 0.00006BTC/Mhs/day as a payout benchmark we can do the sums like this
250Mhs * 0.00006 = 0.015 BTC per day and 0.015 * 1200 = $18USD per day (either L3 or A4 @ 250Mhs)
If an L3 uses 400w and an A4 1050w, the diff is 650w or a 0.65kw saving. 0.65 x 0.075 (my cost of electricity in USD) x 24 = $1.116 saving per day.
Lets assume that Longsnowsm's failure rate is a very conservative MTBF, so we assume that 20% of miners have some kind of failure per 6 month period, and it takes 30 days for the repair.
If we have 5 x 250Mhs miners (either L3 or A4), the gross profits should be 5 miners x $18 per day x 182 (days) = $16,830
The power savings in choosing an L3 = 5 x $1.116 x 182 days = $1018.35
The cost of lost mining production due to card failures = $18 x 2 (failed miners) x 30 days = $1080
Wow, those figures are pretty close, but the difference would fall in favour of electricity efficiency if you used 10c per KWh, however I also haven't factored in your time to deal with repairs, or any shipping or repair costs once they're out of warranty, and frankly I think 20% per 6 months looks too low, so at current mining profit margins I believe reliability is of greater economic value than electricity efficiency, but that's just my opinion.
If you have relatively stable electricity prices they key question in a 2-3 year ROI plan is 'where are scrypt mining profits heading?'
Well, my belief is that GameCedits and Syscoin will explode when the Game App-store and Syscoin BlockMarket are both released later this month. Look at the last 24 Hrs.
I don't understand where the value is in BitconnectCoin so it could implode any day, but overall I think scrypt profits will remain in a range between 0.00005 and 0.00008 btc/mhs.
BTW, I'm assuming Inno match the price of the L3 shortly so up front price differences are removed. Looking at the L3+ price, it does look attractive but we have no idea of the MTBF yet (it really could be worse), and honestly I like the clockwork set and forget reliability of my A4s. Also just remeber that profit loss due to a failed miner will be double with the L3+ as you're losing twice the hashrate. If Bitmain insist on returning the whole miner for a single card failure and it has more cards, the failure rate per miner will actually increase by the % of additional cards it has over the L3, even with the same per card MBTF. Ouch.
Disclaimer: I own quite a few A4's so this response is somewhat biased, but the maths don't lie. ;-)
