Not that you have a lot of control over the pricing, since you want to make enough money to stay in business, but the Avalon 6 is STILL way too high.

 On the other hand, so is the S7.


 Any solid news on the B-Eleven yet?

 There's no reason to go up at this time - the last pump was due to a one-time event, and there's no solid support at the current level.
 Sometime next YEAR, we'll get the halfing - unless something newsworthy for Bitcoin happens between now and then in a POSITIVE way to counteract the "EU moving to tighten down on Bitcoin" NEGATIVE news, prices will mostly likely continue to drift down between now and May-June 2016 timeframe.

Lketc / Innosilicon won't be the first to 14/16nm - KnC announced tapeout of their 16nm Solar chip a long time ago, and announced production of it a month or two back (though yeilds must be very low or they're having major design issues AGAIN given their hashrate shows no sign of major increases lately), and Bitfury also announced tapeout of their 14nm design a while back (I'm inclined to trust Bitfury announcements more, even though getting hard info about them is a pain in recent years given their long-time focus on large accounts only).

Full custom 14/16nm design will be needed to get significant efficiency gains over the best of the current 28nm ASIC though - which is why BitMain and Avalon are likely to be late to the game, they want to maximise the return on their current gen chips.

PACMIC v3 was a good deal, I made 2.8+% on my first contract (which closed out today) in 84.4 days, worked out to about 12.3% annual return rate (though Bitmain claimed 25% at one point).

 Their V4 contract is very risky, some probability of not paying back in total and WAY too low of return.



 Hashnest IS legitimate though, been around a while and proven.

I don't anticipate an S8 release.

 I WOULD bet on an eventual "S7+" release, once they get the bloody things working in a stable way.


 I also anticipate the Bitmain will join the 14/16nm mining chip crowd but they'll go straight to "full custom" with BitFury and KnC have claimed to have tape-out or production on rather than doing the non-custom design like the Lketc / Innosilicon chip design(s) appear to be.

 Pretty much zero as you likely won't have the miners within those 30 days from when you posted.


 As to your real question, your chance of achieving a block solo-mining with that kind of hash rate is very very low.
 You would almost definitely be better off going with P2Pool or going with one of the major pools.

 Probably not. You need to get your electric cost down under 5c/KWH (us) and more likely down to at least 3 to be competative long-term with the "big farms".


 Probably, but not certainty.


 Did you take difficulty increases into account?
 Did you take the reward halfing (in July 2016 most likely) into account?

 My electric cost most of the year is 7.5 cents/kwh (US, approximate) and MY calculations show that no current miner is likely to achieve RoI at ANY timeframe due to those factors and current cost per TH on any available miner - and that I'd have to get to somewhere with 5 cent/KWH or less electric to have a reasonable chance of RoI on anything currently available to the public.

 At your amount of investment capitol, you MIGHT be able to get a deal with BitFury or Spondoolies for more efficient or lower cost/TH gear, but I can't figure based on information not generally available.

Not to mention many major farms are in VERY CHEAP electric places, and a lot of them run hardware not available to the home miner (Bitfury gear, KnC, SP50 Spondoolies, etc) some of which is probably more efficient than the S7.

June 2016. Possibly May, possibly early July.

 Halfing Runup.


 Unless some major new POSITIVE news for Bitcoin shows up before that, I don't see any chance of a substantial price gain before then, and a high probability of continuing drift down to the $250ish level over the next month or two.

I'd not bet on $300 that soon, but I AM betting on under $300 by the end of the year and more likely $250ish.

 The BTCC-inspired buying frenzy is over, now it's back to pretty much no new news (and that news is bad, given the apparent intent of the EU to "crack down on Bitcoin" somehow).

 Might see a short Christmas-time inspired bump driven by short-term demand, but I doubt it'll be enough to bump the price a lot or for any lasting period.


 $400 in June won't shock me though, as I anticipate a substantial runup (but NOT to double the current pricing) during the month or two prior to the halfing, and some settle-back after to a somewhat higher "long-term support" floor.

"Technical speculation" on something as news-driven as Bitcoin is a joke.

 Without any new news to drive it, Bitcoin pricing will continue to drift down for the rest of the year and into next year, with a new runup starting a month or two before the halfing (which counts as news, it's just not "NEW" news).

 I'd bet on under $300 and likely more like $250ish by the end of the year, when the current "settle back down to the real support floor" move should be finished.

 I'm still laughing at that one idiot poll that STARTED at $600 and went up from there on options.

Peak is already past, and did not get to even $500.

Without any new news to drive it, I don't see the price getting back anywhere near $400 for the rest of this year, or managing to stay above $300. I'm inclined to bet on it dropping back to the $250 ballpark by end of the year, as the BTCC-inspired "buying frenzy" is over.


 I don't think it'll drop back under $250 much, if at all - enough new blood bought in that the real support floor probably shifted up a little.



 I anticipate the next major move will start a month or two before the halfing, but I'm tending to think it won't peak out past $500 - close I'd guess though and wouldn't be shocked to be wrong by a little.



 Anyone that bought at $450 is hurting, they never had a chance to sell at a profit and likely won't see such a chance 'till at least the middle of next year.
 

The bubble is already over, and never got to $500.


Without any new news to drive it, I don't see it getting back anywhere near $400 for the rest of this year, or managing to stay above $300. I'm inclined to bet on it dropping back to the $250 ballpark by end of the year, as the BTCC-inspired "buying frenzy" is over.


 I don't think it'll drop back under $250 much, if at all - enough new blood bought in that the real support floor probably shifted up a little.



 I anticipate the next major move will start a month or two before the halfing.
 I don't anticipate $500 in that move, but won't be totally shocked to see $500 or a bit over at the peak of that bump, and I DO anticipate the post-halfing floor will kick up quite a bit - likely $350-$400 range.

Without any new news to drive it, I don't see it getting back anywhere near $400 for the rest of this year, or managing to stay above $300. I'm inclined to bet on it dropping back to the $250 ballpark by end of the year, as the BTCC-inspired "buying frenzy" is over.


 I don't think it'll drop back under $250 much, if at all - enough new blood bought in that the real support floor probably shifted up a little.



 I anticipate the next major move will start a month or two before the halfing.

Bitcoin pricing has pretty much zero connection to any stock market.

 Without any new news to drive it, I don't see it getting back anywhere near $400 for the rest of this year, or managing to stay above $300. I'm inclined to bet on it dropping back to the $250 ballpark by end of the year, as the BTCC-inspired "buying frenzy" is over.


 I don't think it'll drop back under $250 much, if at all - enough new blood bought in that the real support floor probably shifted up a little.



 I anticipate the next major move will start a month or two before the halfing.

Avalon has a direct price now - but 10 bitcoins for a single Avalon 6 is just insane.

 They obviously have ZERO interest in selling direct.

 That "magazine" article was not very good one, too many inaccuracies and the pricing they quoted seemed to be pretty outdated AND WRONG, as the Avalon 6 has NEVER been quoted at a price less per TH than the S7.


 I'm still in "wait and see" mode, my current estimate on RoI at the current pricing on the S7 or the Avalon 6 is ZERO chance to ever come close to RoI on either machine at my 7.5c/KWH electric cost - and as bitcoin price seems inclined to continue to slowly drop back down to the well-under-$300 range that probability is dropping with it.

If you aren't in an area with VERY low cost electricity, the best thing you can do with that $40000 is spend some of it to move to an area with VERY VERY cheap electric, THEN start building a farm.

 Next best option for now is ti WAIT a month or two, and see what Lketc/BW.COM come out with for a miner, and perhaps Innosilicon.

The V3 contracts for me returned appx. 2.8% in a little under 90 days (I still have a few running, don't have final figures on those yet).
Bitmain CLAIMED something like a 25% annual return on those, but the real return was closer to 12%, which wasn't bad.

 The V4 contracts are CLAIMED to be 12% or something like that, actual number crunching shows more like a 6-7% return IF THE HALFING DOESN'T KILL THEM OUTRIGHT PREMATURELY (and right now, it looks like the halfing WILL kill them given recent diff increase trends).
 I strongly suggest IGNORE THE BLOODY V4 CONTRACTS, way too high a risk of LOSING money and very poor return.


 Right now, it looks like the BW.COM B-Twelve contracts are the best option of that type.

There has been zero chance to turn a profit doing GPU mining of SHA256 (Bitcoin etc) for a few years.
There has been zero chance to turn a profit doing GPU mining of Scrypt coins (Litecoin etc) for a couple years.

 Other altcloins might be profitable to mine with a GPU, but count on an RoI measured in years at best.
 Dash is usually a good bet on that, others vary a lot more.

 UnderCLOCK does not improve efficiency to any noticeable degree on ASIC Bitcoin mining gear.

 UnderVOLT can do so, for gear that is capable of running undervolt (some S5s for example) OR has undervolt control/capability built in (SP20 for example).


 "String" designs generally have no capability internal to the unit of undervolting - you have to reduse the supply voltage comming from the external power supply, and some string designs will NOT work at all if you undervolt them enough to matter.