More like late spring, JUST before the flood from Litecoin miners hit when the Gridseeds started getting widely deployed and killed GPU mining on Litecoin (in combination with the Litecoin price collapse).

Yup, just like the Bitmain and Baikal ASIC for Cryptonight coins will probably never make enough back to pay for themselves or even CLOSE.

But will Ravencoin fork if an ASIC does appear for it?

More central than eastern, and under 4.6c/kwh
Unfortunately I've not found a good location (YET) in the land of Super Cheap Electric in 2 of the adjacent counties.

Upside is rents are cheaper where I'm at, so right now it sorta balances out.

 X11 has 11 different algorithms in it, but the original Baikal miner was quite efficient at mining X11 and also high efficiency at mining several of the component algorithms as a "seperate" thing.

FAIL logic.

I won't be selling my rigs (except perhaps the A2 units), I'm just hoping profitability doesn't slip enough more that I end up having to go back to working some sort of a day job to make ends meet.
There is a big advantage to living in a land of Very Low Cost Electric as a full-time miner....

Hint - it's not ETH prices that are dropping, it's CRYPTOCOIN prices that are dropping.
ETH held up BETTER than Bitcoin did 'till the last couple days, compared to it's all-time high - and it's STILL pretty close on a % basis between the two.

Things are a bit tight for me, but no worse than 2 summers back right after I moved - though I have a LOT more hardware working to keep me at this level vs what I had then.
Litecoin dropped enough that I shut down the A2 farm - still debating if I want to go ahead and try to sell those machines or keep them for "just in case" I manage to get into a lower-cost area soon enough to matter, *or* pricing kicks back up/diff drops enough to make them profitable again.

Still working on the "reconfiguration" project, which should help a little - just got my first Mark II shelf/rack unit fully populated, now I need to build the second one and redo my airbox setup then I get to start reconfiguring some more.

*THOUGHT* I had a pair of my R9 290 cards go bad this weekend - but one was just a riser flaking out (I have spares) and the other the fan DID go bad but a left-over Delta 80mm 6800 rpm "screamer" from my Athlon(Thunderbird)/Alpha 462 heatsink days is proving to be "barely" enough to keep the card reasonably cool.
It's amazing how QUIET that screamer is compared to the MASS AIRFLOW through the cards though. 8-O

Turns out that while AMDGPU "supports" the R9 290, you have ZERO clock control and I'm not 100% sure the modded BIOS undervolting is working - fan control works though.
I may have to redo that particular machine with Xubuntu 14.04.03 and fglrx instead (mostly a matter of grabbing the HD out of one of the machines the R9 290 cards came OUT of and adding settings for more cards).

The AMDGPU + multi Polaris machine is working reliably now, with some undervolt and underclock via modding one of the driver kernel modules saving a LOT of power.

Whitepaper has issues.

It claims equihash is the Ethereum algorithm (an easy confusion to make, but a WHITE PAPER should have had plenty of time to get checked before it was published).

It claims "ASIC resistant" but there really is nothing in the component algorithms that makes it more ASIC resistant than any of the OTHER "X series multi-algorithm" algorithms that ALREADY HAVE BAIKAL ASIC for them.

The "rotate component algo order based on the last hash" is not going to be hard to set up in an ASIC - Baikal is already most of the way there with how their EXISTING multi-algo miners are set up, just need to add a matrix switch and control logic for it to connect the "process each individual algo" parts in the right order.

I'm not saying it's a bad coin - but some of the underlying assumptions don't stand up to fact checking.

Wouldn't be the first "close relationship" Innosilicon has had with another manufacturer.
Consider the "Dragon Miner" series and their ancestory to the A2.

Partly wrong - they DO have part of Casper up and running on testnet, though not the full blown implamentation yet.
Last comment I saw from one of the devs was "2 to 3 months for full "hybrid" implimentation on testnet" but they've been known to miss estimated deadlines a LOT.



POS will not kill ETH, just ETH mining - most INVESTORS could not care less about the difference between POS and POW.

And still a high MOQ with zero distributors in sight.

Probably a waste of time, I doubt that Ethereum is going to bother forking to change the algorithm when they're finally getting in sight of the move to POS - Casper is already in action on their testnet, though I suspect it's still a few months away from "live" deployment in the initial "hybrid POS/POW" mode and I still don't anticipate a move to full POS before the end of the year.

There is also the open question of "is this Bitmain ethash ASIC" that has been announced going to be significantly more efficient than current GPU mining is, given the very large RAM requirement to mine ethash coins at all.

2x as efficient isn't going to kill GPU mining, 10x will hurt but will take a while to deploy ENOUGH ASIC miners to make a major impact, but I doubt they're going to be 10x as efficient.


(puts on the "Oracle of Central Washington hat")

I predict the Bitmain Ethash ASIC will turn out to be specified for 800 Mhash/s give or take 10% while drawing 1300 watts "at the wall" using one of their 93% efficiency power supplies, but in actual practice will manage 775-820 Mhash/s while drawing 1350-1400 watts.

Remember you saw it here first!


Price?
The usual recent Bitmain "if diff don't change, 10-12 times estimated monthly gross earnings of the miner".

Many miners run our cards at 60-70% of the "factory TDP" - without issues.
Some very high TDP cards end up at closer to 50% and sometimes less - but not all cards will let you set the power limit THAT low.

Worst case is generally that the card throttles the core clockrate a LOT and you lost a lot of hashrate as a result.

They're still profitable for MANY folks, since a lot of us don't pay 10 cents/kwh for electric.

But the "first batch" folks definitely made out better than anyone since.

Probably a waste of time, if they stay on track for Casper initial release in a few months and full POS deployment ballpark end of the year, any ASIC miner sold will probably NOT make back what it costs even if the "forks" like ETC/ECL don't change their algorithm.

It's kinda scary though to realize that there probably IS somewhere in the $3-$5 BILLION range invested into Ethereum mining rigs.
There are at least 8.5 million GPUs (that number ASSUMES they are all GTX 1070/R9 290 level cards at 30Mhash/s OR MORE per card) and more likely 10-12 million, mostly RX 470/480/570/580 in the $250 AND UP range when purchased, then add in the "cost per slot" of the underlying rigs, YOU do the math!

Ethereum probably won't bother changing their algorithm, since they're getting fairly close to POS at which point ALL mining goes "poof".

Any Bitmain ASIC for ethash algorithm better be bloody LOW in price for anyone to have a prayer of making their money back on them, especially if the estimated "still a couple months off" release date I've seen mentioned a couple places proves to be correct or on the LOW side.

I would be inclined to call this "mistake #2 Bitmain has made this year", following in the footsteps of their Cryptonight miner.

Too bad they're still making money hand-over-fist on their SHA256 and Scrypt miners and can AFFORD to make a few multi-million dollar mistakes like this.

1 amp at 3.3 volts would be 3.3 watts - but even at 80% conversion efficiency that would be 0.66 watts of conversion loss, not enough to worry about.

2 650 watt power supplies from a QUALITY maker alone are going to be $200 give or take $20 - which already uses up your $25 per slot claim.
There is also the VERY POOR COOLING of those 8-slot motherboards to keep in mind, unless you're running water-cooled or hybrid MORE EXPEN$IVE cards you're going to have MAJOR issues with high heat on anything past a RX 460 or GTX 1060, or you're going to have to add a lot of high-power fans (and MORE power draw to run them, as well as the cost) to keep the GPUs even close to cool - adding to the "per slot" cost of the rig.

I don't see $220 Polaris cards around here, the lowest I've seen for decent mining GPUs in the last 9+ months was $200 for one seller with a couple used R9 290 cards (and a phone number they NEVER got around to answering, so I don't know if the posting was legitimate) while the Polaris cards have been listing for MORE than Newegg NEW pricing the past month ($339 lowest current offering on Newegg, a Sapphire Nitro+ model, while most of the Craigslist postings have been over $450 with a very few between $400 and $450 on ANYTHING Polaris).
I suspect that's due to this area having VERY CHEAP electric though - people are NOT shutting down rigs here - while your area has quite a bit higher electric pricing.

On ZEC, a 1080 ti will easily exceed 680 sol/s with good efficiency and can get to almost 800 on good-cooling cards if you push them, while 1070s struggle to beat 400 sols with good efficiency and you're VERY lucky to beat 450 sols/sec by much when pushed hard - the performance vs cost numbers USING YOU PRICE FIGURES work out pretty bloody close.

8 GHash for Ethereum is well within the range of NORMAL LUCK shifts, there is no need for "44 6-card GPU rigs" to have been brought online in 2 hours to account for it.
It's only about 3% of the current total network hashrate after all.
AUTOMATIC PROFIT SWITCHING mining rigs could account for a lot more than THAT in less than 5 MINUTES - or a single medium-sized Nicehash order starting up could account for it in less than *1* minute.
Watching short-term variations and trying to make something monumental out of them is a fools game.

Casper won't kill ETH mining immediately - that's the "1%/99% first step for TESTING PoS" that *MIGHT* finally go live in a few months if they don't have any MORE delays, making it very unlikely ETH will move to full POS before the end of the year even IF there are no more delays.

There is a much bigger market for SHA256 miners than for anything else - far more likely to be able to pay off the HUGE design/engineering/chip costs than for any other algorithm, even if you only manage to get 5-10% of the market.
Most algorithms can't support the sale of enough miners to be able to pay off the $MILLIONS of up-front costs - note that quite a few of the recent "small algorithm" miners have been 28nm to keep the cost down.