The price on these things have really come down.  I just bought a used one for $200 and an APW3++ for $200 off Amazon.  This will be my first foray into bitcoin mining and from the research I did there really isn't any cheaper way to get 4TH/s than an Antminer V9.

My electricity rate up here in Canada is 0.04 kWh so I should definitely be seeing profit.

3 months?
Damn that makes BitmainWarranty look REAL bad.
They go on about their high prices coz they do things fast ... so basically that's not the case?

I found this in my mailbox today:
https://imgur.com/d4kGG1L

Thank you, sidehack and crew!

So, I felt the urge to ramble on about Bitcoin again since this block is pissing me off probably as much as everyone else here

Rather than try to edit in and out of the various posts ... here's how mining works

Firstly, mining is exactly the same as rolling a dice ... except for the number of possible solutions!
When you roll a dice, you have a 1 in 6 chance of getting a 6.
There's no progress to getting a 6, ever. You either get a 6 or you don't.
One try = yes or no. No history involved.

With Bitcoin the dice actually has 2^256 sides - yeah that's a pretty big number ~1.579x10^77
And ... there's more than one solution that's valid.
Any "side of the dice" that has enough zeros on the front, and is less than a specified value, is a block.
Of that ~1.579x10^77 number, the current difficulty decides how many sides mean we found a block.

The current difficulty is actually "0x176c2146" which is called "bits" and is 4 bytes of the data we hash.
"0x17" = 23 decimal and 32-23 = 9 bytes of leading zeros (i.e. 9 times '00')
The rest is "0x6c2146"
So, what those 2 numbers mean is that any block hash with a hex value of
H=0x0000000000000000006c21460000000000000000000000000000000000000000 or less is a block

and if you divide H by 0xffffffffffffffffffffffffffffffff and divide it again by 0x100000000 (2^32) you will get that network difficulty number:
Nd = 2,603,077,300,218.593
which means that every 2^32 times you roll the Bitcoin dice (i.e. hash a full nonce range) you have a 1 in 2,603,077,300,218.593 chance of finding a block.

... and the reason we divide by 0x100000000 again is that the network difficulty number of 1 represents 2^32 hashes

So ... that's how hashing works, but what are we actually hashing?

I wrote a document about it once, I'll repeat a little bit here:

The structure of a block header is an 80 byte binary data set, defined as follows:
Version                   4 bytes
Previous Block Hash      32 bytes
Merkle Root              32 bytes
Block Time                4 byte Unix Timestamp
Required Hash Difficulty  4 bytes
Hash Nonce                4 bytes

So looking at that we can vary 3 things:
Merkle root, Block Time and Hash Nonce

Normal hashing is to setup everything except the Hash Nonce and then count the Hash Nonce from 0 to 2^32-1 and hash each one.
Hashing the full Nonce range from 0 to 2^32-1 is also called 1 difficulty.

With stratum we can modify the Merkle Root to generate a different block header to hash the Nonce range.
The Merkle Root is a hash tree of the transactions we include in the block, however there's one transaction that we can change with stratum, and that is the coinbase transaction - or the transaction that pays the miner.
In this transaction we only need to make sure that:
1) It's a valid transaction (which is actually pretty random for the coinbase transaction due to the "sig" being allowed to contain almost anything as long as it starts with 'height')
2) It pays out to the expected address we want to receive the generated Bitcoins
3) The amount it pays is correct

So for stratum we put a bunch of other 'nonces' in the coinbase sig, that includes a unique number for each miner, and run the others from 0 up to some specified limit to generate a different coinbase transaction each time, that will generate a unique Merkle Root, to be put in a block header, to be hashed over the full Nonce range.

Thus with the unique number for each miner, and the miner itself following the stratum rules for creating coinbase transactions, the miner can keep generating a large amount of work that isn't expected to run out before the pool sends it new work.

Lastly, every time the miner finds a hash value that is higher than the difficulty specified by the pool, the miner sends the nonce and the "sig" nonces back to the pool.
The pool hashes those nonces with the values it sent to the miner, and thus verifies it's valid, and thus rewards the miner with the difficulty of the work sent to  the miner.

Now one more thing I'll add, that got mention a little while back, is that the miner can't modify the payout address to 'steal' the block.
The reason is that the pool wont get a valid hash of the nonces sent back, so the miner will get no reward. Simple

Here is why the recent criticisms of the pool sound a little like half baked frustration than anything else.   Detailed math really isn't even necessary to prove it, and here is why:

We get all transaction fees(better than some!)
We get all bitcoins
We only pay .9% admin fee(among the lowest).
Every share is treated equally(better than most who kick back to large operators)
Data provided on the website provides full transparency necessary to confirm the first 4.
Website is not a "pretty" as some but rock solid in my experience.

There are no other sources of bitcoin in mining we are missing out on and there are no other fees - hidden or otherwise - detracting from our profits.  In fact the only fee is occasionally waived.   Just this high level analysis proves there is no "there", there.

And PLEASE lets all stop talking about the 5Nd ramp and all that.  Its simply a weighted average to discourage pool jumping.   Kano is not "stealing" anyone bitcoins.   My ramp-up went exactly like the math predicted and I am sure my ramp-down will also when I turn off the machines.  

Being a newbie, I'm glad to provide my first defense of the pool.   

So nothing left to see here - MINE ON!  and dance too if you must - no one wants to see me dance though  (-:

This makes me want to vomit just thinking of having to report to the IRS. I need to meet someone that is experienced in tax law and crypto so I can make sure I do things right, or at least as right as possible. I'm going to be saving all of my electric costs over the year. It just makes me have to nervous shit thinking about dealing with all of this.

However, I also tried this controller, but I can't even find it on my network.
https://www.parallelminer.com/wp-content/uploads/2017/11/IMG_3848.jpg

I currently have one self-sufficient miner operating.

That means I've got a Pi (Minera with cgminer-vthoang compiled) powered off the Terminus aux USB jack, running cgminer for the Terminus itself. Only running 200MHz for 88GH, drawing 12V 3.6A so right around 0.5J/GH including the power to drive the Pi. I could turn down the voltage if I wanted, but eh.

If it keeps running well, that should be pretty much everything I need to know. I've already got an updated PCB layout and I'll be ordering a bunch of those later today.

And will probably make some with the existing layout and the jumper hack just a little bit so's we can get these in a few hands. VH will get one pretty much immediately so he can improve software support, specifically handling trip recovery.

Once we get to doling out full kits, I'll see about having an optional Pi with Minera and custom cgminer already installed.

And with that, BM1384 product development is pretty much officially closed. The next miner will use better and faster chips for, hopefully, a much more reasonable price efficiency.

And don't tell anyone I said so, but I'm going to look into compressing some of the onboard features to a future stickminer - trip recovery should greatly reduce "zombie" states, and if we can figure out communication maybe an onboard temp sensor will give readouts in cgminer?

On the left?

Also, the avalon6 is only flashing a yellow light