Thank you zOU!

Just what I needed. Especially enjoyed the warning not to attach the fan the wrong way so it pushes air into your miner. That would be both tragic and hilarious.

Just started with an S7 with a single fan. TOO LOUD, and there is an annoying whine sound.

I like the enclosure ideas, but wanted to try quieter fans.

Ordered 2 Noctua fans (http://noctua.at/en/nf-f12-industrialppc-3000-pwm) for push-pull config.

I see there is a second 4 pin connector on the control board.

My question is, does anyone know whether I can just plug in the new fans and the firmware will control them properly? Or should I just wire the fans straight into the PSU?

Thanks,

Combine this with the ability for non-miners to delegate their bitcoins to an address of a miner that they like, and you have something like a system of voting. One bitcoin, one vote.

Also, you could lower the difficulty even further, so that more than two candidates solve each block and are evaluated with regards to Proof of Stake. There is a trade-off here with network propagation, I suppose.

What do you think of my Proof of Stake idea?

Block 1:
Miner A mines block #1 and attaches an address that proves they control 100 BTC.
Miner B mines block #1 and attaches an address that proves they control 150 BTC.

Transactions that are agreed on by Miner A and Miner B now have 1 confirmation. Transactions that Miner A and Miner B disagree on are now disputed.

Block 2:
Miner C mines block #2 (including the next round of transactions) and attaches an address that proves they control 125 BTC.
Miner C also attaches a reference to two different versions of Block 1. The version with the higher POS becomes authoritative. In this case, the version of Block 1 mined by Miner B becomes authoritative. Disputed transactions resolve in Miner B's favor. Miner B collects the block reward for Block #1. The block mined by Miner A constitutes a dead fork.
Miner D mines block #2 and attaches an address that proves they control 50 BTC. (Miner D also attaches a reference to two different versions of Block 1.)

Block 3:
Repeat the above process to compare the block mined by Miner C and the block mined by Miner D and give the block reward to Miner C.

Ok, so this is what I was talking about.

I don't mine, and it did not occur to me that you did not need to run a full node to mine. (However, I think since de-centralization is the goal, I would prefer full nodes.)

It is not enough to me to simply accept the answer, "the water-heater-asic will pay for itself, don't worry about it."

Let's take a wild stab at calculating this.

There are about 100 million households in the USA alone. Let's assume for the sake of argument that 10% of that number world-wide are equipped with asic-heaters.

Today the mining revenue was $406,223. This figure may go up over time, sure. Let's assume that the water-heater asics are able to capture 10% of the mining reward. I think that is generous.

So that's $40,000 bucks, spread over 10,000,000 households, which is $.004 per day, or $1.46 per year. I don't think that will motivate anyone to differentiate their water-heaters by adding asic chips.

If you feel that my math is off, feel free to offer your own numerically grounded argument. No one ever offers any numbers to back up the "asic heater" story -- it just gets repeated and people say "sounds good" and stop thinking.

I keep hearing about how ASICs are going to be built into routers and water-heaters and such. I think that this meme is dangerous. This is a catchy thought that stops people from doing further analysis.

Sure, there is no technical reason why people could not mine with their water-heater for a trickle of satoshis. The important question is whether they will do this. It seems most likely to me that this practice will never catch on, because amateur miners will not be able to profit by doing it. That trickle of satoshis from your water heater will not be enough to pay for the cost of adding the chips to your water heater. Furthermore, to be a miner will require not just SHA2 chips, but a big hard drive and internet bandwidth. I know that the hardware can be cheap, but it seems unreasonable to expect cheap hardware to perform well enough to pay for itself when the difficulty is raised by cutting-edge mining farms.

I think people are in for a rude awakening by relying solely on the current POW to protect against unhealthy levels of centralization. It bothers me that almost every time the topic comes up I hear about the proverbial curling iron that mines Bitcoins and that is supposed to stop any further inquiry into the matter.

It seems clear to me that we need to add proof-of-stake to the protocol. POW has this very unhealthy non-linear relationship between the amount of money you spend on your rig and the amount of influence you get over the network. The ideal would be that the more money you spend you asymptotically approach a hard limit on how much influence you can acquire. I can think of no way to implement this, however, I will settle for a linear relationship between how much you spend and how much influence you get. This can be done with proof of stake, I think, and must be done.