A bitcoin miner in every hand

[klondike_bar]

Here's a list of things I really doubt:

- someone would waste the time/effort to hack a 21 inc device holding fractions of a dollar
- an ethernet port + cable would be less expensive than wifi
- people want extra cables running though their house
- people want a phone charger that has to be within X feet of their router for it to function

At this point I can easily ignore the 35$ USB charging hubs. I don't like them so much, but the routers and mining cores into regular chips seem a very good idea with minimal capex.

I actually agree with you on this one.
I wouldn't say putting mining cores in regular chips is a great idea, but it's definitely not absurd like putting chips in a toaster/phone charger/microwave/etc.

Again wrong. How on earth can you compare the capex of these 2 situations? Any large player that wants to deploy much hashrate must invest into a good datacenter and into all the required infrastructure which means power and cooling. That adds up fast in terms of capex while having "miner in everything connected to a wall outlet" makes those datacenter costs obsolete.

How much do you think that a 100MW datacenter would cost to be ready for miners deployment? I think that the costs to have 500MW ready for miners are less than the costs directed towards "miners to all" program..

Based on my research, I'd say ~$60/kw for a Bitfury style "hashing center". (including everything but the miners)

+1. Its a nice idea, but the fact is that most mining now occurs only where power is <$0.10/kwh, which rules out these devices in almost half the world (including the majority of europe). That $35 hub would be earning meager scraps while dumping ~10W of heat without proper heatsinks or cooling. imaging plugging it in in an outlet behind a couch, or if clothing is dropped on/around it while plugged in (its extremely common for most people to do that with thier chargers already). its a fire hazard unless theres temperature shutdown built in (extra cost)

re:datacenter: I think $60/kw is pretty reasonable. A 30A 208V (6kW) circuit costs about $100-150 to install, $50 for additional power bars and ethernet, $5 of rack space and networking supplies, and the remaining ~$100-150 is spent on the building space (assuming $30/kw for this in scale)

-> its important to note that a datacenter will pay at least 30% less for power than most residences where this little wallwart is going to be used. couple that with economy of scale and mining will almost certainly centralise wherever power is <$0.07usd/kwh in the next decade.


all that said, if bitcoin price jumps up, and increases faster than the difficulty (remember theres probably 100-150PH of gear currently turned off because its >0.8w/gh) then we could see mining be profitable for anyone and everyone involved (like when it jumped to $1000 and even the guys with 0.25/kwh in the UK were mining)

[JorgeStolfi]

I was excited about the potential of the new thing called the web when Andresseen was working at Netscape, and I am excited about the potential of micro mining on embedded devices now that Andreessen is at AH and involved in 21 inc.

Just because a person's  idea was once a success, it does not mean other ideas by the same person will be successes too.

Also note that Marc did not invent the WWW or graphical browsers.  He was the leader of a team who created NCSA Mosaic, an early (but not the first) graphical browser, and later one of the founders of Netscape, a company that sold the first commercial browser.

[s1gs3gv]

I don't think we can make much progress before we have answers (or minimally fundamented guesses) to these basic questions:

- What is the power consumption of their chip?
- What is its hash rate?
- How many of those chips do they have in stock?
- How much did those chips cost (icluding design)?

Your question casts me back in time to the day I read that Intel had introduced the 4004 microprocessor. If I had relied on the answers to your questions to understand the impact of this new technology I would have been at a loss. Instead, I said wow. I get that.

[s1gs3gv]

I was excited about the potential of the new thing called the web when Andresseen was working at Netscape, and I am excited about the potential of micro mining on embedded devices now that Andreessen is at AH and involved in 21 inc.

Just because a person's  idea was once a success, it does not mean other ideas by the same person will be successes too.

Also note that Marc did not invent the WWW or graphical browsers.  He was the leader of a team who created NCSA Mosaic, an early (but not the first) graphical browser, and later one of the founders of Netscape, a company that sold the first commercial browser.

All in the history books now for anybody to read. I recall advertising my new website on NCSA's 'Whats New' list which was, at the time, the only way to discover new resources.

[JorgeStolfi]

Your question casts me back in time to the day I read that Intel had introduced the 4004 microprocessor. If I had relied on the answers to your questions to understand the impact of this new technology I would have been at a loss. Instead, I said wow. I get that.

I don't know the 4004 but surely it was a competitive commercial product.  The point of my questions is to know whether 21.co is building a competitive commercial product, or just looking for a way to recover some of the money that they invested in an unprofitable mining chip.

[s1gs3gv]

I don't know the 4004

http://en.wikipedia.org/wiki/Intel_4004

[sidehack]

Your question casts me back in time to the day I read that Intel had introduced the 4004 microprocessor. If I had relied on the answers to your questions to understand the impact of this new technology I would have been at a loss. Instead, I said wow. I get that.

I don't know the 4004 but surely it was a competitive commercial product.  The point of my questions is to know whether 21.co is building a competitive commercial product, or just looking for a way to recover some of the money that they invested in an unprofitable mining chip.

My guess is they're looking for the most profitable way to recover money they invested in a pretty good mining chip. They figure self-mining they have 100% of the expenses and 100% of the revenue, but if they ship them out with their plan, they have 0% of the expenses and 75% of the revenue. Especially if they sell "services" which get them most of that last 25% of the coins. I'm also assuming the person buying the router pays the extra $8 for the chips.

I wonder if they'll include options to turn the miner off without having to power down the entire device. If that option is not present, and they also run the software-locked pool, then the user really does have no control. Unless you put it behind another router that redirects requests from their pool to your own? But that's kinda stupid to have a mining router behind a router. Just get a miner.

[s1gs3gv]

1947 - the transistor
1969 - unix
1971 - the first single chip microprocessor
1973 - Cerf/Kahn TCP paper
197x ... 199x - emergence of ARPAnet, internet and WWW
1991 - linux
2008 - Nakamoto bitcoin paper
2013 - $USD 1200 bitcoin
2015 - bitcoin micro mining introduced by 21 inc.
2020 - total collapse of large scale centralized bitcoin mining

[JorgeStolfi]

My guess is they're looking for the most profitable way to recover money they invested in a pretty good mining chip.

The chip surely was profitable (or was expected to be) when they designed it.  They had two big mines in the US, using oil cooling, and seem to have taken a small but significant slice of the block rewards.

However, I read that they have not mentioned those mining istallatons in the recently leaked investors slides.  If the chips are now unprofitable, and they have millions of them in stock, their new "business plan" would make a lot more sense.

[JorgeStolfi]

1947 - the transistor
1969 - unix
1971 - the first single chip microprocessor
1973 - Cerf/Kahn TCP paper
197x ... 199x - emergence of ARPAnet, internet and WWW
1991 - linux
2008 - Nakamoto bitcoin paper
2013 - $USD 1200 bitcoin
2015 - bitcoin micro mining introduced by 21 inc.
2020 - total collapse of large scale centralized bitcoin mining

Good timeline, but some are missing:

2013 - Dogecoin created
2014 - $USD 260 bitcoin

 

[s1gs3gv]

You should buy some BTC while it is cheap Jorge.

[Biodom]

1947 - the transistor
1969 - unix
1971 - the first single chip microprocessor
1973 - Cerf/Kahn TCP paper
197x ... 199x - emergence of ARPAnet, internet and WWW
1991 - linux
2008 - Nakamoto bitcoin paper
2013 - $USD 1200 bitcoin
2015 - bitcoin micro mining introduced by 21 inc.
2020 - total collapse of large scale centralized bitcoin mining

I hope so...
Most people don't realize that centralized mining=bitcoin demise or its transformation into something worse than regular banking.

[RoadStress]

Here's a list of things I really doubt:

- someone would waste the time/effort to hack a 21 inc device holding fractions of a dollar
- an ethernet port + cable would be less expensive than wifi
- people want extra cables running though their house
- people want a phone charger that has to be within X feet of their router for it to function

Ok you have a point, but since we have agreed to rule out the USB hubs/chargers let's move on from this subject because this will not be their main product. This is only their testing phase and yes they can afford to sink a couple hundred thousand for testing stuff. From what I know the R&D from Apple eats up to 1B$ per year. I am sure that Apple had some failed products/tests. Same thing can apply here.

The opex is clearly stated to be 25% of mining earnings. If they plan to give users enough btc to cover electricity costs then it will be 50-100% by time they start mining.

You lost me here. Let's clarify things. Isn't opex an ongoing cost for 21 inc? If consumers are paying for power that's not a cost for 21 inc. Unless you see it as they give up 25% in order to gain 75%. For me the opex is nil. It's less than 1 beer per month in most decent countries.

If your view is different please explain.

At this point I can easily ignore the 35$ USB charging hubs. I don't like them so much, but the routers and mining cores into regular chips seem a very good idea with minimal capex.

I actually agree with you on this one.

I wouldn't say putting mining cores in regular chips is a great idea, but it's definitely not absurd like putting chips in a toaster/phone charger/microwave/etc.

Based on my research, I'd say ~$60/kw for a Bitfury style "hashing center". (including everything but the miners)

Fine. That makes it 600k for 100MW and this also comes with a huge monthly opex that is non-existent in case of 21 inc. 21 inc will mine "on the cloud" with minimal costs 

I completely reject your theory that miners will continue to mine past the point that revenue < opex and that there is some mystical service that can somehow make unprofitable mining profitable.

Your analogy doesn't really make sense because internet providers have always had massive profit margins not negative ones.

Well I think it's more than obvious that the consumers will mine at a loss. If they keep only 25% of the earning and they are still making a profit then everyone will start mining just like in 2013 and difficulty will spike again.

Why can't you accept this fact? That consumers will mine at a loss? I have days when my TV is open all day, but I don't watch it. I have days when I forget the lights open and they stay open the whole day and night. Those are money burned that will bring me no benefit, but it's just spare change. I don't care about an extra 1-2$ per month. Nobody will care. This subject is really useless. There are far more important and more dangerous things to discuss than this.

[RoadStress]

+1. Its a nice idea, but the fact is that most mining now occurs only where power is <$0.10/kwh, which rules out these devices in almost half the world (including the majority of europe). That $35 hub would be earning meager scraps while dumping ~10W of heat without proper heatsinks or cooling. imaging plugging it in in an outlet behind a couch, or if clothing is dropped on/around it while plugged in (its extremely common for most people to do that with thier chargers already). its a fire hazard unless theres temperature shutdown built in (extra cost)

re:datacenter: I think $60/kw is pretty reasonable. A 30A 208V (6kW) circuit costs about $100-150 to install, $50 for additional power bars and ethernet, $5 of rack space and networking supplies, and the remaining ~$100-150 is spent on the building space (assuming $30/kw for this in scale)

-> its important to note that a datacenter will pay at least 30% less for power than most residences where this little wallwart is going to be used. couple that with economy of scale and mining will almost certainly centralise wherever power is <$0.07usd/kwh in the next decade.

My guess is that the hubs will be built just as a pilot project. It would be absurd to work on deploying millions of routers without testing the waters first with a couple of thousand of hubs. Imagine them as mining USB sticks. Very bad $/GH, but people still bought them even when they were clearly unprofitable.

The power bill for a 100MW datacenter at 0.05$/kwh is 500k$ right? And that's only one datacenter. Why have this burden every month when you can migrate it to someone else with less headache?

My guess is they're looking for the most profitable way to recover money they invested in a pretty good mining chip. They figure self-mining they have 100% of the expenses and 100% of the revenue, but if they ship them out with their plan, they have 0% of the expenses and 75% of the revenue. Especially if they sell "services" which get them most of that last 25% of the coins. I'm also assuming the person buying the router pays the extra $8 for the chips.

I wonder if they'll include options to turn the miner off without having to power down the entire device. If that option is not present, and they also run the software-locked pool, then the user really does have no control. Unless you put it behind another router that redirects requests from their pool to your own? But that's kinda stupid to have a mining router behind a router. Just get a miner.

Exactly! I hope jimmothy acknowledges your post.

They must include the option to turn of the miner because otherwise it would be stupid.

[sidehack]

My guess is it won't just be $1-2 per month. I could see $2 per month per device, and they're going to want you to have as many devices as possible. If it's ten things per household (the router, the TV, the DVR, the other TV, both game consoles...) that's $20 per month. For some people that adds half again to the power bill, and it yields them what, $10 in coin disbursements they can use for spamming the blockchain with microtransactions for 21e6-sponsored value-added services? And that's $10 now; in a year they'd still be spending $20 but the yield would be more like $2 in coin (unless the exchange rate starts going back the other way, so maybe $3). Sure it decentralizes the network, but it decentralizes the network in such a way that everyone mining mines at a loss and 21e6 takes all the coin.

It's definitely possible that a lot of people will find the "services" favorable and will jump on the bandwagon of BTC-enabled household devices. But it's also possible that a very wealthy entity will have no qualms at all about increasing its profit margins by convincing its customers that they're getting a good deal while actually making things worse for everyone.

I won't be too surprised if they allow you to turn off the mining functions. I won't be surprised if the device with this option costs more than without it. I will be incredibly surprised if they don't charge even more for a device that also allows you to pick your own pool.

[Biodom]

Bitfury developed a light bulb that automatically mines Bitcoin....:
from reddit
http://www.reddit.com/r/Bitcoin/comments/37hrbg/bitfury_developed_a_light_bulb_that_automatically/
https://twitter.com/chijs/status/603599293602234368

[RoadStress]

Bitfury developed a light bulb that automatically mines Bitcoin....:
from reddit
http://www.reddit.com/r/Bitcoin/comments/37hrbg/bitfury_developed_a_light_bulb_that_automatically/
https://twitter.com/chijs/status/603599293602234368

Was just about to post this. Ok. The popcorn is on you now!

My guess is it won't just be $1-2 per month. I could see $2 per month per device, and they're going to want you to have as many devices as possible. If it's ten things per household (the router, the TV, the DVR, the other TV, both game consoles...) that's $20 per month. For some people that adds half again to the power bill, and it yields them what, $10 in coin disbursements they can use for spamming the blockchain with microtransactions for 21e6-sponsored value-added services? And that's $10 now; in a year they'd still be spending $20 but the yield would be more like $2 in coin (unless the exchange rate starts going back the other way, so maybe $3). Sure it decentralizes the network, but it decentralizes the network in such a way that everyone mining mines at a loss and 21e6 takes all the coin.

It's definitely possible that a lot of people will find the "services" favorable and will jump on the bandwagon of BTC-enabled household devices. But it's also possible that a very wealthy entity will have no qualms at all about increasing its profit margins by convincing its customers that they're getting a good deal while actually making things worse for everyone.

I won't be too surprised if they allow you to turn off the mining functions. I won't be surprised if the device with this option costs more than without it. I will be incredibly surprised if they don't charge even more for a device that also allows you to pick your own pool.

What if having a router, TV, DVR, consoles and so on can make you get rid of the internet&tv subscription? I bet that many people would like to stop paying for some kind of subscription package while having these devices plugged in even if they have to pay a bit more for power? I think there are at least a couple hundred thousand households that would take this deal.

[alh]

Something tells me that Comcast won't be real keen on the idea of forgoing my $60+ per month tribute to them for Internet. Of course maybe 21e6 will just send them a check for $59 a month?

As for the mining light bulb, I'll just walk on by at Home Depot and get the 8.5W LED bulb that runs cooler.....

Is there a sub-forum for "Absurd Mining Discussions"?   

[Biodom]

Most successful complex systems are linked in a way that works to the benefit of both (or multiple parties).
example:
plants use energy of the sun, CO2 (carbon dioxide) and H2O (water) and produce O2 (oxygen) as waste plus carbohydrates (sugars). Plants are producers in general.
animals consume plants (and/or other animals) and produce CO2 and water as waste, which then can be used by plants again.
As such, one kingdom of life is linked to another in a relative balance (at least chemically)

Going back to bitcoin, everyone strives to produce bitcoin (mining, etc), but there is no "natural" sink for it.
Something has to be there to "consume" and recycle bitcoin into the ecosystem, as pure numerical accumulation in the vaults/wallets would not be sufficient.
A long term plan of 21 inc in basically that such "sink" would be automated or semiautomated systems trading bitcoin for services, but in order for these systems (and humans) to be engaged in such trading, they have to be endowed with a lot of bitcoin (and bitcoin eventually has to be much higher in value, obviously).
So, these fractions of bitcoin that gadgets might accumulate is in part for them to be able to exchange services in a network that is 1000Xlarger.
The equivalent of the unexpected (or expected?) benefit similar to animals use of oxygen (a plants waste product) would be LARGE VALUE accumulation (btc appreciation), which would benefit the existing owners of 2/3 of bitcoin mined so far, aka us. As per Metcalfe's law, we should expect at least 10^6 bitcoin appreciation if network expands 1000-fold (square of number of nodes)

Conclusion: In my opinion we should buy/mine while we still can and then allow companies like 21 inc create an ecosystem which might not result in more numerical bitcoins to us as they will be used by devices, but will result in a large value appreciation that would be beneficial in the final analysis.

[sidehack]

But does that work if one entity controls the majority of the new coins generated? I'm very much in favor of the value of bitcoin increasing per its utility as a currency exchangeable for goods and services increasing, but I don't trust heavily-endowed venture-capital-backed entities to not seek more immediate profits than strive somewhat selflessly to build up the economy as a whole.

My skepticism is not that what they're doing could be beneficial. My skepticism is that, because they want to see a sizeable return on their hundred million dollar investment, they'll try and control every available revenue stream back into their own tributaries and it'll have a more embittering effect as millions of customers realize they're basically being tricked into paying someone too much for a nonessential service. Mass distribution of mining gear is a good thing. But a single entity controlling it is not actually mass distribution of mining gear, it's mass dispersion of the cost of one entity's farming operation. That's where my hangup lies. If 21e6 can prove it actually has the bitcoin economy as a whole in mind (rather than greed and/or profit), which includes educating its customers on the actual costs and benefits of the hardware and services they'll be handing out instead of duping them into burning electricity to line 21e6's pockets, I might be able to get behind the idea.