Topic: Qualcomm funds startup to use Bitcoin for Internet of Things (Read 5552 times)

I'm skeptical of that claim of 16nm considering most consumer devices aren't even at that scale. But if it even has a remote merit I can't see how 21inc could compete.

I thought 21 was developing application layer for those smart devices or ICs (Qualcomm in particular). I really do not know what kind of hardware is in the pipeline but judging by KnC they are already on the way to 16nm with 0.07 / GH efficiency (ref: https://www.knccloud.com/blog/archive)

That only works out if 21inc can ensure providing the most efficient Bitcoin ASIC in the long run, if at any point the competition catches up it's over for them.
I don't even think it would work in the short run because as I recall the competition is already pretty close to the latest production node at 28nm. Thinking that they surpass the competition by a significant margin is unrealistic imho.
So what's left is a level playing field where electricity costs make up more then half of the value of mined Bitcoins in the latest design. I already posted about the economics of scale, meaning that in principle a dedicated larger Bitcoin miner will always be more efficient than a small hybrid design.

But lets say they pull a rabbit out of their hat and arrive at >2gh/w wall plug efficacy. How long do you think would it take the competition to catch up and even surpass them? How can they hope to remain providing the most efficient design considering the economics of scale? Keep in mind that, right now the break even point for new mining hardware is close to 8 years. (If price and difficulty stays the same)

I hear you, but we are not talking about anything this powerful... Built-in IC could hash at 1GH and consume less than 0.5W, and with new tech around the corner to be even faster and more efficient. Now, multiply this by millions of smart devices around the world and you have one powerful decentralized network. IBM, CISCO, WHIRLPOOL, etc can now create their own mining pools those devices will hash to and get extra stream of revenue, perhaps sharing some with their customers. I just do not see anything technologically unreachable here I guess.

Cuz it has to be a dedicated ASIC on the SOC which would compete with silicon area of the rest of the device. That makes this thing especially uneconomical in comparison to dedicated devices.
The problem is the whole hybrid design aspect which doesn't live up to the promise of more than the sum of it's parts. Take hybrid laptop/tablet PCs for instance. They are both heavier than real tablet PCs, have less performance than laptops and are less durable than both. A set top box mining bitcoin would have less Gigahash/Watt while having for instance a slower GPU than a standard box.
Worse, if that set top box should have any chance of ever recouping the hardware costs it should have active cooling and a larger power supply to enable at least some reasonable hash rate both which are superfluous for a regular set top box consuming <10W

Toasters are probably a bit of a stretch but I do see micro miners / Bitcoin nodes being built-in on some home smart devices that already have CPUs and wi-fi controllers. Why not?

That's really... awful? I'm finding it hard to believe that article.

Loool

http://ftalphaville.ft.com/2015/04/30/2127543/meet-the-company-that-wants-to-put-a-bitcoin-miner-in-your-toaster/

tl;dr They want to offer cost-free consumer devices, mining bitcoin to pay for the cost of the device. All, of course based on difficulty and bitcoin prices from last year.

I really don't know how this system works. But I have heard people argue that blockchain technology (not the bitcoin blockchain specifically) can be used to handle cell phone identity in a way that would not require sim cards. Apparently sim cards are not liked by cell providers because they require a complex distribution. The cards must be shipped and stocked at stores then sold and reconciled online. The holy grail is a system that can all be done from the phone, one that cuts out the store and sends all profits to the provider. But I am not sure how this happens using a blockchain.

But why using the block chain? I don't think that this makes any sense, cell phone providers don't need to store identities in a decentralized manner. I mean, it would be great if someone could described a very useful use case for this but I don't see one at the moment. Also, that's quite some money for a record keeping technology.

Another theory about what they are doing in not specifically bitcoin related. There are ways of using a blockchain to verify identity and perhaps do away with the need for sim cards. Getting rid of those sim cards is a desire of the phone industry.

 

Yup, this will still take quite some time. But there have to be people who know something. Otherwise this is just some guys throwing an incredible amount of money onto a company they don't even understand correctly,.... wait a minute... never mind.

I agree, this is long-term material. If this really materialises in some sort and becomes successful, this may prove to be beneficial to Bitcoin in maybe 3 years or so. I don't know their exact business model, but if they really invested this ton of money they better have a really really decent plan in place!

Well secrecy is usually kept just in case someone else tries to imitate them
That and if they will take a year or two they probably don't have all the specifics yet

I'm more interested to know when 21 will actually reveal what exactly they are doing? Anyone have any idea on this?

Also, why the secrecy in the first place, what's the purpose?

Well they're still doing mostly very basic research, I believe... But if I'd have to guess....? Maybe late 2016? Definitely not this year. These things could be still quite far down the road from here. You have to think of it as something that may impact Bitcoin in a very slow but big way (if it works out), rather than a slow pump that's over before the daily candle closes!

Any guesses on when the first product or service will be released by 21 inc?

Again, this may be *the* killer app in the making. The thing we've been waiting for for years now. They wouldn't have invested that kind of VC money if there was no plan ready to put into action!

Here is another thought on the subject (Reddit) : http://www.reddit.com/r/Bitcoin/comments/336f1w/its_not_about_21_inc_its_all_about_qualcomm/

Maybe they are entering the ASIC market but that seems risky and short sighted. If I were them I would leverage my power to open new markets with hardware wallets and integration. Even a better ASIC is only a short term money maker.
Since they are playing this close, we might not know until they make an announcement.

Baking that into an ASIC would be nonsense since all that can be done easily on the phones cpu. Why am I even telling you this, do you pretend to be tech illiterate?
Yeah you guessed it too, mining on a phone doesn't make sense either.

proof by reductio ad absurdum: It's a Bitcoin mining ASIC company, just like the others we've seen.
And why the long faces? Perhaps you get double the power efficiency or something.

Agreed. Phone mining is an absolute non-starter. My guess is hardware wallet chips for phones and other wired devices. In the future perhaps your TV will auto-pay when you select a movie from a streaming service. But to mine from a phone you are going to have to back to the past. Back to 2009.

Mining on a smartphone scale just doesn't make any sense at all. It'll degrade the user experience by a ton (battery), also it doesn't make sense to implement a hard-wired one-purpose mining chip in a phone.

I hate having to quote myself but here you go, inca.

e: wtf, it was in this very thread

In practice you'd probably have to pay off everyone in line, if there were that many people, which would be expensive but some would likely do it. My guess is all vehicles would be in some kind of OpenBazaar-style web of trust anyway, and that would likely smooth out a lot of the edge cases.

it'll be a chip. for crypto identity, hardware wallet with supporting nodes and private, priority mining.

to be installed in Samsung phones 2016

 

Source? Or just doing your usual trick of presenting opinion as fact?

Also it's Bitcoin ASICs get over it.

Wow I almost missed that nugget.

Hard to come up with a better argument against randian fantasy babble than simply displaying it.

...so long as the guy in front of you isn't blocking your paid for fast lane.  Maybe in a generation or so, but it's going to take a while.  Look at toll booths in London.  It will take a lot of time to implement that around the world.  Great concept though for wealth distribution (and being an intentional sunday driver).

I hope they will make customer-friendly hardware wallet / SIM card for bitcoin
I wonder if bitcoin price can rise up after this news ?

Exciting visions of the future right there!

Wired and monetized everything, blockchain identity for cell phones; my mind is blown also. There are so many directions this could go that I don't know which way to look.

Whatever they plan on doing, the fact that this guy is near the helm is pretty damn exciting.

http://www.wired.com/2013/11/software-is-reorganizing-the-world-and-cloud-formations-could-lead-to-physical-nations

I don't think this is about bitcoin miners...at least, I hope so. It would be pretty anti-climatic.

That's actually not a bad idea. I also believe that the key (no pun intended) to security in the Bitcoin ecosystem lies in hardware-secured wallets. Trezors are the prototype, but it would be really interesting if this secret element (right now used to store fingerprints, etc.) is integrated into a phones' SoC or something.

Just imagine what's still to come! It's only March and we're up quite a bit. If this year goes well - also price-wise - we may be in for an even better 2016 (with reward halving, man!!!)

I wonder if this 116M is counted as investment in the 'Bitcoin ecosystem', like the 75M for Coinbase was. That will make 2015 going with a sick start, on paper.

A Bitcoin ASIC doesn't need to be a mining ASIC

They really should embed a secret element in the phone that enables the device to store BTC safely, man. This could work without any risk of the keys/seeds being exposed, even if the phone is rooted.

They just want to sell devices probably.

Yeah pretty much this. Already checked it out? They're hiring.

100 mil for a new asic chip?

 

 

with that kind of money they can own 50 % of the network right now.

https://www.cbinsights.com/investor/qualcomm-ventures


They are funding it (21 Inc, whatever the will be doing), not taking an active part in it, so no (insert buzzword salad) inside qualcomm chips.

Also that (insert buzzword salad) 21 inc would be doing looks to be probably mining:
https://www.cbinsights.com/company/21e6

Imagination combined with new "toys"  ==  A great time to be in the right place. 

Source is WSJ so gave it a read

21’s lead investors include U.S. venture-capital heavyweights Andreessen Horowitz and RRE Ventures, along with Chinese private-equity firm Yuan Capital, with a strategic stake going to chipmaker Qualcomm Inc.QCOM -1.13% through its venture-capital unit.

Additionally, Khosla Ventures and Data Collective have invested in 21, as well as chief executives and founders from various tech companies, including PayPal co-founders Peter Thiel and Max Levchin, eBay Inc. co-founder Jeff Skoll, Dropbox Inc. CEO Drew Houston, Expedia Inc. CEO Dara Khosrowshahi and Zynga Inc. co-founder Mark Pincus

This is an interesting set of backer companies and will be something I will keep track of as well.

Perhaps they are going to develop a more consumer-friendly Trezor

Qualcomm makes chips for phones.  One innovation of the Bitcoin blockchain is identity via private key.  This would make SIM cards for phones obsolete.

http://cointelegraph.com/news/113181/what-if-bitcoin-could-replace-sim-cards

Yes, that is exactly what is meant by "internet of things".  There is a cost to running every machine in the world, and there is a cost to allocating resources.  Objects are managed by their owners.  It can take a great deal of time to allocate your resources efficiently.  With an internet of things, your objects compete to provide services within your guidelines, but without your explicit micromanagement.  This frees humanity to focus on the bigger picture: their goals in life.

This can also be extended to the relationships between managers and employees.  A decentralized corporation will be owned by shareholders.  They will provide the currency that ultimately goes to the only employees in the corporation: individual contributors.  Management is eliminated completely, and replaced by a hierarchy of management software objects that work toward the goals defined by shareholders.  Anything that cannot be done by a machine is offered to human employees by the management software objects, probably all on a contract basis.  That includes design, implementation, verification, and support of new products.  Elimination of human managers is Dilbert's dream come true.  Shareholders set out goals and provide money.  Their corporation's IoT manages those resources toward the goals.

This is so mind-blowing!

Absolutely, and that's a great example.

Every rivalrous resource that is now being allocated dumbly, because it's not "on the market," can be allocated smartly, reflecting each person's preferences and priorities. A subset of that enormous set of possibilities is idle resources. The endpoint is zero idle time. Every machine working at full capacity at all times. This is so incredibly far from what we have now that the efficiency gains are staggering to think about. And if every machine of a certain type can go from, say, 9% utilization to 90% utilization, that means the cost to use or own those machines drops tenfold. The technology gets 10x cheaper, just like that. If this were to happen across the board, it would mean a massive increase in the standard of living and far less waste.

Holly $#!+ Could it be... maybe...

What if they are looking at producing chips to conduct BTC transfers that could go into ANYTHING?
Want to rent that car? Send from your phone to start your 100 mile rental. You could meter about anything, you could put such chips in everything from soda machines to hotel room doors (these have to be replaced anyway since they are now 100% hackable).
One could envision a world of devices ready to serve you as soon as payment is received. These devices could be anywhere. You could design a new shifter knob for your car then send your file along with payment to a 3D printer. There is no end to the ideas.  

Impressive if this is working out!

That's kind of fucking awesome.  Really cool thoughts in this thread.

I think you are spot on. Smart electronics are all connected now and they have no way of transferring value to make a deal.

Another easy to understand example would be "lease of computing power". Supposed I have a very powerful computer idling and someone needs a burst of computing power to crunch some numbers or to convert a video. Smart machines can find computing power auction sites and look for the cheapest rate for CPU time. Machines can complete the deal in bitcoin.

Looks like you are guessing right.

Bit_Happy, I think your on to something.  This looks like a new phase. If I'm guessing their intent correctly, they are creating hardware for the coming bitcoin economy. That's a big investment and vote of confidence over just accepting them for payment. It's all happening so fast!

The basic use case I imagine is the situation where two machines with two different owners are drawing from the same utility, say electric power, being offered by a third party. Suppose there isn't enough to power both of them. How are the machines to decide who gets priority in receiving the power?

Well, if the owner of Machine 1 has a high preference for the machine to run uninterrupted (say a server), even if it were to be expensive, he can pre-allocate some spending money to the machine, as an insurance policy in case of power shortages. Meanwhile, the owner of Machine 2, maybe a poorer person or maybe someone who cares less if their machine's operations were temporarily interrupted (say a clothes dryer), would not do this. In fact, they may even allow their machine to sell excess power if someone is paying enough for it.

Then the person who values the resource more can pay for it, enriching others who value the resource less.

This could be for electricity, for driverless cars negotiating right of way on the road for users of different income classes or different time preferences (rushing to the hospital vs. someone just going shopping), or for any other rivalrous good.

So it's basically a way to make machines more effective agents on the owners' behalf. Whereas before any situation that required a monetary transfer in order to effectuate a mutually beneficial deal would have required the owners to negotiate directly. If the matter is too trifling, it would just be annoying and a waste of time to negotiate. Instead, have your machines do it. They couldn't do this before, not just because they weren't interconnected, but also because they have no way of transferring value to each other.

Bitcoin enables this. And the scale can be extremely small, so that even the slightest possibility of mutually beneficial trade is capitalized on.

IoT by itself enables this to happen within one's own property, like the different appliances in your own house, because no other owners are involved so no money needs to change hands. The dishwasher doesn't mind taking a 5-minute break to give extra power to another appliance that needs it. It needs no monetary compensation, because the owner owns both appliances.

It's when this kind of exchange happens with someone else, especially a total stranger, that it becomes necessary for money to change hands.

Now some people imagine this as heaven for the rich, like if for example a rich driver could flip all the red lights green. How sucky would that be if you had to wait at extra-long reds? But no. You had already preset the amount you were willing to pay in bidding for the traffic light to change, and it was far less...but more importantly you get paid for your waiting time. If it's not worth your while to wait and get paid today, just set your asking price higher (a dial just for this purpose is on your dashboard). You only wait if the other driver is offering enough to make you interested in waiting.

This guarantees that everyone benefits from the deal. If the rich driver ultimately wins out and gets the light flipped, everyone else - by hypothesis - gets at least enough money to make it worth their while. Of course the more crowded the intersection the more the rich driver would have to pay, but some will. Everyone is better off, because goods (in this case road space and right of way) are allocated most efficiently by the price system.

This argues for fixing the block size limit in a scalable way asap.  Don't wait until the protocol is embedded in an immutable chip.  Andreas has insight into some of these things, and he is under NDA and thus cannot share his reasoning for pushing so quickly.

I am still curious about what exactly is the company going to come out with ?
The site only indicates miners for now.

He knew it.
Listen 25:00 to 26:35!

https://www.youtube.com/watch?v=9npQ5f74Nr4#t=25m00

At 26:20 he said:"Bitcoin protocol being embedded in chips.Chips that can't be modified very easily!"

Assuming BTC can scale then you'd be pretty brave/confident/insane to stake your operations on any other chain than that assuming it might be vulnerable in any way. It's the bulletproof daddy of them all.

I really don't know how many of these companies really want to use the *Bitcoin* blockchain, or just the general idea of a blockchain technology... But you need some incentive to keep the Blockchain decentralized, so a monetary reward (BTC) seems fitting!

The way IoT is described and understood today IS stupid, kind of like the "information superhighway" description in the early 1990s. But there is real, almost infinite potential in pieces of code doing business with other pieces of code (not toasters) and the blockchain is almost ideal for that kind of app.

Be careful what you type. In a few years time if your home appliances don't like your attitude they'll be able to put out a hit on you before you can turn them off.

The Internet of Things is probably the stupidest idea today, I hate to see it get associated with Bitcoin

This could be one of the biggest stories in the history of Bitcoin.
We are seeing Bitcoin explode onto the global stage!

http://blogs.wsj.com/digits/2015/03/10/secretive-bitcoin-startup-21-reveals-record-funds-hints-at-mass-consumer-play/

Qualcomm's chips represent a large portion of the cell phone market.  They intend to bring their cellular chip technology to all manner of interconnected things, and use the Bitcoin blockchain to do so.