Dreaming about open sourcing an ASIC chip

[cedivad]

Let's suppose that i get rich one day. (This is an hypothetical me, i'm nowhere near that, so don't pm me trying to scam me). Let's say that i have 2M$ (20kBTC). I have all invested into Bitcoin (because i'm crazy) and i want the Bitcoin network to be as strong as possible. I decide that i need to create and to open source to the community a low-nm ASIC chip. I'm ready to invest up to 1M$ into it.

I do that for free, because thanks to Bitcoin i'm rich and i want to give it back to the community.

Do you think that this will actually happen?
If you own that much, why don't you do it?
Is 1M$ nothing in this business?

[1713946298]

1713946298

[1713946298]

1713946298

[1713946298]

1713946298

[tigerfree]

Go China contact any Chip Designer there  that is all .I think .

[SlyFoxy12]

Actually, find a good designer in your own country and at the same time shop around in china for a chip factory with some basic specs of manufacturing requirements from the designer, then get the final design and pump out the prototypes. the problem with ASIC is that to get the best grade of power efficient to hash cracking ratio you need to use more and more complex techniques in that process which makes it harder to just 'do it yourself'. On the other hand, if you look at things like the Raspberry Pie boards, they wanted to originally manufacture in the UK but couldn't because ultimately in VAT it cost more to import each component then have them sent to china and then imported as a whole, they some how got around this later on but I guess it was only due to price drops in demand that made it more feasible.

So ultimately, you could come up with an 'open source' design but chances are the design wouldn't matter if it was open source as it's access to manufacturing and distribution is the difficult bit.

[cedivad]

 , received via pm:

Hi..we're IC designer team.
we've been already complete the ASIC miner chip.
But we havn't enough money to start it.
W just want to find somebody pledge us.
if you can.we are happy to open the source and be the cheapest chip supplier.
Because we want the bitcoin stronger too.
Very thanks

[Dabs]

Well, you go ahead and start designing an ASIC chip with low-nm. Something like 28nm or even lower.

Then, with your money, you order a run of 100,000 chips.

Make a working miner with these chips, that does maybe 100 GH/s. Sell them at cost, or give them away. (If you do this, please remember me, I will be happy to receive a sample unit.) Make a small unit that does 10 GH/s, and a medium unit that does 50 GH/s, and a large unit that does 100 GH/s. Numbers are all examples, you use whatever numbers seem appropriate when the time comes.

Release the source code, the software, the schematics, everything, and how to use the chips.

Then allow people to order just the raw chips from you, you order from your foundry.

People will receive the chips, and they also order separately all the other parts from everywhere else.

You keep supplying chips as long as people keep ordering. The foundry will not care as long as they keep getting paid. You could make chips for the next 10 years, and that would be a worthwhile endeavor because it will be 10 years of bitcoin hashing that will drive difficulty to the stratosphere and secure the network and all that hashing power is distributed world-wide, in 100 countries, in all 7 continents.

Then your next project is to work on lower-nm chips, like 10nm or even lower.

Look at your typical calculator. They still make the chips, but they're not much smaller compared to 10 years ago. I mean, the calculator can be small, but it has to be large enough to hold in your hand to be practical to use.

You just need ASIC chips that do more than any GPU or FGPA can ever do, and maybe keep making the next generation as soon as you can afford to.

[SlyFoxy12]

, received via pm:

Hi..we're IC designer team.
we've been already complete the ASIC miner chip.
But we havn't enough money to start it.
W just want to find somebody pledge us.
if you can.we are happy to open the source and be the cheapest chip supplier.
Because we want the bitcoin stronger too.
Very thanks

Did they even state how much they are looking for?

[KS]

, received via pm:

Hi..we're IC designer team.
we've been already complete the ASIC miner chip.
But we havn't enough money to start it.
W just want to find somebody pledge us.
if you can.we are happy to open the source and be the cheapest chip supplier.
Because we want the bitcoin stronger too.
Very thanks

I can do it for 2M $ in 4 weeks. I'll send you my Paypal address. Just be sure to include your details for the invoice

[Silverpike]

, received via pm:

Hi..we're IC designer team.
we've been already complete the ASIC miner chip.
But we havn't enough money to start it.
W just want to find somebody pledge us.
if you can.we are happy to open the source and be the cheapest chip supplier.
Because we want the bitcoin stronger too.
Very thanks

Looks legit!

[tom_o]

On the other hand, if you look at things like the Raspberry Pie boards, they wanted to originally manufacture in the UK but couldn't because ultimately in VAT it cost more to import each component then have them sent to china and then imported as a whole, they some how got around this later on but I guess it was only due to price drops in demand that made it more feasible.

Nope it was assembly price/bulk assembly.

UK VAT is 20% whether it's individual pieces or the whole product. Think about it, the materials would be cheaper than the final product so the VAT on materials is LOWER than the VAT on the finished product. Doh.

[tvbcof]

If I wished to give something back to the community, it would not be 'low-nm' mining chips.

I think it would be both cool and effective to have individuals who need heat for one think or another (like a hot water pot or rice cooker) be able to mine and have this class form the backbone of distributed crypto-currency hashing (for those distributed crypto-currencies which use massive hashing as a backing.)

What I would see happening in this case is that people doing mining are not really even thinking much about power.  It is either otherwise wasted, or so small and expense in their personal economics that it is written off.  Large commercial mining operations need to compete against in an areana where they are at a dis-advantage on a major expense.

In that case, it makes more sense to focus on chip packaging designs than on the core fab technology.

[tom_o]

If I wished to give something back to the community, it would not be 'low-nm' mining chips.

I think it would be both cool and effective to have individuals who need heat for one think or another (like a hot water pot or rice cooker) be able to mine and have this class form the backbone of distributed crypto-currency hashing (for those distributed crypto-currencies which use massive hashing as a backing.)

What I would see happening in this case is that people doing mining are not really even thinking much about power.  It is either otherwise wasted, or so small and expense in their personal economics that it is written off.  Large commercial mining operations need to compete against in an areana where they are at a dis-advantage on a major expense.

In that case, it makes more sense to focus on chip packaging designs than on the core fab technology.

You forget that these hot water pots or rice cookers would have to be connected to the internet and contain silicone that is somehow temperature hardened to mitigate electromigration and breakdown of the chips.

Also, the wattage output from processor cores aint gonna have an useful output.

Datacentres maybe, but not on a commercial scale!

[SlyFoxy12]

On the other hand, if you look at things like the Raspberry Pie boards, they wanted to originally manufacture in the UK but couldn't because ultimately in VAT it cost more to import each component then have them sent to china and then imported as a whole, they some how got around this later on but I guess it was only due to price drops in demand that made it more feasible.

Nope it was assembly price/bulk assembly.

UK VAT is 20% whether it's individual pieces or the whole product. Think about it, the materials would be cheaper than the final product so the VAT on materials is LOWER than the VAT on the finished product. Doh.

Actually... because sadly I hate to lose I went and looked this up, a quote from the raspberry pi blog:

"I’d like to draw attention to one cost in particular that really created problems for us in Britain. Simply put, if we build the Raspberry Pi in Britain, we have to pay a lot more tax. If a British company imports components, it has to pay tax on those (and most components are not made in the UK). If, however, a completed device is made abroad and imported into the UK – with all of those components soldered onto it – it does not attract any import duty at all. This means that it’s really, really tax inefficient for an electronics company to do its manufacturing in Britain, and it’s one of the reasons that so much of our manufacturing goes overseas. Right now, the way things stand means that a company doing its manufacturing abroad, depriving the UK economy, gets a tax break. It’s an absolutely mad way for the Inland Revenue to be running things, and it’s an issue we’ve taken up with the Department for Business, Innovation and Skills.

So we have had to make the pragmatic decision and look to Taiwan and China for our manufacturing, at least for this first batch. We are still working hard on investigating UK possibilities; at the moment, we’re investigating an option which would mean that all the Model As (whose demand we expect to be much lower than that of the Model Bs) will be built in the UK, and at the moment that’s looking quite do-able, although it’s not as efficient economically as doing it in Asia. I’ll fill you in on how that goes later on."

Link here if you want to claim foul http://www.raspberrypi.org/archives/509

And before it, yes the reason I stated was slightly different to what it actually was but it was tax related any generally I was just making a point that the difficulty with ASCI chip production is making it efficient while competing with companies who can use proper factories and use investment money to make them in bulk.

, received via pm:

Hi..we're IC designer team.
we've been already complete the ASIC miner chip.
But we havn't enough money to start it.
W just want to find somebody pledge us.
if you can.we are happy to open the source and be the cheapest chip supplier.
Because we want the bitcoin stronger too.
Very thanks

Looks legit!

Well, it could quite easily be a scam but if they want to give their design away for free then surely they'd do it already or at worse look for small investors for prototypes to then use for a kick starter scheme.

[tvbcof]

If I wished to give something back to the community, it would not be 'low-nm' mining chips.

I think it would be both cool and effective to have individuals who need heat for one think or another (like a hot water pot or rice cooker) be able to mine and have this class form the backbone of distributed crypto-currency hashing (for those distributed crypto-currencies which use massive hashing as a backing.)

What I would see happening in this case is that people doing mining are not really even thinking much about power.  It is either otherwise wasted, or so small and expense in their personal economics that it is written off.  Large commercial mining operations need to compete against in an areana where they are at a dis-advantage on a major expense.

In that case, it makes more sense to focus on chip packaging designs than on the core fab technology.

You forget that these hot water pots or rice cookers would have to be connected to the internet and contain silicone that is somehow temperature hardened to mitigate electromigration and breakdown of the chips.

Also, the wattage output from processor cores aint gonna have an useful output.

Datacentres maybe, but not on a commercial scale!

Water boils at 100 centigrade at sea level pressures.  It's not commonly used to cool chips for fairly obvious reasons associated with how most chips are used, but it has one of the highest specific heats around and is super common for cooling many other things because it is really good in that capacity.  I doubt that it would be technically all that challenging to develop or produce a packaging form factor which employed water cooling if gathering the waste heat was an actual goal.

For something like a water pot one would need to be producing only as much thermal waste as is needed to account for insulation loss.  Most devices would be built with traditional heating elements to augment thermal production at higher demand times.

Something like a rice cooker would probably not be cost effective until truly commodity elements were around since one would be at idle to much.  But an American style tank hot water heater, or a swimming pool heater and so on would be really well geared for such a thing.

As for communications, that technology is already quite commodity and wifi systems are getting pretty ubiquius in homes these days.  At least in developed countries.

[cedivad]

Using electricity to warm up things (be it a house or swimming pool) isn't cost effective at all, and this is why it's not being used.
I don't think that we will see mining chips in something we use everyday and with other purposes, because it doesn't make economical sense to put them there.
I don't agree with your "we don't need low-nm chips because we will use the heat for other things" point of view.

[tom_o]

Using electricity to warm up things (be it a house or swimming pool) isn't cost effective at all, and this is why it's not being used.
I don't think that we will see mining chips in something we use everyday and with other purposes, because it doesn't make economical sense to put them there.
I don't agree with your "we don't need low-nm chips because we will use the heat for other things" point of view.

Nail -> Head

[tvbcof]

Using electricity to warm up things (be it a house or swimming pool) isn't cost effective at all, and this is why it's not being used.
I don't think that we will see mining chips in something we use everyday and with other purposes, because it doesn't make economical sense to put them there.
I don't agree with your "we don't need low-nm chips because we will use the heat for other things" point of view.

I'm not sure where you live, but in the US it is pretty common to use electricity to heat things.  In fact my house is well above outside air temp due to electric heaters as I sit here today.  And my coffee is warm for the same reason.

You are free to disagree of course.  I don't have a crystal ball, so we'll just have to wait and see on this one.

I will say that I just vastly overpaid for a Block Erupter and would do the same for any other devices which had the potential to distribute hashing power widely since I feel that the 'peer2peer' nature of distributed crypto-currency solutions is what gives them their strength.

[tom_o]

Using electricity to warm up things (be it a house or swimming pool) isn't cost effective at all, and this is why it's not being used.
I don't think that we will see mining chips in something we use everyday and with other purposes, because it doesn't make economical sense to put them there.
I don't agree with your "we don't need low-nm chips because we will use the heat for other things" point of view.

I'm not sure where you live, but in the US it is pretty common to use electricity to heat things.  In fact my house is well above outside air temp due to electric heaters as I sit here today.  And my coffee is warm for the same reason.

You are free to disagree of course.  I don't have a crystal ball, so we'll just have to wait and see on this one.

I will say that I just vastly overpaid for a Block Erupter and would do the same for any other devices which had the potential to distribute hashing power widely since I feel that the 'peer2peer' nature of distributed crypto-currency solutions is what gives them their strength.

In everywhere other than the US electric is significantly more expensive. For example - in the UK I pay 15pence per kwh and that's quite cheap here! Electric heating is about 2-3x more expensive than gas or oil.

[Rampion]

, received via pm:

Hi..we're IC designer team.
we've been already complete the ASIC miner chip.
But we havn't enough money to start it.
W just want to find somebody pledge us.
if you can.we are happy to open the source and be the cheapest chip supplier.
Because we want the bitcoin stronger too.
Very thanks

Epic. Wild west style

[tvbcof]

Using electricity to warm up things (be it a house or swimming pool) isn't cost effective at all, and this is why it's not being used. ...

I'm not sure where you live, but in the US it is pretty common to use electricity to heat things.  In fact my house is well above outside air temp due to electric heaters as I sit here today.  And my coffee is warm for the same reason.
..

In everywhere other than the US electric is significantly more expensive. For example - in the UK I pay 15pence per kwh and that's quite cheap here! Electric heating is about 2-3x more expensive than gas or oil.

I live in the Western US.  Hydro-electric is common which brings prices down.  At the end of the day, a lot of people use electricty to heat a lot of things, at least where I live and that's not likely to change.

A neighbor of mine is lucky enough to have a waterfall and has set up a hydro system.  (Got it from Vietnam where they are in common use.)  It's just kind of a hobby system and he just has some baseboard heaters simply for the purpose of doing something with the electricity even though he, like most of us, primarily use firewood for space heat.  (I've got a home both in the country and in town.)

Any solar-electric systems are probably going to have periods of time when they are producing an excess of usable electricity.  I guess it can be sold back to the grid, but I'd personally just as soon mine Bitcoin...and ideally get some water heated up for later use at the same time if possible.

Many people feel that fossil fuels will become less and less desirable going forward.  For various reason.  Until there is a 'hydrogen grid', using electricity to transfer energy around is probably not going anywhere.  And again, if there are places to insert a hashing device into the system 'for free', devices which are designed for that will, hopefully, appear.

[KS]

Using electricity to warm up things (be it a house or swimming pool) isn't cost effective at all, and this is why it's not being used. ...

I'm not sure where you live, but in the US it is pretty common to use electricity to heat things.  In fact my house is well above outside air temp due to electric heaters as I sit here today.  And my coffee is warm for the same reason.
..

In everywhere other than the US electric is significantly more expensive. For example - in the UK I pay 15pence per kwh and that's quite cheap here! Electric heating is about 2-3x more expensive than gas or oil.

I live in the Western US.  Hydro-electric is common which brings prices down.  At the end of the day, a lot of people use electricty to heat a lot of things, at least where I live and that's not likely to change.

A neighbor of mine is lucky enough to have a waterfall and has set up a hydro system.  (Got it from Vietnam where they are in common use.)  It's just kind of a hobby system and he just has some baseboard heaters simply for the purpose of doing something with the electricity even though he, like most of us, primarily use firewood for space heat.  (I've got a home both in the country and in town.)

Any solar-electric systems are probably going to have periods of time when they are producing an excess of usable electricity.  I guess it can be sold back to the grid, but I'd personally just as soon mine Bitcoin...and ideally get some water heated up for later use at the same time if possible.

Many people feel that fossil fuels will become less and less desirable going forward.  For various reason.  Until there is a 'hydrogen grid', using electricity to transfer energy around is probably not going anywhere.  And again, if there are places to insert a hashing device into the system 'for free', devices which are designed for that will, hopefully, appear.

You're pointing out a lot of special use cases. Most ppl don't have a waterfall, or wood, or solar panel installation, or cheap electricity.

If I'd decide to give back to the community, I'd ask the community what it wants.