It's called Novello Technologies Limited, we're based in the UK and we hope to make an announcement in the next two weeks. The USB miner isn't the only product, but it will probably be the one that gets most interest.

I'm sure that the contributors to this thread have amply answered your question, but I'd still assure you that there is a lot of life left in the USB miner concept yet, especially in view of the comment made by WetSeals about people that would like to mine, but have very limited funds, or that made by MiningSensei about how he built up his hashing power ....

Our company will shortly be offering a 100GH/sec capable USB miner (to get that it has to use the included AC adapter) at substantially under US$75, and no, that's not a misprint. 100 GH/sec at under US$75.

Does that proposition make sense to you?

Well, it's nice that you remembered us. Our project is alive and well, but this forum belongs to another company, so it would be quite incorrect to discuss anything relating to it here.

I was absolutely astonished to read your post effectively attacking a competitor. It's very easy for anyone to hide behind an 'open secret' as you don't have to name a source .... I find it hard to believe that whoever is dealing with KNC's business at their chosen foundry would risk legal action for them violating the NDA's that will be in place, let alone risk losing their job for breaking company rules on confidentiality. And where on earth do you get your timescales from?

I have no interest in your product or KNC's, so have no axe to grind here, I think what you have done is very unprofessional and may well cause your competitor and their customers unnecessary added stress, and for what? Do you hope to pick up some of their cancelled Neptune orders? And why not mention the 'other asic vendor we won't disclose here' when you're happy to 'disclose' KNC's stuff?

Just remember that secrets work both ways: how would you like it if someone said it was an 'open secret' that you're using asicminer chips? I think you owe the KNC people an apology.

Investment is a good thing, but many times it also equates to a substantial risk of losing your money. Experienced investors understand this, miners would have great difficulty in parting with their money for the promise of a potential future return. If company X says to you, "put $1000 into my company and I'll give you 1 share of the 10,000 the company will offer" the first question most miners will ask is "and what do I get for it?". The range of potential answers are mostly not what miners want to hear - they want a return, and in as short a period as possible. A conventional investment model simply won't work for most miners.


I'm sure that forum members appreciate what you're trying to do, and your philanthropic nature - I certainly do - but greed is always going to make people take more risks. Any experienced asic engineer looking at the Hashfast or Cointerra initial offerings would be alarmed at their underlying arrogance and lack of hard technical detail in their pitch. But many other simply see $ signs and the chance to get one up on their competitors, so take the risk. You'll never stop them, but I agree that it's worth the effort.

In this rather peculiar and specialised marketplace, normal rules don't apply. If you do away with all companies that start up with pre-order money, then you'll be left with a few that have past revenues to sustain their future development.

That would dilute any notions of competition - if there were, say, three left then there would be no real incentive to compete against each other, not when there would be such rich pickings, and like it or not, miners would end up paying a lot more for their hardware.

Just remember that buying is a choice, no one is forced to do it but they do because, as gmaxwell has pointed out, they have grand visions of making millions through mining. For most, that's simply never going to happen - you need too much money to start with to buy enough equipment, etc. That's not to say you can't make a reasonable return from a reasonable investment, especially if the chip/rig vendors stop being so greedy.

It will be very interesting to see what happens with the little ASICMiner chip - it's a nice, elegant idea but I'm sure it'll end up being hijacked by the middlemen and that end customers will never get anything anywhere near 1$/Gh let alone the headline $0.5/GH.

20nm technology is still not fully product engineered, and yields are unpredictable. The NRE's and unit prices are horrendous too, so there aren't going to be too many manufacturers queuing up to buy it, unless they've got money to burn and a similar amount of vanity.

Look at what ASICminer have done - smart use of existing, cheap technology with efficient design, and NRE costing about 1/10 of 20nm. Oh, and it consumes a lot less Joules/gigahash than current 20nm designs under construction.

Very good question as the gross margins TSMC and Global Foundries charge for Hashfast type volumes are closer to 70%. An 85% yield on a device of this size at this point on the learning curve is totally unrealistic - more like 60%. Taking these two together, the price per die should be closer to
$100, with testing and packaging on top of that.

It's not likely to change much in the near future unless huge volumes (Qualcomm scale) of wafers are ordered

I would hazard a guess and say 5 - 10,000. It really doesn't make sense to supply chips in the 10's or 100's, and if they did, it would be through a third party that is buying 1000's, i.e. a distributor, so it's extremely unlikely that you'll be able to buy 10 of these for $64 ($0.5 per gigahash) probably more like $200 ($1.5 per gigahash). That doesn't in any way make them bad value, quite the contrary, but I think that a lot of miners have expectations that they're going to be able to put together a rig for $0,5 per gigahash, and that's simply not going to happen.

Perhaps Asicminer could have worded their initial pitch slightly differently, but good on them for going against the herd and making the most of existing, cheap technologies. True Value Engineering.

Contrary to what you suggest in your rather stupid and rude reply, I do know what I'm talking about and you quite clearly don't. Haven't you anything better to do with your life, or is your homework too hard for you to do tonight?

Looks like a controller board using a PIC single chip microprocessor. It's most certainly not a Bitcoin mining solution of any kind as the PIC 33EP generally runs off it's internal 7Mhz oscillator giving 70 MIPS - way too slow for SHA256 by several orders of magnitude.

What's also not mentioned here is that everyone is not trying to solve the same 'puzzle', it's very, very unlikely that any two miners will actually be trying to solve the same problem, as individual 'puzzles' are based on a unique set of transactions : everyone has a different block to hash to find the solution.

However, overall the same rules of probability apply, and often it's not the exact solution that wins, just one that is equal to or less than the target. It's still a matter of more firepower = more chances of solution.

Try looking at the Getwork or Getblocktemplate (https://en.bitcoin.it/wiki/Getwork , https://en.bitcoin.it/wiki/Getblocktemplate) entries, the latter gives more explanation of how a block for solution is 'formed' and also explains what a Merkle Root is. It took me bloody ages to try to get a straight answer about how hashing actually works, so I can sympathise with the confusion over the whole process.

 

This makes no sense at all, unless they have their own asic hardware, or need subscribers to 'lock in' to equal monthly payments for a year. Either that or their fees are very high?

You can put a lot more than 5W through this size of package depending upon how it's mounted. PCB via's won't give a particularly good thermal path, but it is possible to use a copper slug to connect to the die pad (with a suitable square hole in the PCB)  and then connecting to a larger heatsink. That way over 25W can be removed - assuming die size of 5.3 x 5.3mm - with a large enough sink, the trick is in coupling the die, slug and heatsink together.

[/quote]

Cointerra, who has engineers who previously worked at NVIDIA, Intel, Samsung, Qualcomm and Nortel, currently cannot get their 2TH past 1.72TH after tweaking, and these unknowns have it working "24 hours straight never going below the rated Ghs" on the first run. Who is going to believe that.
[/quote]

Thing is, just because you have worked at a company doesn't necessarily mean that you've done any hard core asic design, or seen it to production. Yes, it's all very nice to read peoples 'public' CV's, but they tend to tell you nothing of any substance if you read them closely enough. I'm not referring to Cointerra here specifically, by the way. If I was to buy a rig, I couldn't care less about CV's, but I'd sure as hell want to know a lot about how they intend to implement their design as that will tell me an awful lot more about it's chances of success.

Asic designers nowadays rely heavily on the design tools, and you might have noticed that all the current or proposed 28nm solutions have remarkably similar w/gh figures, and there's a good reason for that - they all design in Verilog or C++ and get it compiled to silicon by the tools -there is next to no chance for any hand crafting, and that results in near identical designs at the chip level.

As regards the possibility of a 5GH mining rig in a reasonably sized desktop unit, it's perfectly feasible if you stop being lazy and complacent about the design process and go back to basics. And I don't mean using 20nm technology.

If you really wanted to get your money back, surely you have better things to do than this? It doesn't help your situation in any way, just gives others - who my not be HF customers - an opportunity to hijack your thread.

If I was part of HF's management, I'd be watching this thread in total disbelief.

Good question. It very much depends on why they are refunding you. If they say, honestly, that they cannot guarantee a delivery date due to technical problems, then they're in the clear. However, if they ship to other customers the following week, then that reason could clearly be challenged, but you have to realise that this is commercial litigation, not criminal, and on both sides of the Atlantic it can take ages, and cost a lot to prosecute.

(Just a thought, by the way for US buyers - thought about applying for a  slot on 'Judge Judy'? It's not as stupid a suggestion as you might think, and I'm sure the networks would love it. Only problem is that the maximum you can sue for is $5000, and the other party has to agree to appear. But think of the publicity! Not sure if someone from, say Germany or the UK could apply to sue a US company?)

As for them hashing with 'your' equipment, I have discussed this at some length with my QC friend, and he is of the opinion that it would be nigh on impossible to prove that Hashfast were mining at all without 'inside knowledge', ie Bitcoin Wallet addresses, IP addresses and the like, and even if you could prove they were mining they might claim they were using other equipment. He thinks your chances of success in a legal battle are near zero - a judge would have great difficulty even understanding the concepts, as would a jury and there are too many variables in the arguments. Again I'm sorry to bear bad news, but it's best to be realistic.

And speaking of being realistic, have the Hashfast buyers thought of puting together a 'fighting fund' for their cause? If you could elect someone  that you all felt you could trust to represent you to Hashfast - not a lawyer - then they might talk to that person rather than trying to fight on hundreds of different fronts....and you'd pay the person for their travel/time etc. I'm sure there is someone among you who could fit the bil.

Despite the negative comments, I still think Josh is your best bet, purely and simply because Hashfast know he's in a senior position in his company and has already expressed very unbiased views concerning the situation - he could just as easily have slated them, but he didn't and I think any reasonable person would recognise that this was a very fair and mature position to adopt. I don't know him and know little about him, but anyone that doesn't jump on a bandwagon to get one over on a competitor deserves my respect. I'm also guessing that despite BFL's early problems, an awful lot of people made an awful lot of money from mining with their chips. Just an observation. I do think Hashfast would talk to him as they would have an 'audience' that wasn't going into a discussion with a prejudicial attitude, and ultimately, he's on a similar seniority level.

Put it this way, if you thought he could get a solution - even if you think he's the devil incarnate - would you agree for him to represent you?

Thanks for taking the  time to post this. I understand your concerns, particularly with the possible increase in difficulty a $1/Gh miner might bring, but there's an underlying point here - dedicated mining companies already have very cost effective hardware and can deploy said for around $1.5/GH (or less).

That's what you're up against, so like it or not the difficulty is going to keep going up whether or not we make rigs, although not anywhere near as much as some of the more extreme posts would suggest. By August, our behavioral model reckons it to be about 20-22000 MM, about 10x where it is today (or 140,000 TH). At that level a 250GH setup would earn about $150 a month, or $5 per day, and pay for itself in about 8 weeks. If you can get a better return for that on your investment then go for it.

The average schlub who can't even get a seat at the mining table right now might see it as being a rather good deal for his $250.

There's nothing magic about our miner, just good design and a good dose of research. The engineers behind this are all in their 50's and want to do this as a professional challenge, money is not their primary goal and so the notion of a competitor luring one away is unlikely. They already have been offered a stellar deal by a VC group, but turned it down because the VC wanted to keep it 'in -house' and do exactly what you said and use it to mine for their own enrichment. The engineers didn't intend this design to make more money for those who already have it, they genuinely want to democratise the mining process and open it up to everyone.

Ultimately, like every other potential buyer, you pay your money and make your choice. I do hope you might consider us when you get to that point.

As an impartial observer in this situation, I'm seeing lots of contributors turning against one another over a problem that's been caused by a third party not delivering what they promised. To make matters even more confusing, a senior employee of another supplier is trying in his own way to bring a sense of reality to proceedings and possibly to help people come to terms with the situation.

I've been discussing this ongoing show with a friend of mine who's a QC (Queens Counsel or Barrister - basically a very senior lawyer who has a Right Of Audience in the highest courts in the UK) and sometimes stand-in judge, particularly over the T&C's of Hashfast and the thorny subject of refunds. Now, the first thing to recognise here is that American and UK contract and consumer law are not the same, however they are similar enough that a judge in either jurisdiction could make sense of and rule on the terms of a contract. The exceptions are the notion of 'class actions' which are not (usually) recognised in UK law, and 'fair terms and conditions' in UK consumer law which does'nt seem to have a parallel in the US.

I don't intend to bore you with the details but he had some very interesting observations.

The first and most important one was that his was not a normal sales agreement - it was more akin to a 'futures' contract for delivery of a piece of electronic equipment at a fixed price at some future time. When buyers agree to purchase the contract, it was - by the nature of the equipment and it's application - fully understood by the buyer that the equipment did not as yet exist and would only do so if sufficient buyers purchased contracts to fund it's development and manufacture. You might at this stage argue that people preorder new model BMW's or Iphones, but in these cases the development and manufacturing costs have already been met by those suppliers. It is also clear that the buyers knew that there was a risk that things might not go right, and were willing to accept that risk in return for a possible future commercial advantage.

Where this gets muddy is not in what was purchased, but rather what would happen if Hashfast failed to deliver the equipment, especially with their MPP where they are unable to supply the promised additional chips to make up for lost hashing power. I couldn't find an original copy of the original Hashfast T&C's but the current ones clearly distance Hashfast from any consequential damages from their non-delivery and limit liability to the purchase price only. For the original equipment, this would be the most likely ruling from judges on both side of the atlantic, ie neither party would gain or lose any benefit. For the MPP it's a lot more complex, but basically it's another 'futures' contract although in this case the damages could be quantified in terms of lost BTC's -regardless of whether a court accepts that they are legal currency- if Hashfast cannot deliver the missing chips. It seems they haven't built in an escape clause for that eventuality, and it may be their undoing.

Had Hashfast been contracted by a third party to build rigs for them and them  only, the contract would no doubt have included T&C's from the buyer about consequential damages or remedies in the event of non performance by Hashfast. With regards to the thorny question of refunds in BTC, my friend pointed out that Hashfast advertised products for sale in US$ OR Bitcoins, so there was a clear equivocacy between the two. It would be entirely reasonable for an 'average person' to know that being a US based company, Hashfast would pay for it's supplies and services in US dollars, and sell products for said. The fact that it was willing to accept a token of value, ie Bitcoins does not dilute this and a buyer would be under no illusions that Hashfast operated in US dollars and would have to exchange Bitcoins for said to function normally. Following on from this, it is reasonable for Hashfast to refund in US dollars. Courts in the US or UK would both rule this normal and reasonable. The fact that people might prefer BTC would not come into it, Any loss claimed by a buyer because of the rise in Bitcoins value would be swept aside - they would be no worse off with a US diollar refund now that they were when they made the purchase, and for the reasons cited above, neither (under normal circumstances) would Hashfast. Whether or not they promised refunds in Bitcoins is a moot point. If the value of Bitcoin had bombed, you wouldn't mind getting refunded in $US -that's what a judge would say.

This might not be what many of you want to hear, but it's what the end result of a court hearing is likely to bring. Hashfast can't 'force' you to take a refund but a refund is their only remedy if they can't deliver their side of the bargain. If they are refunding you or terminating the contract for convenience, then that's a whole different matter and ventures on fraud, although it would be extremely difficult to prove. You might be able to track what happened to your Bitcoins, ie did they go to an exchange when or soon after you paid, but even if they didn't it wouldn't prove Hashfast was acting in any way fraudulently.

So where do you go from here?

Well, it would help if you stop attacking one another and focus oh the real source of the problem. I don't know to what extent Hashfast are or aren't communicating, but at the risk of getting hundreds of abusive posts, might I make a suggestion?

Josh is a very prominent member of the Bitcoin community and, love him or hate him, he is an important part of a leading rig supplier. He has continued to post in this thread despite the abuse he gets, and there is some sense in what he says about the development process and things that can (and do) go wrong and some buyers attitudes. I'm sure from an entirely pragmatic perspective he doesn't want Hashfast to crash; competitor or not, it's not good for business.

If asked, he might be willing to try to talk to the Hashfast management, if only to try to get them to communicate with their buyers. He has a much better chance of doing so than any of you. I know it's not how these things normally work, but both companies and Hashfast's customers could stand to gain from a bit of pragmatism.

How about it, Josh?

Thanks for you good wishes, and for taking the time to explain your thoughts.

I can tell you that we do indeed have some tricks up our collective sleeves and have some truly devious engineers who seem to think that electronic devices - especially asics - have some inate grudge against them and their profession. I would say 'we' but it's really 'they' that have tried to design all unnecessary costs out, I'm sick to the back teeth of endless arguments and debates about heatsinks, for example, just to save a dollar here and there but they seem to view it as some kind of entertainment.

We know prices will keep coming down, and our way to survive will be to become a true volume supplier. We're in this for the long term, not as a one shot deal to make the principals overnight millionaires. I can appreciate your comments about prices coming  down, but if you've set your expectations high, that's going to be very difficult to do. There is a limit to how cheaply you can make rigs profitably, but we're a long way from that point at the moment, I'm glad to say.

I'd also like to point out (as has been pointed out to me many, many times over the past few months) that our strategy is to assume that things will fail and go wrong, and devise ways to ensure that if they do, we have a ready made plan to recover and recover quickly. Better still, try to make sure that the design, procurement and manufacturing systems are engineered to ensure that nothing fails. Unlikely, but you gotta try.

There's no denying that it would be nice if we got funded and brought our products to life, but as you said there's not exactly a great track record of on time delivery by the rig companies, so it's perfectly understandable that buyers might be reluctant to part with their hard earned money to fund yet another start up. That's a very important first lesson to learn. If the project doesn't happen it's not the end of the world for any of us and I know the engineers have already had a huge amount of fun trying to squeeze a quart into a pint pot. It's more like my idea of torture, but each to their own.