Theres an old saying that a backup doesnt exist untill its in 3 separate places  thats a good saying. Additionally, rjk's Third Law Of Backups states that if your backup procedures are rigorous and the backups are stored in at least 3 locations, you will never ever have to recover from them. If however they suddenly disappear due to a freak accident, that is the day you are guaranteed to need them the most. Kind of like the Law Of Extra Parts also by rjk: if you need 5 screws to complete a project, and you bring only 5, you are guaranteed to lose one of them. If however you bring 6 screws to a project that needs only 5, you are guaranteed to always have 1 left over.  Lol! You're just taking credit for Murphy's Law!  In rjk's defense, Murphy's Law states that anything that can go wrong, will go wrong. This is similar but with a slight modification. In other words, "Any problem that you anticipate won't happen and every other one will." Surely we're not talking about Hanlon's Razor on this thread, are we? Never attribute to malice that which is adequately explained by stupidity. ~BrunO~ No. Its a special corollary of Murphy's Law that I don't think has a name but most people are aware of: Murphy's Law is efficient, if you're sufficiently guarded against it in one way, it'll just strike you some other way. |
|
|
|
Theres an old saying that a backup doesnt exist untill its in 3 separate places thats a good saying. Additionally, rjk's Third Law Of Backups states that if your backup procedures are rigorous and the backups are stored in at least 3 locations, you will never ever have to recover from them. If however they suddenly disappear due to a freak accident, that is the day you are guaranteed to need them the most. Kind of like the Law Of Extra Parts also by rjk: if you need 5 screws to complete a project, and you bring only 5, you are guaranteed to lose one of them. If however you bring 6 screws to a project that needs only 5, you are guaranteed to always have 1 left over. The generic version of that is just Diablo's Rule #1: Redundancy in planning is not paranoia. |
|
|
|
Yes, believe it or not, but your posts and PMs on this forum are actually safer than your current balance at Bitcoinica.
I lol'd. |
|
|
|
Yeah I already saw that. I'm going to see it when I believe it. There has been several attempts at ASICs, all of them failed except for ArtForz's, and he wasn't too happy with his attempt. |
|
|
|
Only a matter of time right. I have no idea how to run a business but I think one of you business savvy people should do it and put some shares on GLBSE to raise your funds, and so the little people can be a part.
Regular cash can be accepted but you also accept bitcoins.
Menu Items can include things like, The Block Chain Burger ect.
You can have a nice pool out back and call it the mining pool.
If you want to go the sleezy way you can have a strip bar, and instead of putting money in their underwear you scan the QR code on their underwear to send the strippers bitcoins.
Just throwing it out into the universe.
I demand it has a burger called "The Diablo". It must contain bacon. Preferably, thats all it should contain. |
|
|
|
If you have excess land you could run alpacas on it, they are pretty much self sufficient lol.
Alpacas are naaaaasty animals. They bite and spit and smell and produce entire metric tons of shit. I mean, the shit I probably can sell to a local farmer, but the rest of that? Nothx, do not want. |
|
|
|
What if you can earn more renting out the DC for purposes other than bitcoin mining ? You might get more return for the electricity generated just by selling back to the grid...
Actually, I've considered that. Such a thing isn't off the table. However, I would probably rent it out to other mining operations. High bandwidth things like webservers and such would be inappropriate for this facility. |
|
|
|
No, not "completely played off". This isn't a bank loan. I own the panels in full on day one.
It's not a loan, but the shareholders who gave you that money want something in return. The shareholders entrusted me the money to follow the plan. This IS the plan that we all agreed to at the start of this venture. And yes, by my logic, if cost of electricity is your largest problem... then lower the cost of electricity until its no longer a problem.
But solar isn't a way to lower your cost of electricity! Solar is a way to have a very predictable cost of electricity for the next 10-15 years. So again, I'm not disagreeing with you that solar might be a good idea. It is a good idea if electricity cost is going up and that seems likely. I would just like to see the cost of solar be correctly compared to other options. (And also not claim that it is an insurance against high difficulty - it's not, unless shareholders don't care that their initial investment is managed less than optimal by having solar equipment run on an inefficient process, instead of having it sell to the grid.) But it seems we are unable to find a common understanding. :-/ I guess we will have to agree to disagree. In any case: Good luck with the project! I think we're both saying the same thing but with different contexts thus different words. The plan was designed to deal with pretty much every threat to the continued operation of the company. As I said, I am not an oracle, I cannot tell what the future is, but I can tell you with certainty that there will be situations that will be very difficult for miners. I do not know when they will happen, I do not know what they entail, but I do know what variables I can control to minimize their impact. One of the variables that I can control is that I can pay a large sum of money now to make sure that even in the worst conditions the mining hardware is never shut off. Operational efficiency is a measurement of how many dollars per time unit I get per dollars spent on operating costs: not mh per time unit, and not BTC per time unit, and not either of those per dollar spent (that is a different efficiency calculation used for evaluating TCO and not minimum operational viability, for that you look at mh/$ and damn everything else) and this is not a fixed number. If operational efficiency drops to a point that operating costs exceeds dollars produced, the hardware is shut off. This can happen no matter if I buy 28nm hardware and no matter if 28nm hardware is still the best hardware available. If I have to shut the hardware off because its too expensive to mine, then I am losing money because the hardware is off and cannot do anything to stop it. Yes, that means I am spending more money up front and spending more money on infrastructure that is not mining hardware. However, anything I can do that buys me more time to not have to shut the hardware off during adverse conditions MUST be done. If I have a chance to turn known unknowns and unknown unknowns into known knowns, then I would be a fool if I do not do it. If you want to view this whole thing as prepaying electricity to get guaranteed cheaper rates, then fine, thats what it is. Even if solar takes 10-15 years to pay off, panels are typically guaranteed for at least 20 years, and its most likely going to be 30 or more years before they start failing to the point I need to replace the entire installation. And that is just for current generation hardware: several companies are developing technology to reduce panel size (which is really the largest cost of panel manufacturing) and using manufacturing processes that are less complex and rely on fewer expensive ingredients (such as rare earth metals); next generation hardware may end up taking in the 5-10 range of payoff instead of 10-15 and use a much smaller physical space. I really want to know why we're not seeing eye to eye on this. Am I explaining this with less clarity than I intended? |
|
|
|
You need $10 million minimum to do ASIC right and still have to sell them with a high markup.
$10M might be the ballpark if you go for a leading edge process like 32nm, but if you settle for an older process you are between one and two orders of magnitude wrong. Let me put it this way: write me a check for $1M and Ill provide you with 5TH worth of bitcoin asics within the next 12 months. Was thinking more like 45nm. Either way, I'm not interested in it. If someone comes up with a cost effective ASIC solution in time then I'll buy them as first gen hardware, but I don't want to run that kind of company myself. |
|
|
|
You misunderstood what I said. By trading bonds with me, you are cutting your dividend yield in half.
If I am only buying the bonds to trade them with you then no, I'm not really cutting my dividend yield in half. In the same way that when I buy an mp3 player to swap for a laptop, it's irrelevant what features the mp3 player has (and I subsequently "lost" access to). I only bought it because it was cheaper than buying the laptop outright. The people who are flipping cannot continually flip because it will drive bond prices up to unrealistic levels.
That's true. But just because this rebate can't go on forever, doesn't mean it's not a rebate.
Yes, but if your goal is short term profit... that doesn't work here. DMC does not have a high enough volume to flip quickly.
There is a difference between flippers and investors. If an investor buys bonds just to trade with me and intends on holding DMC long term, fine, you can call it a rebate of sorts, just a very thin one. Problem is, most of the bonds traded to me were not purchased originally for that purpose.
Out of the 1200 shares sold, around 200 were purchases straight out, around 200 are up for sale by the flippers, and the other 1000 were traded with existing bonds not purchased for the purpose of trading with me.
Once I own the panels, there is no continued cost of operation.
Only if by "own" you mean completely payed off. (Is that what you meant?) So in 10-15 years you have no continued cost of operation. By your logic a miner that becomes unprofitable could just go out and buy a solar panel and then say: Hey look, no continued cost of operation anymore, I can fire up my miners again. No, not "completely played off". This isn't a bank loan. I own the panels in full on day one. There is no money further removed from the DMC bank account after day one. I exchange money I have now for money I don't have to pay in the future. I cannot predict the future: I know I have the money now and I do not know if I will still have sufficient money in the future. And yes, by my logic, if cost of electricity is your largest problem... then lower the cost of electricity until its no longer a problem. |
|
|
|
You can get ASIC designed for $1 million and then sell the resulting hardware which distributes the hashing power more evenly.
Nope. $1 million would effectively get me a SASIC solution on 65nm or a very badly designed non-SASIC solution that would perform worse. You need $10 million minimum to do ASIC right and still have to sell them with a high markup. I doubt you'd break 5mh/$, not an efficient use of the $10m. |
|
|
|
The market isn't very efficient here, sadly. The market is not factoring in the fact that I only pay out 50% of the bond dividends, the other 50% is going into the growth fund (which is now just buying more bonds).
The market isn't very efficient regarding what? The price of DMC? or the price of other mining bonds? It still doesn't change the fact, that you are giving a rebate to people who do a share swap. It's kind of a early adopter discount, which is fine I guess, except that it didn't apply to the very first buyers, who got in at 1 BTC. All those differences between DMC and other mining bonds are really not relevant, if you simply compare two situations with the exact same starting position: Namely starting out with 3 BTC and do either: a) buy 10x YABMC, then swap for 5x DMC b) buy 3x DMC at IPO price In both cases I had 3 BTC and afterwards I either have 5x DMC or 3x DMC, showing that the share swap gives a discount. (And since I started out and ended up with no other mining bonds, it's really not relevant what benefits they have or how they compare to DMC.) You misunderstood what I said. By trading bonds with me, you are cutting your dividend yield in half. Over the short term, you're losing money, approximately 1/3rd of your total value. Over the long term, you're profiting. The people who are flipping cannot continually flip because it will drive bond prices up to unrealistic levels. It'd only take about 500 BTC to drive all the bonds up to 1 BTC of bonds == 1 DMC share. If they want to flip a small number of shares, thats their choice, but its no way to seriously invest in DMC. Its about 200 shares being sold by people who think they're flipping shares by trading me bonds and selling their shares at a slight premium. They cannot effectively do this without raising bond prices (although, for now, there IS a profit to be made). Existing investors, both those who bond traded and those who bought straight out, have been buying them as fast as possible because they see it as a good deal and want to take advantage of the flippers.
No, it's not a good deal to buy a share from a flipper. Only if you don't want to put in the effort to get the share swap discount yourself makes it sense to buy a share from someone who already did that, but then of course adds a premium for his 'work'. There are shares sometimes for sale in the 0.60s and it just encourages the flippers to flip more. In the long term, it actually is a profitable move if you're a big DMC investor. Just figure out what your best deal is with bond prices vs shares in trade, and put a bid up for a bitcent lower than that and catch all the falling shares. Now, at the moment? I'd flip tygrr.tech, seems to be the best deal. Cost efficiency isn't the issue. It already is efficient enough that it only takes 10-15 years to pay for itself in Maine. The problem is we have no fall back if mining becomes unprofitable: that insurance policy is a very important part of DMC, it is something we offer that no one else does; additionally, having that second revenue stream is also a very important part of DMC.
I have to disagree here as well. So much disagreeing today, sorry. ;-) But yes, I think cost efficiency is the issue. Solar isn't some magical way to make cheap energy - and you have pointed out as much, quoting a payback time of 10-15 years. So there is a guesstimated price to your own solar electricity and it's a rate which does not greatly differ from the rate you get from the electricity grid. So if mining becomes unprofitable, then you will also reach the point where it falls below your "internal" solar electricity rate at which point it is better to the sell the electricity back to the grid than waste it in an unprofitable mining process. That said, I don't necessarily think that it's a bad idea. Bitcoin mining is an ideal electricity consumer in the sense, that it can quickly scale down and back up to match supply. So it's a good fit for any fluctuating renewable energy source. And of course consuming your own electricity will usually be preferable to selling it back to the grid where you are paying middlemen and possibly other fees/overhead. So combining a solar farm with Bitcoin mining might give a unique advantage compared to other solar farms which just feed back into the grid and pay this overhead. I just want the economics of this clearly spelled out. Its funny how you disagreed, then agreed in the second paragraph. Once I own the panels, there is no continued cost of operation. Electricity is paid in dollars not bitcoins. If the dollars produced by the mining operation is less than the cost of electricity, we're sunk. Solar/wind is an insurance policy in which if I consume 100% of it... it costs me nothing to mine in adverse conditions. Lets say, for example, tomorrow, difficulty suddenly changes to 10 million. Most of the existing large scale farms will shut down, and if this 10 million becomes the norm, they will shut down FOREVER. Lets say BTC prices plummet to 50 cents, and this becomes the norm. Farms shutdown forever. Electricity prices double. Farms shutdown. Forever. DMC, however? We're still up and running as if nothing happened. This is an insurance policy that is well worth its cost. The fact it can also generate revalue is also a useful aspect. |
|
|
|
Since that you have only got 1/1000 of your original IPO. How are you planning to scale up the investment.
Most of the shares bought on GLBSE are being offered for prices less-than the IPO price.
Only once all of those shares are bought, then you will get more IPO investment.
Most? Your math has much to be desired. Its about 200 shares being sold by people who think they're flipping shares by trading me bonds and selling their shares at a slight premium. They cannot effectively do this without raising bond prices (although, for now, there IS a profit to be made). Existing investors, both those who bond traded and those who bought straight out, have been buying them as fast as possible because they see it as a good deal and want to take advantage of the flippers. Currently, DMC is using the existing funds to hold hash power at existing mining companies using the bonds issued by those companies. First dividend day is on the first (or approximately half a month into operation). |
|
|
|
I'm not sure that I would touch solar until its cost efficiency goes way up. Current panel designs are expensive and inefficient, and that doesn't look like it is changing a lot any time soon.
I agree. Solar is too expensive. Edit: The upfront cost of solar is very expensive. Also you can only run your miners at day-time. Making the initial cost of the miners effectivly 50% more expensive. Unless you decide to use both solar and paid electricity. (maybe buy off-peak power, that is cheaper). I'm not sure where people are getting confused here. At no point have I said I'm going off-grid. I will purchase power at night at off-peak commercial/industrial rates just because I have to for obvious reasons. This doesn't mean there is not a net profit. The Plan includes example numbers for a break even installation, which means selling excess power to the grid at daytime and effectively buying back only what I need cheaper at night. EVERYTHING has a large upfront cost. I can't think of anything in the world in any industry that doesn't. |
|
|
|
I used to mine with phoenix on Eligius. The hash rate reported by both of them did agree (in average). Now I am trying Diablo. This miner reports much higher hash rate (by 15%). However, Eligius reports only 53% of the hash rate reported by Diablo. When I count shares/blocks submitted in last 15 minutes it actually does roughly match. So it seems that despite apparent higher hash rate less shares/blocks are submitted to the pool.
I got Diablo from binary zip archive from link at the very front of this topic. It seems to be quite up to date. Timestamp on DiabloMiner.jar in archive is May 23. So it is recent version.
I was searching this forum and googled it. However, I am still in the dark abyss (and Diablo is lurking somewhere around ...) Does what I described ring a bell?
Never use the pool hash rate. It is trying to count using an random event (share finding). Eligius comes close on the 3hr meter, but its still usually ~5-10% +/- off As I said it agrees in average. It is +- 10% but not 47%. Now I am testing cgminer and pool reported average is consistent with what miner displaies. As I wrote, I checked number of accepted shares and what pool was reporting (60 per 15 min) was in agreement with what miner reported (when I checked it was 51 shares in previous 15 min). Now with cgminer I am averaging 124 shares per 15 min. I will try DiabloMiner again. I am puzzled by what has happened. If you think DM is not producing shares correctly, start DM with -d and count the number of attempts (NOT accepts), you should get roughly 10000 attempts every 43000 ghash. |
|
|
|
So how about starting a wiki on the framework for recoverable and resilient systems? For instance, binlogging to an encrypted disk on a server in a different datacenter, because that would fix issues with deleted databases. And HSM devices or some equivalent for storing private keys. And daily and hourly backups to systems outside of the core network. Encrypted of course.
What other ideas do people have?
Reason there isn't a consolidated resource is because people and companies make a lot of money in the computer security sector. You can find a lot of good information piecemeal but the "how" of putting it all together has extra value from the hoarding of that knowledge. Very true. And, the "how" often changes so rapidly that maintaining a comprehensive resource on it is not workable. What I want to see is a list of minimum standards that should be expected from businesses dealing with our money. Simply a framework of technologies that you have to stir together in the right proportions, and a list of potential consequences of omitting one or more of them. You are describing financial regulations, for that you should deal with financially regulated companies. I personally have no wish to see BTC taken over by any government's financial framework. It's a tough problem, imo this is a great opportunity to see if open sourcing security completely would actually be effective. If the whole process is completely transparent then people can decide for themselves whether to use a BTC service. Not even government based, but just a wiki somewhere. Many of the FIPS-140-2 guidelines are extremely applicable though. If you're just talking about data storage, there are commercially available FIPS-140-2 level 3 devices that can be unlocked under Linux such as IronKey: http://www.imation.com/en-US/Mobile-Security/Mobile-Security-Products/Secure-Data/ (the S200 series is SLC, D200 is MLC, and Enterprise includes remote wipe next time its plugged in performed by the unlocking software). |
|
|
|
I'm not sure that I would touch solar until its cost efficiency goes way up. Current panel designs are expensive and inefficient, and that doesn't look like it is changing a lot any time soon.
Cost efficiency isn't the issue. It already is efficient enough that it only takes 10-15 years to pay for itself in Maine. The problem is we have no fall back if mining becomes unprofitable: that insurance policy is a very important part of DMC, it is something we offer that no one else does; additionally, having that second revenue stream is also a very important part of DMC. If wind turbines are not feasible for first gen power gen hardware, then we do solar. Its not like we have a choice, I can't go to Walmart and buy a fridge sized fusion reactor, there isn't many choices. |
|
|
|
at 0.1$/kWh you can buy 2 500 000 kWh at 250 000$ or enough for running those BFL mini rigs 50 years. I bet the solar panels is dead way before 2062.
If you promise use wall power instead of solar your project actually looks good.
Solar sucks, but I think he is looking at wind power, not solar. And, an additional revenue stream from selling the wind power to the grid. I am still looking at both equally. Large scale wind requires permits and licenses, and even if I had all the paperwork done right now it would take about 2 years start to finish to get the first turbine up. Solar is fine in Maine, we generate about 1200kwh/yr per 1kw of panel measured on real panels installed in Maine. There are better places in the US for the solar (areas in AZ, NM and TX can do 1600), but there are also worse areas in the continental USA and moving to AZ, NM or TX for superior solar means we're also wasting money on air conditioning and the electricity to run the air conditioning. Basically, even though I'd generate about 31% more electricity, the cost and overhead of needing air conditioning would cost much more than the 31% I'd be gaining. TINSTAAFL. http://www.ecotechnousa.com/Portals/0/Media/News/SolarMap-US.gifThat map is overly optimistic and lists Maine as ~1500 kwh/yr and those areas in AZ, NM and TX as 2100, but it illustrates the point well. |
|
|
|
at 0.1$/kWh you can buy 2 500 000 kWh at 250 000$ or enough for running those BFL mini rigs 50 years. I bet the solar panels is dead way before 2062.
If you promise use wall power instead of solar your project actually looks good.
I think you missed something rather important. Buying power from the power company does not involve the power company paying me for generating power. Notice the large discrepancy between what you want and how the plan was written. In addition, difficulty will only go up, electricity prices will go up, and those two in combination with prices going down we can end up in a situation where it is no longer profitable to mine. If I remove the only constant operation cost (electricity), then we can mine when everyone else can't. Green power generation is both a second revenue stream and an insurance policy. |
|
|
|
|
Show Posts
