Teknohog, could you please try with Stream 2.1. I can't see why 'vectors' would give invalid results.
With SDK 2.1, the miner does not even compile with vectors  /tmp/OCLQa31hL.cl(128): error: mixed vector-scalar operation not allowed
unless up-convertable(scalar-type=>vector-element-type)
sharound(E,F,G,H,A,B,C,D,R(36),0x650A7354);
But it gets stranger. After reinstalling 2.2, I am getting accepted results with vectors, using the latest version of poclbm. So this looks like an error in my system, and poclbm is fine  (Possible sources of error: overheated GPU, which is now cooler after the SDK changes; pyopencl, which is the only component I have to manually compile and install with every SDK change.) I'm encountering this same issue (python 2.6.6 on Ubunu 10.10 with SDK 2.1 and the latest pyopencl (as of this writing)...I've tried uninstalling/reinstalling pyopencl...nothing seems to work. Wondering if anyone else has had this problem and solved it. Bumping this out of the poclbm thread in hopes someone might know of a quick fix. |
|
|
|
|
That's interesting...I was having all sorts of trouble with 10.10 (cloned from a 5970 based rig...but with sdk 2.4 and catalyst 11.4) and a 6990 (it wouldn't even read the clocks or temps using aticonfig)...I upgraded Ubuntu to 11.04 and suddenly everything started working.
|
|
|
|
With the markets being volatile as hell, it is a nightmare for people actually selling goods.
I came up with the following idea, but would like to know how other people think of this.
Some merchants already provide goods against a fixed price, some adjust against the markets quickly.
My focus is on the ones that provide against the fixed price. One could value their goods in USD and see
how many bitcoins they want for that. From that you can calculate a price per bitcoin. If you would average
all these rates (from the different merchants) out, you would IMHO get a more realistic price of how much
a bitcoin would be worth.
If more and more merchants would stick to that calculated price it would reduce its volatility even more and the
exchange markets might move towards that same value.
Every month the price would be recalculated depending on the trades that took place during that last month.
Is this doable and would this work?
I posted an idea similar to this a few times...rather than use the dollar or any inflated fiat currency, make up a new unit of account that is indexed to some basket of basic goods (I suggested indexing basic foods you'd find at a grocery store that one would need to survive for a day). 1 or 10 of these could roughly equate to what the average person would need to feed themselves for a day. You could base your pricing on this unit but use bitcoins or anything else for actual payment. This unit would also have utility for contracts (loans could be made in terms of this unit (but settled with bitcoins) to avoid the risks that bitcoin volatility poses to such contracts). |
|
|
|
Teknohog, could you please try with Stream 2.1. I can't see why 'vectors' would give invalid results.
With SDK 2.1, the miner does not even compile with vectors /tmp/OCLQa31hL.cl(128): error: mixed vector-scalar operation not allowed
unless up-convertable(scalar-type=>vector-element-type)
sharound(E,F,G,H,A,B,C,D,R(36),0x650A7354);
But it gets stranger. After reinstalling 2.2, I am getting accepted results with vectors, using the latest version of poclbm. So this looks like an error in my system, and poclbm is fine (Possible sources of error: overheated GPU, which is now cooler after the SDK changes; pyopencl, which is the only component I have to manually compile and install with every SDK change.) I'm encountering this same issue (python 2.6.6 on Ubunu 10.10 with SDK 2.1 and the latest pyopencl (as of this writing)...I've tried uninstalling/reinstalling pyopencl...nothing seems to work. Wondering if anyone else has had this problem and solved it. |
|
|
|
|
One thing that has not been mentioned in this thread is the first amendment defense. To spend bitcoins, you basically announce to the world that you have done so. To restrict bitcoin is to limit free speech.
|
|
|
|
If bitcoin will be a success (e.g. 10% of US economy), than i think it will demonetize gold relatively quickly, so gold's price will be around industrial use price. This is the same effect, that happened to silver in 19-20th century.
I don't think gold will ever be demonetized. It's the best physical form of money and the utility of that combined with a need to be diversified means it should always remain valuable as money (short of some kind of cost effective alchemy). Bitcoins or something substantially like bitcoins should take a similar role as gold in the digital world. Government bonds on the other hand are a different matter.  |
|
|
|
|
What motherboard are you using that will let you run with 3 cards? Are they spaced far enough apart to allow airflow between?
|
|
|
|
MtGox' version of trading curbs  |
|
|
|
It is the imposition of the software rules. Try to change bitcoin's software rules, and you'll realize it's a fiat currency.
This is a valid point, what you are really arguing here though is that the software development group of bitcoin are somehow a recognised "authority" ... if you can classify them as an authority, then yes you are right, it is by their decree, or rules, that bitcoin is fiat. Now how do you go about arguing that the software developer group of bitcoin are a legitimate authority? Definition of authority, as used in the context of fiat, not quite a match but maybe ... The software development group isn't an authority in the manner in which you think. They are only an authority in the sense that people have a certain amount of trust in the software they produce. Also, there is no one software development group...there is different bitcoin related software and different bitcoin clients that completely unaffiliated people are creating. They do not possess any unique position to influence the operation of bitcoins than anyone else. The software development group does not have the ability to unilaterally change the bitcoin protocols without convincing the network of people using bitcoins of the merits of such a change. It is in fact mathematics and physics that govern the bitcoin system. You only have to trust that your computer runs the code correctly, that the mathematics are sound and that there aren't major flaws in the code. And, you don't necessarily have to trust those things...to the extent you are knowledgeable in those subjects, you can verify it. And, here are a few definitions of fiat money in case you are still unconvinced: http://en.wikipedia.org/wiki/Fiat_money ("Fiat money is money that has value only because of government regulation or law.") http://www.investopedia.com/terms/f/fiatmoney.asp ("Currency that a government has declared to be legal tender") http://dictionary.reference.com/browse/fiat+money ("paper currency made legal tender by a fiat of the government") http://www.answers.com/topic/fiat-money ("Legal tender, especially paper currency, authorized by a government") http://www.thefreedictionary.com/fiat+money ("Legal tender, especially paper currency, authorized by a government") I think it would be a shame if you didn't at least devote a paragraph to the definition of fiat money (and preferably go on to explain how bitcoin is not a fiat currency, but has no commodity backing...but at a minimum, mention that there are many that would vehemently disagree with its characterization as a fiat money). |
|
|
|
|
The one major issue I have is in calling bitcoin a fiat currency. That is completely inaccurate...you might think people view that term to mean that it is not backed by a commodity, but even if that were the case (and I would strongly argue that it is in fact not the case), it does not make it correct. A fiat currency is a currency that has value only because of government regulation or law. Period. And that absolutely is not the case for bitcoin. In fact, one of the essential attributes of bitcoin is that no one is forced in any way to ascribe any value to bitcoin. I would encourage you to start with a definition of fiat currency. Then talk about commodity backed currencies. And then describe bitcoin, which is neither a fiat currency, nor backed by a commodity (but which does have certain properties that, based on the evidence, does give it value).
Please correct this mistake. Otherwise, I think it's a very good paper.
|
|
|
|
Bitcoin is more versatile, but gold will probably remain more stable for a long time. Gold also has the advantage that it's shiny. If I ever get rich, I might trade some for gold as fallback. Not a lot, just a few percent. For the case some madman cuts the trans-Atlantic cables or fucks up power supply. Can't pay in BTC without functioning internet. Diversification is always a good idea. It also happens to be one of the things the deflationistas overlook. Even if you think bitcoin has a bright future, it's not very prudent to put everything you have saved into it. Just as it wouldn't be prudent to put everything into gold, silver, or some stock or a particular bond. Bitcoin being a deflationary currency isn't a problem because people won't want to put everything they have into that currency...and if everyone did put everything into it, it's a sure bet that it will experience a period of devaluation as people realize that fact and start to diversify out of bitcoin. Deflationistas also overlook the fact that loans don't have to be made in terms of bitcoins...you could just as well loan someone an amount of bitcoins sufficient to purchase a loaf of bread with the repayment being enough bitcoins to purchase two loaves of bread (the repayment might even be fewer bitcoins than were loaned if there is sufficient appreciation in the value of bitcoin...but there's still value in that loan for the lender in that the lender has diversified some risk out of bitcoins and into the borrower and to the extent that the borrower is a good risk, they lender has locked in a real gain that is not dependent on what the value of the currency might do). |
|
|
|
Nonsense...that's just the deflation argument again...I think that question has been sufficiently debated in other threads. |
|
|
|
An interesting thought I had earlier today: anyone with large gold reserves could singlehandedly peg the exchange rate as high as he likes by putting on the market, say, 10,000 ounces of gold at 600 BTC/oz.
Why would anyone want to do that though? Also, I don't think it would work for long. Gold is a very large and established market... Bitcoin is far smaller and its use may very well grow at a rate that makes it impossible to hold the price of bitcoin at just 1/600 oz of gold. A person trying to hold that exchange rate may very soon find themselves holding 10,000 oz of gold and no bitcoins. So, they lose nothing and bitcoin takes off as a medium of exchange for subdividing gold, skyrocketing in price at the same time. Sounds like a good plan for someone with a lot of bitcoins and 10,000 oz of gold. They are now wealthier and can spend their gold more easily in the form of bitcoins. Wrong...by trying to peg a rate of exchange they'll simply create an arbitrage opportunity for everyone else and quickly go broke. Yes. He would end up with 6,000,000 btc and no gold. Well, maybe not, but with a lot of bitcoins and very few gold. The market price for BTC would rapidly respond as soon as the first few successful trades with the gold seller. If he was well invested in bitcoin before the gold offer, he could sell off and then retract his offer. The price would drop and he could walk away with a nice profit or reinvest. If he actually wants to sell it all instead of just selling a few ounces to convince people it's legit, he may well end up with quite a few bitcoins and no gold. After the gold was gone, the price would move toward what the market sans gold can support, and he would be at a loss. If he has a very large portion of the bitcoins, that value might be very low. So, if he follows through he goes broke. If he takes advantage of people's assumption of continuity he could profit. Arbitrage opportunities that are too good to be true usually are. It may only work long enough to get people to invest in it, and not long enough for them to even get back what they put in. Then that isn't really backing bitcoins. Backing bitcoins by anything is really a silly idea. With gold and bitcoins, you're dealing with two currencies whose quantity is fixed. So attempting to peg the value of one to the other is like trying to mandate how everyone should value one vs the other. It can only work on a short timescale (until you lose all of your bitcoins, all of your gold, or both). Pegging or backing a currency can only be done if you control the supply of that currency (like the way mtgox backs mtgusd with bank deposit usd redeemable via wire transfer). |
|
|
|
An interesting thought I had earlier today: anyone with large gold reserves could singlehandedly peg the exchange rate as high as he likes by putting on the market, say, 10,000 ounces of gold at 600 BTC/oz.
Why would anyone want to do that though? Also, I don't think it would work for long. Gold is a very large and established market... Bitcoin is far smaller and its use may very well grow at a rate that makes it impossible to hold the price of bitcoin at just 1/600 oz of gold. A person trying to hold that exchange rate may very soon find themselves holding 10,000 oz of gold and no bitcoins. So, they lose nothing and bitcoin takes off as a medium of exchange for subdividing gold, skyrocketing in price at the same time. Sounds like a good plan for someone with a lot of bitcoins and 10,000 oz of gold. They are now wealthier and can spend their gold more easily in the form of bitcoins. Wrong...by trying to peg a rate of exchange they'll simply create an arbitrage opportunity for everyone else and quickly go broke. |
|
|
|
Thus bitcoin has the same flaws that gold has.
And what might those flaws be? |
|
|
|
An interesting thought I had earlier today: anyone with large gold reserves could singlehandedly peg the exchange rate as high as he likes by putting on the market, say, 10,000 ounces of gold at 600 BTC/oz.
Why would anyone want to do that though? Also, I don't think it would work for long. Gold is a very large and established market... Bitcoin is far smaller and its use may very well grow at a rate that makes it impossible to hold the price of bitcoin at just 1/600 oz of gold. A person trying to hold that exchange rate may very soon find themselves holding 10,000 oz of gold and no bitcoins. |
|
|
|
|
I agree with asdf. But a more interesting question to me is whether bitcoins (or something very similar to bitcoins) would ever be worth more than gold (meaning the total value of all bitcoins has more purchasing power than the total value of all gold). Bitcoins are more convenient and much easier and cheaper to secure than gold, so I think it's possible. IIRC, I calculated that if you took half the value of all gold and assigned it to bitcoins, a bitcoin's value would be in the realm of $150,000.
|
|
|
|
|
You can use the -datadir option to specify a different location for the data files for two different bitcoin instances. That will let you run two instances concurrently, each using its own wallet.
|
|
|
|
|
When will you be coming through Atlanta? I could meet you and sell you things you need for BTC.
|
|
|
|
|
Show Posts
