Sorry. That's profit I am making. I thought gross was before expenses and net was after.
Dude, are you serious?
http://wiki.answers.com/Q/Difference_between_profit_and_revenue
The ONLY way a X% change in revenue leads to the exact same X% change pm profit is if expenses are zero. But expenses are not zero, even for people with free electricity and air conditioning, you still have depreciation eating away at the value of the mining hardware over time. If revenue falls, profit always falls faster if expenses are greater than zero.
Sorry. That's profit I am making. I thought gross was before expenses and net was after.
My numbers are net profit.
Ah, I think I know what I did wrong, sorry about that. Difficulty went up 28% but to get revenue change you need to divide the old difficulty by the new difficulty and that's more like a 22% drop than 28% (something like 12,153,000 divided into 15,605,633 = 22.1%). So your profit should be going down more than 22% and that is what we see; so your profit did drop by more than 22.1%, they dropped by almost 28% which is coincidentally the rise in difficulty but doesn't have to be that way and it could be different depending on your electricity costs.
I was using the wrong baseline. Like if a $100 item goes on sale for 50% off and then increases in price 50% from there, what is the final price? It's not $100, it's $75. I used the wrong basline, sorry.
The point remains though that profits decline faster than revenue. But to get decline in revenue you need to divide the old difficulty into the new, not treat the decline in revenue as equal to the percent increase in difficulty.
An interesting little factoid.
In late Nov 2012, difficulty was 3,438,909 while bitcoin was only worth $12.36 each.
(278k difficulty per $1)
After the next estimated difficulty increase, difficulty will be 15,191,765 and value will presumably still be around $120 per bitcoin.
(127k difficulty per $1)
With a value of $120 per bitcoin, difficulty would need to be at 33,934,951 to match the (low) level of profit from back in Nov 2012. If difficulty increases by 10% per adjustment, we should be there in eight or nine more adjustments after this next big one. Of course if difficulty increases at the more insane 25% per adjustment rate we will be there in 4 difficulty adjustments.
Why is this interesting to me? After November 2012, difficulty actually dropped on three out of the next four adjustments. Apparently that was the point where mining just wasn't profitable for a significant number of people such that they actually quit mining, at least for the time being. It'll be interesting to see if the same occurs again, I would assume it will.
Interestingly enough, it was also immediately prior to the bitcoin "correction" which in the end resulted in value increasing by about 10X. Yeah, it's a tired argument about how bitcoin value doesn't follow difficulty, yet historically if difficulty increases high enough it allows value to increase as well. ASICs might break this pattern because they allow for a far cheaper way for miners to increase difficulty.
An interesting little factoid.
In late Nov 2012, difficulty was 3,438,909 while bitcoin was only worth $12.36 each.
(278k difficulty per $1)
After the next estimated difficulty increase, difficulty will be 15,191,765 and value will presumably still be around $120 per bitcoin.
(127k difficulty per $1)
[insert wishful thinking here]
Without doing the arithmetic, it looks like the window has closed on buying GPUs for mining profit. I'll keep my GPUs running for a few more months since they're bought and paid for, and circumstances allow me to have an effective KWH rate of about $0.07. Time to look into Litecoin?
Isn't Ripple centralized? The really interesting feature of bitcoins is that it is decentralized. If I want to use a centralized currency, why would I not use my VISA or MasterCard for purchases online? I do not see anything in Ripple so far that would make it a serious alternative to solutions that we have already.Further, for the first time bitcoin faces serious competition: ripple.
This you need to explain better. I understand that it is certainly easier to buy BTC and turn a profit than mine and turn a profit in the scenario that the prices keep increasing. But how is it more profitable? The cost to mine one coin is considerably smaller than the cost to buy it. Of course, if you need to buy all hardware first then your statement will be true (on the short time scale, or the long time scale if difficulty skyrockets), but I assume that the people you are addressing here already have their mining rigs set up, right?Lastly, note that if you believe that BTC prices will rise in the long run, it's more profitable to simply BUY the BTCs directly.
Chiro you estimates usually have at least one foot in reality but this is too completely wrong analysis.
First, we just got through the whole revenue-is-not-profit arc in this thread again. Your analysis treats revenue as if it were the same as profit with a tiny disclaimer that ASICs may be able to mine longer for cheaper. Well yeah, ASICs are literally 100 times more efficient than GPUs. GPUs are 100x costlier to operate per watt, so the revenue vs profit discrepancy is much greater for GPUs than for ASICs. The bottom line is that your analysis is way off... ASIC miners can withstand a much worse difficulty to BTC price ratio because their operating costs are 100x lower.
Second, you need to account for the fact that ASICs (and realistically, also FGPAs) can not leave mining (no SHA256 alt-coin is going anywhere). ASICs in particular have one purpose in life: bitcoin mining. That's their sole purpose in life. You can't realistically expect FPGAs and ASICs to stop mining until it becomes very unprofitable for them.
Third, due to the above, GPUs represent an ever-shrinking portion of bitcoin hashing power. If you took every single GPU off bitcoin mining today, you would still have Avalon's first two ASIC batches + all existing FPGA + ??? BFL Jalapenos worth of hashing power + anyone who kept quiet and made some FPGAs and ASICs without selling them to others + ASICminer + that guy selling USB sticks of ASICs in batches of 300. That's already the majority of the system, given that Avalon's first two batches alone represent roughly 60 TH/s, and total bitcoin system hashing power is roughly 125 TH/s right now. Avalon third batch and chip orders and BFL chip orders after that, plus AM and other shipments or internal mining. GPUs will be a very small, trivial portion of hashing power by this fall.
Fourth, there was at least one news article last summer or fall where they interviewed some big-time GPU miners who were losing thousands of dollars but kept mining anyway because they were "true believers." There are lots of GPU miners who will mine no matter what, even if it costs them. So the impact of GPU miners leaving BTC will not be as big and fast as you might think.
Fifth, you owe taxes on mining profits anyway.
In summary, difficulty will only go up every two weeks for as long as people keep buying FGPA/ASICs, which is already the majority of hashing power and quickly growing. You will lose a lot of GPU miners, but some will cling to mining at a loss, and GPU hashing power will be an irrelevant portion of hashing power by fall this year anyway.
The rest of your post is the old reversal of difficulty and pricing, but that isn't why bitcoin prices exploded a few months ago. As usual, price drives difficulty, not the other way around. And demand drives price. In this case, the media outlets focused the world's attention on bitcoin. In particular the Cypress bank banditry, and the rise of media exposure in China which, on a per-capita basis, was almost absent prior to 2013. When you have such big media exposure some people get excited and buy lots of btc thinking they are early adopters.
Also, you may want to look at the threads about legal issues at bitcointalk, it's pretty nasty how expensive it is to register as a money transmitter in all fifty states, then the even huger expense of actually complying with financial regulations and demanding personal information from everyone using your exchange.
But reward dropped from 50 coins per block to 25.
So that virtually "doubles" actual difficulty.
I don't disagree with anything you wrote here, but I can't for the life of me figure out why you wrote it. What did I say in my post that made you think I thought ASIC were going to fail or something?
I don't think I said they would quit mining. Why are you trying to put words in my mouth? There are at least 20Thash worth of GPU miners still mining away. When they all stop mining, difficulty will drop, unless BFL shipments exceed all expectations enough to completely erase the effect. I know what BFL has claimed recently, but I will believe it when I see it. Network strength has been steadily increasing by 10TH or so each difficulty adjustment, this particular one is an anomaly with a 20+ TH increase. If 20TH worth of GPU miners quit at once the difference would be noticeable, either an actual difficulty drop, or a period with no significant difficulty increase.
Again, I agree with what you say. GPU accounts for about 20-30THash IMO, which is a minority when you consider the entire bitcoin network is estimated to be nearing 108THash. But a 20-30Thash DROP would certainly still cause a difficulty decrease, regardless of it being less than the majority. I'm not saying difficulty is going to be cut in half or anything close to that. I'm saying difficulty will probably decrease, although I will accept that that may not happen if BFL actually starts shipping 4000 units per week.
Uh, my theory about GPU miners leaving is based on GPU miners who actually did leave the network back in Nov 2012, before presumably returning when increased value made mining profitable again. The network lost 4THash worth of miners in that time.
What does this have to do with anything? I feel like this is a very random statement to make, that doesn't really have anything to do with my post which you are responding to. You pay taxes on profit, yes. If there is no profit there is no tax. If there is less profit there is less tax. As the equation works, there is no situation where a profit becomes a loss due to tax.
"Irrelevant" is really a matter of opinion, isn't it? I think you are making a lot of assumptions. Mining is still profitable on GPU today, and will still be profitable after the 26% difficulty increase, at least with my cheap east coast electricity. In the fall? At the current rate of difficulty increase, it will probably be unprofitable by then, yes. But if value spikes for ANY reason that could easily not be true. If BFL has further delays, difficulty might not increase as fast as you think. If the network loses 10TH worth of GPU miners and gains 8TH worth of ASIC miners in one adjustment, difficulty will actually go down a bit.
Bitcoin price started increasing long before the Cypress news. It started increase right around the reward halving, actually.
I've read all about that on reddit and bitcointalk, but what does it have to do with the point I was making? I feel like you decided to go off on a rant about everything you hate about bitcoin and you missed my point.
As far as ripple, nope. Can't agree at all that it would ever compete with bitcoin. It's completely different idea, and VERY flawed, and it's centrally controlled. I hope it dies soon, but even if it doesn't I can't see any sane merchant or individual using it instead of bitcoin, though a few will use it *with* bitcoin.
Good catch, I was thinking it had already occurred but it looks like the halving happened a few days after that particular date. Unfortunately the halving is a rather large jump in "effective" difficulty, so it's hard to pin down the exact number where miners started quitting. It does look like we will certainly reach that point again soon, if value doesn't increase.
I agree with most of what you said, with some exceptions:
Isn't Ripple centralized? The really interesting feature of bitcoins is that it is decentralized. If I want to use a centralized currency, why would I not use my VISA or MasterCard for purchases online? I do not see anything in Ripple so far that would make it a serious alternative to solutions that we have already.
This you need to explain better. I understand that it is certainly easier to buy BTC and turn a profit than mine and turn a profit in the scenario that the prices keep increasing. But how is it more profitable? The cost to mine one coin is considerably smaller than the cost to buy it. Of course, if you need to buy all hardware first then your statement will be true (on the short time scale, or the long time scale if difficulty skyrockets), but I assume that the people you are addressing here already have their mining rigs set up, right?
Anyway, I think that the ASIC miners are very bad for the future of bitcoins. For bitcoins to succeed in the long term, they need to expand the currency aspect as opposed to primarily be a commodity for speculation. A condition for this to happen is that the wealth is spread among a large user base, but if most of the coins are in the hands of a few thousand ASIC miners the trade of BTC for goods and services will die. Ultimately this could mean the end for bitcoins.
It is in this light I see the alternative currencies. We need a currency that has as large number of holders as possible, and a currency that does not deflate like bitcoins tend to do, since deflation implies that it is more profitable to hoard than to spend. A viable currency needs as many miners as possible, and this means it should run on as many hardware configurations as possible.
We have recently seen the advent of a CPU only coin, and while this gives a very large potential user base, it also opened the door for bot nets and server parks to take most of the profits. This is a smaller problem for GPU mined coins, since that takes the power to individuals PC's. A currency that is ASIC resistant and runs well on both NVidia and AMD cards, and that improves other bitcoin issues (such as slow confirmation times, deflation etc) would be ideal.
I turned my back on the coin price and we are now at $92 USD.
Any ideas why they have dropped so much over the past week?
Also, with Dwolla being frozen how are you guys getting cash in and out of MTGOX?
don't know why the drop. I am now using coinbase seem to be okay so far.
Who knows, but from the recent history I would say that it will be back to $120ish before long.
Wait, I mean, the end is here! Sell your mining equipment! Yeah, that's it!
It's no big deal to me, free money basically. I wish I'd gotten in sooner, it's been pretty fun. If a lot of GPU miners are getting out en masse, I'd expect to see some deals on cards. I haven't seriously been looking, yet nothing has popped up to make me think this is happening yet. I'd love to have a matching mate for my 7970, lol.
I agree... no regrets for me either, as it's been a lot of fun. Although I do wish I'd taken the plunge a bit earlier. I'd known about BTC since before the 2011 bubble (I've been reading this thread with interest since it was created), but I couldn't be bothered to learn how to do it at the time.
If the next difficulty increase is over 25%, I'll probably sell my extra 5870 and move on. The rest of the cards I've bought all have a practical purpose in my various machines.