As a rough estimate, I'd say that probably 30-35% of the current hashrate is produced by gear in the 0.2-0.3 w/GH range.
I'd also guess that a small amount is produced by gear in the under-0.1 range (KnC small amount, Bitfury probably more) but doubt that's a HUGE number yet since neither Bitfury nor KnC have been showing hugh hashrate increases over the last few months.
I doubt there was a TON of old gear that had been turned off due to not being profitable last summer that is now turned on. 1-2% MAYBE.
I suspect there is more gear that was "turned off for summer" or "undervolted/underclocked for summer" but still not a ton of hashrate. 2-3% perhaps.
Topic: Network percentage of new (0.25 J/GH and better) vs old ASICs? (Read 683 times)
December 15, 2015, 08:52:32 PM
This kind of analysis assumes that folks are entirely "rational" in terms of profit and loss. A rising BTC suggests that it might go higher, and hence I'll mine for a period of time beyond the point of strict profit. Yes, I expect the larger mining farms are more disciplined, but they also a fair amount of inertia in terms of what to do with gear that's only marginal in terms of profit. They might not have the money to replace it, and their electric rate increases if they fall outside of the commercial rate they currently have, or are contracted for.
There are humans still controlling things, so the decisions may not always be as cut and dried as you suggest.
There are humans still controlling things, so the decisions may not always be as cut and dried as you suggest.
you're absolutely right. And many people behave under "average cost" assumptions rather than "marginal cost".
For example if you made +10% then +5% then 0% a marginal cost (rational) actor would stop.
A human actor may continue at -5% and then -10% so that over 5 cycles he is "even money"... (and theoretically irrational).
And then yet others may mine at a loss hoping for some future bonanza payoff (1 BTC = $1 million!!!!!).
Also by hoarding and not selling newly minted bitcoins, less supply is available for sale = a tendency towards a higher price than "equilibrium" would suggest.
Well said. Anecdotally, in the past I would intentionally continue to mine Bitcoin or Litecoin at a loss, but only for periods of time where the cost was less than say, 20-35%, of my previous profits.
Being a day-trader I found it much easier to have that extra BTC (mined at a loss) to trade with, and usually making a higher profit margin than I would have if I had shut down my miners on a more strict accounting doctrine.
Dollar-cost-averaging is key to this game, and it goes a few ways -- accidentally buy high, then you better also buy low -- or if you day-trade and you mine at a loss you better make up the margin to average back to a higher profit than you otherwise would have.
December 15, 2015, 08:11:53 PM
This kind of analysis assumes that folks are entirely "rational" in terms of profit and loss. A rising BTC suggests that it might go higher, and hence I'll mine for a period of time beyond the point of strict profit. Yes, I expect the larger mining farms are more disciplined, but they also a fair amount of inertia in terms of what to do with gear that's only marginal in terms of profit. They might not have the money to replace it, and their electric rate increases if they fall outside of the commercial rate they currently have, or are contracted for.
There are humans still controlling things, so the decisions may not always be as cut and dried as you suggest.
There are humans still controlling things, so the decisions may not always be as cut and dried as you suggest.
you're absolutely right. And many people behave under "average cost" assumptions rather than "marginal cost".
For example if you made +10% then +5% then 0% a marginal cost (rational) actor would stop.
A human actor may continue at -5% and then -10% so that over 5 cycles he is "even money"... (and theoretically irrational).
And then yet others may mine at a loss hoping for some future bonanza payoff (1 BTC = $1 million!!!!!).
Also by hoarding and not selling newly minted bitcoins, less supply is available for sale = a tendency towards a higher price than "equilibrium" would suggest.
December 15, 2015, 08:08:34 PM
This kind of analysis assumes that folks are entirely "rational" in terms of profit and loss. A rising BTC suggests that it might go higher, and hence I'll mine for a period of time beyond the point of strict profit. Yes, I expect the larger mining farms are more disciplined, but they also a fair amount of inertia in terms of what to do with gear that's only marginal in terms of profit. They might not have the money to replace it, and their electric rate increases if they fall outside of the commercial rate they currently have, or are contracted for.
There are humans still controlling things, so the decisions may not always be as cut and dried as you suggest.
There are humans still controlling things, so the decisions may not always be as cut and dried as you suggest.
December 15, 2015, 08:06:54 PM
Considering 1.1 J/GH is still profitable here and i live in Canada, which is not a place with effectively free electricity like Venezuela, we still have a long way to go before 0.75J/GH become unprofitable. After all, many industrial setups have even cheaper electricity than be, where they can run 2 J/GH at the moment profitably.
With the BTC price rising, the difficulty will not only be sustainable at 95B but will most likely raise by another 30-50% during the next few months.
At a difficulty of 95,000,000,000 and with a Bitcoin worth around $450 you will earn roughly 10 cents per hour with one Terahash.
So a 1.1 J/GH system will be economic to keep running if you pay less than about 9 cents per kWh.
With a 2 J/GH system it would only be worth running if you pay less than 5 cents per kWh.
The issue over the next few months is how far difficulty will rise as the large mining farm roll out their 16nm miners. I believe that with sub 0.1 J/GH 16nm systems appearing and systems such as the SP50 that a tripling in difficulty is possible before the halving.
I'm at 0.06 so 2J is no good for me. However if i had industrial electricity the price would be 0.02~0.03~. So i'm guessing many place in the world has it for even cheaper and thats where most of the hashrate is.
So it does not matter if all the home users stop using 0.75J/GH (its not actually the case) because is such a minority of the hashrate anyways.
The big jump is causing old gear to be profitable again. But many like me I'm guessing have sold old gear and already invested in new. The difficulty is also something to watch.
We could get a 20 percent difficulty. That would be huge and effect old machines if your barley making profit. I mean 1/5 change is massive.
December 15, 2015, 08:05:11 PM
until the difficulty rise, the new more efficient machines (esp. at higher hash rates) actually have a marginal advantage over "old" machines.
For example, back in the day with say 1,000 TH/s in the whole network then you come along and add 1,000 TH/s more, you will have a more than 50% chance of finding blocks until the difficulty re-targets taking into account your load..
For example, back in the day with say 1,000 TH/s in the whole network then you come along and add 1,000 TH/s more, you will have a more than 50% chance of finding blocks until the difficulty re-targets taking into account your load..
December 15, 2015, 06:48:35 PM
Considering 1.1 J/GH is still profitable here and i live in Canada, which is not a place with effectively free electricity like Venezuela, we still have a long way to go before 0.75J/GH become unprofitable. After all, many industrial setups have even cheaper electricity than be, where they can run 2 J/GH at the moment profitably.
With the BTC price rising, the difficulty will not only be sustainable at 95B but will most likely raise by another 30-50% during the next few months.
At a difficulty of 95,000,000,000 and with a Bitcoin worth around $450 you will earn roughly 10 cents per hour with one Terahash.
So a 1.1 J/GH system will be economic to keep running if you pay less than about 9 cents per kWh.
With a 2 J/GH system it would only be worth running if you pay less than 5 cents per kWh.
The issue over the next few months is how far difficulty will rise as the large mining farm roll out their 16nm miners. I believe that with sub 0.1 J/GH 16nm systems appearing and systems such as the SP50 that a tripling in difficulty is possible before the halving.
I'm at 0.06 so 2J is no good for me. However if i had industrial electricity the price would be 0.02~0.03~. So i'm guessing many place in the world has it for even cheaper and thats where most of the hashrate is.
So it does not matter if all the home users stop using 0.75J/GH (its not actually the case) because is such a minority of the hashrate anyways.
December 15, 2015, 06:46:43 PM
I think its impossible to predict this. I think maybe around 33% of all the total network hashrate is 0.25J/GH or better.
December 15, 2015, 06:46:23 PM
Considering 1.1 J/GH is still profitable here and i live in Canada, which is not a place with effectively free electricity like Venezuela, we still have a long way to go before 0.75J/GH become unprofitable. After all, many industrial setups have even cheaper electricity than be, where they can run 2 J/GH at the moment profitably.
With the BTC price rising, the difficulty will not only be sustainable at 95B but will most likely raise by another 30-50% during the next few months.
At a difficulty of 95,000,000,000 and with a Bitcoin worth around $450 you will earn roughly 10 cents per hour with one Terahash.
So a 1.1 J/GH system will be economic to keep running if you pay less than about 9 cents per kWh.
With a 2 J/GH system it would only be worth running if you pay less than 5 cents per kWh.
The issue over the next few months is how far difficulty will rise as the large mining farm roll out their 16nm miners. I believe that with sub 0.1 J/GH 16nm systems appearing and systems such as the SP50 that a tripling in difficulty is possible before the halving.
December 15, 2015, 04:32:14 PM
This is a serious question. Obviously the newest round of ASICs is coming on line at 0.25 J/GH or better in the past few months, and we have seen difficulty increase as a result.
At today's price of $460, and an implied forward difficulty of ~95,000,000,000 anybody with over 0.75 J/GH will be mining at a loss and presumably taking those rigs offline (assuming prices stay here or decline).
For the sake of argument, assume everybody over that threshold removes themselves from the network. Does the remaining hashpower still sustain a 95B difficulty? In other words, is the amount of new ASICs coming online over the past few months and in the next few months enough to replace what will be removed?
(I have no idea, but I actually think there is more hashpower coming on than will be removed).
At today's price of $460, and an implied forward difficulty of ~95,000,000,000 anybody with over 0.75 J/GH will be mining at a loss and presumably taking those rigs offline (assuming prices stay here or decline).
For the sake of argument, assume everybody over that threshold removes themselves from the network. Does the remaining hashpower still sustain a 95B difficulty? In other words, is the amount of new ASICs coming online over the past few months and in the next few months enough to replace what will be removed?
(I have no idea, but I actually think there is more hashpower coming on than will be removed).
Considering 1.1 J/GH is still profitable here and i live in Canada, which is not a place with effectively free electricity like Venezuela, we still have a long way to go before 0.75J/GH become unprofitable. After all, many industrial setups have even cheaper electricity than be, where they can run 2 J/GH at the moment profitably.
With the BTC price rising, the difficulty will not only be sustainable at 95B but will most likely raise by another 30-50% during the next few months.
December 15, 2015, 02:25:54 PM
This is a serious question. Obviously the newest round of ASICs is coming on line at 0.25 J/GH or better in the past few months, and we have seen difficulty increase as a result.
At today's price of $460, and an implied forward difficulty of ~95,000,000,000 anybody with over 0.75 J/GH will be mining at a loss and presumably taking those rigs offline (assuming prices stay here or decline).
For the sake of argument, assume everybody over that threshold removes themselves from the network. Does the remaining hashpower still sustain a 95B difficulty? In other words, is the amount of new ASICs coming online over the past few months and in the next few months enough to replace what will be removed?
(I have no idea, but I actually think there is more hashpower coming on than will be removed).
At today's price of $460, and an implied forward difficulty of ~95,000,000,000 anybody with over 0.75 J/GH will be mining at a loss and presumably taking those rigs offline (assuming prices stay here or decline).
For the sake of argument, assume everybody over that threshold removes themselves from the network. Does the remaining hashpower still sustain a 95B difficulty? In other words, is the amount of new ASICs coming online over the past few months and in the next few months enough to replace what will be removed?
(I have no idea, but I actually think there is more hashpower coming on than will be removed).
Jump to: