3 ms·

Computer prices are only falling when you compare processors with the same speed and monitors with the same resolution. Overall, prices might as well increase i

Computer prices are only falling when you compare processors with the same speed and monitors with the same resolution. Overall, prices might as well increase if you look at the prices of the last flagship phones or at least remain stable if you look at main memory. It comes down to the [1] Jevons paradox. >In 1865, the English economist William Stanley Jevons observed that technological improvements that increased the efficiency of coal-use led to the increased consumption of coal in a wide range of industries. He argued that, contrary to common intuition, technological progress could not be relied upon to reduce fuel consumption. PV and Wind fell because they were above the market costs of energy. Once they are below, they will stop falling further. Prices will only go further down if energy supply outpaces demand. Why should that ever happen? Have you seen people stop buying SUVs? Once energy gets cheaper, SUVs will become even bigger. Being well-off essentially means being able to waste energy. To signal status, people will burn all available non-essential energy, however much there is to burn. [1] https://en.wikipedia.org/wiki/Jevons_paradox https://en.wikipedia.org/wiki/Jevons_paradox

pfdietz 7y ago
> PV and Wind fell because they were above the market costs of energy. Once they are below, they will stop falling further As PV and wind fall in cost, they will end up dictating the market price. So, you are technically correct, but otherwise wrong: the current market price is no floor for the cost of PV and wind, any more than (at any point) the cost per unit of computation or storage was the floor on the cost of computers. I expect PV to drive the levelized cost of power below $0.01/kWh, possibly much below (recent utility scale bids in some places have the power not far above this, and extrapolating the experience curve for PV has costs falling another factor of 4 or so by the time PV becomes a globally dominant energy source.) At these costs, resistive heat from PV is cheaper than any fossil fuel heat source, per unit of thermal energy. And thermal energy is very cheap to store, compared to batteries.
- toohotatopic 7y ago
  >the cost per unit of computation or storage was the floor on the cost of computers. What do you mean? That the energy bill for computers is bigger than the material bill? What else but the price of the hardware has determined the price of computers? >I expect PV to drive the levelized cost of power below $0.01/kWh For some time, this is possible. But if there is a sea of energy, people will find a way to use it. Then, demand rises and prices will rise again. Sooner or later, material science, genetics and everything else will simulate everything. Progress will depend on the amount of energy that is spent on those calculations. The cloud will take any energy that it can get. Bitcoin was just the start. Humans take roughly 2kWh worth of food per day. We are willing to pay about $20 for it. It might as well be that energy prices rise to $10/kWh. Not because PV itself is expensive, but because somebody owns the land where the PV modules are standing.
  - pfdietz 7y ago
    What I meant was clear (and not what you wrote there.) Jevon does not say that prices cannot fall. So if you're trying to use it to say prices will not fall, you're doing it all wrong.
    - toohotatopic 7y ago
      >What I meant was clear Unfortunately not to me.

pfdietz 7y ago

> PV and Wind fell because they were above the market costs of energy. Once they are below, they will stop falling further As PV and wind fall in cost, they will end up dictating the market price. So, you are technically correct, but otherwise wrong: the current market price is no floor for the cost of PV and wind, any more than (at any point) the cost per unit of computation or storage was the floor on the cost of computers. I expect PV to drive the levelized cost of power below $0.01/kWh, possibly much below (recent utility scale bids in some places have the power not far above this, and extrapolating the experience curve for PV has costs falling another factor of 4 or so by the time PV becomes a globally dominant energy source.) At these costs, resistive heat from PV is cheaper than any fossil fuel heat source, per unit of thermal energy. And thermal energy is very cheap to store, compared to batteries.

toohotatopic 7y ago
>the cost per unit of computation or storage was the floor on the cost of computers. What do you mean? That the energy bill for computers is bigger than the material bill? What else but the price of the hardware has determined the price of computers? >I expect PV to drive the levelized cost of power below $0.01/kWh For some time, this is possible. But if there is a sea of energy, people will find a way to use it. Then, demand rises and prices will rise again. Sooner or later, material science, genetics and everything else will simulate everything. Progress will depend on the amount of energy that is spent on those calculations. The cloud will take any energy that it can get. Bitcoin was just the start. Humans take roughly 2kWh worth of food per day. We are willing to pay about $20 for it. It might as well be that energy prices rise to $10/kWh. Not because PV itself is expensive, but because somebody owns the land where the PV modules are standing.
- pfdietz 7y ago
  What I meant was clear (and not what you wrote there.) Jevon does not say that prices cannot fall. So if you're trying to use it to say prices will not fall, you're doing it all wrong.
  - toohotatopic 7y ago
    >What I meant was clear Unfortunately not to me.