4 ms·
Their approach of attributing cost per transaction doesn't really make sense for bitcoin. I don't know the edge case of zero transactions within a block timefr
by blunte 9y ago
Their approach of attributing cost per transaction doesn't really make sense for bitcoin. I don't know the edge case of zero transactions within a block timeframe, but for 1+ transactions, a block will certainly be generated. And the generation of that block has very, very little energy cost associated with transaction validation. The vast majority of time spent is on hashing over and over until the outcome has the lucky leading zeroes (and thus winning the privilege of writing the block).
Whatever throughput solution is devised, unless the fundamental block-writing-lottery rules are changed, energy use will not change significantly.
Also, you can't just sum up all bitcoin mining energy costs and project how that energy could otherwise be spent. Energy transmission is complicated and wasteful over great distances. If some mining operations are in areas of low habitation, then they aren't necessarily taking from what could otherwise be given to more productive uses.
Finally, assuming _some_ mining is done with local renewable energy, then that energy is not necessarily being wasted any more than sun that falls on unused dirt is being wasted.
- thisisit 9y agoWhile I can't comment on the exact stats they are pulling up but I don't understand your logic here. Transaction is not "valid" until it is part of a block. So, validation is not separate from block generation, it's a part of it. A good read on this topic: https://bitcoin.stackexchange.com/questions/49531/concept-of-validating-transactions https://bitcoin.stackexchange.com/questions/49531/concept-of...
- blunte 9y agoCorrect, validation is certainly part of block generation. But the act of validating a transaction is a minimal effort compared to the (many times of) hashing that goes on as part of the proof of work. My point is that for all practical purposes, the transaction validation is so insignificant that it makes little sense to try attribute the energy cost of making one block to the transactions within that block. Stuffing more transactions into a block doesn't change the fact that a new block will be created about every 10min, and that the energy cost of writing that block will be about the same. Seeing this cost from a per-transaction perspective is questionable. The reason this whole process is so expensive is because of the design goal of relative immutability and the proof-of-work scheme that aids in this reaching this goal.
- thisisit 9y agoA transaction needs to be part of a block to be valid. So using transaction validation is a red herring. The validation uses same amount of energy it is required to be included in the block. But, I see your point. So let's see how this thing will unfold if we talk about blocks (assuming all data is correct): From the article: > This averages out to a shocking 215 kilowatt-hours (KWh) of juice used by miners for each Bitcoin transaction So, as per the article a house needs 215kwh per week. Data Source for the article here: https://digiconomist.net/bitcoin-energy-consumption https://digiconomist.net/bitcoin-energy-consumption Energy consumption for yesterday shows as 24.25 Twh. Block is generated every 10 mins so 144 blocks in a day. 24.25/144 = 0.17 Twh ~ 168 MWh or 168000 kwh for each block. When divided with per house per week from article - 168000/215 ~ 781 houses So the apt headline, as per you is - "One Bitcoin Block Now Uses as Much Energy as more than 750 Houses in a Week" I am sure people will love to talk about consumption per transaction if that was the headline.