5 ms·

Visa processes an average of 150 million transactions per day [1] (less than 2000 tx per second), so a 256MB block would do it. Of course that blocksize should

by hggh 6y ago

Visa processes an average of 150 million transactions per day [1] (less than 2000 tx per second), so a 256MB block would do it. Of course that blocksize should be incremented with more demand. [1] https://usa.visa.com/run-your-business/small-business-tools/retail.html https://usa.visa.com/run-your-business/small-business-tools/...

littlestymaar 6y ago
150Mtx/day means less than 2000 tx/s on average, but you can't dimension the throughput of a system on the average load… If you read the entire sentence in the link you posted, you'd see that VISA's network can handle peaks way higher than the daily average (which is mandatory if you don't want your payment system to break during Black Friday or the last Saturday before Christmas). > VisaNet handles an average of 150 million transactions every day and is capable of handling more than 24,000 transactions per second And even if a 256MB was enough, it's still way too much to work with the incentives of the bitcoin protocol: you cannot expect people to run full nodes on dedicated machines with several TB of storage, without any retribution whatsoever. And if nobody runs a full node, then the mining cartel have total control of the blockchain, which is obviously not what you want…
- hggh 6y ago
  > 150Mtx/day means less than 2000 tx/s on average, but you can't dimension the throughput of a system on the average load… Keep in mind that BCH transactions are "zero-conf" (there is no return-by-fee issue, so double spends aren't viable in the vast majority of transactions [1]). So, in case of a big peak in, most of the transactions would "settled" without the 10-minute confirmation, and after the peak they would be confirmed eventually. It's not the ideal situation, but the final user wouldn't see any difference (unless they are buying something expensive like a car). > without any retribution whatsoever If someday Bitcoin Cash gets to that level, the price will rise (and the number of tx per block) and so the coinbase will be much higher. For example with 2000 tx/s, around 200 bytes/tx, and 1 sat/byte, the coinbase of a block would be 2.4 BCH in fees + the reward (currently 6.25 BCH), and then you can expect coinbases of tens of thousands if not hundreds of thousands of dollars. [1] https://doublespend.cash/ https://doublespend.cash/
  - littlestymaar 6y ago
    > Keep in mind that BCH transactions are "zero-conf" (there is no return-by-fee issue, so double spends aren't viable in the vast majority of transactions [1]). So, in case of a big peak in, most of the transactions would "settled" without the 10-minute confirmation, and after the peak they would be confirmed eventually. It's not the ideal situation, but the final user wouldn't see any difference (unless they are buying something expensive like a car). Wat? You realize that without the “10-minute confirmation” as you called it (10 min is not enough, there are stale blocks from time to time) you can spend the same money indefinitely? (double-spending) > If someday Bitcoin Cash gets to that level, the price will rise (and the number of tx per block) and so the coinbase will be much higher. For example with 2000 tx/s, around 200 bytes/tx, and 1 sat/byte, the coinbase of a block would be 2.4 BCH in fees + the reward (currently 6.25 BCH), and then you can expect coinbases of tens of thousands if not hundreds of thousands of dollars. Miners get that reward. (non-mining) Full nodes get zero, which is the fundamental problem of the bitcoin protocol (Nakamoto didn't envisioned that there could be miners that wouldn't run full nodes, and that most full nodes wouldn't be mining).
    - hggh 6y ago
      I think this e-mail [1] answers these two points [1] https://satoshi.nakamotoinstitute.org/emails/cryptography/2/ https://satoshi.nakamotoinstitute.org/emails/cryptography/2/
      - littlestymaar 6y ago
        It doesn't. Satoshi simply failed (as you can see in this response) to imagine a world where full nodes and miners would be disjoint entities, and as a result full node receive no reward whatsoever (and the biggest miners don't even bother to verify that the blocks they mine are correct). The reliability and independence of the bitcoin blockchain runs on top of the shoulders of benevolent people running full nodes. Increasing the block size, means increasing the cost for those essential people. Not a good idea at all.
        6y ago

by hggh 6y ago

littlestymaar 6y ago

150Mtx/day means less than 2000 tx/s on average, but you can't dimension the throughput of a system on the average load… If you read the entire sentence in the link you posted, you'd see that VISA's network can handle peaks way higher than the daily average (which is mandatory if you don't want your payment system to break during Black Friday or the last Saturday before Christmas). > VisaNet handles an average of 150 million transactions every day and is capable of handling more than 24,000 transactions per second And even if a 256MB was enough, it's still way too much to work with the incentives of the bitcoin protocol: you cannot expect people to run full nodes on dedicated machines with several TB of storage, without any retribution whatsoever. And if nobody runs a full node, then the mining cartel have total control of the blockchain, which is obviously not what you want…

hggh 6y ago
> 150Mtx/day means less than 2000 tx/s on average, but you can't dimension the throughput of a system on the average load… Keep in mind that BCH transactions are "zero-conf" (there is no return-by-fee issue, so double spends aren't viable in the vast majority of transactions [1]). So, in case of a big peak in, most of the transactions would "settled" without the 10-minute confirmation, and after the peak they would be confirmed eventually. It's not the ideal situation, but the final user wouldn't see any difference (unless they are buying something expensive like a car). > without any retribution whatsoever If someday Bitcoin Cash gets to that level, the price will rise (and the number of tx per block) and so the coinbase will be much higher. For example with 2000 tx/s, around 200 bytes/tx, and 1 sat/byte, the coinbase of a block would be 2.4 BCH in fees + the reward (currently 6.25 BCH), and then you can expect coinbases of tens of thousands if not hundreds of thousands of dollars. [1] https://doublespend.cash/ https://doublespend.cash/
- littlestymaar 6y ago
  > Keep in mind that BCH transactions are "zero-conf" (there is no return-by-fee issue, so double spends aren't viable in the vast majority of transactions [1]). So, in case of a big peak in, most of the transactions would "settled" without the 10-minute confirmation, and after the peak they would be confirmed eventually. It's not the ideal situation, but the final user wouldn't see any difference (unless they are buying something expensive like a car). Wat? You realize that without the “10-minute confirmation” as you called it (10 min is not enough, there are stale blocks from time to time) you can spend the same money indefinitely? (double-spending) > If someday Bitcoin Cash gets to that level, the price will rise (and the number of tx per block) and so the coinbase will be much higher. For example with 2000 tx/s, around 200 bytes/tx, and 1 sat/byte, the coinbase of a block would be 2.4 BCH in fees + the reward (currently 6.25 BCH), and then you can expect coinbases of tens of thousands if not hundreds of thousands of dollars. Miners get that reward. (non-mining) Full nodes get zero, which is the fundamental problem of the bitcoin protocol (Nakamoto didn't envisioned that there could be miners that wouldn't run full nodes, and that most full nodes wouldn't be mining).
  - hggh 6y ago
    I think this e-mail [1] answers these two points [1] https://satoshi.nakamotoinstitute.org/emails/cryptography/2/ https://satoshi.nakamotoinstitute.org/emails/cryptography/2/
    - littlestymaar 6y ago
      It doesn't. Satoshi simply failed (as you can see in this response) to imagine a world where full nodes and miners would be disjoint entities, and as a result full node receive no reward whatsoever (and the biggest miners don't even bother to verify that the blocks they mine are correct). The reliability and independence of the bitcoin blockchain runs on top of the shoulders of benevolent people running full nodes. Increasing the block size, means increasing the cost for those essential people. Not a good idea at all.
      - 6y ago