While technically Ethereum's ~15tx/s is an order of magnitude more than Bitcoin's 7x max, that's not really what I had in mind, they're effectively equivalent in relation to where they need to be (hundreds or thousands/sec). And as expected Ethereum's tx cost is increasing with usage too. And Vitalik to his credit is clear that Ethereum still has significant work to do on this front:
>In a lengthy statement released to TechCrunch, Buterin explained that he plans to spend more time ensuring that Ethereum fulfills its potential as a platform.
"2017 really has been the year where hype in crypto, including financial hype and social hype in general has far exceeded the reality of what existing blockchain systems can offer. There is a lot of attention, and a lot of eager expectation, but as far as reality goes the practical usability of blockchains has in some cases even regressed due to rising transaction fees."
"I expect 2018, at least within the Ethereum space that I’m best able to speak about, will be the year of action. It will be the year where all of the ideas around scalability, Plasma, proof-of-stake, and privacy that we have painstakingly worked on and refined over the last four years are finally going to turn into real, live working code that you can play around in a highly mature form in some cases on testnets, and in some key cases even on the public mainnet. Everyone in the Ethereum space recognizes that the world is watching, and we are ready to deliver."
1 block per 10 minutes with 4000tx per block results in 7tx/s ... well, why not use blocks of 100x size for starters, that would immediately produce 700tx/s. Still not VISA, yet that would be at least something. The main reason it is not going to happen is probably that that the major Bitcoin stakeholders are completely satisfied with the current situation where they are collecting higher and higher fees (reaching 10BTC+ per block!).
The main promise of Bitcoin - the scarcity of coins - has even been much more exceeded by the scarcity of transaction slots.
> On-chain scaling is not sustainable. If you want to handle as many transaction as for example Visa, you would need 1 GB blocks every 10 minutes, which would make the whole blockchain heavily centralized because regular users won't be able to host full nodes to validate payments.
The way things are today and the way they need to be to ensure a censorship-resistant secure public ledger are not the same. Core is working toward that, security is first priority, then optimize performance after that.
Bitcoin core has made significant technical decisions that has hampered the project in these two areas.
*Source: https://themerkle.com/bitcoin-vs-ethereum-transaction-throug...