Conversation

SBF
@SBF_FTX
3) After our panel on effective altruism, Vitalik and I talked about a bunch of stuff; one topic was on different layer 1 blockchains. My main question was: What does he think makes an L1 better, or worse?
2
75
SBF
@SBF_FTX
4) And more specifically: What could
@solana
do that would make it better? This wasn't about Solana in an absolute sense; obviously Vitalik thinks Ethereum is better. This was about what it could do to be more good and less bad than it is now. ------- Some background:
1
66
SBF
@SBF_FTX
5) There are a lot of axes on which blockchains can differ! This will primarily focus on two: a) speed/throughput b) security/consensus My argument: the main advantage of Solana is that it's really fast. Now, *all* geographically distributed chains have block times > 100ms.
4
69
SBF
@SBF_FTX
6) That's the time it takes light to go around the world, roughly. But Solana manages to get roughly 50,000 transactions through per second right now, with the ability to scale up to 1m or so. And all of those are on the same shard: they can all compose with each other.
3
71
SBF
@SBF_FTX
7) The main thing we compared it to was ETH 2.0, coming out over the next few years. ETH2 is closer to 1k TPS *per shard*, but can can have multiple shards. Each shard is independent on short timescales, synching up on longer ones.
2
62
SBF
@SBF_FTX
8) The cost of ETH2 is that you can't have a coherent, composing ecosystem that collectively takes > 1k TPS. A cypto exchange, Facebook, Twitter, WeChat, Alipay, etc. each take ~10k-10m TPS, so won't fit on a shard, or really on ETH2. --------
5
84
toly 🇺🇸
@aeyakovenko
Replying to
@SBF_FTX
and
@SBF_Alameda
> A cypto exchange, Facebook, Twitter, WeChat, Alipay, etc. each take ~10k-10m TPS, so won't fit on a shard, or really on ETH2. Google serves about 350k ads per second. While there are a ton of RPC requests, actual txs per second might not be all that much for the world.
2
8