Content
@
https://onchainsummer.xyz
0 reply
26 recasts
26 reactions
androidsixteen 🌲
@androidsixteen.eth
Base, OP folks — how serious is this concern? Seems like blobspace is the bottleneck to hitting Base throughput targets If Eth mainnet is slow to supply more blobspace (I hate saying this word) and optimizations aren’t enough, does that only leave alt DA? https://x.com/jon_charb/status/1889819894298058960?s=46
6 replies
5 recasts
29 reactions
Roberto Bayardo 🎩
@bayardo.eth
Yes. Advocating for (and helping build out) more blob space is a p0 concern for the Base team. See for example https://base.mirror.xyz/6NDvVKw8x5obo3h1OgQx582hF73m81CnJabgbRneW2Q
1 reply
0 recast
1 reaction
Roberto Bayardo 🎩
@bayardo.eth
There's also more headroom to be had from optimizing blob usage, eg with stateful compression. But ultimately more capacity is needed.
1 reply
0 recast
1 reaction
androidsixteen 🌲
@androidsixteen.eth
How much juice do you think we can squeeze from optimizations alone? One thing I don’t know is if incentives are aligned to reduce back running and other failed txs since the sequencer still collects fees
1 reply
0 recast
0 reaction
Roberto Bayardo 🎩
@bayardo.eth
Because failed txs don't use much gas, they don't generate a whole lot of fees compared to the DA they occupy, so eliminating (most of) them might not be so bad for revenue. (I haven't done any deep analysis here though... just speculating.)
2 replies
0 recast
2 reactions
Roberto Bayardo 🎩
@bayardo.eth
Stateful compression is likely where the bigger wins are, along with signature aggregation. The latter though requires most people start using smart wallets/bundlers since Ethereum doesn't support aggregatable signatures natively.
1 reply
0 recast
1 reaction
Roberto Bayardo 🎩
@bayardo.eth
Last I checked, signatures comprised like 40% of the Base batches posted to the L1, because they don't compress at all.
2 replies
0 recast
1 reaction
Roberto Bayardo 🎩
@bayardo.eth
Good image from Vitalik here:
0 reply
0 recast
0 reaction
androidsixteen 🌲
@androidsixteen.eth
Ideal would be native BLS? I never understood why there wasn't a precompile for it...
1 reply
0 recast
0 reaction
