The reason is because the only point of proof of work _is_ to keep latency high. You don't actually need it for anything else. The entire point of PoW is just to have there be a single value you can check before validating a transaction block. It's DDoS protection.
-
-
Replying to @cmuratori @TimSweeneyEpic
This is (but one) of the reasons "blockchain" is not a particularly good idea. People want low-latency, high-volume transactions, but the designs of these systems preclude that possibility entirely. They are, by design, not able to do the thing you wanted them to do.
5 replies 0 retweets 34 likes -
Replying to @cmuratori @TimSweeneyEpic
Throughput isn't a problem if block size scales (on BTC, notably, it does not). Latency remains, but various chains have addressed it with e.g. opt-in 0-confirmation transactions, where the payee accepts the double-spend risk.
1 reply 0 retweets 0 likes -
Replying to @moistgibs @TimSweeneyEpic
It's unclear what "block size scales" means here, though. Block sizes can't be scaled arbitrarily because they are universally replicated state. The VISA volume would crush most nodes on the BitCoin network, etc., just for storage.
2 replies 0 retweets 2 likes -
Replying to @cmuratori @TimSweeneyEpic
For example, BTC blocks are limited to 1MB. BCH 32MB. BSV blocks are variable and uncapped, and blocks >1GB have been mined. Throughput scales proportionally. This has storage implications for nodes of course - large block advocates contend that storage is cheap.
1 reply 0 retweets 0 likes -
Replying to @moistgibs @TimSweeneyEpic
That is kind of obviously false, as is well-covered in the original Lightning Network paper.pic.twitter.com/LYRhULDJqS
3 replies 0 retweets 2 likes -
Replying to @cmuratori @moistgibs
Those are rookie numbers. We’ve got to pump those numbers up. 15M PCU * 30Hz = at least 450M transactions per second required to support a live Fortnite event.
2 replies 0 retweets 14 likes -
is this true? The 450m / sec number? Not because I give a shit about the original topic but because that’s an interesting number to benchmark against
1 reply 0 retweets 0 likes -
Well, it doesn’t work that way now - today in that scenario there would be 150,000 servers handling 100 players each at 30Hz without communication among servers, and then databases handling at most several transactions per player per minute for globally persistent data.
1 reply 0 retweets 2 likes -
Replying to @TimSweeneyEpic @chriseberly and
But if we want a single metaverse in a shared world, with an open world programming model that’s not a mess, on the scale of a modern social network, then you need something like 250M PCU * 30 Hz = 7.5G transactions per second.
2 replies 0 retweets 4 likes
I don't know that it'd be that hard to build such a network, but it definitely wouldn't involve a blockchain, since they are very inefficient, and get worse as they go.
-
-
ya the number seems pretty realistic. I like having reference points like this for designing things. Good luck with the rest of the thread!
0 replies 0 retweets 0 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
Show additional replies, including those that may contain offensive content
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.