[poll] optimal block time

[adolf512]

assuming that the difficulty can be increased at most 0.5% per block(adjusted every block based on the previous block time).

The advantage with fast block time is
-the difficulty can be adjusted more quickly
-more confirmations can be done for a given time(makes <50% attack harder)
-A solo-miner or small pool will receive a more even income of new coins

the disadvantage with fast block time is that the effective hashrate decreases which makes it easier to do a >50% attack. The typical blocktime for altcoins seams to be 60 seconds, however their is some exceptions such as fastcoin(12 seconds) and geistgeild(experimental, 15 seconds), both these have bean running for a large period of time, wordcoin tried a target block time of 15 seconds but changed to 30 due to a lot of issues with orphans.

The blocktime can sometimes be a lot less than the target block time, especially during instamining

[Orangecoin]

Dont go under 120 seconds.

[adolf512]

Dont go under 120 seconds.

I do not se any good reason to go above 60 seconds(assuming the difficulty is adjusted quickly and not like bitcoin).

[adolf512]

The results is lower than i expected, i voted for 20 seconds since i think it's a good compromise, if there wheren't for p2pool(decentralised mining pools) i would chose 15 seconds.

[adolf512]

still i am the only one who have motivated my decision(20 seconds) anyone else with an opinion?

[SaraMine]

good idea couse fastcoin and other fast block coins are opran factory.

[sumantso]

30 seconds seem to be optimum. Anything less and the orphan problem becomes severe.

Thats why WDC moved from 15s to 30s.

[mrvegad]

That's why mining pools coins are not very good, emunie, exo and qora can do txs in 15 secs or less with no doublespend or 51% attacks. They are database driven and will lead to the future of crypto.

[adolf512]

I added a the options 6, 7.5, 9, 12, 25, 36 and removed the options 7, 10 since they didn't had any votes anyway.

[adolf512]

That's why mining pools coins are not very good, emunie, exo and qora can do txs in 15 secs or less with no doublespend or 51% attacks. They are database driven and will lead to the future of crypto.

51% attacks is overrated and is has never killed a serious coin as far as i know. If a 51%  attack kills a major coin it will most likely be via a mining pool that has a majority of the hashing power, this could have happaned ti litecoin reasently where a single pool had over 50% of the network hashing rate.

[defaced]

that depends on how fast your blocks propagate through the network.. latency is a helluvadrug that orphans seem addicted too.

[adolf512]

that depends on how fast your blocks propagate through the network.. latency is a helluvadrug that orphans seem addicted too.

Yes of course, the latancy should be lower for a local cryptocurrency than a global cryptocurrency

[adolf512]

I guess the 3 persons who voted for 300 and the one who voted for 5 seconds doesn't have any deeper knowledge about cryptocurrencies, when the bitcoin blocktime was set to 10 minutes it didn't exist any major cryptocurrencies and the centralization of bitcoin that have occured wasn't predicted for obvious reasons.

It is easier to justify a target time of 5 seconds since it would allow for ultra-fast transactions but with the current internet latency it would probably work very bad if at all(i am considering releasing a testcoin with a target blocktime of 5 seconds).