--- Log opened Mon Jun 14 00:00:37 2021
01:29 -!- belcher_ is now known as belcher
05:35 < elichai2> If I uniformly randomly pick a point in time, the expected time between that block and the next block should be 10 minutes, shouldn't it?
05:35 < elichai2> Related to https://twitter.com/pwuille/status/967878361782652928 but here it's from the random point to the next block, while sipa's one was from the previous block to the next
05:36 < elichai2> When I tried sampling 100M times, I got ~930 seconds between a random point and the next block, why is that? (I know it's off topic but I use this channel as a bitcoin math channel :) )
06:05 < real_or_random> re offtopic:  "If I uniformly randomly pick a point" .... I was in EC mode when I read this
06:17 < sipa> elichai2: between that *time* and the next block shoukd be ten minutes *if the hashrate matches the difficulty*
06:20 < elichai2> sipa: right, but I didn't think it would be that far off, I even remembered the average block rate was even *lower* than 10 minutes (9.5), so why do I get 15.5 minutes from the random point to the next block?
06:20 < sipa> how are you sampling time?
06:22 < elichai2> I gathered a vector of all the headers timestamps, I generate a random number in [genesis_timestamp, last_block_timestamp] and then binary search for it in the vector and measure the time between that one and the next block
06:22 < sipa> ok
06:22 < sipa> then it's expected i think
06:22 < sipa> during most of bitcoin'd history, the hashrate has been increasing
06:24 < elichai2> shouldn't it then be *lower* than 600 instead of higher?
06:24 < sipa> oh!
06:24 < sipa> sorry, not really awake yet
06:25 < sipa> that i cannot explain
06:25 < elichai2> haha, what time is it there?
06:30 < elichai2> hmmm, I would've showed you the code I used but it's in rust (although it's ~40 lines of code - https://play.rust-lang.org/?gist=98dd234bfa95461729ab20523737e10d) 
07:21 < roconnor> There was a medium-long period in Bitcoin's history where difficulty 1 was too high.
07:22 < roconnor> Maybe your effect goes away if you exclude that period?
07:27 < elichai2> do you remember what period that was?
07:27 < sipa> the first few weeks
07:28 < roconnor> http://bitcoin.sipa.be/speed-small-ever.png suggests it lasted until 2010.
07:30 < elichai2> let me try skipping the first `40000` blocks
07:31 < elichai2> roconnor: Ha! now it's 565 seconds
07:31 < elichai2> `sum: 56526097068 amount: 100000000, avg: 565.26097068`
07:32 < roconnor> Hmm, I'm a bit surprised that accounted for the entire effect.
07:32 < roconnor> 1 year seems so short.
07:34 < roconnor> I mean, I did call it medium-long, but still.
12:37 -!- jesseposner [~jesse@2601:647:0:89:ec38:1bfb:5fdf:9317] has joined #secp256k1
21:34 -!- belcher_ [~belcher@user/belcher] has joined #secp256k1
21:37 -!- belcher [~belcher@user/belcher] has quit [Ping timeout: 252 seconds]
22:48 -!- belcher_ [~belcher@user/belcher] has quit [Quit: Leaving]
--- Log closed Tue Jun 15 00:00:38 2021