Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Feb 28 2019 19:45
    @Arachnid banned @merkle_tree_twitter
  • Feb 17 2019 00:56
    @jpitts banned @Aquentson_twitter
  • Apr 09 2018 04:14
    @holiman banned @JennyJennywren_twitter
  • Oct 21 2017 19:12
    @Arachnid banned @Musk11
James Hancock
@MadeofTin
Great video! I am adding it to my list to remember. Language is a big piece of the puzzle. I wonder if gas prices and state size management are considered higher or lower level systems in the context of a establishing firm foundation to build on.
Hudson Jameson
@Souptacular
So according to https://www.ethernodes.org/istanbul Istanbul is set to occur Saturday Dec. 7th. I am finishing up the blog post for blog.ethereum.org (waited until Monday for max exposure). How should I word in the blog that the time is variable between x and y? and if y is Saturday Dec. 7th what is x?
Danno Ferrin
@shemnon
So…. block times went longer than expected. My initial swag was for one time and I got feedback to use a shorter time. It trended up and past my initial time.
I would think we want to phrase it as “at block 9,069,000, currently extimated at Y” and if we want X say “originally estimated at Wednsday 4, December"
I’m thinking for Berlin we do need to do time based forks. Perhaps if (blocknum % 1000 = 0) and (most recent 256 headers time >= FORKTIME). Only occuring at x000 blocks provides some certanty, as well as provides a backward compatible number we can say “the fork was at block x000"
The fork could still skew off of noon wednesday, but it will skew by hours instead of days.
And then in the next software update clients can just switch to a FORKBLOCK
Arachnid @Arachnid is sure he's suggested gating forks on timestamp instead of block number before...
Danno Ferrin
@shemnon
No “I told you so” gif? here you go - https://media.giphy.com/media/q0gmJhbizzdE4/giphy.gif
Nick Johnson
@Arachnid
I don't think anyone ever claimed the time wouldn't drift a bit. :)
Danno Ferrin
@shemnon
But 4 days seems less than ideal. Especially with difficulty bomb shenanigans.
Nick Johnson
@Arachnid
Actually, a chart showing the total drift between expected block time and actual block time could be really interesting. Power networks do this for the drift between expected cycles and actual cycles for mains freq.
Danno Ferrin
@shemnon
This method would present issues with Fast Sync, you would have to add logic to enable time based forks, not too difficult just more code to write.
Hudson Jameson
@Souptacular

Couple other q's:

@tkstanczak which version of Nethermind should I include in the blog post that points to Istanbul compatibility?
@holiman @Arachnid @cdetrio @jpitts is anyone responsible for http://forkmon.ethdevops.io/?

Tomasz Kajetan Stańczak
@tkstanczak
1.q.
1.1.8
for simplicity
has Istanbul mainnet block number
Hudson Jameson
@Souptacular
Thanks @tkstanczak
Eth-Gitter-Bridge
@Eth-Gitter-Bridge
<M H Swende (holiman)> @Hudson Jameson (souptacular) sure, I am
Hudson Jameson
@Souptacular
Cool! Should I include the same link as last year in the blog I am putting out soon? http://forkmon.ethdevops.io
It doesn't have to be running yet neccessarily, but it would be good for the blog to have the link.
@holiman
Eth-Gitter-Bridge
@Eth-Gitter-Bridge
<M H Swende (holiman)> Sure
Hudson Jameson
@Souptacular
cool ty
Nick Savers
@nicksavers
Is there confidence that the majority of miners have upgraded?
Hudson Jameson
@Souptacular
Not yet @nicksavers
Outreach starts after the blog is posted usually.
Ethereum Cat Herders and others plan on reaching out to mining pools. I'm doing a huge twitter tagging outreach as well.
Paweł Bylica
@chfast
For Istanbul Aleth 1.7.1 is needed: https://github.com/ethereum/aleth/releases/tag/v1.7.1
Holger Drewes
@holgerd77
We just released EthereumJS v4.1.1, this is the first release with stable Istanbul support. An upgrade from v4.1.1 (introduced beta support for Istanbul) is highly recommended.
Hudson Jameson
@Souptacular

@/all Istanbul announcement blog released.

We are recommending people update their nodes before Dec. 1st in order to account for variable block times

Blog: https://blog.ethereum.org/2019/11/20/ethereum-istanbul-upgrade-announcement/
Tweet: https://twitter.com/ethereum/status/1197022601329201153?s=20
Tweet with fun emojis: https://twitter.com/hudsonjameson/status/1197021040305754112?s=20
Reddit Thread: https://www.reddit.com/r/ethereum/comments/dyx7oq/ethereum_istanbul_upgrade_announcement/

Spread the word to your friends, miners, exchanges, and anyone else running a node!

econoar
@econoar

Block times keep rising. 14.39s a block yesterday. Up from 13.1

It’s been 2mn blocks since the last delay so it’s activated as far as I can tell.

The ice age is activated*
El De-dog-lo
@fubuloubu
When is your estimate for 20s blocks?
econoar
@econoar
Last time it took 2 months
IIRC each one gets worse but could be misremembering
El De-dog-lo
@fubuloubu
So, we might just make it in time?
Hudson Jameson
@Souptacular
Can any of the core devs or whoever knows the secret ice age finding formula check?
econoar
@econoar
It’s not reset in Istanbul @bryant
Wow Gitter mobile is horrid
Hudson Jameson
@Souptacular
It is!
El De-dog-lo
@fubuloubu
Welcome to paradise @econoar!
sassal
@sassal
ia3_kitteh.jpg
i would like to prevent this :P
sassal
@sassal

did some more digging:

since the diff bomb delay is 2mn blocks from when it started kicking in last, the increase in block times syncs up perfectly.

~1 month ago the diff bomb kicked in at around block 876xxxx
~1 year ago the diff bomb kicked in at around block 676xxxx

It's rough data but it fits the theory that block times are getting longer due to the diff bomb.

James Hancock
@MadeofTin
It won’t happen like last time because the difficulty is different this time.
Which is both a good thing for us.
The lower the difficulty the earlier it shows up, but the less quickly it climbs in severity
sassal
@sassal
ah okay thats good to know! do you have any estimates of when block times are likely to hit 20s+?