Please note that the Eth core devs have a contingency in place to account for hashrate changes which can affect the merge timing.

The actual merge is triggered by a TTD (total terminal difficulty) value is set in the code, this value is the difficulty of the entire Eth network up to that point in time which when it reaches a specific value triggers the merge to occur.

Due to fluctuations in hashrate (i.e. 5GB DAG size affecting some asics, miners moving on/off the network, etc.) the core devs have included a TTD override switch in the code. This allows the node operators to on-the-fly update the TTD value to a new number to account for hashrate fluctuations such that the merge still occurs around a desired date (i.e. Sep 15 or thereabouts).

The idea is that the core devs are continuously monitoring the hashrate and if it fluctuates to a point were it will affect the merge date to an undesired degree, the devs will publish a TTD override value which node operators will then use to update their nodes and thus adjust the merge timing to account for the hashrate fluctuation.

Consensus layer call wrapped up:

1) Mainnet shadow fork #9 completed
2) Goerli shadow fork to be performed next week
3) Goerli TTD/merge tentatively targeted for the week of Aug 11
4) Assuming no issues with Goerli merge, Mainnet merge tentatively targeted for either the week of Sep 12 or Sep 19

Goerli is the next (and final) testnet to be merged (#3 item). Sepolia was merged this week (#2 item).

Pari is the developer working on the weekly mainnet shadow forks, the other devs work on other merge (and post-merge) items. Pari has not mentioned if he will be discontinuing the shadow forks.

Core devs call wrapped up:

1) Mainnet shadow fork #9 scheduled for next week
2) Sepolia testnet TTD/merge completed
3) Goerli testnet TTD value to be decided on in consensus layer call next week or core dev call the week after

Core dev call just wrapped up:

1) 7th mainnet shadow fork done Wed, some non-merge issues with Erigon/Besu clients, fixes being put into place, 8th mainnet shadow fork set for two weeks from now.
2) Sepolia beacon chain live, TTD/merge targeted for Jul 6.
3) Mainnet difficulty bomb delay fork (gray glacier) still targeted for Jun 29.

Consensus layer dev call wrapped up:

1) Mainnet shadow fork #7 to take place next week.
2) Sepolia testnet beacon chain to also launch next week, TTD/merge the week after.

So Sepolia testnet could be done by Jul 1 with Goerli to follow after. Since Ropsten was done Jun 8, Sepolia planned for Jul 1, if the same one month spacing is followed Goerli could be done by start of Aug. So mainnet could be in Sep.

EIP-5133 updated to change block adjustment to 700,000 blocks. Devs are now targeting mid-Sep for the merge as stated in the edited EIP.

Here's an interview conducted with Tim Beiko a few days ago where he clarifies on this, see at time 53:40 in the interview:

https://www.youtube.com/watch?v=pUA-HzIbWD0

The withdrawal function for staked eth is not part of the merge. Core devs will schedule enabling of withdrawals in a future fork post-merge, but no hard date has been published.

All core dev call has just wrapped up.

Devs have decided to delay the difficulty bomb. Initial block delay adjustment is as published in the proposed EIP-5133 (https://github.com/ethereum/EIPs/blob/master/EIPS/eip-5133.md) but the delay value may be adjusted. Fork to implement the delay is planned to go in effect around Jun 29.

Also, second testnet for merge will be Sepolia, beacon chain for Sepolia will launch around Jun 20 with merge TTD to follow around 2 weeks after.

"Unlocking"/withdrawal of staked eth is not part of the merge. The withdrawal function is planned to be added in a later fork post-merge.

I really hope that 12x prediction for the next btc bull run is correct.

Back in 2019, of the various computations to predict btc bull run peaks one of the simplest ones was to use the low-to-high ratio for each historical cycle, and it went something like:

Nov 2011 low to Nov 2013 peak was an approx 500x multplier.
Jan 2015 low to Dec 2017 peak was an approx 100x multiplier.

The multiplier change between these 2 cycles was 500/100 = 5, so the projected low-to-high ratio for the next cycle (i.e. 2021 bull run) would theoretically be 100/5 = 20.

And what was the actual multiplier in the 2021 bull run: approx 23

If that trend in the decrease in the multiplier by a factor of 5 holds true then the next bull run multiplier could be in the 4x to 5x range only! If the recent 25k low turns out to be the actual cycle low then the next peak could be in the 100k-125k range. That's some depressing sh*t!

I always felt that the low-to-high ratio method was too lowball in it's predictions, but the fact that it was almost spot on for the 2021 peak made it scary.

Do you really think that "layering-out/dcaing-out starting at 1.5k" is "trading the past"?

Back at the start of 2019, looking at what eth did over 2017-2018, were you not possibly thinking: "ok, eth reached 1.4k last time, I expect that it can reach at least 1.5k next bull run", so you end up planning to layer-out starting at 1.5k.

That's not "trading the past", that's just executing a plan you made based on the previous bull run.

In Metroid's case he based his decision mainly on how eth reached certain price levels (or how the market in general reached those levels) and not just on the price level itself. When I made my plan back in Jan 2019, I did not consider how the market reached certain price levels and only looked at the price levels (though I was greedier as my layer-out start point was at twice the previous bull run ath). For the next bull run, I'll probably factor in how the price reaches certain levels and not just the actual price level.

I always found it curious why you did this. I know you bought your eth at around $80 so selling at $1500 was a huge profit, but why only target around $100+ above the previous ath from back in 2018? Why didn't you dca out, ie. layer out, say 25% at $1500, 25% at $2500, etc.?

EIP for difficulty bomb delay has been published:

https://github.com/ethereum/EIPs/blob/master/EIPS/eip-5133.md

Delay has been adjusted by 500,000 blocks (approx. 2.5 months).

To be fair, the concern was they were not comfortable with 20+ sec block times by Aug, moving the bomb offset by 400,000 would make Aug block times stay in the 13-14 sec range and Sep block times could be in the 15-16 sec range, so if they wanted to slip the merge to Sep they'ld still be in the "comfortable" block times range. So merge could be in Aug, but could be in Sep.

Not yet.

In the discussion on drafting the EIP those devs who wanted to delay the bomb were advised to simply copy the last bomb delay EIP and adjust the block offset value to add 400,000 blocks. This value may change, so we'll have to wait for the EIP to be published.

Core dev call just wrapped up, devs sound like they are no longer ok with having 20+ sec block times by Aug.

No decision yet on pushing the bomb, but an EIP will be drafted to push the bomb by 400,000 blocks in order to ensure Aug block times in the 13 sec range. They will decide on that EIP in the next core dev call.

More blocktime estimates have been posted at: https://ethresear.ch/t/blocks-per-week-as-an-indicator-of-the-difficulty-bomb/12120/22 (see last post/graph)

Jun 15 estimated block times: ~15 secs
July 15 estimated block times: ~18 secs
Aug 15 estimated block times: ~21 secs

Vitalik in a previous core dev call said they can tolerate 20-25 sec block times as they've done in the past.

So given the schedule for the upcoming Ropsten testnet merge on Jun 8, plus two weeks running on Ropsten, then a further two testnet merges running for two weeks each, followed by a couple of weeks to schedule mainnet merge, brings it to around mid-Aug.

If their estimates of 21 secs in mid-Aug is correct then they might not delay the bomb anymore (barring any issues during the upcoming testnet merges).

Ouch, this brings back painful memories, I started mining at the start of 2017 but didn't have confidence that eth would last long term so I was selling my mined eth every month. I remember how each gtx1060 mined close to 7 eth over the whole year back then.

Really stupid of me to be selling monthly rather than saving all that eth. Took me a whole year before I finally stopped selling and started saving at the start of 2018. But by then each gtx1060 was only making something like 0.6 eth over the entire year.