Ethereum developers discuss whether to delay the difficulty bomb yet again with some arguing that it should be moved by another 3-4 months.
One of them said there shouldn’t be pressure for the sake of pressure, and some argued that “testnet deployments have been bumpy.”
The launch of the Ropsten testnet merge met some difficulties due to an artificial increase in its hashrate causing the launch time to be far sooner than expected.
“Given the low hash rate on Ropsten (and Sepolia), it is cheap for someone to increase it. On May 26th, someone did that and 20x’d the hash rate,” said Tim Beiko, an eth dev.
Rather than merge related testnet problems, however, that sounds more like someone playing games with a number of devs on the All Core Meeting saying this issue doesn’t reflect on the Merge.
But Tomasz Stańczak, the founder of the eth1 client Nethermind who pushed for delaying the difficulty bomb, argued that the shadoforks don’t tell much in reply to a dev stating they had gone well.
Stańczak instead said that the hive tests had shown “multiple failings” with no further detail provided.
Marius van der Wijden was of the view that a delay would mean another hardfork would need to be scheduled which requires community coordination. “We might lose some people if we schedule another delay,” he said.
He further argued that he is not seeing technical merits to the proposal to delay the difficulty bomb, and that teams should really strive to “hit the timelines we set ourselves.”
But an EIP will now be drafted to be discussed in the next call with it sounding like there’s some sort of consensus for a half-way delay of two months.
The difficulty bomb has not quite began moving yet as can be seen above, yet Stańczak argued block times will rise to 25 seconds in August.
Previously it has taken some months from the difficulty bomb kicking in to blocktimes rising anywhere near 30 seconds, and so it probably wouldn’t be until the end of September if not later.
A delay of two months would instead mean the difficulty bomb would start in early October, and so blocktimes would stay at 13 seconds through August and September.
Some take that difficulty bomb and its delay as the actual target date for the Merge, with Stańczak publicly stating prior to the call that he estimates the Merge “in the second half of October,” with August being optimistic.
The initial target for the merge was June considering they have been testing it on Klin since March. This was pushed back to August, though that probably always was September, and now may be pushed back again to October.
If the difficulty bomb is delayed, then a significant part of the Merge pushback will be the hard-forking of eth1 to implement that delay, with two months costing eth holders $60 million on top of the $90 million for the delay to August.
All the while, no one has quite explained why they spent time on shadow forking when it’s apparently useless according to Stańczak, or why do we need endless testnet after testnet merges when it usually has been one testnet upgrade – klin – running for three months.
In addition, letting the difficulty bomb continue should smooth the transition of miners out of the network as instead of all their income being suddenly cut off, it will gradually reduce due to blocktimes increasing, halving in September and then down to zero once the merge goes live.
But from the discussion, the worst they seemingly plan to do is to delay to October anyway, so it all sounds a bit academic with the Ropsten testnet merge to occur first in about two weeks which should give some indication of how the delay strategy is going.