BIP95: Change proposed Testnet 5 genesis block structure #2347

pull ajtowns wants to merge 1 commits into bitcoin:master from ajtowns:202611-bip95-bump changing 1 files +32 −30

ajtowns commented at 3:36 PM on November 4, 2026: contributor

This moves the anti-pre-mine commitment back into pszTimestamp and keeps it as ASCII text, rather than using it as a fake public key. The genesis transaction's output is replaced by an OP_RETURN to avoid confusion over spendability. Using the OP_RETURN output and nVersion rolling for additional nonce bits are explicitly allowed.

in bip-0095.md:12 in 41c604aae6

   8 | @@ -9,7 +9,7 @@
   9 |    Assigned: 2026-06-22
  10 |    License: CC0-1.0
  11 |    Discussion: 2026-06-02: https://groups.google.com/g/bitcoindev/c/kGUMTxOvdJA
  12 | -  Version: 0.1.0
  13 | +  Version: 0.2.0

jonatack commented at 4:49 PM on November 4, 2026:

Following the merge of #2346, here and in the changelog

  Version: 0.3.0

jonatack added the label Proposed BIP modification on Nov 4, 2026
jonatack added the label Pending acceptance on Nov 4, 2026
jonatack commented at 4:50 PM on November 4, 2026: member

cc BIP authors @fjahr @polespinasa
in bip-0095.md:90 in 41c604aae6 outdated
```
 109 | +  space, as it is unspendable in any event, and this avoids bike shedding
 110 | +  over its content.
 111 | +- `Message` carries a recent Bitcoin mainnet block hash as an anti-pre-mine
 112 | +  commitment, serving the same purpose as the newspaper headline in
 113 | +  Bitcoin's mainnet genesis block.
 114 | +- `Time stamp` is the time of mining or that of the block which hash is
```
darosior commented at 6:40 PM on November 4, 2026:

So, which of the two is it?

Or do you mean "start with the timestamp of the mainnet block, and roll it forward to current time as you exhaust the search space"? But is it really necessary with nonce + 8 bytes of extranonce + version rolling?

ajtowns commented at 6:54 AM on November 5, 2026:

I don't think it's necessary (1M diff should need about 52 bits of nonce space, so 64 bits of extranonce alone should certainly cover it, plus 48-56 bits of header nonce via BIP320 or BIP323 nVersion rolling). But miners designed for mainnet may roll the timestamp by default, and I don't believe either stratumv1 or stratumv2 provides any way of even specifying otherwise, so I think it's better to leave it flexible.

darosior commented at 3:27 PM on November 5, 2026:

Makes sense.
ajtowns force-pushed on Nov 4, 2026

in bip-0095.md:87 in 352bf23c26

 106 | +
 107 | +- `scriptPubKey` is the script public key in the coinbase transaction's
 108 | +  output. It is `OP_RETURN` followed by a push of 8-bytes of extranonce
 109 | +  space, as it is unspendable in any event, and this avoids bike shedding
 110 | +  over its content.
 111 | +- `Message` carries a recent Bitcoin mainnet block hash as an anti-pre-mine

murchandamus commented at 5:03 PM on November 5, 2026:

This previously said “in the coinbase input script”, which still seems to be the intention. I am confused that this part was dropped.

murchandamus commented at 5:06 PM on November 5, 2026: member

Good changes. Using an OP_RETURN output is a neat idea. Having the commitment in the coinbase field also seems more familiar.
stevenroose commented at 5:10 PM on November 5, 2026: contributor

LGTM

BIP95: Change proposed Testnet 5 genesis block structure

This moves the anti-pre-mine commitment back into `pszTimestamp` and
keeps it as ASCII text, rather than using it as a fake public key. The
genesis transaction's output is replaced by an `OP_RETURN` to avoid
confusion over spendability. Using the `OP_RETURN` output and `nVersion`
rolling for additional nonce bits are explicitly allowed.

69ec29d518

ajtowns force-pushed on Nov 5, 2026
murchandamus commented at 5:24 PM on November 5, 2026: member

LGTM, waiting for owner sign-off.
polespinasa commented at 5:58 PM on November 5, 2026: member

Looks good to me ACK :)
jonatack commented at 6:14 PM on November 5, 2026: member

@fjahr any thoughts?
jonatack removed the label Pending acceptance on Nov 5, 2026
fjahr commented at 1:25 AM on November 6, 2026: contributor

ACK 69ec29d5184bdd2cdfd819804598e906ff631526

I would have liked some additional context in the description that this makes the mining of the Genesis block easier with a pool and that there is a concrete plan to do a fair, decentralized bootstrap but I guess the change already got enough attention and agreement without that info :)

We can add an acknowledgement when the actual Genesis block is added. It may also be interesting to add an informational section to the BIP how the fair bootstrap worked when it was successful, so it is documented for potential future interest. But not sure the BIP is the best place for it.
jonatack merged this on Nov 6, 2026
jonatack closed this on Nov 6, 2026
jonatack commented at 1:35 AM on November 6, 2026: member

Thanks @fjahr. Please open a follow-up for your suggestions, if you like.

Contributors

Labels

Proposed BIP modification