CHIP-2025-03 Faster Blocks: gathering statements of support (and dissent)

Approve!

Name or project: The BCH Podcast, BCH Bullet. I will also enquire about Selene & BLISS, but can’t speak for those unilaterally. ConsensusCash has to stay neutral (as a consensus oracle).

Statement: 1 minute blocks is an essential upgrade for BCH. Many users, including myself, constantly suffer a pain point around confirmation requiring services with unexpected & barely tolerable waiting times, need to follow up with services where confirmations don’t arrive in the window they monitor for a transaction and more. Without it, we are well behind the industry standard. 1 minute blocks is a good technical improvement, a good user improvement & it once again emphasizes BCH as committed to making the best money - the best & most usable cryptocurrency - possible.

I will write up a more indepth blog post & approval statement with my reasoning soon.

4 Likes

Disapprove (for now). Rationale:

  • code is not ready and likely to have new and exciting bugs. We can throw AI at it to catch any potential bugs but that has risks too ai is not perfect — moreover bchn has been busy with other assorted tasks.
  • it feels fundamentally wrong to take a chain with like ~200 txns per block and .025% hash divide it by 10x — ~20tx per block and far less secure blocks at 0.0025% hash. Hmm.
3 Likes

Stance: NEUTRAL

As said previously, I always keep my word. Even if I don’t use the word “promise”

3 Likes

Also on a second thought:

Would it be possible to separate the tick mechanism from the change to blocktime and make it two separate changes?

I mean it’s such a great invention and a great idea, we could (for example) have ticks early and then decide on the precise blocktime later.

Just a suggestion.

Yes it’s possible.

But it’s kinda impractical to split them up, because if we aren’t going to change the block time, why even bother creating a new ticks system? Why merge it unless we are already confident that we will change the blocktime at some point. And if we are going to change the block time, why not set it to the currently optimal value (1 minute)?

3 Likes

What would be required to turn this into approve?

Understood, but without demonstrated support where will the motivation to make the code ready come from? It would suck to have social support and then miss the window just because code is not ready. Is it humanely possible to get at least a preliminary code-build ready to fork chipnet, and maybe be slightly late with final binary signed release (December)?

You’re the 2nd to bring this up today. It’s a technical concern, let’s discuss it here: CHIP-2025-03 Faster Blocks for Bitcoin Cash - #179 by bitcoincashautist

5 Likes

Sure, I even have a test branch somewhere that just adds ticks to RPCs, but what’s the point? I fear it’d only be used as excuse to postpone the actual upgrade, so could end up being self-defeating. We ship ticks, then what? Nobody bothers to start using them because there’s no impending activation that would prompt them, then that is indefinitely used as blocker to activation: “stakeholders are not ready yet because still not using ticks.”

3 Likes

Yeah, indeed this is a good point.

Stakeholders could just not adopt ticks and delay all of it indefinitely.

1 Like

Approve
Name or project: BCH-1
Statement: BitcoinCash Autist has done extensive work on mapping all the benefits and fail modes of this project. I believe the architecture change is justified, with the pros far outweighing the cons. The main benefit is developers get an option like a slider to choose the level of security they want vs defaulting to 10 minutes. As he mapped, variance also adds considerable instability to Bitcoin Cash’s UX when apps do not accept 0-conf.

For everything that already accepts 0-conf , nothing changes, it remains a viable path for payments, but wider crypto to be sold 0-conf is a hard sell that BCH does not have outreach resources, Its simply easier for crypto apps to accept the common standard of confirmations as security. We cannot predict whether exchanges and so on will raise or lower confirmations, but we know for sure they cannot lower them under 10 minutes under the current design.

For DEFI and cross-chain applications, a 1 minute block time isn’t even fast compared to other blockchains, but it makes BCH at the range where it’s tolerable, vs 10-30 min waiting for a block. Every user that thinks BCH is too slow, is a user we lose, whilst the proposal does not ensure UX improves, it also has the double effect of apps that realize lowering the confirmation low and being safe, gets them to look at 0-conf eventually.

Optionality is the real win of this proposal, and the ticks design means adjustment to the blocktime in the future will be a non-issue for Bitcoin Cash in the case there is a clear argument for raising it back again.

The trade-offs are well documented in Bitcoin Cash Autists own work, these should not be ignored, for the apps again, I see far more enabled than is disabled.

Furthermore, my support of this is outside of the purview of whether this is engineering work that is achievable, that is not my expertise. I am an apps layers guy.

4 Likes

Disapprove.

If it aint broke, don’t fix it.

3 Likes

It wouldn’t necessarily turn it into approve right away but – it would certainly help to get the code over the finish line ASAP w/ more testing, etc. Maybe have Dagur or someone that has advanced AI skillz throwing their AI at it for review only wouldn’t hurt. That would improve the situation.

BCHN is going to do a release very soon for some updates that were planned for a while now, and then we can focus on ye faster block tyme.

4 Likes

I love you so much right now. :hearts:

2 Likes

1 minute blocks is extremely conservative in the modern era by all metrics.

If the code isn’t quite ready yet we can easily make it ready, we have the resources.

I, Luke Pryor, APPROVE.

4 Likes

The CHIP’s thesis is that 10-minute blocks are broke. 10-minute blocks are (for now) survivable for BTC because it won the “digital gold” market position. We don’t have the same luxury. Excluding BTC, we have the slowest blocks of any top coin, and of the UTXO-PoW set too: LTC 150s, XMR 120s, ZEC 75s, DOGE 60s. We can’t afford to pretend this doesn’t matter for user acquisition and retention, and I was hoping you’d see it, too. Relevant discussion between Brady and me happened here: 1, 2, 3, 4, 5.

Long shot, but what kind of argument could change your mind?

PS: we’re still agile enough to make this change now, but later we may not be. There’s a possible timeline where we get stuck with 10-minute blocks forever. Using this window now (and installing ticks alongside) prepares everyone for doing it again at much smaller cost: the next time, it’s basically one line of code. That means we’d actually get to benefit from future block propagation improvements instead of just reading about them, and one day we could step down again toward Nervos-style ~10 seconds.

PPS: The day before I wrote the above ZCash voted to go ahead with their 2nd target block time adjustment, which should be telling us something about activation costs and risks concerns. CHIP-2025-03 Faster Blocks: gathering statements of support (and dissent) - #23 by bitcoincashautist

6 Likes

Approve, but with caution
Name or project: mainnet_pat, projects I own or maintain
Statement: the idea is good, improvement of defi experience and block inclusion time variance is a must. But I am not sure about the proposed implementation. The ticks approach seems to bring more complexity.

3 Likes

Disapprove.

In balance doesn’t seem to be a good idea.

3 Likes

Noted. Your most recent substantive objection was the BIP-37/SPV one, and I think @nyusternie answered it point by point. I’ll log your Disapprove as stated.

2 Likes

Name/Project: 2qx, vox.cash, unspent.cash, futurebitcoin.cash, awesomebitocin.cash, CatDex, photons, SAFAs, etc.

Statement: Approve of the concept, if implementation is ready in time.

… with notes about future height nomenclature

And a more elaborate explanation of time systems and the advantages of height based contract logic.

3 Likes

FWIW 2 days ago ZCash voted to go ahead with the 2nd block time change, now from 75s to 25s. And they don’t even have ticks to make the 2nd change easier!

Their main concern was orphan rates, and they’re more tolerant than us (5% instead of our 2%): Zcash Block Time Reduction Appears Safe for NU7 w/ Zebra Only Devnet - Zebra - Zcash Community Forum

I did not see any concerns over activation costs / ecosystem migration. If they had a traumatic experience when moving from 150s to 75s back in 2019 I’d have expected some word about it now.

3 Likes

Thanks, and I’ll log these notes too. Some thoughts about the 10x idea: Wallet Default nLockTime Practices - #15 by bitcoincashautist