[lawn]

Sigh. Increasing the block size to 2MB nor 8MB for that matter will not make a centralized coin.

Segwit (besides being a crappy implementation) is by far not enough scaling.

That you blame Jihan and Roger is typical stupid propaganda which doesn't belong here.

[etr-strike]

According to actual research here: http://www.tik.ee.ethz.ch/file/49318d3f56c1d525aabf7fda78b23...

Each additional kb results in an additional 80ms of delay until the majority learns about a block. Increasing to 8mb means you’re adding a 9min delay in propagation. That is pretty much guaranteed centralization.

Perhaps you have some research you can point me to that demonstrates otherwise?

[lawn]

Which does not take into account xthin/compact blocks which cuts down propagation time a lot. In the future graphene will further reduce the time necessary.

How you can conclude a 9min propagation delay for 8MB blocks is beyond me as miners are basically directly connected to each other. This is the only thing that matters decentralization wise.

Bitcoin Unlimited are successfully testing 1GB blocks: https://news.bitcoin.com/bitcoin-unlimited-reveals-gigablock...

[etr-strike]

Your link is a news article and not actual research. The details of their specific test are barely provided. Please provide actual research as I have provided you.

[CryptoPunk]

This overcautious, "let's test everything for 10 years before trying it" approach is exactly what Bitcoin doesn't need when it has 0.001% of world transaction throughput.

[Rmilb]

Bitcoin is the conservative approach to any change that decreases the keystone of decentralization that allows it to survive. The market is deciding what it values.

[CryptoPunk]

The market is reacting to the fact that Bitcoin is in the public consciousness, and that is a result of years of advocacy done for Bitcoin, when its advocates (e.g. Roger Ver, Coinbase, PayPal, and all of the other signatories of the SegWit2x agreement) still believed that it would scale with large blocks and thousands of transactions on-chain per second.

[martindevans]

Is there any data available on how long BCH blocks take to propagate? It's testing big blocks out in the real world so we can see how much delay that introduces and whether that leads to a significant increase in centralisation.

[etr-strike]

The BCH blocks are all very small. Go look at their blockchain stats. Most blocks are 10kb in size. The network is barely used. Once they start actually mining big blocks we’ll get some real data. I think it’ll be very interesting to see, tbh.

[martindevans]

Ah that's a shame it would have been really interesting to see how that worked in reality.

I wonder if it could be measured the _other_ way around - i.e. are the sustained tiny blocks on BCH _increasing_ decentralisation compared to BTC right now? I can't find any good information on block propagation time though :(