Bitcoin Forum

There seems to be a lot of technical problems with increasing the block size in Bitcoin, so I was wondering, could we recalibrate the formula for difficulty so that blocks would be generated faster? This would increase tx throughput while keeping blocks the same size.

This could be done one of two ways.

(1) Keep the block reward the same. Disadvantage of course is that coins would be mined faster reaching the 21 million point much sooner. Advantage would be compatible with previous versions that could mine on this chain. Personally against this version of the idea as I think it would cause a short term drop in prices with decreased future security.

(2) Decrease the mining reward by an inverse proportion to the decrease in difficulty. So if we re-calibrated the mining to be twice as fast, we would cut the coin reward to one half. Currently, this would 6.25 Bitcoins generated every 5 minutes if we doubled the speed.

We could even implement this slowly by not reducing difficulty downwards and keep it the same till blocks are being generated every 5 minutes (or the desired rate).

Wouldn't this have the same problem as increasing the blocksize?
That is, nodes that aren't updated wouldn't accept the new blocks as they'd see your difficulty to be too low and you'd fork the main chain?

Yes, unupdated nodes would reject.

However if 5 times as many blocks are being created, even with only 30% of the hash power of the original main chain, we would still be producing blocks at 150% of that chain. Our chain would be longer.

That's not entirely correct. Bitcoin Core clients judges which chain to follow using the block height and with respect to the difficulty that it was generated at. The client uses a formula and weighs both of the chains to see which one has the higher weight and will follow that. The chain that required the most work will be valid.

By reducing the block time, you would also be increasing the orphan rate and in turn possibly reducing the security of the network. This idea has been thought before and it doesn't work.

It's something that I've discussed several times, and I've even suggested halving the blocksize to improve efficiency. It seems that there are problems with orphaned blocks if you do this, and I'm not sure that much effort is being put into solving this problem.

This would be a hard fork change.  Therefore, it would have many of the same "technical problems" as increasing the maximum allowed block size.


It would, but it would also increase the rate at which the blockchain grows in size.


Correct.


Incorrect.  Previous versions would not recognize any of these blocks as "valid".  It would reject them all and would fork the blockchain. Previous versions would be incompatible and unable to mine on this chain.


It doesn't matter which chain is longer.  Nodes only accept the longest "VALID" chain.  An invalid chain can be 1,000,000% longer, and it would still be rejected.  Since un-updated nodes won't recognize these faster blocks as being valid, it would create an altcoin fork.