Quote from: iCEBREAKER on February 09, 2015, 10:20:30 AM

Quote from: amincd on February 09, 2015, 08:46:01 AM

Yes, a geometric increase is what I meant. Gavin's current proposal is for it to double every two years, meaning it increases 1.4X every year, which gives it a reasonably high chance of staying behind consumer bandwidth growth (and thus keeping the network highly decentralized).

So to answer your question, yes I would support something along the lines of what you're proposing. I think 5 MB would be much better than 2 MB though, because Bitcoin has a tendency to see sudden spurts in adoption, and so it'd be nice for it to have some room to grow quickly in the coming years.

Thanks for posting.

And I think one thing that often gets lost in the discussion on the hard limit, and that I can't stress enough:

There are other ways to stop bloat besides a hard limit in the protocol. A protocol limit is the most crude and inflexible way to counter spam/bloat. If a 40-50% increase per year ends up being faster than connection speed increases, it is very unlikely that other means will not be found to limit block sizes.

It's nice we agree on a geometric increase, but wouldn't it be great to have actual data on which to better determine the optimum initial increase and eventual rate of increase?

Before changing the max_blocksize constant, we should know what happens to the BTC function at (and over) the 100% limit of the tx/block variable.

we already know what will happen

transactions wont confirm and there will be a backlog ,it would render the currency unusable if we had to wait hours or days to get included in a block

localbitcoins would die ,retailers would stop accepting btc for coffee or food etc because they cant force a person to wait all day until the transaction gets a confirmation

miners would include the highest paid transactions so it would mean higher fees for everyone to use btc efficiently and the price would be all over the place if you accepted $1000 in btc for a laptop it might be worth $900 by the time the transaction gets from the customers wallet to your wallet to the exchanges wallet etc

this would be huge to giant retailers

bottom line is we dont need to wait until it breaks and have a mad panic to get it fixed

do you drive your car until the tyres explode or replace them before that happens ?