Topic: moore's law block size (Read 282 times)
I think this is a bad idea, there are many other problems without this one.
The two main reasons for restricting the max block size are bandwidth and storage. Especially bandwidth. Moore's Law only applies to the amount of transistors per square inch, and thus the speed of computers.
Moore's Law hasn't even continued working - it takes a bit longer than a year these days and it certainly wouldn't continue forever.
This would also be continuing from 1MB, but there's no guarantee that 1MB is exactly the correct number anyway. The exact limit is pretty arbitrary, so long as there's a limit which is low enough for running a node to be affordable (that's the main argument of people who prefer smaller blocks).
Moore's Law hasn't even continued working - it takes a bit longer than a year these days and it certainly wouldn't continue forever.
This would also be continuing from 1MB, but there's no guarantee that 1MB is exactly the correct number anyway. The exact limit is pretty arbitrary, so long as there's a limit which is low enough for running a node to be affordable (that's the main argument of people who prefer smaller blocks).
Why not hard fork it just once to add a rule that automatically doubles the block size every 75000 blocks? Block size should automatically keep up with Moore's Law.
Because Internet connection bandwidth is the bottleneck for blocksize, which has nothing to do with the Moore's Law.
Some time ago Luke Dashjr, one of Core's contributors, made a proposal to decrease blocksize to 300 kB and then have it slowly and automatically increase - you can read it here if you want.
But this kind of ideas is pretty bad because it's essentially betting on how technology will be improving in the future, and if you overestimate it you will risk damaging Bitcoin. If you underestimate it, than community might decide to fork from this protocol, just like it would fork without it. And anyway no realistic blocksize would allow on-chain scaling, so it's better to keep the main network stable and secure and build second level solutions.
the cost per gb of storage space halves every 12-18 months so this will have ZERO effect on the decentralization of the network. It will be no more costly to run a full node in 2030 than it is now.
It will affect decentralization when Block sizes increase and the Blockchain bloats with unnecessary data. In the end, only a
few people will be able to afford to run full nodes and this in turn will harm decentralization of the network. Side-channels will
reduce on-chain crap tx's and this will reduce the "bloating" of the Blockchain, making it more affordable for people to run
full nodes and improving the decentralization of the network.
few people will be able to afford to run full nodes and this in turn will harm decentralization of the network. Side-channels will
reduce on-chain crap tx's and this will reduce the "bloating" of the Blockchain, making it more affordable for people to run
full nodes and improving the decentralization of the network.
Why not hard fork it just once to add a rule that automatically doubles the block size every 75000 blocks? Block size should automatically keep up with Moore's Law.
Moore's law is every 18 to 24 month doubled the block size. But this is not a same concept. Exponential growth is happen in block size. But we don't predict the block size. Every year and every month it will change the block size chart. In future it should raise up or fall down. So we don't compare to Moore's law 
1) Stupid idea.
2) Moore's law is not a law, but an observation.
3) The limitation of computing isn't the only problematic factor.
4) Close this stupid thread.
2) Moore's law is not a law, but an observation.
3) The limitation of computing isn't the only problematic factor.
4) Close this stupid thread.
Why not hard fork it just once to add a rule that automatically doubles the block size every 75000 blocks? Block size should automatically keep up with Moore's Law.
Jump to: