Topic: Bigger blocks coming in release 0.11 - page 4. (Read 4732 times)

http://gavinandresen.svbtle.com/why-increasing-the-max-block-size-is-urgent
http://gavinandresen.ninja/time-to-roll-out-bigger-blocks

Very good move , Gavin is strong point of Bitcoin

I don't know enough/haven't been around long enough to know one way better than the other, but I can say that I'm glad something is happening and all the back and forth can stop.  It works or it doesn't, time to find out.

Agreed. This has been an issue for a while.

Here's a Q&A with Gavin and Mike, at some point they discuss the bigger block issue if you're interested.

CoinScrum: QA with Gavin Andresen and Mike Hearn https://www.youtube.com/watch?v=RIafZXRDH7w

Nope, bitcoin works fine now , but lately we are starting to hit the blocksize limit more often.

This is what Bitcoin needs to make it awesome-

https://bitcointalksearch.org/topic/lightning-network-another-proposal-to-make-bitcoin-scale-970822

I don't think i understand your position.
Trading activity has always passed through offchain transactions..

So, do we really need ripple or stellar to make bitcoin work?
Why?

Time to embrace Open Transactions, Ripple, Stellar or whatever for offchain transactions.

Edit:  I say go for it.  Go for the 20mb blocks.  I want to see what happens...  Just hedge your bets.

Increasing the limit won't create bigger blocks on average, but allow for it to happen if it is needed. If we stay at 3-4 tps average the change will have no effect outside of occasionally allowing a 2MB block through because of a large fundraiser through lighthouse or a temporary spike in transactions.

There will likely be a middle ground found because most of the other core developers are wary of 20MB being needed. Perhaps if we give some other projects a bit of time to be released like the lightning network we won't need a 20MB limit.

What is a serious problem already? The one who does not know what he/she is talking about is you ....

Blind following the blind going on!

Right now certain projects like Lighthouse decentralized crowdfunding are limited to 684 pledges (unusually low compared to kickstarter) and we can only process 4-7 transactions per second which isn't enough.

We are hitting the block limit more often these days and it is only going to get worse.

That and by the time blocks are actually 20mb each the price of storage will continue to be lower, just like the price per gigabyte was 1.24 in 2005 and in 2014 it was $.03 every 5 years it is ATLEAST been a third of the price and some years a large jump. 1995 $/gb was $1,120. In 2000 it was $11/gb in 2005 the $1.24 and in 2010 it was .09$/gb . Storage price shouldn't be an issue it is getting cheaper compared to the requirements needed.

BlockChain is  already almost 40 GB of size, but this change is inevitable. There is no problem with our current terabyte hard disks to storage it. And you dont have to run full node wallet like Bitcoin Core at all.
Electrum or MultiBit will do just fine.

Finally it was about time!

If 20MB block size comes with pruning feature, i'm OK for this. 
i want a counterparty ... 

Of course it isn't.  If you don't want to store the full chain, run an SPV client.  If you do: http://www.amazon.com/Seagate-Desktop-3-5-Inch-Internal-ST4000DM000.  Pruning is coming, so the storage "issue" will soon vanish completely.

Increasing the max block size does not mean that all blocks will increase in size - they'll increase when they need to.  Does anyone actually take the position that it is better to let transactions back up than to marginally increase the incremental chain size?

The real block size is up to the miners. The protocol (wired in the bitcoin reference implementation) can suggest a MAX_SIZE and miners can agree on that. However, they can mine blocks with size between [MIN_SIZE, MAX_SIZE] and that means they can include any number of transactions per block always they respect the agreed MAX_SIZE. Note that it is possible to mine blocks with zero transactions for example.

IMO there is an incentive to keep the block size small because it makes the space in them scarce, and we know what happen when something is scarce. Given the block size is 1MB, if your transaction requires 60kb you have to pay more fee than if your transaction is 800bytes. So, bigger blocks could reduce the fees but at the same time miner are allowed to introduce more transactions per block, what means more fees for them.

I really don't fully understand all the implications around this change because even when the current block MAX_SIZE is 1 MB, miners don't mine 1MB blocks. In fact, I read that one of the most common sizes is around 731kb. Clearly, for miners is not the same to broadcast to the network 1MB than 20MB because if I mine a 20MB block and, at the same time, other miner mines a 200Kb block, its block will spread faster than the mine one.

So, I don't see the incentive for miners now. What I mean is that currently they earn more with the 12.5 bitcoins that the protocol gives them that with the transactions fee and for that reason I don't know if more transactions per block (more fees) worth the risk.
      

I think that's right. It could be 5 or 10, but we don't want to have to do this every year.

If we see a mainstream adoption and a large rise in users no 5 or 10 may not be a enough and we would have to fork quickly. Think if for some reason Bitcoin blows up (maybe a stupid celebrity starts using bitcoin like Kayne) and we get 100k users in one day, with more jumping on the band wagon after and they all start trying it out, blocks would start to fill quickly.

but my question is why they choose 20mb as the next size? 5 or 10 weren't enough? did he tested those value?

Trading is going to happen regardless and transactions are going to increase if we want to keep bitcoin how it is today we will need to increase the block size or start expecting a long block queue for everyone. There is a limitation set at 20mb so we won't have to fork off it for a very long time and that gives us time to adapt. Blocks won't automatically be 20mb either just have the option to increase that high and only if we need it, if we have that many transactions and the same 1mb there would be a 20 block queue so you would wait 200 minutes for a single confirmation (over three hours!)