Topic: About the maximum amount of tx per second. (Read 1273 times)

I'd expect it to be 0.10.x

Not directly related, but I remember reading that "bitcoin-qt" will drop the "beta" description after version 1.0.0.
So, will the next major update be 1.0.0 or 0.10.0 ?

This would not be a difficulty change.

Over time technology will make it so it would not be difficult for nodes to handle more TXs per second. Hopefully there will also be an incentive to run a full node (which would increase the incentives to create such technology).

As of now this is not an issue as there are far fewer TXs per second to worry about this.

The first problem is to get people to use Bitcoin enough so that the network has more transactions.

So I learned a valuable lesson, send high amounts not low amounts not get raped by fees.

Yes it is.  My guess is, it will continue to be "beta" for another 5 to 10 years.


This is very unlikely.  Far more likely is that the maximum size of the block will be increased (it is currently limited to 1 megabyte which is what creates the 10 tx per second limit).

Note that even at a limit of 10 tx per second, that still works out to 6,000 tx per block.  If the standard fee that miners (and mining pools) accept is allowed to grow to $1 per transaction, then bitcoin could still be used very cheaply for high value transactions.  It would become too expensive for low value transactions, but could still function.

This would provide $6,000 worth of bitcoins in every block, or $864,000 worth of bitcoins per day, in revenue for the bitcoin mining community.

I guess the question is, "How much revenue do you think needs to be provided to maintain security?"

We can institute changes to the protocol in the future that can allow for increases in the number of transactions per second.

We can make it so Bitcoin-QT comes with a large portion of the blockchain preloaded. This would reduce a large amount of the issues associated with a larger number of transactions per second.

We could also provide a small incentive to run a full node on the network.


Along with the above solution the price of bitcoin will likely increase, making the small amount of tx fees worth more in terms of US dollars.

I dont even... words

https://what-if.xkcd.com/31/

-- better now.

Well this might be a growing market. The blockchain is growing and DVDs can be produced cheap. Esp. in regions where you have a slow internet connection you might be faster waiting 2-3 days for the DVD.

Someone has already started selling blockchain DVD. 
https://squareup.com/market/setupdisc/bitcoin-blockchain-update-dvd-bootstrap-dat-on-disc

You didnt actually buy the distribution but the manual that came with it. Same as today you dont buy the OS but the support for it.

That's an interesting idea.
Like they used to sell GNU/Linux distributions once.

0. Running a full node on bitcoin-qt is not the best option for everyone, and there are alternatives like SPVs.
1. The technology improves quickly. The age of floppy disk and 56K modem isn't that long ago.
2. You could use the bootstrap torrent to speed up the initial blockchain download.
3. If bitcoin becomes more and more popular, I wouldn't be surprised to see someone selling blockchain discs or USBs.

https://blockr.io/charts
The average number of blocks a day is about 160, and the average block size is about 250 KB.
So, the blockchain is growing at about 40MB a day or 14.5 GB a year.

And of course, the growth rate could even be faster if bitcoin adoption improves.

For now it is a small problem indeed, but in 10 years or so it probably won't be, since hard drives and internet connections will most likely be better.
And one can always use a light wallet, or blockchain.info if it is too much of a problem.

The size of Blockchain can be a problem for some people, the blockchain is already very big, it takes time for first timers to download and verify the blocks. The size of blockchain is already 20GB, if the time is reduced, the blockchain may grow to 40GB within 3 years.

Indeed, it is important to remember Bitcoin is still a beta. Perhaps the time between blocks will be reduced, since the size of the blockchain probably won't be a problem in 10 years or so.

And this is a significant part of why bitcoin is still considered "experimental".  There is no way to know if the concepts coded into bitcoin's decentralized consensus and maximum quantity will actually stand the test of time other than to test them out and see.

There are a lot of possibilities.  The blocksize could increase significantly allowing for many more transactions per block (meaning the total block reward from transaction fees could be larger).  The blockchain may become a clearinghouse of sorts where on-chain transactions may become the domain only of significantly large transactions that can afford a substantially large transaction fee.  Other developments that we can't even predict today could occur.  It sure will be interesting to be a part of this and see how it all plays out.

Thanks for taking the time to write that.

However, that model works only if Bitcoins value was to go up proportionally to the block halvings. If it does not, I fear that regular miners will quit the network, which will lead to centralisation of mining worse than what we are seeing right now (because ASIC manufacturers could afford to mine with the machine they built themselves). Or leave a network with a low hashrate proportionally to Bitcoin's market cap, which would make an attack quite profitable.

And sure the last satoshi will only be mined in 2140, but practically, in only 21 years the block reward will have fallen to less than 1 BTC. The problem does not seem so far away.


I doubt this will be true; transaction fees will have to decrease to accomodate a price increase, just like it happened a few months ago. Or else, services will do offchain transaction which won't contribute fees at all.

This has probably been discussed in other threads already; my apologies.

Off blockchain transactions don't matter.

Every cash transaction using local currency today is an "off-chain" transaction.  Every credit card and debit card transaction today is an "off-chain" transaction.  Every barter or commodities trade today is an "off-chain" transaction.  There are billions of transactions that are all "off-chain" and bitcoin doesn't care.

If mining isn't profitible, then some miners will stop mining.  When the total network hash power is reduced, blocks will start taking more than 10 minutes.  When the average block time for an entire difficulty period is more than 10 minutes, the difficulty will adjust downward. A reduced difficulty means those that continue mining become more profitable again.  The mining (or "transaction processing") system is self regulating.

The last fraction of a bitcoin won't be mined until approximately the year 2140.  Most all of us will be long since dead.  So really, I don't care too much how the people of the future decide to handle the situation.  They'll know it's coming and have plenty of time to decide.

Keep in mind that the block subsidy is cut in half approximately every 4 years (50 BTC, 25 BTC, 12.5 BTC, 6.25 BTC, etc). Meanwhile, as bitcoin becomes more popular, the total value of all the transaction fees will increase with the increased exchange rate of the bitcoins.  Eventually (far in the future), the total value of the transaction fees will become larger than the total value of the block subsidy.  As the block subsidy continues to be cut in half, it will become an every smaller percentage of the total block reward.

Yes, the bitcoin price will be higher since no new coins will be mined. The users of bitcoin could be more than what it is now. If miners include more transactions with the current minimum transaction fees. Even if alot of people choose not to mine, those who still mines profits more if they have low electrical fees and ASICs that have already ROI. I do not think most people would want bitcoin to be offchain, it compromises their identity and they need high trust on the operator.

Thanks for your answers.
However, you seemed to have missed my concern; will it still be profitable for miners to secure the network with only the transactions fees when most coins will have been mined? It still seems that it would be unprofitable; even if the transaction-limit can be changed, it would have to be by a lot to make it profitable. And of course the fee can change, but if it gets too high, Bitcoin will lose one of its main advantage over fiat, which would make it relatively irrelevent for commerce so it would bring down its price, and the problem would be worse.

Yes I understand that off-chain transactions do not pay fees; that's part of the problem. In a few years, when the vast majority of coins will have been mined, miners will have no reasons to stay with Bitcoin if most transactions happen off-chain and thus pay no fees; plus the problem mentionned above.

I may very well be wrong though, that's why I'm asking the question in the beginner section