What are the capacity constraints, if any, of Bitcoin?

[Brendan Burgess]

I moved this from the "Problems with Bitcoin" thread to get a better focus.

Brendan

1) Have 1,000 times the number of transactions (resolved)

Hi Gus

Philip mentioned a maximum of 7 transactions per second compared to something like 7,000 per second using Visa.

Here is what Wikipedia says:

https://en.wikipedia.org/wiki/Bitcoin_scalability_problem

"The transaction processing capacity maximum is estimated between 3.3 and 7 transactions per second.[3] There are various proposed and activated solutions to address this issue."

Not sure why they need proposed solutions if they have "activated solutions" or if it has been resolved as you suggest.

Brendan

 

[Gus1970]

Hi Brendan, btc (bitcoin) are now using a second layer named lightning network that will resolve the problem.

Bch (bitcoin cash) are using the approach proposed initially by satoshi to make the blocks larger to resolve the problem

Bitcoin does not have the problem today but the engineers are future proofing it

Bitcoin won’t need 7,000 transaction per second for a good while and if you have appreciation for lean and lean startup you will know that priorities depend on customer needs not on things that might or might not happen in a few years.

HTH

Gus

 

[Gus1970]

Also brendan, i know things from building and doing them myself, not from wikipedia

 

[Leo]

btc (bitcoin) are now using a second layer named lightning network that will resolve the problem.

That solution involves addressing the bitcoin scaling problem by moving transactions off the blockchain. Bitcoin itself is still limited to a theoretical max of 7 transactions per second. The lightning network has other fundamental issues discusses elsewhere that mean it is a long way from the holy grail in resolving bitcoin's issues. Even if LN were to take off, the seeding transactions required to fund channels and retrieve funding from them would again hit that 7tps limitation.

3) Give people privacy (this is outrageous, banks lose their customers data all the time and in bitcoin only with a private key you see data. Privacy perfected)

They lose it all the time? Really? Bitcoin privacy hasn't been true for quite some time. It's at least two years since convictions have started to be secured based on evidence obtained from the blockchain identifying individuals and linking them to illicit transactions. Researchers have also demonstrated how to out Tor users based on their bitcoin transactions, exposing their full transaction history for the world to see.

5-6) block and revert ( this is exactly why bitcoin exists so that banks and governments cannot block, cannot freeze, cannot revert anything that is of a citizen’s possession)

Many criminals previously thought holding assets in bitcoin protected them from the authorities' grasp, but the multiple sales of seized coins proves that it not the case.

11) reduce price volatility (not sure i want that, for now)

If it's to be seen as anything other than purely a means of speculation, encouraging those who think they can outwit others in terms of timing to make a gain, that needs to be resolved.

 

[Gus1970]

Bitcoin itself is still limited to a theoretical max of 7 transactions per second.

WRONG: the limit can be changed by changing the block size. 1MB = 7 transactions; 2MB = 14 transactions etc. Theoretical what?

They lose it all the time? Really? Bitcoin privacy hasn't been true for quite some time. It's at least two years since convictions have started to be secured based on evidence obtained from the blockchain identifying individuals and linking them to illicit transactions. Researchers have also demonstrated how to out Tor users based on their bitcoin transactions, exposing their full transaction history for the world to see.

You confuse anonymity with privacy, we were talking about privacy.Bitcoin is private not anonymous, bitcoin never claimed to be anonymous

Many criminals previously thought holding assets in bitcoin protected them from the authorities' grasp, but the multiple sales of seized coins proves that it not the case.

Uhm, go find my bitcoin and seize them. If you find them they are yours

If it's to be seen as anything other than purely a means of speculation, encouraging those who think they can outwit others in terms of timing to make a gain, that needs to be resolved.

Is it a problem? For whom? How is the volatility affecting you? If you come up with a strong enough case I can make a pull request on your behalf

 

[Leo]

WRONG: the limit can be changed by changing the block size. 1MB = 7 transactions; 2MB = 14 transactions etc. Theoretical what?

Wow! A whole 14tps! An amazing 0.002% of what Visa alone can process, that certainly is progress.

You confuse anonymity with privacy, we were talking about privacy.Bitcoin is private not anonymous, bitcoin never claimed to be anonymous

If a user's complete transaction history can be laid bare for the world to see, how is that private?

Uhm, go find my bitcoin and seize them. If you find them they are yours

Well, I won't but I'm not in law enforcement, but let's not pretend it's impossible or not happening regularly as your point suggested

Is it a problem? For whom? How is the volatility affecting you? If you come up with a strong enough case I can make a pull request on your behalf

It doesn't affect me in any way as I don't see bitcoin adding any value to me in terms of a means of payment or store of value for as long as that volatility continues.

 

[Firefly]

Bitcoin itself is still limited to a theoretical max of 7 transactions per second.

If that read "Bitcoin itself is still limited to a theoretical max of 7 thousand transactions per second." I would still be concerned about its ability to scale. 7 transactions per second reminds me of computing throughput and memory sizes during the Cold War!!

 

[Gus1970]

If that read "Bitcoin itself is still limited to a theoretical max of 7 thousand transactions per second." I would still be concerned about its ability to scale. 7 transactions per second reminds me of computing throughput and memory sizes during the Cold War!!

In fact it is not true, but you can choose to believe the reality you want to live it, I hope you enjoy it.

 

[tecate]

7 transactions per second reminds me of computing throughput and memory sizes during the Cold War!!

Yes, 7tx/sec doesn't cut the mustard but that's where Lightning Network comes in. Leo maintains LN is technically flawed. Perhaps it is - I keep an open mind. Would welcome more comprehensive discussion on that as I don't understand how it falls down.
As Gus1970 points out, block sizes can be increased to facilitate greater tx volumes but my understanding is that there is a downside to this approach also.
@Protocol: LN brings tx costs down to next to nothing. There's another up n coming crypto with NO tx costs.
On volatility, that will take a much longer timeframe to overcome.

 

[Firefly]

In fact it is not true, but you can choose to believe the reality you want to live it, I hope you enjoy it.

Hi Gus1970,

I never said it was true, I said "If"...

However, as you seem to be better versed than I am, can you please confirm the max transactions per second and provide something to back that up such as a real-world test/benchmark?

tpc.org have well known benchmarks for these types of tests and the link below lists the transactions per minute achieved by the top database vendors:

http://www.tpc.org/tpcc/results/tpcc_results.asp?print=false&orderby=tpm&sortby=desc

Both Oracle and IBM have multiple configurations reaching 3m+ transactions per minute or 500,000 per second...

 

[tecate]

you please confirm the max transactions per second and provide something to back that up such as a real-world test/benchmark?

Can't help you with benchmarking (perhaps that's available but not in a position to check right now). However, check the scalability section here - https://lightning.network
- they give an indication of what's possible with LN.

 

[Gus1970]

Hi Gus1970,

I never said it was true, I said "If"...

However, as you seem to be better versed than I am, can you please confirm the max transactions per second and provide something to back that up such as a real-world test/benchmark?

tpc.org have well known benchmarks for these types of tests and the link below lists the transactions per minute achieved by the top database vendors:

http://www.tpc.org/tpcc/results/tpcc_results.asp?print=false&orderby=tpm&sortby=desc

Both Oracle and IBM have multiple configurations reaching 3m+ transactions per minute or 500,000 per second...

There is NO MAX transaction limit, the max transaction depends on parameters that can be adjusted. Bitcoin Unlimited node for example doesn't set a limit to the block size.

BCH on the 15th of May will have 32MB blocks that by inference will be 224 tps. There is no need for more transactions now, NONE. When they will be needed we will make changes. https://www.reddit.com/r/btc/commen...utm_medium=hot&utm_source=reddit&utm_name=btc


The data is out there, everywhere, I am not your benchmark boy, you can decide not to believe me and I am fin e with it.
https://bitcoin.org/en/ BTC
https://www.bitcoin.com/ BCH

 

[Gus1970]

I am curious to get your opinion on why so many people think these things about Bitcoin? Given that Bitcoin is around for 7 years now surely if it was all fine and dandy people would know by now?

The fact that you are commenting on bitcoin topics, but don't know when bitcoin was created I think answers your questions.

 

[Firefly]

The fact that you are commenting on bitcoin topics, but don't know when bitcoin was created I think answers your questions.

Apologies. I was referencing your post above (93). Still, the question stands....why do so many people think these things about Bitcoin? If it was all fine and dandy wouldn't people know by now?

 

[Firefly]

There is NO MAX transaction limit
There is no need for more transactions now, NONE. When they will be needed we will make changes.

But how can you expect Bitcoin to be used as a mainstream currency without the need for a higher transaction count than 224TPS? Amazon alone are processing upto 300 transactions per second and it is estimated than online purchases still only account for 9% of all transactions....

Finally, I am asking the questions out of interest so no need for CAPs

 

[Leo]

Would welcome more comprehensive discussion on that as I don't understand how it falls down.

I've only read a few articles on the LN design, but my main concerns would be

• Two bitcoin transactions required to set up a payment channel between two parties and further one or two to bring funds back to bitcoin if the users wish to cash out. Further funding would require more transactions. So a large scale adoption on LN would challenge the 7tps, or any similarly low limit. Of course, if there are many pre-existing channels available, then a path between the two parties may be discoverable without the need for a new channel to be set up and verified, but the challenge is getting adoption to that point. What's the benefit for early adopters that outweigh the additional hoops and risk to get us there? How many channels will each user require for the network to find routes between all users? How much funding needs to remain committed by each user to LN to keep it viable?

• The reliance on people leaving these channels funded in order for payment paths to be established. Many on here quite sensibly endorse keeping any bitcoin or similar holdings offline to ensure they are secure, the same approach to LN would mean it is doomed to failure. So you want to send $100 to another user, you need to have $100 or more available in one of your existing channels between you and another party from where the network can establish a route to the payee. Say that takes 20 hops, all 20 the nodes along that route must have at least $100 available for the transaction to succeed.

• The potential for fraud (users prematurely closing channels and pocketing the funds therein) and the limited time the impacted user has to do anything about it. I believe the current limit for action is 24 hours after which point there is nothing that can be done to retrieve funds. There are some suggestions for the establishment of a Watchtower network to identify such activity and ban misbehaving nodes from the network to prevent them from trying the same trick again, but the suggestion is that infrastructure be funded through increased transaction fees.

Maybe you or someone else here knows, but how long are the time-locks in place while payments are in the transmission process over multiple hops? Say I want to transfer a payment to another party, but the funds in my channel have already been committed to complete a route between two other parties? I can't use them until that time-lock is cleared whenever the payee confirms that they have received the payment or the transaction fails and is cancelled. Is there a time limit within which a user must acknowledge the transaction?

I'm not suggesting these issues are insurmountable, just that my reading to date suggests they're quite a way from finding solutions.

 

[Gus1970]

But how can you expect Bitcoin to be used as a mainstream currency without the need for a higher transaction count than 224TPS? Amazon alone are processing upto 300 transactions per second and it is estimated than online purchases still only account for 9% of all transactions....

Finally, I am asking the questions out of interest so no need for CAPs

As i said 224 tps is the limit for 32mb. I also said that tests on GB block size have been rub successfully, hence the scalability problem will be resolved when the system will need to scale. Is it clearer now? Any other questions?

 

[Gus1970]

Finally, I am asking the questions out of interest so no need for CAPs

I am not sure what the acronym CAP stands for in this sentence. If you mean "cap" as maximum size of the blocks, then for clarity you can look at the work done by Andrew Stone on Bitcoin Unlimited, if other meaning please explain.

 

[tecate]

Firstly, thanks for taking the time for such a comprehensive response on Lightning Network. Much appreciated.

Maybe you or someone else here knows, but how long are the time-locks in place while payments are in the transmission process over multiple hops? Say I want to transfer a payment to another party, but the funds in my channel have already been committed to complete a route between two other parties? I can't use them until that time-lock is cleared whenever the payee confirms that they have received the payment or the transaction fails and is cancelled. Is there a time limit within which a user must acknowledge the transaction?

Regrettably, I don't. In the first instance, I'm not a techie - have an interest and skirt around the edges but struggle to keep up. Secondly, despite having a year free, I don't have enough time to check stuff out as thoroughly as I'd like. I intend to spend some time on studying crypto & conventional trading in the coming weeks (thank you BTC!).
It would be great if some of the others here could chime in on this if they have the knowledge.

I'm not suggesting these issues are insurmountable, just that my reading to date suggests they're quite a way from finding solutions.

BTC has faced a mountain of problems over the guts of 10 years. I hope it can overcome this too as scalability is essential. Folks on here will know me as a crypto enthusiast but I'm not so blind as not to recognise an issue highlighted by a sceptic.

 

[Gus1970]

Maybe you or someone else here knows

It seems like a non trivial question, I am not a supporter of lightning network and don't know enough to give you an answer that doesn't include my assumptions, but I can point you to an extremely rich documentation set where you can probably find it by yourself by either reading or asking the question directly.

https://www.reddit.com/r/Bitcoin/comments/7pwna9/lightning_network_megathread/

Thanks,

Gus