Bitcoin Forum

Is it true that Bitcoin is limited to 7 transactions a second which are recorded in 1 block every 10 minutes? Does that mean 1 block can only hold 7 x 60 x 10 = 4200 transactions? With the growth of Bitcoin usage this has to be a drastic limitation oversight by Satoshi? How can Bitcoin be scaled for a global audience?

Opinions.

Proof : http://statoshi.info/dashboard/db/transactions?from=1441785552673&to=1442573582938

It's not true.

Only if it was intended to be permanent.  Most people don't think that was the intention.  It's likely there will be a change in 2016 to alter it.  



The exact number of transactions you can squeeze in will depend on the size of each one.  Some transactions take up more space than others.

There are no fixed and absolute limits. The number of transactions per block can vary hugely. What you shall consider is that the Visa credit card network can handle hundreds of transactions per second while BTC can't.

now ... yes.
in the future ?




no, not with the moore law of storage device.  ;D





We Will Win, We're WORLD WIDE WEB of money.

Normally, I ignore people that beg for money on this site, but I'll answer your question this time.

The limit is the size of a block in bytes. The 7 tps is determined by taking the average size of a transaction and seeing how many can fit in a block. Whoever came up with the 7 tps number apparently assumed that the size of an average transaction is somewhere around 250 bytes. You can fit up to 4200 250-byte transactions in 1 MB block so that means that the limit is 7 tps. Sizes of transactions vary, so that actual limit (in terms of numbers of transactions) can also vary. It turns out that with multisig transactions, the average size has gotten a lot bigger and 2 - 4 tps is a more realistic limit.

The idea of a 1 MB limit was originally designed to prevent a particular attack. It is not necessarily a permanent limit.

Apologies, it's an old signature that I've been meaning to remove from when I first joined and didn't know how to get Bitcoin.

Thanks for insightful post. Clearly to compete with the likes of visa etc then block size needs to be increased to process more tps? If so, what's the disadvantage to larger block size? Slower confirmations? I'm surprised satoshi wrote nothing on upscaling this. He might have however I'm not aware of his writings.

Nobody disagrees that 7 tps is not enough. The disagreement is on the best way to achieve a higher transaction rate.

There have already been thousands of posts on this site and others on the pros and cons of increasing the block size. I don't see a reason to start a new thread.

No. It is not true. 7 transactions per second is an estimate that was calculated by making some assumptions about transaction size.  Those assumptions are unlikely to be true in most cases.  In reality the average number of transactions that bitcoin can confirm per second with the current block size is likely to be significantly less (closer to 3 transactions per second).


That depends on the transactions.  If the transactions are small enough, then the current block size can actually hold more than that.  Realistically with the typical transactions, blocks can hold about half of that.


People have been discussing this very topic for over 5 years.  There have been hundreds of threads about it on bitcointalk.  Several potential solutions have been proposed with various levels of support and technical merit. Perhaps take a little time to read through some posts on this forum before making assumptions and declarations


Oversight?  What makes you think that Satoshi didn't have any plans regarding this?  What makes you think that his plans are still the best possible solution 8 years later?  What makes you think that there is nobody else in the world after 8 years of study and use of the protocol that can come up with any better way to handle growth?


There are many people that don't have any desire for bitcoin to "compete with the likes of visa".  There are others that feel that bitcoin could compete, but that many transactions can be handled "off the blockchain".  Then there are others that feel that the protocol should be updated to handle more transactions per block (or more blocks per hour).  I really can't believe that I've chosen to get involved in yet another of these pointless discussions just to re-state all the things that have been stated so many times before.


Greater consolidation of mining power.  Increased orphan blocks.  Increased cost in running full nodes. It is assumed by many that scarcity of available block space creates an incentive for users to include transaction fees which will eventually be needed to maintain hash power for transaction confirmation.


Not sure why it would lead to slower confirmations.  Care to explain?


I'm not sure why anything he wrote about this matters, but since it seems to be important to you, here you go:



Well, if you are not aware of his writings, then why do you make assumptions about what he has (or hasn't) said?

Here's an idea.  Try reading the white paper.  Then try reading Satoshi's writings.  Then try searching for discussions about the things that you want to know about.  Then after you've done all that, see if your opinions about the technical details of bitcoin have changed.

White paper:
https://bitcoin.org/bitcoin.pdf

Satoshi's writings here at bitcoin talk:
https://bitcointalk.org/index.php?action=profile;u=3;sa=showPosts

When Satoshi lowered the max block size to 1MB network conditions were different; and, it was intended only as a temporary measure. Then he vanished with an estimated 1 million bitcoins. If he were around today, he would simply raise the limit to 8MB.

The reason this has not been done is mainly because some of the core developers have conflicts of interests involving outside paid projects.

Hi OP! Just popping in this thread quickly to say everything this guy just wrote is shameless lies.

Enjoy the rest of your discussion  :)

Nailed it.

Totally agree! 1MB was used in the experimental stage. With mainstream adoption in sight, the community should be getting ready and be more ambitious!

You are not him, you are not them. Please stop putting words in other people's mouths. Even if Satoshi wanted to raise the limit, he may not be able to, he doesn't have dictatorial powers.

I do agree that the 1MB was not intended to be permanent. It was set in place to prevent spam and DoS attacks.

http://labi.org/images/made/assets/media/documents/Bush_350_263.jpg


If you will look carefully at my statement, you will clearly see I didn't put words into anyone's mouth. I simply shared a perspective on the situation. :)

This is the most logical conclusion that one would draw if/after studying the entire contents of Satoshi's postings on this board. He would have raised the limit to 8MB (static) and everyone would have gone along merrily; because after all - he's Satoshi. Gavin's BIP101 idea is insane.

Now, we're stuck in a circle jerk with core developers that have excessive self-interest and over-blown egos.

I really don't understand why you are being so blunt and, frankly, borderline rude. Maybe it's just your style or you have issues? I'm just asking general questions on what I've read online in an article about the 7tps limit. I'm not constantly on here to follow all threads, nor just to catch up on what may or not have been posted. FFS. I'd really appreciate if you and others would simply either answer my questions directly if you so wish and know the answer, rather than post some diatribe 'this has been posted x times before read through the forum.'

Yours may do, but my day job doesn't permit me such liberty.

Ok, I'm looking carefully:


Correct


Correct


Correct


Wrong


Wrong



It's interesting, you know what they say about casual reading? That the points made at the end are the ones that people tend to internalise the most.

Take my advice; if you're going to make egregious mistakes in your assertions, don't put them all together at the end of your post. It makes it look like you're making true statements, only for the purpose of convincing people of the false statements. I'm sure you wouldn't want that.

It's not limited to 7 tx / sec. Depends on how much tx can the network handle, the 7 tx / sec is just an estimate.


Classic.  ;D

I have no idea how long you weren't active for in a past (I don't actually have time to check your profile and post history), but I guess that you have missed the whole block size increase debate that was raging during the past three months.

Much debate has been taking place and it's still not decided about this 1MB limit. Satoshi has designed this 1MB limit just as a temporary solution. We will see what will be decided.

Now when I think off, it's not bad that you have missed this whole debate. You have saved yourself a lot of nerves!

Where are the satoshi quotes where he said 1mb was temporal? what did he think about off chain transactions as a solution for this? I wonder what would he think about 8 mb raise, let alone 20..

yes but thats because visa has separate databases, one for each country.. so their total TPS is based on adding all their database(ledgers) together.. not based on the individual subsidiary systems

The Fedwire system which handles transactions between thousands of banks in whole US works at 4 tps. It's much more than enough right now if you don't fill the blockchain with coffee and candy transactions (they usually do not end up in blockchain anyway). In fact the larger the transaction, the greater the benefit of using blockchain

There are many quotes from which we can see that the limit was supposed to be temporary. Like I said, all of this was already talked about many, many times in the many threads over the last 3 months. Go and dig through them. I don't have time to do such a thing. But the quotes do exist from 2010 if my memory serves me well.

You took my post out of context. I was replying to someone who was speculating on what Satoshi would have done.

But I personally don't see any harm in speculating what Satoshi would do since he was after all the original creator. It doesn't mean his views on a particular subject would have been ideal, but in a way it helps us stay true to the spirit of his original vision.

Yes the number of transaction that can be included depends upon the transaction size not on the no. of transaction only. So what you said is absolutely incorrect. Bitcoin will be future, without limitation and complete ananomous.

I'm confident that after the second scaling conference in december, we'll be gravitating towards some sort of break in the deadlock.  Probably the 2-4-8 thing, but we'll see if anyone comes up with something better in the meantime.  I'd still like to see something a little more flexible and dynamic.



We will.  We're just waiting for the right moment.  We're going to take the merchants, miners and vast majority of users with us and leave you to run a hollow shell of a network all by yourselves.  A few MP fanboys acting in isolation does not constitute a consensus network.  You need the rest of the world to agree with your abhorrent views and they're not going to do that, so have fun trading your altcoin amongst your fellow converts in complete obscurity while the rest of the world happily ignores you.

Yeah, that seems like a pretty sober assessment of what's actually likely to happen here and now.


Eh?

How can you switch from agreeing to the most significant consensus position right now (2-4-8), and then still argue in favour of a minority blockchain fork?

You did the switch in the same post, am I missing something? Are we both crazy?

I'm saying if anyone tries to fight the tide and go against the 2-4-8 or whatever we end up agreeing on, they'll be in the minority.  If hdbuck wants to keep fighting and try to dismiss even moderate and reasonable increases, he's going to get left behind.  I bet a few of them desperately try to cling on to 1MB.  At least watching them marginalise themselves will be a fun spectacle, though.

brg444 reports from the scaling workshop that there are a whole plethora of angles to come at the transaction throughput problem from, without changing (aka 1MB). I'd be surprised if it's possible, but I'm not gonna write it off without hearing the details. That even Adam Back doesn't mind compromising with the 2-4-8 schedule suggests otherwise, but I'm gonna have to look at it all alot more closely from here on in

The problem I'm having there is that anything a few select individuals (https://bitcointalk.org/index.php?topic=1175596.msg12375776#msg12375776) think is a great idea is something I'm going to be inherently skeptical of because I don't trust them or their rather extremist views of what the future of Bitcoin should look like.  The more they express their preference towards an elitist and exclusive chain, the less I want to share a chain with them.  I want consensus, but not if exclusivity is the ultimate outcome.

Depends on what kind of exclusivity you're talking about. I can see how keeping low value transaction off the main chain would work, so that's one type of exclusivity.

But making the Bitcoin userbase exclusive? Bad idea. Some other crypto will just fill the gap.

Sounds like more that Bitcoin has 7 transactions per second, not limited to .

Unfortunately Bitcoin will never be able to scale as a world global payment network without compromising at a certain level, either by handling transactions into a private company (blockstream) whose will deal with transactions that don't pay an enough high fee, or by becoming totally centralized since the node will be way enough for the average joe to deal with, so I would rather have the blockstream solution. This is all better than what we have now.

It's not handing processing to Blockstream at all. All Blockstream are doing is writing software protocols, so the actual operating of all side chains or lightning hubs will be open to competition. We'll see how competitive the incentive structures are when the details get firmer.

Exactly.

To anyone with a hint of foresight it was always clear that Bitcoin would not support all of the world's financial infrastructure on its own blockchain. In order to provide us with a decentralized protocol layer Satoshi had to make considerable design compromises, often at the cost of efficiency and flexibility.

By enabling the creation of this quasi-trustless source of monetary value Satoshi allowed for an unprecedented, permissionless development of financial tools creating true market competition for more open-source, efficient & trust-minimizing platforms supported by Bitcoin's liquidity.

While the notion of "being your own bank" is a cool marketing gimmick it really isn't practical considering the effort one has to go through to make actual, proper use of Bitcoin. Yes, that involves running a full node. No, "bitcoinating" has nothing to do with web wallets, consumer-friendly "killer apps" & "merchant processors".

Consequently, if you decide to trust your "Bitcoin" business with such entities then it would seem to me other similar "users" should have no problem relying on second-level layers for their VISA payment network pipedreams.

Fortunately the trade-off really isn't as bad as I make it sound. The market-lead demand for these solutions will necessarily lead to valuable innovations in true open-source fashion. Lightning, side-chains are early demonstrations of this and more is to come.

I think one day VISA is going to integrate bitcoin payment for micro transactions. From their view, it is just another foreign currency that they need to integrate into their payment network, nothing more. And from bitcoin's point of view, it will save lots of work that VISA has done. VISA is currency independent, it works with all the currencies

Since it is bitcoin, they don't need to cooperate with banks, but bitcoin exchanges/online wallet providers. Since most of the user in each country usually only use domestic currency, the merchant need to install a special bitcoin POS terminal so that you can get a bitcoin quote to charge your VISA bitcoin cards (???, maybe people would still like to be quoted in local currency, then using bitcoin works like traveling abroad: you only get a quote in local currency anywhere around the world)

Well, actually I don't know how Visa manages its systems, but I've used my cards in many different countries, so the databases must be linked somehow. Besides, even when compared to one single country, BTC is substantially inferior to Visa. I'm sure that in the UK only, Visa handles way more transactions that the blockchain could handle.

use Litecoin to pay for the coffee and candy, or Dogecoin, it confirms faster too  ;D

No. Even if a transactions is broadcast, there is no guarantee that it will confirm and thus become irreversible. Sometimes transactions have some issues like low fees or dust outputs which prevent that transaction from being confirmed. With the upcoming Opt-In RBF feature, it may not be safe to take unconfirmed transactions if it has opted in to be able to be replaced by another transactions with a higher fee.

imagine it as if funds where debited out of your account instantly. but the recipient cant use them yet as they havnt been credited to him until its confirmed.
many people can still mess around and cause issues before its confirmed. and so for large purchases its not enough to rely on the publishing of a tx, but rather wait till its confirmed.

bitcoin is still not perfect in regards to the limbo between publishing a tx and having it confirmed.. however there are many idea's in the woodwork that are all aiming to sort that out.
that way payment systems can have 100% trust that once a transaction is published, its then locked in the waiting list to be confirmed. and any attempt to respend the same origin funds elsewhere will be ignored

http://imagizer.imageshack.us/a/img910/3716/9bPb6Y.gif





Proof : https://bitcointalk.org/index.php?topic=1096789.msg12391362#msg12391362

http://imagizer.imageshack.us/a/img903/9325/1Wp7av.png





Pay the network, Pay the fees ... and all will be good.
http://bmkbucket.objects.cdn.dream.io/wp-content/uploads/2014/12/free2playgif.gif

transactions relayed is a meaningless statistic, especially if it takes 3 hours until they are confirmed.

so even if there are 60,000 tx in a 10 minute timeline, does not mean bitcoin can handle 100tx/s

because those tx's are left in limbo for hours. because if only 4000 can be confirmed per 10 minutes. it will take them 60,000tx's, 15 blocks (2hours 30minutes) to confirm

which 60,000 over 2hours 30 minutes (9000seconds) is still 7tx/s

... and VISA/MASTERCARD take 8 days to confirm transaction ...  ::)
yes, i have read my 200 pages of bank note for electronic payment. :P

which is where visa has also failed in its calculations..
especially when visa is not one database, but infact several databases.. so each database which can take days to confirm means it actually has a very lower tx/s..

bitcoin is actually closer to visa tx/s than we think

I'm not sure I understand you both. Knightdk, correct me if I'm wrong but I think you may have meant to say that even if a legitimate transaction is broadcast, because of the possibility of a bad transaction within that same block there is no guarantee the block will confirm and thus become irreversible.  Franky1, I think you may have meant to say that many people can still mess around with their own transactions and cause issues within their blocks, thus causing them to be orphaned. But as I understand it, should there be a bad transaction, the revised block belonging to the first miner to solve the new hash will be the one used thereafter and the fork will continue from there, orphaning the block with a bad transaction or miner's error. But the legitimate transactions within that orphaned block will surely not destroyed or ignored -- even the slightest practical possibility of such an outcome would invalidate the BTC system completely and it would never have been tried.

What I think I am saying is that because an institution will have full control of its own BTC transactions and will ensure that each one obeys the rules, once they are published they are as good as confirmed for the institution's own purposes, and so can immediately be used for a range of internal reasons, including initiating or concluding smart-contracts.

Maximum no of transaction depend on the volume or size of transaction. You can see in blockchain no any block is over 1mb, that is the size limit for every block. You can find lots of development are going on to increase this size limit. I think satoshi haven't thought bitcoin will go this big.  8)

Your are wrong. First, a bad transaction would completely invalidate a block, and if that happens then so properly functioning bodes will not remove the UTXOs spent in that block from their UTXO set.

Secondly, it is completely possible for a legitimate (meaning valid and confirmable) transaction to not be confirmed. Miners could simply choose not to add a legitimate transaction to a block. If could be because they don't want to, the transaction having a low fee, or having a dust output, among other reasons. All of those simply means that a valid transaction may not be confirmed once it is broadcast. You can be fairly certain if a transaction will confirm by checking a few things, but it is not guaranteed. You can check the fee and for dust outputs, but even if the fee is high and there is no dust, there is nothing that guarantees that it will confirm. All you can say is that it probably will.


Yes, they could. And actually most clients do just that. Most bitcoin clients will allow you to spend unconfirmed UTXOs if the transaction they come from originated from you because it knows how it made the transaction and it can trust itself to not double spend against itself.

this mean that every merchants that accept visa, is accepting an uncofirmed transaction each time, so why they should not do the same with bitcoin?

double standard much?

and thats my point, people can send millions of unconfirmed bitcoin tx's and then let the blocks bottleneck for 8 days.. which would actually be on par with visa..

but we bitcoiners are too honest. our 7tx/s is based on fully settled transactions where the recipient is fully funded(confirmed) to then spend them funds.

infact with Visa USA being one database for the dollar. visa UK being another database for the pound. Visa europe for the euro, Visa asia, blah blah blah..

what we could do is claim bitcoin as equivelent to the US ledger, feathercoin for uk, litecoin for europe and dogecoin for asia... and then add up all the tx/s of all them chains..
but no.. we like to be honest and stick with the main principal of bitcoin and the real statistics of international use

Very true. You can't tell how many transactions are going to be in 1 block. Hopefully in the future, bitcoin will be able to handle more transactions a second.

Bitcoin transaction to my knowledge there is no limit.
Maybe it was just the issue that has not been accurately

Yeah. I think that there is a limit, but it is very high. If I am wrong please tell me. This is what I have heard, I may be wrong.

Limit to what part of transactions? Confirmation speed? Transaction size? Please specify.

I think they are talking about number of transactions accepted in a block per second to be confirmed...
If this is it, then it's a very less number as compared to previous numbers, and even the fees have now been increased...

is that not one of the great things about bitcoin it can be programmed to be what the community agree it should be...... ;D

I didn't mean that for certain purposes (such as archiving and smart-contracts) the institution would try to spend the bitcoin prematurely or indeed at all. They might be sending a minimal amount of bitcoin (with a sufficient/generous fee) to an address just to get that address on the BTClockchain. That done, the address might represent a database of Visa transactions (basically acting as a notary); or the address might be part of the computer code that a smart contract would look for before starting or completing.

On the matter of transaction confirmation by individual miners, I found two sources which contradict your position. At http://bitcoin.stackexchange.com/questions/16607/if-a-block-chain-is-considered-invalid-how-about-transactions-in-this-block-cha?rq=1 (http://bitcoin.stackexchange.com/questions/16607/if-a-block-chain-is-considered-invalid-how-about-transactions-in-this-block-cha?rq=1) look for the answer to the top question "If a block chain is considered invalid, how about transactions in this block chain?"  At http://bitcoin.stackexchange.com/questions/30452/how-long-does-it-take-for-a-transaction-with-0-fees-to-be-confirmed-or-rejected (http://bitcoin.stackexchange.com/questions/30452/how-long-does-it-take-for-a-transaction-with-0-fees-to-be-confirmed-or-rejected) look for the statement "No transaction can be rejected though, there's just no such thing in the protocol."

My original point was that, for certain purposes -- purposes that will continue to expand in volume and scope -- the alleged 7-transactions- per-second limitation does not and will not exist as a practical matter. Instant or near-instant publication (broadcast) will be as good as confirmation for those in no hurry to spend the token bitcoin that was sent for the sole purpose of establishing an address on the BTClockchain.

It is so declared: All legit and properly-fee'd transactions will be instantly broadcast and routinely confirmed.

Bang the gavel.

How do they contradict me?

I thought they did, but since you are apparently in agreement with them, the matter is settled: the 7-transaction-per-second limitation is and will be irrelevant for many present and future users of the BTClockchain.

 

No, I never said that. How do those articles lead to your conclusion? Please explain how I those contradict me (if in fact they do, then I probably missed something or you misunderstood something) and how those lead to the conclusion that the 7tps limit is irrelevant.

Sorry for the holi-delay. Earlier you painted what was to me a rather alarming picture of the process of transaction-confirmation:


But to further make my counter-assertion that the 7-transactions-per-second limitation is and will be irrelevant to many users; and that all legitimate and fully fee'd transactions will be always be routinely confirmed, go to:

http://allcoinsnews.com/2015/12/28/uproov-first-to-offer-mobile-proof-on-the-blockchain/  (http://allcoinsnews.com/2015/12/28/uproov-first-to-offer-mobile-proof-on-the-blockchain/)

and look for the statements "There is virtually no time lag between the taking of the photo, video, audio and its encryption to the Blockchain for any manipulation to be performed, it is instant," and "This ground breaking technology means it will be very difficult to argue against a Uproov photo or video, as an encryption key is permanently time-stamped instantly into the blockchain, making it unchangeable and the perfect court evidence if need be." What is lacking is any discussion of a delay in confirmation-time or a fear of a random chance of non-confirmation. Clearly what matters to this for-profit business -- what is essential to their service -- is the instant broadcast (publication) on the bitcoin ledger of a transaction that will -- not probably -- be confirmed. (You can also go to the website https://uproov.com/ (https://uproov.com/))

Other business are already depending on the same thing. I think that if your description was accurate, then these businesses could not exist because they could not guarantee instant and permanent incorporation of a transaction onto the blockchain. Claiming they could might make them liable for damages caused by a randomly failed confirmation and the possible excision of a critical transaction from a majority of miners' block-hashing.

There is no guarantee about anything, only a probably. With a high enough fee and a standard transaction, then it is extremely likely, almost guaranteed that a transaction would be included in the blockchain. However, in theory, if a miner had a thing against you, he could choose to not include your transaction in a block. That doesn't mean that other miners wouldn't (or that that would happen at all) but it is a possibility (however still unlikely).

Also, it isn't instant incorporation of a transaction into the blockchain, you still need to wait for a block to come out with the transaction for it to actually be there.