Topic: Bictoin: Only 7 Transactions Per Second. Not gonna fly... (Read 3007 times)

given bitcoin black friday peaked at 1.18 per second (if you take the day average, it may have peaked higher in reality) and the daily average is around 0.7 per second, we're only a factor 10 away from 7, and less so on peaks.

i'd say it's about time to upgrade the limit.

by the way, is the max. blocksize of 1Mb ever going to cause problems when there's a lot of transactions?

The code change itself is trivial (a handful of lines of code) but making the change which prevents a lasting hard fork requires establishing a consensus and that is never trivial.  So no it can't be changed overnight, the "issues" are more social than code related.  There currently is some debate on how high the cap should be raised, when it should be raised, and how it should be implemented (should it be a new temporary fixed cap, a new permanent fixed cap, a deterministic cap, or no cap).

so, upping the limit to say, 100 tps is just trivial? they could do it within minutes?

or does it take days/months to change it?

because if bitcoin does boom overnight, and the transaction limit does prove to become a bottleneck, then the negative publicity and trust in bitcoin may severely damage it's reputation.

Real transactions are tiny in my view. Perhaps 1% or less.

BTC trading volumes spike massively as the BTC/USD rate improves (see blockchain), this suggest this is being driven by speculative activity rather than spending on products (which should reduce in BTC as the rate increases).

You seem to have a view.   Maybe you can tell me this.
How many BTC are trading per day (real trading, not daytrading).
in 1998, when the day trading fad was kicking in, many internet stocks had 70-80% of their volume in NON board lots.   And day traders did not have bots.

agreed so keeping a few hundred in BTC today may pay back 10 fold or we may lose it outright. Im definitely willing to take that risk and am fully aware of the consequences.

I completely agree with all your post.

The issue will be whether people holding as a store of value are patient enough and whether some form of crash or problem causes confidence to fail either in Bitcoin itself (to be replaced by another medium) or in all crypto-currencies. Time will tell.....

It might take decades.  For a currency to be used as a unit of account in commerce both parties have to be willing to use it.  Not just some mechant and some customer but the specific customer looking to use a specific merchant.  For an individual to use a currency as a store of value they just need to believe that SOMEONE, anyone will accept it in the future.  That is somewhat simplified but you can see the network effect requires for direct commerce are much higher than for a store of value.

Imagine a hypothetical store.
Does the store accept Bitcoins? No? Then no Bitcoin commerce is possible.
If the store accepts Bitcoins does the customer have Bitcoins to spend? No? Then no Bitcoin commerce is possible.

However if you believe that Bitcoin will be increasingly used as a store of value then that means in the example above the second requirement is more likely to be fulfilled in the future even if it isn't today.  100K Bitcoin holders means 100K potential customers.  1B Bitcoin holders means 1B potential customers.   If adoption as a store of wealth continues to grow (not just value of the currency by the number of individuals holding it).  More currency holders means it is more likely merchants will accept it as payment to gain access to those currency holders.

Imagine I am a merchant that makes widgets and all my competitors make widgets.   At the current time none of us accept bitcoins as payment.  If someone has wealth stored in bitcoins and wants a widget (lets assume these widgets are expensive) they need to trade it for local currency and buy indirectly.  This means they face additional time, cost, and risk.   It would be much easier for this bitcoin holder if at least one widget maker accepts bitcoins.   If I accept bitcoins then I can gain an advantage over competitors even if my prices are marginally higher (as long as bitcoin holding customers see value in a direct transactions).  If you believe in free markets and economic competition then the end game is that eventually there will be large enough critical mass of potential customers that at least one widget maker will accept bitcoins to tap that market. 

A currency can be used as a store of wealth or a unit in transactions.   The store of wealth use requires less of a network effect so it will grow quicker but baring some failure of the system that will boost the growth as a transaction unit use as well.  It however will take longer and require more potential customers holding the currency.  We have seen this effect somewhat already.  The earliest holders of the currency were miners so products and services offered for bitcoins that met that demographic were more successful than ones which didn't.  Think used computer parts vs cosmetics.

Bitcoin has not yet proved it can be used for trade in volume, but the hope remains, over time that will either be proven or interest will dissipate.

As for China they're only buying it for speculation, they have a lot of money and are very active gamblers/investors, they cannot use it to buy anything. As for it being a reserve currency that'll only be the case if it has global usage and a belief system that supports it's value. Right now we're still in early Facebook 'I believe' territory, it remains to be seen whether Bitcoin ends up being Facebook or Friends Re-United.

Well how long will it take to prove that it doesn't work for commerce?? Surely there has been plenty of time already for something like that to be proved true or false.

Bitcoin has value because it is set to become the world's reserve currency replacing the USD and also for global trade replacing SWIFT. That is where it is headed. The signs are clear from the adoption of Bitcoin in China. They've been looking for a way out of the USD trap they are currently in and this is it.

Bitcoin is only a store of value whilst people think it has a value, which is because it can or 'will' be usable to buy things.

If it's ever proven that it doesn't work for commerce (and I believe it doesn't) then it's value will be fraction of what it is now.

What you're experiencing is not a network overflow of some sort, but a normal confirmation process.
Your transaction must be included in a block to become part of the "blockchain".
Blocks are generated every 10 minutes, so the time for the 1st confirmation can be as long as 10 minutes. It can take even longer, depending on fees, transaction size and other factors.
This block should be "covered" by yet few more blocks on top to be considered fully confirmed. Current clients require 6 confirmations for the transaction to be considered as "final / done"; this can take an hour or more.

This might appear a very long process, but considering that credit card companies can cancel transactions months after they happened, the finality Bitcoin offers in an hour is a winner.

Litecoin is not quicker. It generates more confirmations in a shorter amount of time, but those confirmations are less secure and more subject to change than Bitcoin's ones.

The UX should be improved indeed, though I'm not sure how. There's probably an opportunity for payment services on top of Bitcoin to offer this.

Possibly but the default QT client will prevent you from sending tx without a fee if they will be dropped because they don't include a fee.  High priority tx are always relayed even without a fee, low priority tx are never relayed unless they have a fee.  Unless modified the QT client enforces no fee on high priority txs and does enforce the minimum fee on low priority txs.

The more likely scenario is the OP client never broadcast a transaction for 15+ minutes.

It is the other way around. Litecoin penalizes you if coins are spent too quickly after they were received (takes about 2 weeks for that fee penalty to go away). Litecoin transacion fees are also much higher.

Litecoin advocates, please enlighten me if this is wrong. Currently, it is a pain to move LTC.

Oh, learned something. Could a missing fee have prevented the TX from being relayed?

Bitpay doesn't even require a confirmation to save the invoice.  The invoice will be considered paid pending confirmations if the tx is seen on the network before it expires.  In other words you only need 0-confirm to avoid an invoice expiring.    Now most merchants* will still require 6 confirmations but as soon as the tx is seen on the network it stops the "expiration clock" and bitpay waits for 6 confirms before notifying the merchant that it has been paid and confirmed.


* Merchants can set the level of confirmation they require (0-confirm, 1-confirm, or 6-confirm) but regardless the invoice will not expire once the tx is seen on the network it just will take longer to eventually be marked as confirmed.

Excellent question.

Thanks Barek - I guess that proves my suspicions wrong.  

QT still took 25 minutes to go from a " ? " to a clock icon ...

I suppose this thread could be a discussion about the 7 transactions per second limit anyways.... lol

Its going to be an issue if Visa has to handle 2,000 per second ...

I thought Black Friday - maybe we'd be hitting 7+ transactions per second.


No, just did what an average consumer will do.  Hit "Pay", then Send, and waited.

Plenty of room in the blocks.

Did you include a transaction fee?

I was fully sync'd to the block chain.

Definitely made sure of that first.

I sat here and watched.  It was a " ? " on the QT transaction thing for roughly 25 solid minutes.  

So this is a network speed issue, is it not?