Bitcoin Forum

I'm so tired of seeing this complaint on block times.

If you are one of the complainers, then here's an experiment I want you to perform and then explain to me how it is possible:

Transfer some bitcoin to your blockchain.info wallet and watch/listen for how long it takes to get there. That's right-- It takes about 5 seconds until you hear the *Beep*. Now think for a second, how did they do that? If they aren't waiting for the entire 10 minute block confirmation time, then how do they know money was sent?

Any merchant can use this method to see a transaction propagating through a majority of the nodes! This whole "10 minute" thing is given way too much importance because it doesn't matter for 99% of transactions. It's just something that people can focus on to complain about because they don't understand this. You can see any transaction on the network nearly instantly.

So the problem must be with the wallet you are using if you can't see an instantaneous transaction. So, rather than complaining about 10 minutes confirmation time which frankly doesn't matter in 99% of situations, then the solution is to get your wallet to support the same technology that blockchain.info is using to SHOW the transaction occurring in near-realtime. Not to complain about 10 minute block times.

And you could argue that double-spends could make this method irrelevant, but if you are THAT concerned for the particular transaction, then go ahead and wait for the full confirmation. But that level of verification is a very small % of the time. Not many people want to steal a $5 cup of coffee bad enough to do a double spend on it. And yes, I'd wait for a few confirmations if I was transferring 100 BTC.

The point is: Block time doesn't matter for 99% of transactions. And for the 1% where it might matter (bigger purchases), 10-30 minutes is an acceptable waiting period!


A Misconception:
I think that the complaints about the block confirmation time stem from a misconception of what is happening with credit cards, which causes a person not to compare apples to apples:  Your credit card and debit transactions are not instantly confirmed either. At stores these transactions are given credibility "immediately" but still take 3-5 days to actually move through the payment processors between bank accounts. 3-5 days is WAY slower than 10-30 minutes for a block confirmation if you want to compare apples to apples. The problem is people aren't comparing apples to apples when they compare 10 minute block times with credit card transactions. The credit card transactions aren't "confirmed" either at this point. The payment processor is simply fronting the money (just like blockchain.info shows the transaction instantly) while the transaction goes through. These credit card payment processors are basing this method on the same logic that blockchain.info uses to instantly display transactions: It's the fact that most transactions aren't fraudulent. And who wastes their time trying to defraud a system to steal a $5 cup of coffee?



Correct comparisons:
Bitcoin purchase = appears within seconds on the network
Credit card purchase = appears to be completed within seconds
(TIE)

Bitcoin money is ACTUALLY transferred = 10 minute average time
Credit card money is ACTUALLY transferred = 3-5 days (or longer)*
(BITCOIN wins)

*If I remember correctly, Andreas Antonopoulos said in one of his videos that credit card transactions aren't settled for up to 30 days. Now how does that compare to 10 minutes?  Compare apples to apples.


- - - - - - - - - - - - - - - -


Lastly, from satoshi's own mouth on the subject of verifying fast payments:

Or send a bank wire transfer.

Compare the speed of the ironically-named SWIFT system to the speed of the bitcoin network.

Exactly. That is the true comparison to a confirmation on the blockchain.

good post Colin.

but .... yeah.... a noob is likely to complain
again in about a week.

Where do you get the 99% figure? In my experience, every merchant waits for at least a confirmation. Bitpay min is 1, max 6. Exchanges: 6. Poker sites: 4, etc.

The details of the protocol are irrelevant to the end user.

Side Chains will hopefully solve the block confirmation road block to a great degree when they come on. Still a nascent but promising space.

Thanks. I've watched a lot of Andreas Antonopoulos videos. I highly recommend this.



If so, please just send them to this thread.

I'd really like to get the community awareness up on this area so we can move into building on bitcoin rather than spending time complaining about things that don't really matter. Let's get this topic understood more widely.

That's cool to think about. What is the basic theory on how this might work? I get how the coins are "killed" in the main blockchain and brought to life on the side chain. But how could this be used to do anything with confirmation times?

Do sidechains have their own blockchain and thus could have their own confirmation times?

My knowledge of side chains isn't as good as my bitcoin knowledge.

If I remember correctly (from my last BitPay transaction), doesn't BitPay send you (the buyer) a confirmation as soon as you have sent the payment? They don't wait for a full block confirmation to send you your email receipt. The merchant gets the full payment after it has been fully confirmed in a block. But this is another good comparison to what is happening with credit cards and merchants who accept fiat/credit.



Exactly. Which is why (eventually) when enough Bitcoin user applications have been built to make the use of bitcoin super user friendly, easy and seamless, you won't even need to know what a "block confirmation time" is, nor worry about it. It will be so buried beneath the surface of the whole transaction procedure that it won't matter.

A comparison would be like you and I fully understanding how a cell phone works right now. We don't need to know. We just need to know how to use a phone and have a use to find it useful. 99.9% of the population is never going to understand the technology and mechanics of bitcoin. They are just going to use it (once the applications and use are easy enough). We are definitely the geeks in the space and will remain so.

I think you misunderstood what I meant. You're saying that block confirmation times shouldn't be considered and instead we should look at the time it takes to relay a transaction.
It's all good if merchants were actually accepting 0-confirmation transactions. For the most part, they aren't. Maybe they are too uncomfortable with the risk of double-spend, the bottom line is that as a user I have to wait for several confirmations.
To follow your metaphor, today most merchants require wire transfers. Until they change their policy, we can't ignore the block confirmation times.

There could be an opportunity for a service that covers the risk of a double spend and acts as insurance. I heard that Coinbase (no affiliation or recommendation implied) is basically doing that.

Actually I think we are saying the same thing more than you realize. I understand what you mean completely.

You said it yourself, "It's all good if merchants were actually accepting 0-confirmation transactions". This is one of the [several] gaps to bridge in making bitcoin more usable. Just like if there was no web browser it would be pretty hard to use the Internet. Well, there are many features and enhancements that can be made to make the existing bitcoin network more user friendly, and more widespread use of "instant transaction recognition" (to coin a new term) will be one of them.

Also, I'm not at all saying that block confirmations shouldn't ever be looked at. A block confirmation is an obviously vital part of the bitcoin blockchain technology and absolutely has its place. They just aren't always required right away to process a transaction in the real world. And that is the point of this thread-- to point out that 10 minute blocks are not the end of the world, when implemented properly.

Isn't it quite simple to double spend if no one waits for 1 confirmation?  I don't know how the double spends work but it seems like it would be quite easy to pull off double spends once people allow spending of funds with not even a confirmation.

I've done BTC transactions and waiting 6 confirms sux.  I go use Bitshares and 10 seconds unless network isn't healthy but worst case is always far far faster than BTC.

In fact the reason I liked BTC to begin with turned out to not even be true due to these limitations of the block times and POW.

^ Thank you! That is a great post by satoshi that is saying about the exact same thing!  I should go read all his posts. There is some wisdom there!

The problem is for daily usage. Like if you want to go out and make some groceries with Bitcoin, most people aren't safe until the confirmations go through.. for some reason they think that when they pay with Credit Cards, it gets automatically confirmed when the same thing happens.

I have never complained about confirmation times, but I see cases when it becomes important.
First is what is quoted.
Second, when you go to sites which require you to deposit bitcoins, like you have a poker tournament at 11, and you send the funds at 10:30. Now those sites require 2 confirmations to confirm your deposit and allow you to play. In those cases, the over wait becomes a problem.
You could always say, then send before, but sometimes it has happened that I saw a tourney in the last half hour, and my deposit didn't go through.

agreed 99% of Tx's dont really require confirms..

but for all of those that do need confirms.. they cry when they wait over 10 minutes.. but i have never ever seen them post when their TX's have taken them just a couple minutes to confirm due to fast hashing

Other people argue differently in that same thread.  Spitting out 2 spends from the same time is easily done.  It can work enough to be a valid attack.  That is why everyone insists on at least a confirmation for transactions of any significance.

The problem is that for people are fascinated by the pure technical aspect of being able to spend digital money, this wait time causes a severe problem.  The answer of the koolaid drinkers is not better technology, it is layering on kludges to fix the problem.

\

It isn't fast hashing, it is just luck.  I've had to sit around a BTC ATM machine an extra 30 minutes because hashes were coming through that were longer than 10 minutes.

Credit card confirmations take 180 days.

its a valid complaint. don't try to explain how things aren't instant in an instant wanting world.

The only reason why most transactions would need a confirmation would be because it is large (in terms of the amount being transacted), and large value transactions tend to take a long time anyway because of various features being explained, questions being answered, ect. If someone were to pay and then continue a conversation with the merchant about the product then any TX that needs to be confirmed will likely be confirmed long before the conversation is complete/winding down

Yes but I think you've missed the point. Bitcoin is more instant than the current technologies (debit cards, credit cards and bank-wire) of the "instant-wanting world."

From your post, I assume that you think credit card and debit payments are instant. But they aren't. Bitcoin can do what you think is happening with credit card/debit card transactions.

Take the time to read the entire OP and satoshi's post and maybe you'll see what is being said and why.

True, that's how long you can call up and get the transaction reversed, right?

Yes, but if two simultaneous spends get detected then both transactions can be cancelled. And if one of the transactions was first, even if only by seconds, then it will propagate a majority of the network more quickly (exponentially) and be accepted by miners while the second one is rejected. This is as satoshi explains in his post (see link in OP).

This would be coded by payment processors to make the threat of double spends a non-factor. satoshi explains this and it makes sense if you think about how it could be implemented.

It helps if you can think like a programmer to imagine how this would be implemented in a workable fashion and all its ramifications.

I don't complain about confirmation times. There are situations where confirmation times are important. Like sending funds into exchanges for trading, or sending money from cold wallet to hot wallet for an online purchase. Most of the time, people are just impatient. Bitcoin is already very very fast compared to wire transfer, but users are used to "credit card" speed.

If you pay a digital item through bitpay or coinbase it is automatically given to you without any confirmation.

Confirmation times aside, already been well explained in the OP... I read a comment today that mentioned Visa can do 100,000 transactions per second whilst Bitcoin is capped at 7 max per second.

On that basis btc would allow for max 400 (approx) transactions per minute and 604800 per 24 hours.

Just wondering what peoples thoughts are on this (perhaps it's not even true)?

http://pastebin.com/Vf6XkDm3 (http://pastebin.com/Vf6XkDm3)

Litecoin is faster.
If you need to send money quick: use not bitcoin

Just did a quick google search and found an interesting Bitcoin wiki about scalability.

https://en.bitcoin.it/wiki/Scalability (https://en.bitcoin.it/wiki/Scalability)

Scalability targets

VISA handles on average around 2,000 transactions per second (tps), so call it a daily peak rate of 47,000 tps. [1]

PayPal, in contrast, handles around 10 million transactions per day for an average of 115 tps. [2]

Let's take 4,000 tps as starting goal. Obviously if we want Bitcoin to scale to all economic transactions worldwide, including cash, it'd be a lot higher than that, perhaps more in the region of a few hundred thousand tps. And the need to be able to withstand DoS attacks (which VISA does not have to deal with) implies we would want to scale far beyond the standard peak rates. Still, picking a target let us do some basic calculations even if it's a little arbitrary.


Network

Let's assume an average rate of 2000tps, so just VISA. Transactions vary in size from about 0.2 kilobytes to over 1 kilobyte, but it's averaging half a kilobyte today.

That means that you need to keep up with around 8 megabits/second of transaction data (2000tps * 512 bytes) / 1024 bytes in a kilobyte / 1024 kilobytes in a megabyte = 0.97 megabytes per second * 8 = 7.8 megabits/second.

This sort of bandwidth is already common for even residential connections today, and is certainly at the low end of what colocation providers would expect to provide you with.

When blocks are solved, the current protocol will send the transactions again, even if a peer has already seen it at broadcast time. Fixing this to make blocks just list of hashes would resolve the issue and make the bandwidth needed for block broadcast negligable. So whilst this optimization isn't fully implemented today, we do not consider block transmission bandwidth here.

I consider Bitcoin, due to its irreversibility and aspects of trustlessness, better suited for larger, big-ticket items than small-change transactions. A merchant selling yachts may have more confidence in Bitcoin's finite supply and blockchain technology anything else. I foresee part of Bitcoin's future as being either a means of international remittance, or a preferred currency for luxury goods such as bullion, luxury cars, diamonds, etc. And as I mentioned previously in this thread, one of Bitcoin's direct competitors is the SWIFT system. Bitcoin is faster and more secure than SWIFT, the de facto standard for wire transfers. Banks or Western Union could easily adopt Bitcoin as a means of transferring money.

I'm mystified that so many people see Bitcoin as a direct competitor to the likes of Visa or Mastercard. Goods and services aren't denominated in "visas" or "mastercards." No one says, "That'll be 22.50 visas, please."

Absent a (preferably decentralized) payment processor layer atop the blockchain, and assuming Bitcoin users push for its use as a small-transaction payment method, then the issue of network strength vis-à-vis confirmation times take precedent over other areas of development.

Here is a good post from one of the core bitcoin devs that explains how he / they are actively seeking to deal with foreseen scalability issues.

https://bitcoin-foundation.ghost.io/a-scalability-roadmap/ (https://bitcoin-foundation.ghost.io/a-scalability-roadmap/)

Great link. Somehow I'd missed this. Thank you.

I assume a cancelled transaction does not meant funds are lost, just sent back.  This means the cost of the attack is too close to 0.

There are reasons people who think like programmers and want funds fungible ASAP still rely on confirmations.

Bitshares fixes this and the scalability factor at the expense of managed centralization.  This centralization is in fact less than the current bitcoin centralization which centers around a handful of mining pools.  Effectively meaning that the BTC network is controlled by as many people as I can count on one hand.

The network is more centralized than we think it is. Once the first transaction reaches the main mining pools, it's practically impossible to do a double-spend. A merchant could maintain a direct connection to them and make sure that it was propagated there before handling the goods. It would only take a few seconds of waiting.

'Centralized' isn't the word you're looking for. Perhaps 'mature' would best describe that observation. Well connected nodes don't add anything to the hashrate.

I was talking about the main mining pool. I meant centralized - not mature. You are free to disagree though.

not really luck.. in short if i hashed 1 number a minute and wanted an average 10 minute hash time.. each hash would be a number between 0 and 20..
10 times it would average below 10 minutes.. 10 times it would average over 10 minutes.. but on average.. you will get a 10 minute hashtime when repeated the process a few thousand times..

its not lucky that a block is found in 1 minute or unlucky if it too 19 minutes.. its all based on maths, not luck. so its just a standard feature of randomness that some take alot of time some take very little time.

but i do laugh that you complain about 30 minutes, whilst a western union takes much longer. and then for instance trying to get paypal or apple pay to transfer money to your bank and then to a FIAT ATM... even longer..

As I understand it, there are centralized mining farms, but not centralized mining pools. Miners on pools are free to jump to other pools. Mining farms don't mine on pools, but compete with them instead. Miners don't generally allow a pool to get too large. I realize that pools can conspire. As cypherdoc is fond of saying, it's not a fault with the Bitcoin protocol, it's a human choice.

True, but still looks pretty centralized to me. https://blockchain.info/pools (https://blockchain.info/pools). If a merchant connects to the top 5 pools, he has access to more than 51% of the total hashrate. I'm not criticizing the network, it's working as intended. I'm saying that it wouldn't take much for a merchant to accept 0-confirmation transactions while getting reasonable security.

I think side chains are inevitable as a necessary evil. Kinda like lawyers.

Franky.. why have I never forgotten your account ?  Oh I am too lazy.

Hitting a hash is analogous to a lottery where every hash is an entry.  It is luck as much as averages as much as math.  When I'm expecting the 6 confirms to take an hour and they take 1.5 hours, I got unlucky.

I don't use Western Union as an ATM.  I have received payments from WU and when I show up it doesn't take over an hour.  Oddly the ATM fixed it by having your xfer go to an account before you show up but their solution has me leaving some random amount under $20 USD in the account that didn't fit their fiat dispersal denominations.  Typical BTC solution !  SMH

And much more volatile than Bitcoins. Alot of merchants uses Bitcoin payments only, not litecoin. Due to the fact that most merchant services which companies prefers offers only Bitcoin payments. When using Bitpay, they use special algronithm to determine the risk of a transaction being a double spend.

You are right and a small Bitcoin transfer is usually safe without waiting for the confirmation anyway.

In this case the merchant would not have control over the pools. They would only have a connection to the nodes the pools broadcast from. If a merchant is broadcasting a double spend transaction to these nodes then others should notice as other nodes should know and understand the importance of these nodes and alert anyone relying on a transaction of the possibility of a double spend attempt