Hacker News Reader

Comment by mootothemax | Hacker News Reader

the correct fee should have been approximately 2BTC.

At current rates isn't that a $5,000ish transaction fee?

Have I misunderstood something? That seems like a crazy amount, especially given the hype surrounding Bitcoin.

Bitcoin fees are based on the amount of space that the transaction takes up in the block. A block contains on average 1,000 to 2,000 transactions. The average transaction only has a couple inputs and outputs. But you can make a transaction that has as many inputs and outputs as you want, which increases the size (in bytes) of the transaction. The person who spent the 80BTC on fees in this block used up about half of the space in the block, hence the high fee. It was also spread over several transactions, not a single one.

I think cryptocurrency have a future but things like that will prevent it from mass adoption. Good luck explaining this to a non-CS, why a transaction cost .5$ one time and 5000$ another time.

It's not a cryptocurrency issue in general. For example, Ethereum doesn't have this problem because it keeps track of balances (instead of unspent transaction outputs).

If your bank charges you $0.50 to make a wire transfer, and you make 10,000 wire transfers at once, they will charge you $5000. Making a large number of transfers at once with cryptocurrency is essentially the same thing. This is not difficult for most people to follow.

Except your coins can become fragmented (into dust) through no fault of your own. You might not even realize that many inputs are needed to complete your single transaction. I don't see how this would happen with traditional institutions.

I hadn't really considered that; I was only vaguely aware it was possible. Yes, that's a problem.

I'm inclined to think of Bitcoin as the beta for widespread adoption of cryptocurrency. It has some issues, including this one that mean it will probably never be what most people use for everyday money. Whatever succeeds it and eventually assumes that role will avoid those problems.

> It has some issues, including this one that mean it will probably never be what most people use for everyday money. Whatever succeeds it and eventually assumes that role will avoid those problems.

I wouldn't go as far as to say this is a show stopper. Wallet developers should be (and probably are) aware of this and take it as a challenge optimize the way their wallets spend money to avoid this issue, even in the face of receiving btc in a way that may create dust.

I think the built-in deflation and transaction volume issues are bigger problems, but I only have a passing knowledge of the subject.

Try explaining any technical feature of the incumbent monetary system to any normie...

Because those features are generally safe and unsurprising, people don't need to understand them.

I haven't taken a very close look at the transactions, but it seems like the 'appropriate' 2BTC fee that btc.com is talking about pays for 11 extremely large transactions, with around 100-160 inputs each and a half dozen outputs.

A typical bitcoin transaction is made up of 1-2 inputs and 2 outputs (destination and change), so this person is doing the equivalent of thousands of transactions on the network.

That being said, bitcoin transactions are fairly expensive, (I paid $1 the other day to buy a steam game, eww), but there's a scaling roadmap being implemented to lower fees going forward, likely within the next month or so.

But yeah, right now the equivalent of a few thousand transactions will cost a few thousand bucks.

ELI5 what type of transaction could have been done which is that large/complex that also would be 100% on the up-n-up and not trafficking [something], laundering [something]?

If you've received 10 $1000 payments from your job in bitcoin and now want to buy a brand new $10 000 toaster, you'll have to create a transaction that has 10 inputs and one output.

A more typical example is a miner who receives very frequent payouts from a mining pool over the course of a few months, and then wants to send all of their money to a bitcoin exchange to pay for their power.

Some people create massive transactions just to mess with the bitcoin network, too. ¯\_(ツ)_/¯

Edit: Someone lower in the comments asked why people create large transactions to mess with the network, and I don't really know for sure--some people have unknown motivations, or want to prove a political point. Some are public, and I assume some are done secretly. Here's an example of one that was public: https://multibit.org/blog/2015/06/23/bitcoin-network-stress-...

>Some people create massive transactions just to mess with the bitcoin network, too.

Why?

I replied to you in an edit to my post... I don't really know what was going on but when I first saw your comment there was no reply button.

Is that something that's normal on this site?

If I reply to a comment you made, there is a short time where you cannot reply to my post, I believe this is an effort to either make people read the reply post before knew-jerk replying... or to breathe for a sec and slow spamming down...

Oh, neat! Thanks.

Imagine that Stripe has received a few hundred Bitcoin transactions in the last hour and they want to move all that BTC onto an exchange.

Be your own bank they said, it'll be awesome!

Don't you know Bitcoin is a system with really low fees and fast transactions??

this must be sarcasm, right?

Fees are around 80c and average confirmation time is ~150 minutes. Pretty good for the first implementation of a completely new technology which has scaled 100,000x in 6 years. I don't really get the cheap shots but ok.

The problem is there are multiple new implementations that are better suited to do fast cheap transactions. Bitcoin is the Myspace of crypto.

When you use "crypto" as an abbreviation for "cryptocurrency", you promote confusion about what cryptography is.

Seems pretty clear in context to me.

The cheap shots are because people are upset they missed out on some of that sweet, sweet bitcoin early on.

Aka, nocoiners. They will always be mad

We'll see.

Well, the Bitcoin supporters are composed of a wide range of people. Some of them are too optimistic about the technology. Unlike some people thought, transactions are not cheap (although there are scaling plans to bring the current high cost down). Being your own bank is not easy. All the things that the current banking system does is not useless overhead. So, you can't just throw it all out. OTOH, crypto-currencies potentially allow nearly anyone to be a bank if they want. That could lead to innovation and reduced banking costs. Right now, it is the "wild west" and so caution is advised.

You raise some great points. I would love to see an banking insider document regarding a true assessment of how they could compete with bitcoin. Or in other words, what is bitcoin missing that banks account for in overhead.

I want bitcoin to succeed as much as the next but I think a lot of people are naive and convinced its ready just because they can transfer btc from one party to another.

For one, you can't undo a transaction if you made a mistake. I.e. Put in a wrong amount or sent to an incorrect address. You can't call customer service if you don't see your transaction. I think those are big and come with trade offs. Some others are the huge volatility of the value id bitcoin ..

Bitcoin is currently dealing with a scaling problem. People are creating more transactions than can fit in the blocks. So if you want your transaction included in a block, you need to be willing to pay a higher transaction fee. the metric for this is satoshi/byte (s/b). Not long ago (Christmas time) you could easily get a transaction in a block for 60 s/b, now it's like 300-400 s/b. There is an update in place that will go live August 1 that should fix this.

From what I hear, this update make bitcoin no longer decentralized. Is Segwit centralized?

SegWit is not centralised, at all. It's just a way of separating signatures from transactions so that they can be stored more efficiently, and accounted for more accurately.

Segwit just separates (segregates) out the transaction data from the signatures (witnesses). The "centralization" talk stems from worries about the Lightning Network, but the "centralization" idea is a fair bit more nuanced than that.

I haven't looked too closely at this transaction, but here's some info on how normal fees are calculated: http://bitcoinfees.com/ Very new and very small transactions are charged extra to discourage spam.

As noted at the bottom of the article, each miner really gets to set their own fee, below which they will just ignore transactions. But the "satoshi client" is the default and most other clients mimic it.

Edit: one of the comments mentions that "multi-sig" transactions are more expensive, and I don't know how to explain that part.

Multi-sig transactions are larger, and fees are based on tx size.

That information is out of date - in particular, "priority" is long gone. Nowadays most miners (and Bitcoin Core) maximize for highest fee profits - basically select transactions highest fee/kb first.

I didn't realize Bitcoin Core had switched. I mean it's kind of an obvious idea if you're trying to make money, but it could make the network weaker if some kinds of transactions take days to get verified.

The fee is normally proportional to the size of the transaction (i.e. mostly the number of input and output addresses), not the amount of bitcoins you move. That is, it's more expensive to move 1BTC from 10 addresses than 1000BTC from a single one.

So if you make a huge transaction and you want it confirmed quickly it can get quite expensive.

This is why it's a good idea to consolidate dust during off-peak hours when the fees are lower, so that if you do need to spend a lot urgently, the total fee you pay is much less.

It's kind of like the electricity company giving you lower rates for doing your laundry at night.

Aren't they doing this because not enough electricity being used at night is bad for them anyway ?

Depends on the generators in the system. Certain types of generators don't like / can't really / doesn't make sense to spin down much (e.g. nuclear, hydro). A nuclear reaction is going to generate as much heat as it does, you don't dial down a knob. A dam has (usually) a given amount of water flowing through based on external circumstances, not electric demand. Compare that with a coal generator, where you can just stop shoveling in more coal to reduce heat output.

Sorta like that.

I wouldn't really classify it as "bad for them", more like it's gonna be wasted anyways.

Note: been a decade since I studied power systems, so... inaccuracies abound.

That's been around for decades. Electricity at peak hours is more expensive than at any other time, so anything to move demand from peak hours to off-peak hours is good. Laundry is especially bad because electric dryers are very energy intensive and people tend to run them in early evening hours after getting home from work, when electrical demand is at its peak.

I am with you on this one. BTC is supposed to solve the problems current systems have, the key ones being; high tx fees and slow tx.

It was never intended to solve that. As described by the bitcoin paper, it solves the problem of decentralized cryptocurrency which is minting and preventing double-spending without a central authority.

Maybe the paper didn't address this but it sure is all over bitcoing.org. "As these services are based on Bitcoin, they can be offered for much lower fees than with PayPal or credit card networks." https://bitcoin.org/en/faq

But 'not having a central authority' is only valuable if the alternative solves the problems that having a central authority causes. The main problems I have heard articulated about a central authority is that they have high costs, are slow, and can block transactions or withold funds with no recourse. Bitcoin doesn't solve the first two issues; does it really solve the third, either? Maybe, but there are also a host of NEW issues that it brings up, too.

On the contrary, centrally authorized systems can be extremely fast and cheap. Postgres outperforms any blockchain. The issues are about authority and power, and also the barriers to innovation and access which that power imposes.

Payment channels can solve scaling issues in blockchains just like various settlement layers solve the slowness of bank transfers. Some people don't want such a solution, but it's still a solution.

Development doesn't happen the way we want. Some people want cryptocurrencies to stop existing, which seems unlikely. Bitcoin's development is really thorny because of diverse interests and general incentivized hostility; that's just reality.

Maybe a faster blockchain will gradually take Bitcoin's market share. Maybe Bitcoin will remain quite valuable. If you know a certain answer to this, there are large profit opportunities!

BTC solves the problem of preventing double-spending without a trusted central ledger.

Maybe the paper didn't address this but it sure is all over bitcoing.org.

"As these services are based on Bitcoin, they can be offered for much lower fees than with PayPal or credit card networks."

https://bitcoin.org/en/faq

From that same paragraph: "There is no fee to receive bitcoins, and many wallets let you control how large a fee to pay when spending. Higher fees can encourage faster confirmation of your transactions." The more you're willing to pay in fees, the faster your transaction is likely to get processed. That's what it says.

Why does Bitcoin have to solve those problems? Is it completely failed even if it doesn't provide fast and cheap transactions? What if it turns out that Bitcoin has a unique role to play as the first PoW blockchain and the only successful blockchain with an anonymous, nearly legendary creation narrative? Gold is slower to transport than bank notes, but it's still uniquely valuable.

I think the future for cryptos is bright but Bitcoin, and other coins, have some issues. I'm also happy for the miners since you'r happy paying 5k$ for a single digital transaction.

I assume that's for the entire block. Recent blocks hover somewhere between 1000 and 2000 transactions per 10-minute block, which gives a modest 2.5 USD fee per transaction.

By what measure is a 2.5USD fee modest?

2.5USD is not very competitive with the old guard unless you are moving more than $100.

I had to pay $20 to sweep some old wallets(circa 2011). In my case it was worth it, but there is so much dust out there which is not worth collecting any more.

People used to tip using BTC.. those days are gone.

Reply on news.ycombinator.com