Transaction Cost Problem

[tradulant]

Hey,
what is the current status regarding the transaction cost problem?
Is the team working on a solution?

[floaks]

Yes. The lightning network.
 
https://lightning.network/

[Anti-Cen]

Hey,
what is the current status regarding the transaction cost problem?
Is the team working on a solution?

Yes this is a major problem on BTC and has put me off using the coins when the cost can be over $20 per transaction and at the start of the year the cost was
something like $0.00001 and the miners should take profit by mining new coins and competing to add new blocks the the block-chain.

Unlike 99% of people I am not a big fan of the block chain because it's too slow and complicated to stop double spending when you have to process and check
history on all the inputs to a transaction in a database that is already 200gb in size and have all the energy and GPU power being wasted as machine compete
with each other to find out who is fastest at counting grains of sand to win a prize for the miners.

Cost something like 120KWH of electric just to confirm a single transaction so yes the block chain solves one problem and made ten more as I see it

[pawanjain]

The current satoshis needed per byte to make a normal transaction successful is around 293 which give around $10 for an average transaction.
The high fee is an issue which everybody is pissed of. Bitcoin is not meant for any micro payments at the moment. Although it will be done in the future with the help of the lightning network it is just a waste of money to spend high fees on micro transactions. It is better to avoid small transactions and just hold the BTC until the lightning network is fully established. The development is going with a successful testing recently. I think people should stand together and join hands towards the development of BTC. Why should just a small group of people do all the hardwork and let the other enjoy it's benefits.

[tradulant]

Is there any estimate how long it will take to establish the lightning network?

[DannyHamilton]

Yes this is a major problem on BTC

No, it's not.  It's a designed feature.

the cost can be over $20 per transaction and at the start of the year the cost was something like $0.00001

Yes. Bitcoin has gained popularity in the past year.  Higher demand for a limited resource results in higher costs.  This is the way the world works with ALL resources.

and the miners should take profit by mining new coins and competing to add new blocks the the block-chain.

They do.

However, the miners receive the transaction fees of the transactions that they include in their blocks.  If someone is willing to voluntarily pay a higher fee, then why should their high-fee transaction be ignored to include a cheaper transaction instead?  That wouldn't be fair and wouldn't make sense. You are suggesting that low fee transactions should confirm quickly and that high fee transactions should wait longer?

Unlike 99% of people I am not a big fan of the block chain because it's too slow and complicated to stop double spending when you have to process and check history on all the inputs to a transaction in a database that is already 200gb in size and have all the energy and GPU power being wasted as machine compete with each other to find out who is fastest at counting grains of sand to win a prize for the miners.

It is clear that you do not understand how the system works.

Cost something like 120KWH of electric just to confirm a single transaction

Check your math.  You seem to have made an error in that calculation.

That electricity is being used for much more than just "confirming a single transaction".

[Anti-Cen]

No, it's not.  It's a designed feature.

Yes sky high transaction fees we agree but you need to check your understanding on how much electricity it costs per transaction
and bring yourself up to date and if you think that is bad then did you know that the cost of mining a coin is between $1000 - $2000

High fees might suite miners but it will drive the public away and they have plenty of alternatives to pick from

It is clear that you do not understand how the system works.

I know about IOTA using Tangle instead of block-chains and that the Lightning Network being used to plug the speed issue has many critics
and i cut code for a living so maybe I am closer to the truth than you think

[HappyMod]

Honestly its kind of the same question as why has Cryptokitties broken the ETH chain.
People use it and everyone loses their minds...

[DannyHamilton]

No, it's not.  It's a designed feature.

Yes sky high transaction fees we agree but you need to check your understanding on how much electricity it costs per transaction

Are you dividing the number of transactions in a single block by the amount of revenue that block earns for the miner?  If so, then you are assigning too much value to "how much electricity it costs per transaction".

Mining a block serves a purpose beyond the transactions that are in that block, and you are failing to account for that.
 

did you know that the cost of mining a coin is between $1000 - $2000

An AntMiner S9 uses 1372W to generate 14TH/s.

At the current difficulty and a mining rate of 14TH/s it would take approximately 451.914 days to mine a block.  That's about 10846 hours.

The average bitcoins earned per block is the 12.5 BTC subsidy plus about 2.43 BTC in fees for a total of 14.93 BTC.

Dividing the 10846 hours by the 14.93 BTC gives us about 726.46 hours of mining per bitcoin earned.

726.46 hours at 1.372 kW is:
726.46 X 1.372 = 996.7 kWh per BTC.

The "cost" of mining that bitcoin will depend on what you pay for electricity.

Assuming that you can get electricity rates around $0.09/kWh...
996.7 X 0.09 = $89.70

Looks like the "cost of mining a coin" is a LOT less than "$1000 - $2000".

Perhaps you need to take a closer look at your own maths?  Did you calculate that "$1000 - $2000" yourself?  Or are you just parroting some nonsense that you read on the internet somewhere?

High fees might suite miners

High fees also suits the users.  If it didn't, they wouldn't pay it.

but it will drive the public away and they have plenty of alternatives to pick from

If the public moves away, then the quantity of transactions will drop.  If there are less transactions, then it won't cost as much (in transaction fees) to get your transaction confirmed.  If it doesn't cost as much, then the public will choose to use Bitcoin.  If the public chooses to use bitcoin, then the nuber of transactions will increase.  If the number of transactions increases, then it will cost more (in transaction fees) to get your transaction confirmed.  If it costs more to get your transaction confirmed, then the public will move away.

And with that we've come back to the beginning.

You see, there is a feedback loop built into the incentive structure of Bitcoin.  The amount of the transaction fees will ALWAYS be as high that the users are willing to pay, and will NEVER be higher.  The usage will bounce around a bit, but will eventually settle on a price where there are enough users willing to pay that price to keep the blocks full, and bitcoin will continue to operate successfully.

It is clear that you do not understand how the system works.

I know about IOTA using Tangle instead of block-chains and that the Lightning Network being used to plug the speed issue has many critics
and i cut code for a living so maybe I am closer to the truth than you think

Well, if you "cut code" for a living, you should perhaps take some time to look at the Bitcoin code and get a better understanding of how it actually works.

There are, and have been for a while now, thousands of alternatives out there.  They all do something different.  Still, Bitcoin seems to be fine.

[9kek]

Blocksize is simply too small.
2nd layer solution will help, but also with LN the 1mb limit (or about 1.7 in a full-segwit scenario) is too small. Also, 2nd layer solutions will need to do some on-chain transaction to work (ex. transactions to open/close lightning channels).
If bitcoin will get 10x users then today, opening a LN channel will cost something like 100-200$....not very pratical.
Devs must be realistic, there sould be a rule like "blocksize is doubled every x years" or something similar, so only 1 hard fork will be needed.

PS not a fan of Bcash and similar, just realism and math

[talkabout]

Hi Danny (and all the others that are well informed)

Hope you can help spread some light on a few questions.

From what I understand of the Bitcoin and cost of transaction:
1. Any transaction has to be hashed in a new block to be considered valid.
2. The amount of resources needed to mine a new block will increase exponentially in the future.

Will this not make it really cumbersome to use bitcoin as means of payment on a larger scale?

If your calculations are accurate about the cost of mining a bitcoin aprox $90 USD, how come they are trading at approx $17000 now?
Even if we say the cost is $2000 its still alot less than $17000
Maybee Im missing something obvious but please enligthen me.

Im very new here and its not my intention to offend anyone but the topic seemed right for my question, hope you don't mind.

[DannyHamilton]

1. Any transaction has to be hashed in a new block to be considered valid.

No.  A transaction is checked to be "valid" by each node as the node receives it before the transaction is ever relayed to any other node.

I think what you meant to say is that a transaction must be included in a block in the blckchain to be considered "confirmed".

2. The amount of resources needed to mine a new block will increase exponentially in the future.

No. A block generally requires the amount of resources that are used.

If it takes more than an average of 10 minutes to complete blocks (due to insufficient resources being used) then the protocol automatically reduces difficulty so that future blocks will be completed in an average of 10 minutes with the available resources.

If it takes less than an average of 10 minutes to complete blocks (due to extra resources being used) then the protocol automatically increases difficulty so that future blocks will be completed in an average of 10 minutes with the available resources.

Will this not make it really cumbersome to use bitcoin as means of payment on a larger scale?

No. 

If your calculations are accurate about the cost of mining a bitcoin aprox $90 USD, how come they are trading at approx $17000 now?

My calculations were based on a few assumptions...

1. That an intelligent miner would be mining with the most efficient equipment available.  Some people may choose to spend extra money on mining, but that isn't a requirement.
2. That there are places in the world where electricity is less than $0.09 per kWh.  Some people may choose to mine with more expensive electricity, but that isn't a requirement.

In addition, the exchange rate for bitcoins is based on supply and demand, and not based on the cost of production.  The demand for bitcoins has grown much faster in the past few years than the supply has, therefore the exchange rate has increased substantially.

[jnano]

Cost something like 120KWH of electric just to confirm a single transaction

Check your math.  You seem to have made an error in that calculation.
That electricity is being used for much more than just "confirming a single transaction".

Do you mean in the philosophical sense?

If the actual, it's not too far off. Rounding some numbers:

Current network hash rate = 13.4 EH/s
13.4 EH / 14 TH per S9 = 960K units
960K * 1.4 KWatt = 1.34 GWatt
Average block time = 1/6 hours
Average transactions per block = 3000

1.34 GWh / 6 / 3000 = 74 kWh.

I assume not everyone uses the S9, so let's add 25% to get 90 kWh/tx.

[peter0093]

yes it a problem in trisection of bitcoin

[Anti-Cen]

Looks like the "cost of mining a coin" is a LOT less than "$1000 - $2000".

No i stand by my numbers and your calculation did not deal with miners who did not win a coin

One factor that may help limit the increasing energy consumption is that if the price of bitcoin should drop drastically mining might become a losing proposition. Currently, at the average cost of $0.12 for electricity in the U.S, it costs $1,567.88 to mine one bitcoin. And at the lower cost of $0.04 available in some locations, the price to mine is just $522.62

See http://www.digitaljournal.com/tech-and-science/technology/op-ed-bitcoin-mining-consuming-more-and-more-power/article/509257

You will also find calculations for the estimated cost of each transaction using google (fixed search results)

[Anti-Cen]

If your calculations are accurate about the cost of mining a bitcoin aprox $90 USD, how come they are trading at approx $17000 now?
Even if we say the cost is $2000 its still alot less than $17000
Maybee Im missing something obvious but please enligthen me.

The actual cost is way over $2000 per coin and the reason for this is the capital cost of the equipment to compete for coins that is often
forgot about when miners talk about the profit they have made.

Typical array of 8 1080 GPU rigs tend to take 2-3 years to pay back the hardware costs that go on top of the electrical costs
and that is without factoring in GPU's over heating and going wrong since many of them are over clocked to run faster and this
results in more heat being produced.

Too me mining is like getting lots of computers to count grains of sand in a cup and the first one to get the correct result wins a price
after the other computers reach a consensus on the correct number. Block-chain has solved a few problems but has created more in the process
not that i am saying that I could come up with a better plan without taking years to do so   

[Cluster2k]

Always a pleasure for me to read @DannyHamilton's reply on the forum!

[9kek]

If the public moves away, then the quantity of transactions will drop.  If there are less transactions, then it won't cost as much (in transaction fees) to get your transaction confirmed.  If it doesn't cost as much, then the public will choose to use Bitcoin.  If the public chooses to use bitcoin, then the nuber of transactions will increase.  If the number of transactions increases, then it will cost more (in transaction fees) to get your transaction confirmed.  If it costs more to get your transaction confirmed, then the public will move away.

But this way, fewer people can use it. Isn't it better allowing more people to use bitcoin?

[MoonJeina]

Hey,
what is the current status regarding the transaction cost problem?
Is the team working on a solution?

The current status regarding the transaction cost problem is going bad .
The current scenario is pretty bad . My last transaction took like 3 days to get confirmed .  This is all due to the raging load on the block chain and sudden increase in  the number of transactions . The transaction fees is so high and it is quite  obvious when the load too much . This is the time where there are way many transactions then the number of mining pools to handle . I hope the team is working to build some more pools to divide the mining load .           

[DannyHamilton]

Looks like the "cost of mining a coin" is a LOT less than "$1000 - $2000".

No i stand by my numbers

You can "stand by your numbers" as much as you want.  It won't make them correct.

and your calculation did not deal with miners who did not win a coin

Yes. It did.

Perhaps you don't understand how proof-of-work works?

One factor that may help limit the increasing energy consumption is that if the price of bitcoin should drop drastically mining might become a losing proposition. Currently, at the average cost of $0.12 for electricity in the U.S, it costs $1,567.88 to mine one bitcoin. And at the lower cost of $0.04 available in some locations, the price to mine is just $522.62

See http://www.digitaljournal.com/tech-and-science/technology/op-ed-bitcoin-mining-consuming-more-and-more-power/article/509257

Did you think I wouldn't bother clicking your link?

From that same article:

At least one critic finds Digiconomist estimates much too high. There is an extended critique here. The critic argues that the estimates inflate the energy use by bitcoin mining considerably: "The author of the Bitcoin Energy Consumption Index makes fundamentally flawed assumptions, causing it to demonstrably overestimate the electricity consumption of Bitcoin miners

You will also find calculations for the estimated cost of each transaction using google (fixed search results)

Don't believe everything you read on the internet.

Learn math.  Learn how Bitcoin works. Do the calculations yourself.

You'll find that there are a lot of false reports on the internet that make outrageous and demonstrably false claims about the amount of electricity and/or cost involved in mining.

Are you here to learn?  Or are you just here to parrot nonsense that you've read without giving it any critical thought?

If the public moves away, then the quantity of transactions will drop.  If there are less transactions, then it won't cost as much (in transaction fees) to get your transaction confirmed.  If it doesn't cost as much, then the public will choose to use Bitcoin.  If the public chooses to use bitcoin, then the nuber of transactions will increase.  If the number of transactions increases, then it will cost more (in transaction fees) to get your transaction confirmed.  If it costs more to get your transaction confirmed, then the public will move away.

But this way, fewer people can use it. Isn't it better allowing more people to use bitcoin?

Perhaps Lightning Network will make that possible.  The design of Bitcoin since 2010 was always going to limit the number of people that have direct access to the blockchain.  This was explained to Satoshi when he added the 1 megabyte limit.  He didn't care.  He decided that it should be limited, and we are stuck with his decision now.  If we don't like it, our only choice is to create an altcoin with a better solution.

This is the time where there are way many transactions then the number of mining pools to handle . I hope the team is working to build some more pools to divide the mining load .          

Nonsense.  This has nothing to do with the number of transactions a mining pool can handle. Building more pools will not make any difference.

Bitcoin is designed to limit the number of transactions per block. That limit is exactly the same regardless of whether there is 1 person mining with 1 AntMiner S9 or if there are a billion people mining with a billion AntMiner S9 units each.

Bitcoin is also designed to limit the speed of blocks to an average of 10 minutes per block. That limit is exactly the same regardless of whether there is 1 person mining with 1 AntMiner S9 or if there are a billion people mining with a billion AntMiner S9 units each.

Check your math.  You seem to have made an error in that calculation.
That electricity is being used for much more than just "confirming a single transaction".

Do you mean in the philosophical sense?

If the actual, it's not too far off. Rounding some numbers:

Current network hash rate = 13.4 EH/s
13.4 EH / 14 TH per S9 = 960K units
960K * 1.4 KWatt = 1.34 GWatt
Average block time = 1/6 hours
Average transactions per block = 3000

1.34 GWh / 6 / 3000 = 74 kWh.

I assume not everyone uses the S9, so let's add 25% to get 90 kWh/tx.

When a miner adds a block to the blockchain, it doesn't ONLY add 1 confirmation to the transactions in their block.  It ALSO adds 1 confirmation to every transaction in the blockchain.

New transactions are only possible because the earlier transactions are immutable.  The earlier transactions are only immutable because of the continuing proof-of-work process.  This is why empty blocks are still valuable and useful.  The electricity being used is sustaining an international permissionless trustless immutable decentralized record keeping system.  It is securing every transaction that has ever happened, and making possible all future transactions that ever may happen.

So, if you want to use:
Average transactions per block = 3000

Then to be intellectually honest about it, at a minimum, you need to also use:
Total number of blocks in the blockchain: 499273 (as I'm typing this)

Therefore the electricity used should be divided by 3000 X 499273  and not just by 3000.