Post: #1

The "Lightning Network" is a proposed solution to the Bitcoin (and by default Litecoin) scalability problem where transactions are processed off-chain through "payment channels" that only need to hit the blockchain when they are open and closed.

I researched how it works and believe it will not help scale Bitcoin. Here's a couple videos on it:









The example often used is paying for coffee. You open a payment channel with the coffee shop, buy an X amount of coffee, and then close the payment channel at which point the coffee shop gets his coins. It essentially acts as a gift card. The problem is that you will need to hit the blockchain twice, and at current fees, that means a huge transaction cost just to buy coffee. The theory is that once everyone is connected to LN, the opening and closing of payment channels will be infrequent, but merchants still have to close those channels somewhat regularly to get paid.

I'm not convinced that this will relieve existing congestion on the network, because who currently does micro transactions on Bitcoin today? Lightning Network seems to be most useful in a commercial retail environment IF bitcoin gets used in that respect, which it's currently not.

Because there are no other scaling proposals for Bitcoin other than LN, which may not be even implemented for years, I don't think Bitcoin will solve its scaling problem.

As for Litecoin, it's fast now, but if it reaches the transaction usage of Bitcoin, which may take years, it will have the same congestion problems.

Post: #2

Monero will never have the scalability problems Bitcoin and other coins have:

Quote:When transactions are announced to the Monero or Bitcoin networks, they appear as part of a ‘block’. Monero blocks are produced on average every 2 minutes, and Bitcoin blocks are produced on average every 10 minutes. Bitcoin blocks have a maximum size, so if there is no room then your transaction will be delayed. If you are desperate to have your transaction included in a Bitcoin block promptly, you will have to increase the transaction fees that you pay to the Bitcoin network.

Monero, however, has been designed to have an automatically adaptive block size limit. This means it will automatically be able to handle future increases in transaction volume by automatically expanding the size of blocks to accommodate higher future transaction volumes.

https://www.monero.how/why-monero-vs-bitcoin

Post: #3

(12-14-2017 10:14 PM)Samseau Wrote:  Monero will never have the scalability problems Bitcoin and other coins have:

Quote:When transactions are announced to the Monero or Bitcoin networks, they appear as part of a ‘block’. Monero blocks are produced on average every 2 minutes, and Bitcoin blocks are produced on average every 10 minutes. Bitcoin blocks have a maximum size, so if there is no room then your transaction will be delayed. If you are desperate to have your transaction included in a Bitcoin block promptly, you will have to increase the transaction fees that you pay to the Bitcoin network.

Monero, however, has been designed to have an automatically adaptive block size limit. This means it will automatically be able to handle future increases in transaction volume by automatically expanding the size of blocks to accommodate higher future transaction volumes.

https://www.monero.how/why-monero-vs-bitcoin

If adaptive block size limits are the solution to the scalability problem, why haven't the major blockchains (Bitcoin, Bitcoin Cash, Ethereum) implemented it?

Post: #4

Great question.

Post: #5

(12-15-2017 12:44 AM)Samseau Wrote:  Great question.

Here's the answer

https://cointelegraph.com/news/satoshis-...block-size

In a nutshell: there are two issues:

1) dynamic block sizes can allow miners to "mine" blocks that other miners may not accept:

i.e. say I'm miner A and you're miner B. There's a huge log of transactions that need to be put in the next block. I end up "winning" by mining the next block (remember mining is really a game of probability, sometimes I get the block, sometimes you do and the proportion is based on our respective computing power). I mine a huge block, 100MB, and take all those transactions and I get the associated fees. You get pissed off because I basically got all the transaction fees from my massive block, while you might not get any in the next block you mine. You can end up with some serious infighting, and a potential split in chains.

Having a fixed block size (or block gas limit in Ethereum) means everyone knows the agreement and mines a limited sized block despite the incentive to add more transactions to the block (block size = # of bytes in a block, each transaction is a certain amount of bytes).

2. Propagation speed - the network needs to catch up and verify the block you mined. Bitcoin blocks are 10 minutes on average. Imagine you mine a massive block and you send it to the network, but it takes 40 minutes to download the new block information. In that time, someone else could've mined a different block and started propagating that block. At some point you might end up with chaos as you've basically created different chains because older blocks aren't propagating fast enough.

Satoshi, the creator of Bitcoin, put in a 1MB block size limit for these reasons. In the years since, computing power and internet speed have improved to allow block sizes up to 8MB (or at least that's what the Chinese miners are comfortable with, in Bitcoin Cash).

In the future, block size limits may go to again, but it can really only go up at the same rate as download speed goes up.

Hence the challenge of finding other ways to scale.

Post: #6

^ These seem like minor issues. The miners suffer a bit more, but the users have a much better experience. Since users are more important than miners, the potential conflicts for minors are outweighed by the benefits for users.

Quote:In the future, block size limits may go to again, but it can really only go up at the same rate as download speed goes up.

Internet providers are always getting faster. How is it worth compromising the blockchain for the sake of technical limitations that may be improved within a few years, such as bandwidth speeds?

I remember reading in the 90's about how no one knew how the internet would be able to accommodate all the new users signing up. Turned out to be paranoia over nothing because ISPs were able to create better solutions.

Adaptable blockchain sizes might be problematic for slow bandwidth speeds, but ultimately as ISPs continually improve why would this be an issue?

Post: #7

(12-15-2017 01:46 AM)Samseau Wrote:  ^ These seem like minor issues. The miners suffer a bit more, but the users have a much better experience. Since users are more important than miners, the potential conflicts for minors are outweighed by the benefits for users.

Tell that to the miners who are actually run the network. With proof-of-work, no miners means no blockchain.

I agree that users are more important than miners, but that argument won't fly with the miners who care about their own self-interest (and why wouldn't they). Why do you think Ethereum for example is pushing towards of Proof-of-Stake model, where "mining" a block is done through staking your Ether and not by computing power? Partly it's because they've realized PoW means miners control the blockchain.

My guess with Monero is that they haven't processed anywhere near the same number of transactions as Bitcoin or Ethereum. It'd be quite fascinating to see what happens if they do.

Quote:

Quote:In the future, block size limits may go to again, but it can really only go up at the same rate as download speed goes up.

Internet providers are always getting faster. How is it worth compromising the blockchain for the sake of technical limitations that may be improved within a few years, such as bandwidth speeds?

Bitcoin does about 3-7 transactions per seconds. Ethereum 10 transactions/s.

For these blockchains to truly scale, they'll need to be able to process thousands of transactions per second at least. Visa can do about 50k/s. That's what they're aiming for, and they're aiming for it in the next 2-3 years.

Bandwidth speed isn't going to increase by 4 orders of magnitude in the next 2-3 years.

Quote:I remember reading in the 90's about how no one knew how the internet would be able to accommodate all the new users signing up. Turned out to be paranoia over nothing because ISPs were able to create better solutions.

What do you think (and I'll speak just for Ethereum) Raiden, Proof-of-stake, Plasma, and sharding are if not solutions to accommodate more transactions?

I don't disagree witth your argument here though. The scalability problem is indeed reminiscent of the bandwidth problem ISPs had. And many people are working on scaling blockchains, but I don't think it'll come from just perpetually increasing block sizes. If it really was that simple, Satoshi never would've patched Bitcoin with a 1MB block size limit, Bitcoin Cash would probably have an even larger block size limit (the fact that they went from 1MB to 'only' 8MB should tell you something), and Ethereum wouldn't even bother with sharding.

Quote:Adaptable blockchain sizes might be problematic for slow bandwidth speeds, but ultimately as ISPs continually improve why would this be an issue?

It's a good question, though I'm not sure it'd be worth the infrastructural investment for ISPs to upgrade their speeds by several orders of magnitude in the next few years. Or for that matter if it's even technically possible to increase bandwidth speed by that much.

Additionally, there's a security/centralization risk if you increase the block size too much: only nations where speed is fast enough can keep running the blockchain. This could mean that a blockchain may end up heavily concentrated in 2-3 countries, maybe even less. All you need then is have those countries make it illegal to run said blockchain and it's over. Chinese miners already dominate mining. In a few years, they might control the vast majority of mining - and the Chinese government isn't exactly known for its love of freedom.

Btw it's not just internet speed. It's also the capabilities of the computers that work as nodes (verifying transactions). Running a full node means you have to download the entire blockchain, which often runs in the 10GB range. I think Bitcoin is already 25GB and it's had an exponential growth. If it keeps up, only people with TBs worth of free space and the processing power for it will be able to run a node. Keep it up and at some point it'll only be large entities that can run a node.

Post: #8

Bitcoin transaction fees are outrageous. These days you have to transfer at least $1000 to be reasonable.

Lightning Network is the best scaling solution I've seen but has it flaws (supposedly centralization). I'm not technically apt to say if it'll work fine in a real world scenario but just increasing the blockchain size won't do.

Post: #9

Quote:i.e. say I'm miner A and you're miner B. There's a huge log of transactions that need to be put in the next block. I end up "winning" by mining the next block (remember mining is really a game of probability, sometimes I get the block, sometimes you do and the proportion is based on our respective computing power). I mine a huge block, 100MB, and take all those transactions and I get the associated fees. You get pissed off because I basically got all the transaction fees from my massive block, while you might not get any in the next block you mine. You can end up with some serious infighting, and a potential split in chains.

Another thing I forgot to mention is that Monero block sizes are prevented from increasing more than a certain amount to keep things fair for the miners.

So the block sizes increase evenly over time, not simply just making one block huge and the rest small.

https://getmonero.org/resources/monerope...ility.html

Quote:Monero has no hardcoded maximum block size, which means that unlike Bitcoin it does not have a 1 MB block size limit preventing scaling. However, a block reward penalty mechanism is built into the protocol to avoid a too excessive block size increase: The new block's size (NBS) is compared to the median size M100 of the last 100 blocks. If NBS>M100, the block reward gets reduced in quadratic dependency of how much NBS exceeds M100. E.g. if NBS is [10%, 50%, 80%, 100%] greater than M100, the nominal block reward gets reduced by [1%, 25%, 64%, 100%]. Generally, blocks greater than 2*M100 are not allowed, and blocks <= 60kB are always free of any block reward penalties.

Monero uses Proof of Work and has the scaling figured out.

As for this:

Quote:Btw it's not just internet speed. It's also the capabilities of the computers that work as nodes (verifying transactions). Running a full node means you have to download the entire blockchain, which often runs in the 10GB range. I think Bitcoin is already 25GB and it's had an exponential growth. If it keeps up, only people with TBs worth of free space and the processing power for it will be able to run a node. Keep it up and at some point it'll only be large entities that can run a node.

Monero devs are on it:

https://www.reddit.com/r/Monero/comments...y_84_with/

Quote:Basically. Thus, it could be that for the March fork we get single-output bulletproofs and for the September fork multi-output bulletproofs. For a typical transaction (2 ins + 2 outs), single-output bulletproofs translate to a ~85% reduction in transaction size, whereas multi-output bulletproofs translate to a ~90% reduction in transaction size.

The devs at Monero seem to be the smartest of the bunch. They are keeping Satoshi's vision for private decentralized banking by focusing on making the code of the Blockchain as efficient as possible instead of the shitty workarounds other coins make to their chains or miners.

Post: #10

Monero seems like they figured out scaling. With bitcoin you have to keep forking as the network grow to make bigger blocks. Bitcoin Cash has 8mb blocks but eventually it will have to fork and increase the size of blocks again. This creates more and more coins and resembles inflation.

Post: #11

To answer the thread's original question:

Many of the people in RVF slack and discord know my position on LN. I won't comment too much on the technical aspect, I'm assuming the tech will eventually work. What I am concerned about are the potential negative consequences. My understanding is that you'll have to deposit your Bitcoin with a third party. This means you'll have to disclose all your personal information to the third party - like a bank. They can freeze, lose, do what they want with your money - like a bank. If they see that you are politically affiliated with the wrong people, or you say things they don't like they can cut you off - like PayPal. Obviously the third party is going to be charging you monthly subscription fees for letting you deposit your bitcoin with them - like a bank. They might even lend out Bitcoins to other they don't even have - like a bank. They will report your transactions to the IRS - like a bank. You can see where I'm going with this, centralisation leads to the same problems we've always had, no thanks!

Others have told me that you don't have to use the LN and can send BTC the old fashioned way. Fees will be much higher and speeds will be much slower. It'll be so bad that Bitcoin will be virtually unusable without LN in the future.

The real reason for this push for LN centralisation is to remove power from the miners whom traditionally had the last word on what happens to Bitcoin to Banks. The Core developers get funding from the banks and are their puppets. That's why they have never innovated Bitcoin. An outside group of BTC holders eventually got mad and forked the coin, creating Bitcoin Cash which has 8mb blocks and has cheap and fast transaction speeds. Once the LN happens, these same core devs will be sitting back collecting monthly fees and destroy the only value proposition Bitcoin has which is decentralisation and real ownership - power over governments of the world.

Post: #12

^^

Q 7: Will there be any form of custodian risk in a Lightning Network?
Do I need to trust anyone to hold my money on my behalf?

A:
No, this system is not based on trust; you remain in full control of your money.
If anything goes wrong, you simply broadcast the latest state of your channel as a normal on-chain bitcoin transaction.
All your money will be returned to your address, and it will be recorded on the blockchain as normal.

Source:
https://medium.com/@AudunGulbrands1/ligh...bd2b957d70

Post: #13

A broadcast which costs on average $25 today and is likely to be thousands by the time LN comes around, as well taking probably days or weeks to process.

Post: #14

From that page I've got it that you either have to run a full node or "outsource" the running of the full node to a third party. That doesn't sound good, if the current blockchain is already 25GB.

So you either run your own full BTC node, or you hire a 3rd party to do it. Perhaps a good money maker for separate entities, or might it be possible that exchanges take on this type of responsibility?

Post: #15

(12-15-2017 11:37 AM)Samseau Wrote:  The devs at Monero seem to be the smartest of the bunch. They are keeping Satoshi's vision for private decentralized banking by focusing on making the code of the Blockchain as efficient as possible instead of the shitty workarounds other coins make to their chains or miners.

Oh, sharding is a shitty workaround? Do tell us why.

Quote:

Quote:Btw it's not just internet speed. It's also the capabilities of the computers that work as nodes (verifying transactions). Running a full node means you have to download the entire blockchain, which often runs in the 10GB range. I think Bitcoin is already 25GB and it's had an exponential growth. If it keeps up, only people with TBs worth of free space and the processing power for it will be able to run a node. Keep it up and at some point it'll only be large entities that can run a node.

Monero devs are on it:

https://www.reddit.com/r/Monero/comments...y_84_with/

Quote:Basically. Thus, it could be that for the March fork we get single-output bulletproofs and for the September fork multi-output bulletproofs. For a typical transaction (2 ins + 2 outs), single-output bulletproofs translate to a ~85% reduction in transaction size, whereas multi-output bulletproofs translate to a ~90% reduction in transaction size.

Very misleading. Monera transactions are ~12kB in size, compared to ~600B for Bitcoin. In other words, Monero transactions are already 20x the size of Bitcoin transactions. Reducing them from ~85 to 90% means Monero transactions will still be 2-3 times larger.

Quote:Another thing I forgot to mention is that Monero block sizes are prevented from increasing more than a certain amount to keep things fair for the miners.

So the block sizes increase evenly over time, not simply just making one block huge and the rest small.

https://getmonero.org/resources/monerope...ility.html

Quote:Monero has no hardcoded maximum block size, which means that unlike Bitcoin it does not have a 1 MB block size limit preventing scaling. However, a block reward penalty mechanism is built into the protocol to avoid a too excessive block size increase: The new block's size (NBS) is compared to the median size M100 of the last 100 blocks. If NBS>M100, the block reward gets reduced in quadratic dependency of how much NBS exceeds M100. E.g. if NBS is [10%, 50%, 80%, 100%] greater than M100, the nominal block reward gets reduced by [1%, 25%, 64%, 100%]. Generally, blocks greater than 2*M100 are not allowed, and blocks <= 60kB are always free of any block reward penalties.

Implementing this in code would take at most 20 lines. It's something a freshman in Computer Science could write. The logic is extremely straightforward. So I'll ask the question again: if scaling was really that simple, why haven't other blockchains done this yet?

And the answer is because it won't work. Your block size is limited by the median of the last 100 blocks. So you can't adjust if you suddenly have a huge number of transactions that need to be processed (for example during holiday shopping times).

Imagine your average block size is running around 100 transactions per minute. What do you think happens if there's a sudden surge in transactions? Say it unexpectedly goes up from 100 to 10,000 per minute. Since blocks with more than 200 transactions per minute get rejected, the next 50 blocks are going to be 200 transactions per minute. Then finally when the median hits 200, you can now have blocks with 400 transactions per minute. You have to keep doing this, waiting for at least 50 new blocks with maximum transactions (2*M100) to be created, before you can double again.

That's not a scalability solution, that's a scalability disaster waiting to happen.

(12-16-2017 03:28 PM)[email protected] Wrote:  Monero seems like they figured out scaling.

It hasn't. No project out there has figured out scaling. Any project claiming to is flat out lying (either to themselves or their investors/token holders).

Ethereum can process ~1.2 million transactions per day (https://etherscan.io/chart/tx)
Bitcoin can process ~400k transactions per day (https://blockchain.info/charts/n-transac...mespan=all)
Bitcoin Cash (in theory) should be able to do ~3.2 million transactions per day.

Scaling, at the level people want in order to compete with Visa and do smart contracts, would require 50k transactions per minute, or a whopping 72 million transactions per day. We're at least 2 orders of magnitude off, and that's a minimum.

How many transactions does Monero do per day?

https://bitinfocharts.com/comparison/mon...tions.html

~10k transactions per day.

Monero transaction costs:

https://www.monero.how/monero-transaction-fees

Monero (median) ~ $5
Bitcoin (median) ~ $77
(snapshot of when I looked at the website, it updates constantly for the last 100 transactions)

So Bitcoin does ~40x more transactions per day, but "only" costs 15x more.

Monero better hope their network doesn't scale too much in the next year, before they reduce their transaction size. If Monero does anywhere near the same number of transactions as Ethereum or Bitcoin, you're looking at average fees in the 100s of dollars.



From a purely point of curiosity, I'd love to see what would happen in such a scenario. Maybe, just maybe projects like Ethereum (and others) are working on the scalability problem because they realize there's a tradeoff between efficiency and privacy (why Monero transaction sizes are so large) - so you're better off solving the scalability problem before increasing the size of your transaction by a factor of 10 or even 100 for quantum-resistant algorithms in order to get privacy. Btw I think everyone is unanimous about eventually having all private transactions. Bitcoin and Ethereum will be dead-in-the-water if they don't reach that point. But scale first, then make your transactions larger because of privacy. Monero is approaching the problem from the other side: fix privacy first and then scale. That may end up biting them in the ass if their network scales too fast.

Also from:

https://steemit.com/bitcoin/@dragosroua/...o-it-seems
https://www.monero.how/monero-transaction-fees

Quote:Note that to support "microtransactions", Bitcoin and Monero are likely to support systems like the "Lightning network" in order to radically reduce the cost of small payments in the future.

I'm getting more and more convinced a major component of 'the' solution will end up being on-chain scalability, something Vitalik Buterin and Ethereum have taken the lead on with sharding.

Post: #16

What's the RVF slack channel?

Post: #17

Monero.how is not an official source of info.

Quote:Imagine your average block size is running around 100 transactions per minute. What do you think happens if there's a sudden surge in transactions? Say it unexpectedly goes up from 100 to 10,000 per minute. Since blocks with more than 200 transactions per minute get rejected, the next 50 blocks are going to be 200 transactions per minute. Then finally when the median hits 200, you can now have blocks with 400 transactions per minute. You have to keep doing this, waiting for at least 50 new blocks with maximum transactions (2*M100) to be created, before you can double again.

You've calculated median sizes incorrectly. Each new block will grow exponentially bigger.

Costs will increase linearly.

From the dev:

https://getmonero.org/2017/12/11/A-note-on-fees.html

Quote:From combining the penalty formula and the dynamic block size formula with the dynamic fee formula we can infer that a higher minimum block size limit (for example, 300 kB) leads to lower initial default fees, but fee reduction (by the dynamic fee algorithm) being somewhat "slow". By contrast, a lower minimum block size limit (for example, 150 kB) leads to higher initial default fees, but faster fee reduction.

In conclusion, whilst fees are currently too high, they, most likely, won't be anymore in the future. In addition, more research has to be conducted on the topic of the minimum block size limit, because, preferably, we'd like to use a limit that doesn't require future intervention anymore.

Fees won't be a serious issue in the long run. Nor will transaction speed due to no max block size.

Quote:Monero (median) ~ $5
Bitcoin (median) ~ $77
(snapshot of when I looked at the website, it updates constantly for the last 100 transactions)

So Bitcoin does ~40x more transactions per day, but "only" costs 15x more.

That $5 is entirely due to Tx size, bulletproofs should fix this.

Post: #18

(12-23-2017 09:43 AM)Samseau Wrote:  Monero.how is not an official source of info.

What about the tweets by Monero developer Ricardo Spagni (see the steemit link). Is that not official confirmation Monero is implementing Lightning for Monero?

Quote:

Quote:Imagine your average block size is running around 100 transactions per minute. What do you think happens if there's a sudden surge in transactions? Say it unexpectedly goes up from 100 to 10,000 per minute. Since blocks with more than 200 transactions per minute get rejected, the next 50 blocks are going to be 200 transactions per minute. Then finally when the median hits 200, you can now have blocks with 400 transactions per minute. You have to keep doing this, waiting for at least 50 new blocks with maximum transactions (2*M100) to be created, before you can double again.

You've calculated median sizes incorrectly. Each new block will grow exponentially bigger.

Doesn't seem to be the case right now:

https://bitinfocharts.com/comparison/mon...ze.html#3m

Block sizes seem to fluctuate quite a bit over time, though overall trend is upwards (no doubt as more and more people use Monero).

Quote:Costs will increase linearly.

From the dev:

https://getmonero.org/2017/12/11/A-note-on-fees.html

Quote:From combining the penalty formula and the dynamic block size formula with the dynamic fee formula we can infer that a higher minimum block size limit (for example, 300 kB) leads to lower initial default fees, but fee reduction (by the dynamic fee algorithm) being somewhat "slow". By contrast, a lower minimum block size limit (for example, 150 kB) leads to higher initial default fees, but faster fee reduction.

In conclusion, whilst fees are currently too high, they, most likely, won't be anymore in the future. In addition, more research has to be conducted on the topic of the minimum block size limit, because, preferably, we'd like to use a limit that doesn't require future intervention anymore.

This doesn't negate what I argued earlier - that if Monero sees a sudden spike in transactions, its dynamic block size won't be able to keep up and will most likely create a huge backlog.

Quote:Fees won't be a serious issue in the long run. Nor will transaction speed due to no max block size

Quote:Monero (median) ~ $5
Bitcoin (median) ~ $77
(snapshot of when I looked at the website, it updates constantly for the last 100 transactions)

So Bitcoin does ~40x more transactions per day, but "only" costs 15x more.

That $5 is entirely due to Tx size, bulletproofs should fix this.

That's conjecture. As I mentioned before, right now Monero is doing significantly less in terms of total number of transactions. If it scales to Bitcoin level volume before bulletproofs get implemented you could very well see $1xx.xx transaction fees.

This isn't to say Monero is a bad project or the people behind it are idiots. But as with all crypto projects, much is still up in the air. It still remains to be seen which project(s) come out on top.