Topic: How many tx/second can we handle now? (Read 305 times)

I'm not certain it is that high. According to Visa, they can only handle 65k tps. https://usa.visa.com/dam/VCOM/download/corporate/media/visanet-technology/aboutvisafactsheet.pdf. A p2p network is usually less efficient than a centralized network, so I suspect lightning to be slightly less. However, Visa does not even go near their capacity at any time. LN will probably be able to handle more than enough to be competitive. The main hurdle LN is experiencing atm is it is far from being safe for idiots. In fact, someone should only tinker with it if their computer literacy skills are on at least a computer help desk employee level.

on block, nine would be the answer but off block with the lightning banker network, well who knows, millions I guess

I read in the posts here an underlying assumption, that bitcoin was developed to make it possible for unlimited coffees be timestamped on a blockchain. I think this was never the goal, and doesn't even make sense. So the use cases for technical or "non tech savvy" people might vary. My point of view is, that "non tech savvy" people should have at least an understanding of what they use ("a bitcoin wallet in a bitcoin universe"). This is fairly easy to use, but has it's limitations. You don't need to understand the underlying technology, but you should know, what game you are playing. One cannot expect to use bitcoin, and have same conditions as FIAT money. This would be dumb.

I do not understand fully the logic of bitcoinfees site. I would go this route:

- I buy a coffee with bitcoin, then I don't care too much on the fees, and send a std tx with one Satoshi per byte, at 4000 US$ a Satoshi is 0.00004 US$. Multiplied with the length of the tx (~300 Bytes) one would get 0.012 US$, or 1.2 cents.

- alternativly, if there is s.th. of low value, that I repeatedly order, I would use lightning. Then I have an "open channel" tx: this is not time critical, as it is done only once (maybe even two or three channels) - so I could open a channel with s.th. between 10 and 50 Satoshi/Byte.

- I buy something valuable in the range of 1000 US $. I want to make sure tx goes through within half a day, cause delivery is usually next day. Half a day is ~72 blocks, so I look into actual statistics of my full node (or https://jochen-hoenicke.de/), and I see there are several "low" level peaks per hour, and all is "fairly blue" - so I choose 10 Satoshi/Byte.

- I want to do some regular trading or exchange of amounts, therefor I transfer Satoshis/BTCs into a side channel. I open a transaction (similiar to the lightning) for a single time, and from there I'm on a different channel, and fees on bitcoin blockchain get less relevant...

As answered by ETFbitcoin, there are still "annoying weakness" (*like*), yes, I think this is true. All is still new, and as you state, probably not that easy to use, especially when going to side chains. It is hard to say, when maturity is achieved.

So yes, "for non tech savvies", maybe not the immedeate solution. For these people there is BCH (or ABC or unlimited or whatever it is now). And one shouldn't forget Ethereum, which also allows for "higher throughput" (aka tx per block per hour).

Oh, yes, both off-chain solutions are implemented. For sure there is a whitepaper for lightning (http://lightning.network/docs/), and lots of links on the net (and here in the forum). To start your own lightning node: https://medium.com/@dougvk/run-your-own-mainnet-lightning-node-2d2eab628a8b. Also interesting the graphs: https://rompert.com/recksplorer/

On the liquid side chain: https://blockstream.com/2018/10/10/liquid-launch/

I disagree since off-chain scaling solution such as LN still require on-chain transaction to open and close channel. With current 1MB 4 million weight unit and assuming everyone use LN, it's still not enough when more people uses Bitcoin.
But again, there's better alternatives such as compress transaction (such as Schnorr  Signature) and increase block weight/size should be considered as one of last solutions.


Most off-chain/side-chain solution is still young and barely used, so it's hard to to use at first. But however, i'm sure it's UI/UX would be better in future.
For example, using Bitcoin was difficult on earlier stage (2009 - 2012) as there's no friendly mobile/desktop wallet, but now using wallet is far easier.

Off-chain scaling solution such as LN already implemented, but it's still far from finished and have some annoying weakness.

I think the unlimited theory is true and will stand true, but also, not only ignoring the on-chain requirement to open/close channel and settle final books, the amount in the channel will somewhat limit the #s of simultaneous transactions... for example, if the channel only has say 1 million satoshi, presumably it can handle a max of 1 million txs each of 1 satoshi going in one way.

Practically also, LN channels have trouble with higher amounts, but this exact amount is increasing daily.

Are you sure?  Check this out https://bitcoinfees.info/


If the avg fees are almost 0.40$ now that the tx volume is very low, imagine what can it be if we get back to the same situation as last winter.
Segwit or not, that is what i read from the stats on that site

p.s. maybe you were talking about off-chain only, while the linked site only takes into account on-chain.
But are offchain transactions easy to use for non tech savvies ?
Is that solution already implemented?
And if so, post a link thanks

I'd like to add two small things:

• Lightning is having more than 300 Bitcoins in it's channels, so just to give an imagination, what can happen in volume and size off-chain.
• And then there is already the first sidechain ("Liquid"), another on/off chain layer (requires a in and out transaction on the blockchain), which allows for additional transactions outside of the bitcoin main blockchain.

So scaling bitcoin isn't a point of discussion anymore (only for those people, who want to make sure, that the coffee they paid for a dollar is forever visible in a blockchain - for tose people it is better to use bitcoin unlimited, abc or cash chains).

That depends whether we're talking about on-chain/off-chain and use SegWit/not. If we're only talking about tx/s on-chain, then it would be :
1. about 7 tx/s if all transaction don't use SegWit
2. between 12-20 tx/s if all transaction use SegWit

But if we're talking about off-chain tx/s, then in theory it's unlimited (ignoring on-chain transaction to open and close channel)

The main problem with transaction fees used to be that they can skyrocket when the volume of transactions got serious.
We had a taste of this in winter 2017/2018, the fees got too expensive because the miners could not process all of those transactions fast enough, so they gave the priority to the ones paying a high fee.

What happens now, a year later, if the transaction requests get overwhelming like in the 2017 ?
Can we handle the transactions faster now , so that the fees would be cheaper as well?