Bitcoin Forum

Bitcoin fees have been quite a problem and even though it's not that bad at the moment. However, the bitcoin network gets overloaded again during next bull run, fees will rise again. If bitcoin truly becomes mainstream will SegWit be enough to keep fees down? I know the lightning network is coming, but that only fixes payments after you have funded a channel. If bitcoin truly becomes mainstream all over the world, an obscure amount of new channels will have to be funded through the bitcoin network all the time. Thoughts?

You are asking quite complicated question in itself.

Will adopting segwit solve scalability problem - definetly NO.
Will it reduce fees - yes. However keep in mind main purpose of segwit was not in reducing fees, but rather in solving bitcoin transaction malleability problem https://en.bitcoin.it/wiki/Transaction_malleability . Fixing this problem is a necessary step to implement lightning network https://lightning.network/ that suppose to solve scalability. But SegWit itself (without lightning) won't be enough to keep low fees if Bitcoin goes mainstream.

However in today's scale, adoption of SegWit will likely to have massive impact on the fees (there is an opinion that if Coinbase alone will adopt SegWit fees will drop because Coinbase uses ancient method of sending transactions that puts huge load on the network https://bitcointalk.org/index.php?topic=2738284.0 )

Keep in mind that Lightning network is just first step of Bitcoin scalability solution. After Lightning Network will pass through excessive testing and prove it's robustness and usefulness I expect Bitcoin block size will be further raised to account for raising demand.

Fees won't go anywhere but I'm sure that community, driven by need of expanding and growth of BTC will make sure network is functional and affordable for those who want to use it.

Raising Bitcoin block size is one possible option.

Alternatively we have concepts like this, that could vastly improve Lightning Networks scalability:
https://www.tik.ee.ethz.ch/file/a20a865ce40d40c8f942cf206a7cba96/Scalable_Funding_Of_Blockchain_Micropayment_Networks%20(1).pdf

Then there's Schnorr signatures, that could increase transaction throughput by another 25% without increasing block size:
https://medium.com/@SDWouters/why-schnorr-signatures-will-help-solve-2-of-bitcoins-biggest-problems-today-9b7718e7861c

So there's still a lot of improvements that can be implemented before BTC needs to resort to a blocksize increase.

Most of the transactions during bull runs in which people see bitcoin on the national television news and FOMO to the nearest exchange (Coinbase, which is where all the noobs go) so now that Coinbase will have segwit support, and other big exchanges like Bitfinex are already ready for segwit. I guess we will need to watch and see how it goes.

But don't discard the spam factor: Roger and Jihad will be back spamming the network during the next bull run for the ATH once again, to make things worse. Hopefully with segwit they will just end up lose money for nothing.

By next year we should have lightning network pretty established as well. We are about to hit 1000 nodes already.

I think we will have a different situation, going into the next Bullrun. The MemPool is basically empty at the moment and

Coinbase is rumoured to have finalized the implementation of SegWit. {We saw 2MB blocks with SegWit, so it will alleviate

some of the scaling problems that we had.} We also have loads of LN channels added lately and they might handle some of

the smaller transactions.  :D ... Oh, it is also a bit more expensive to spam SegWit.  ;D

For now, yes and it's been proved by the fact we can use lowest possible fees (1 satoshi/byte or 0.02$ for normal TX) and get confirmation with good speed.

But no worry since other solutions such as LN will coming soon and other solutions such as MAST, Schnorr signatures/MuSig & other which can reduce transaction size and increase privacy on some conditions.
Also, i'm sure the community will agree to raise block size/weight once there's proof that majority users still can run full nodes even after increasing block size/weight, especially it will be needed if we want to see bitcoin ready for mass adoption.

Actually, I think there could be a case of serious lack of support for another round of sustained spam factor. Roger and co doesn't seem to any longer include Jihan's support, if his pinned Tweet exhorting BCH supporters to stop the fighting and hate would suggest (this came up shortly after Segwit2x was dropped).

BCH itself will be wondering if they need to any longer carry out those deliberate overload attempts: they've already got their own fork to bolster on one hand, but on the other, if they let off now, it's further evidence that they were, indeed, the ones behind all the network spam of 2016/17.

Indeed raising block size is a viable and logical option but my point it's not the only one. In fact I can't really understand seeming opposition between raising block size and Lightning network because those measures don't really in any opposition. Neither of those solution won't be crippled by implementation of other so I don't see why people keep fighting over what is best. The only reason why (my oppinion) LN should be implemented before block size increase is because LN is reversible, optional addition to network, where BlockSize change is irreversible.

Also Schnorr signatures would be a bliss in my opinion. Probably won't solve scaling but will improve performance greatly. Sadly I don't see how it can be implemented while maintaining backwards compatibility.

Given the fact that their previous spam attacks didn´t have the desired effect and Bitcoin
Cash hasn´t gained any popularity I doubt that they will try a similar attack again.

There is just no reason for them to waste lots of money - even though they get back a
share of the transaction fees that a spam attack costs - if even a congested Bitcoin
blockchain didn´t cause people to switch to using Bitcoin Cash.

Therefore I don´t think that we will see a spam attack of this proportions again.

If that would have been the alternative, people would be considering the use of BCH instead of BTC.

---

LN is one of the most difficult concepts to be obtained yet as the development is still ongoing. But it will prove to be one of the cheapest ways to get our transactions done in a matter of flash! I believe that scalability issues will still be there, but SegWit has maintained the mempool in an awesome way compared to previous days, and I will undoubtedly compare those days to have spam attacks to block the mempool with less-fee transactions so that higher fee would be paid. This looks like getting serene after the hubbub that made the miners listen to what the community wanted.

The awnser to this question heavily depends on the definition of 'enough' and the meaning of 'reduced fees'.
If segwit, for example, reduces the fees 50%.. is this enough?
Or is 1$ per transaction 'reduced enough' ?

Depending on how the lightning network will be accepted (or whether there will be different implementation of HTLCs [1] will be superior)
and on the estimated user base segwit may or may not be enough to keep the TX costs 'low enough'.


If you consider it 'safe' to leave a channel open, and not closing it too often.. the transactions fees will hardly attract attention.


[1] https://en.bitcoin.it/wiki/Hashed_Timelock_Contracts

I guess it all depends on the use the network will have. Right now we have low fees again, because there are no spam attacks going, and most users don't really do anything with their coins. The fees problem was greatly caused by exchanges like coinbase and gemini, but they seem to be finally correcting the problem by adoptiing segwit and batching their transactions. So for the time being, I think we will have no problems with fees.
In the long run, and if BTC starts getting the support from merchants again, and people start using it, then of course it wont be enough, but that was never the point. Segwit is suppose to help in that matter, but it's "true" purpose was to solve the malleability problem, so that LN would work properly (don't know the exact technical details, but I do know it's related).

I don’t think the current fee decrease is due to exchanges adopting Segwit and batching transactions.

1) Segwit adoption has not increased.
http://segwit.party/charts/

2) Num transactions has been going down but transaction value is also going down.
https://blockchain.info/charts/estimated-transaction-volume
If more transactions were batches shouldn’ t transaction value being going up?

I believe it’s just lower transaction volume.

The fees now can be much less of what they were before, you can send for just a few satoshis now the mempool and blocks are bigger. Of course there is the waiting as before, but not for days now, not for hours but some minutes on the four transactions in the last days. Be brave and take the plunge, you will be surprised.

But isn't transaction batching basically lowering the amount of transactions? so this could be part of that too.

Segwit penetration is still slow, and will remain slow until we have Core 0.16 and all major exchanges are supporting it. We are closer to that. So I would give a year after Coinbase finally accepts segwit to estimate if segwit adoption is being a failure or not.

We'll have to evaluate impact of LN now that it's on the mainnet.

And like Andreas A said... scaling is a never ending battle. And sometimes not scaling is a better move than rushing things and fucking things up.

What do you mean, that LN will hold back segwit adoption?

yes it will!

According to research released in late 2015, U.K.-based Magister Advisors, a financial institution which focuses on facilitating mergers and acquisitions in the technology sector, explained that bitcoin could become the sixth largest reserve currency by 2030.

Bitcoin’s biggest hindrance for achieving mainstream adoption is still scalability. Once the Bitcoin Core development team’s scaling and transaction malleability solution Segregated Witness (SegWit) becomes integrated by major wallet platforms and the majority of transactions are facilitated as SegWit-enabled payments, bitcoin block size and transaction fees will likely decrease significantly, making the network more attractive to new users.

The article above is from [https://coinjournal.net/can-bitcoin-become-major-reserve-currency-2030/].

A reserve currency (or anchor currency) is a currency that is held in significant quantities by governments and institutions as part of their foreign exchange reserves. The reserve currency is commonly used in international transactions, international investments and all aspects of the global economy.

I believe if above happens the hardware cost and performance would change also - resulting in low fees.
Even if the fees still high, there is no doubt bitcoin becomes main stream currency.

Two digital currency exchanges, Coinbase and Bitfinex, have already announced upgrades to SegWit.

SegWit reduces transaction costs and adds more space to the blockchain.

gemini has always been tiny so they wouldn't have any effect. they've also never batched and don't appear to be planning to and i haven't seen any mention of segwit from them.

coinbase doing segwit is cool, but them batching would be the single biggest reduction in blockchain load. again, i haven't seen them say anything about implementing this. this is bizarre as they must be throwing away huge amounts on gdax withdraws.

segwit is a step in the right direction. it's not the magic bullet. better use of the blockchain by the major polluters would help the most.

Yes Segwit can reduce fees upto a great extent for many next upcoming bull runs. But it has some of its limitations while reducing transaction fees and times because it is not created to solve these problems. While LN on the other hand which is made to solve such problems is in itself a very complex concept of taking many transactions off chain. I think we need more solutions to reduce the traffic by increasing block size on the main chain. We need things which can be implemented using soft forks. Hard forks in today's times will be useless.

SegWit has two big pros:
1) SegWit transactions are way smaller in size compared to legacy transactions
2) SegWit was the first step towards HTLC's (https://en.bitcoin.it/wiki/Hashed_Timelock_Contracts (https://en.bitcoin.it/wiki/Hashed_Timelock_Contracts))




LN does only exist because of SegWit. The lightning network is just one (of many possible) implementations of HTLC's.
If the LN gets developed further and finally adopted and used, there won't be any reason to increase the block size anymore.
Additionally basically 'increased' the amount of TX's which can fit into a block x4.




You can't increase the blocksize in a soft fork. That would result in a hard fork.
And to be exactly.. in a pretty shitty one (look at bcash).

How could that be possible, if one of segiwt main reasons, if not the mot important, was to allow for the lightning network to be implemented with no issues. Not an expert, but it was something related to the network malleability. So I think that the LN will never hold back segwit adoption. Anyway, since segwit is implemented by default on the latest core wallet release, I think it's adoption will only grow from now. As for the question in this thread, "is it capable of solving the fees problems alone"? I don't think so, since that's the reason LN is being implemented.

I do have some doubts regarding the LN and segwit. Will exchanges use the LN for trading transactions? Wouldn't they need to run very "large" channels, and be the target of hacks? I know the channel needs to be funded, so where does the coins stay? Will they be able to hold the majority of their funds on cold storage, as they should?

There are and should be doubts. There will be adversaries again probing for weakness and faults in the bitcoin, and now the LN protocols.

Meanwhile there is a huge and increasing lack of education and understanding on the protocol and the LN layer. In spite of the apparent slowness that we see in the implementation of LN, it seems that public understanding is not tracking with our technology but falls further behind.

Thus while we discuss issues with LN, BTC has just barely became something most or some people have heard of. Something that's vaguely like a stock. Maybe it's some magical internet money.

Is that gap between common understanding, and these technologies, increasing? Where does that lead?

Does it lead to some moment in which the common guy realizes there's a way he uses that bitcoin stuff to buy things with?

If so where is the technology by then?

Although I think that investors should have some decent knowledge about bitcoin technology, and this of course leads to some knowledge about segwit and LN (otherwise how will they decide to invest in bitcoin). I completely understand that it's users (and I'm separating users from investors here), won't have a great understanding on how it all works, and just want to be able to use BTC as a payment method it in a simple and easy manner.

I actually think that some of the bitcoins greatest problems to be solved, in order to get global adoption, will be related to how "common people", with no interest from tech stuff, will be able to use bitcoin. I mean, some people don't even know how to use windows well, and that's very user friendly. I know some old people that don't enjoy using debit cards, because they think it's to hard. Of course these might be extreme cases, but the gap you were mentioning will always exist, and it's perfectly normal.

Currently, segwit is enough to reduce the fees. The fees for segwit transactions is going as low as 1sat/byte and I think its reasonably fair to say that we won't have to get into the topic of looking for a solution to reduce the transaction fees for a long time. Also, with lightning network coming there isn't much need for fees to get lower as LN fixes that issue.

segwit would be enough for the time being
you can set fee less than 10s/B and still get confirmation relatively fast now
but LN and other improvements still needed for the never ending scaling problems

What do you mean with 'trading transactions' ?
The moment you are trading coins on an exchange no transaction happens.
The coins are owned by the exchange at that moment. The next TX happens when people withdraw.

An exchanges uses transaction to 1) pay withdrawals 2) move funds from/to cold wallet 3) batching/consolidating.




Just like they need to have 'large' amounts in a hot wallet to let people withdraw. They are the prior target of hacks regarding crypto currencies, regardeless of LN.




You might want to read the LN whitepaper: https://lightning.network/lightning-network-paper.pdf (https://lightning.network/lightning-network-paper.pdf)




Yes. Just like they have their hot and cold wallet now, they will have their cold wallet and fund their channel (LN hot wallet) with it.

Segwit have never promised for transactions to have lower fees, its whole pourpose is to get faster confirmation time, but it may have an affect on it some way or another, segwit promises that transactions will have their size decreased wich means it will take less space in the block making room for more transactions, so less transaction to be stuck and be unconfirmed, which is the reason behiend the fees going up.

SegWit's purpose is not to get 'faster confirmation time'. The confirmation time (aka timeframe between 2 blocks) conforms ~10 minutes (on average).
This is hardcoded into the bitcoin protocol. Only a hardfork could change this value (which would create a new coin).




SegWit doesn't 'promise' anything. SegWit basically, among other things, reduces redundant data from transactions, making them smaller in size.