SegWit (26.8%) vs Bitcoin Unlimited (32.2%)

[AliceWonderMiscreations]

(2) If BU hardforks whether at 51% all the way up to 100%, and you wish to follow their
chain, whether it becomes the main chain or is one of two, you will need to download
their BU client in order to transact on that chain.

Or rebuild any of the other clients changing probably just one line of code so they accept 8 MB blocks.

It is pretty evident that if you want to use an alt coin, that you need a wallet talking the protocol of that alt coin...
You can't transact litecoins with a monero wallet either...

Same port, and non BU blocks are valid on BU so no - you don't need a new client, just remove the 1 MB restriction from existing clients and they will work.

[dinofelis]

Same port, and non BU blocks are valid on BU so no - you don't need a new client, just remove the 1 MB restriction from existing clients and they will work.

That is another client.

You could just as well change the hash of the genesis block.  That's how you make a bitcoin clone.  You have another coin.

[AliceWonderMiscreations]

Well what I'll probably do if there is a fork is move all my BTC to a cold address on the BU chain, and once it confirms, fire up core where that same value hasn't moved and immediately sell them.

I bet a lot of people will.

It will be interesting to see what happens.

[AliceWonderMiscreations]

And if after I sell I buy again - its possible that the value I buy will show up on both chains. That would be cool.

[freedomno1]

(4) If the hardfork occurs, your coins will be safe as long as you do not transact until it has
been shown to be safe. You will have the same amount of coins on both chains, both
associated with your single privatekey. It is best for noobs to wait and hold their coins until
the community and devs determine it is safe to move and how to move if there are any
special precautions.

There is one area that makes me wonder though in a hard-fork scenario what happens to the unconfirmed transactions.
Unless you delete the unconfirmed transactions and put the balance back into the forked wallets balance they would be stuck but that pretty seems complex if your not running your own client and using a service instead unless they keep rebroadcasting it.
https://bitcointalk.org/index.php?topic=35214.20

[AngryDwarf]

And if after I sell I buy again - its possible that the value I buy will show up on both chains. That would be cool.

Not unless you buy a pre-fork UTXO I wouldn't have thought.

[SvenBomvolen]

Dont let the cartel from Roger Ver and Bu altcoin take over! Fight. Of course chinese miner wants bigger blocks so they can mine more blocks then people outside the china internet wall.

Yes, I even have heard some ideas of Chinese monopoly for bitcoin mining. I hope it will never happen, cause it's just not fair.
And if to talk about this results that there are in the head of the topic - the difference is just 0.3% and we can't count it as something serious.

[AliceWonderMiscreations]

And if after I sell I buy again - its possible that the value I buy will show up on both chains. That would be cool.

Not unless you buy a pre-fork UTXO I wouldn't have thought.

If the inputs are valid on both chains then the transaction may end up being relayed to both chains, is what I'm thinking.

I suspect they will have measures to prevent that though, it looks like Jihan Wu offered a bounty for someone to write a guide for exchanges on how to avoid that happening in case of a UASF prompting the need for the fork to happen early.

[Etemind]

This is probably due to AntPool starting to signal for BU, if they go trough with it we'll see a significant rise in BU hashrate. It may even cause others to jump on and get the train rolling. I agree with Amph, we need something quick, it doesn't really matter which of the two.

I agree with that. We have to make sure that problem is resolved quickly and ease the network congestion.

[dinofelis]

And if after I sell I buy again - its possible that the value I buy will show up on both chains. That would be cool.

Not unless you buy a pre-fork UTXO I wouldn't have thought.

If the inputs are valid on both chains then the transaction may end up being relayed to both chains, is what I'm thinking.

Yes.  Because a signature of a transaction of valid pre-fork UTXO is valid in both chains.  It can even be picked up from one chain (not network) and incorporated in the other one by a miner wanting your fee on his branch.

I suspect they will have measures to prevent that though, it looks like Jihan Wu offered a bounty for someone to write a guide for exchanges on how to avoid that happening in case of a UASF prompting the need for the fork to happen early.

There are two ways.  The first way is to make the signatures incompatible.  That would be a serious protocol change.  It would be the best solution, but it would be a deep modification of bitcoin's protocol, much more than simply the block size.  The simplest trick I can suggest is to flip somewhere a little-endian in a big-endian convention.  That would make the signatures, hashes, and everything different, avoid backwards compatibility and be a full, clean hard fork.  You make a signature according to one, or the other branch, and the signature is invalid (and cannot even be computed without your secret key) on the other chain.  Because, be careful.  If you only WANT to transact on one chain, it is important that not only your transaction is incompatible with the other chain protocol, but mostly, that it is cryptographically NOT POSSIBLE to have others make a transaction in your name on the other chain with the information you provide on the first.  So just superficially changing some aspects that don't change the cryptographic signature, won't do.  One can still DEDUCE your (unwanted) transaction on the other chain, and produce it - to get the fee.  So you need to modify the cryptographic signature for both.

But I think that BU will not dare to do that.  So we stay with a very messy, dangerous and backward compatible hard fork.  
Bitcoin not being able to make smart contracts, the ONLY way to separate transactions is to include some dust of a coinbase transaction with a coin mined AFTER the fork.  This UTXO not existing on the other chain, you also have a non-valid transaction on the other chain.
But it is difficult to get to these new coins.  Only exchanges can probably do so, but how are you going to get your pre-fork coins to the exchange ?  If exchanges are friendly, they let you have a "joint wallet", and they do the mixing and split for you.

[Xester]

I'm not entirely sure of what happened but It looks like Bitcoin Unlimited is seriously catching up with SegWit while SW is pretty much stable and not changing. I want to know now, If BU get 95%, is it when It will get activated and If yes, we will have to download another client or they will take control over the current Bitcoin Core Repository because It looks pretty much confusing... and should we switch our coins somewhere else after that?

Its not yet time to panic since the Segwit and Block unlimited is having a tie and there is no consensus yet. But i still bet on the 8mb increased on blocksize and hoping that there will be a miracle during the last hour. But if another choice will be approved then it is still okay for me since what we need today are solutions to the long confirmation and higher fees on transactions.

[dinofelis]

I'm not entirely sure of what happened but It looks like Bitcoin Unlimited is seriously catching up with SegWit while SW is pretty much stable and not changing. I want to know now, If BU get 95%, is it when It will get activated and If yes, we will have to download another client or they will take control over the current Bitcoin Core Repository because It looks pretty much confusing... and should we switch our coins somewhere else after that?

Its not yet time to panic since the Segwit and Block unlimited is having a tie and there is no consensus yet. But i still bet on the 8mb increased on blocksize and hoping that there will be a miracle during the last hour. But if another choice will be approved then it is still okay for me since what we need today are solutions to the long confirmation and higher fees on transactions.

My idea is that nothing will change.  Or that bitcoin will hard fork in two coins, but I don't think that will happen before it loses its first mover advantage.

[lordoliver]

I think any kind of change of the current protocol would mess up the payment systems. Doesn't matter what.
We all committed to 2 major bitcoin laws:
1. The protocol was said to be 1MB per block. Period!
2. There are only 21 Million Bitcoins. Period!

While the first one may be discussable and unimportant for the majority (I talk about users not miners!) the second one is severe.
We will never reach 100% consensus with any kind change. NEVER! That means, we will have a forked chain for sure.
And if there was a fork, we would automatically have 2 chains, with a maximum of 21 million EACH!
That means, we more or less DOUBLE the amount. Yes I know, its not a double actually, but the majority of investors don't understand that and will just think bitcoin was hacked. Consumers just don't understand that! "Which one is the right one now?"'s will be everywhere ...

We have similar problems with ETH already. The fork harmed ETH a lot. And this was although people invested in ETH are probably a lot more technical understanding already than the ones invested in Bitcoin.
If we fork bitcoin the trolling will be gigantic!

Keeping that in mind I have only one "solution" for Bitcoin as "unique Bitcoin":
We have to create a second chain (BITCOIN 2.0), where Bitcoin can MOVED to by burning it from the original chain. Only this way the number of bitcoins stays the same. Consumers will move their coins, if there is an easy way and everyone accepts the new protocol. Laggers can easily be forced by stopping support of 1.0 payments...
We can enhance the protocol in any consensus way (of course here is a lot of work hidden) and anyone can decide to move. Decision is only made by stakeholders not by miners.

There are only a few more or less acceptable problems (at least more acceptable than to double the coins):
- new and old coins are not fungible (new ones can not be converted to old ones any more)
- the bitcoin generation on the old chain will still go on. Thats why there should not be any coin generation on the new one, because that would increase the amount. For the new chain is probably only a POS/POI consensus acceptable. (I will not go into a discussion about POW against POS here. Let's just say, that with PoW we will run into a mess again)
- as a result the old chain will sooner or later just run for generating bitcoins that can be transferred. Transactions will not generate fees for miners any more so the chain will run out, if there can be generated any bitcoins any more.
- The payment providers have to implement 2.0 in advance in order for any change.

[dinofelis]

I think any kind of change of the current protocol would mess up the payment systems. Doesn't matter what.
We all committed to 2 major bitcoin laws:
1. The protocol was said to be 1MB per block. Period!
2. There are only 21 Million Bitcoins. Period!

While the first one may be discussable and unimportant for the majority the second one is severe.
We will never reach 100% consensus with any kind change. NEVER! That means, we will have a forked chain for sure.

I fully agree with you, except that 1. is now also important: it determines the fee market.


However, you proposition is not going to be accepted, because the second coin (obtained by burning bitcoins) will not be seen as bitcoin, but as an altcoin, and the first chain (true bitcoin) will keep on living.  Nobody is going to burn a real bitcoin to obtain an altcoin.

[lordoliver]

I think any kind of change of the current protocol would mess up the payment systems. Doesn't matter what.
We all committed to 2 major bitcoin laws:
1. The protocol was said to be 1MB per block. Period!
2. There are only 21 Million Bitcoins. Period!

While the first one may be discussable and unimportant for the majority the second one is severe.
We will never reach 100% consensus with any kind change. NEVER! That means, we will have a forked chain for sure.

I fully agree with you, except that 1. is now also important: it determines the fee market.


However, you proposition is not going to be accepted, because the second coin (obtained by burning bitcoins) will not be seen as bitcoin, but as an altcoin, and the first chain (true bitcoin) will keep on living.  Nobody is going to burn a real bitcoin to obtain an altcoin.

I know that its hard, but I think it can work like this:
- You make a proposal for 2.0
- You ask major payment processors and exchanges, if they support it. They have to implement in advance and sign a contract, that they will stop support for 1.0 after some time
- You invite the major stakeholders (vinclevoss, roger ver, major exchanges) to a consensus party where everyone has to click on the button so that most of the coins are moved...

I mean its a change of a protocol. Its hard but not impossible...

[Kprawn]

I cannot see how a important change like this can be done without 95% consensus... I see people are saying that consensus can be reached at a

50%+ consensus for BU? { Please explain, because I am missing that point } ... I just love the fact that it is not easy to change important code in

the protocol, otherwise miners would have dominated these decisions... we still have a say. 

[lordoliver]

I cannot see how a important change like this can be done without 95% consensus... I see people are saying that consensus can be reached at a

50%+ consensus for BU? { Please explain, because I am missing that point } ... I just love the fact that it is not easy to change important code in

the protocol, otherwise miners would have dominated these decisions... we still have a say. 

We don't have a say. If 51% miners decide to change to a software with a different protocol the original chain will be changed. The 95% was just implemented in the segwit software. And BU has implemented 75% as far as I know...

[dinofelis]

I cannot see how a important change like this can be done without 95% consensus... I see people are saying that consensus can be reached at a

50%+ consensus for BU? { Please explain, because I am missing that point } ... I just love the fact that it is not easy to change important code in

the protocol, otherwise miners would have dominated these decisions... we still have a say.  

We don't have a say. If 51% miners decide to change to a software with a different protocol the original chain will be changed. The 95% was just implemented in the segwit software. And BU has implemented 75% as far as I know...

No, this is not true.  You only have no say with Segwit.  If 51% of the miners implement segwit, all of them have to follow or get orphaned, and there will be only one chain.  Whether your node supports segwit or not, only one single chain will be produced.  No choice.  Only miner majority.  Core is nice to not activate before 95%, but they don't have to.  51% is enough.

However, with BU, you have a choice.  BU being a HARD fork, (but backward compatible), it can make a NEW CHAIN forking off the normal bitcoin chain if it has 51% mining hash rate or more (if it has less, it will, being backward compatible, always be orphaned as it makes invalid blocks for the majority, but the standard chain is accepted by BU miners, which is the longest one).

So if BU goes beyond 51%, it makes a new chain WITH A NEW COIN, call it butcoin.  Former bitcoin holders now hold their former stash of bitcoin, and a new, equal stash of butcoin.  Bitcoin has forked.  If exchanges include this butcoin, people can exchange butcoin for bitcoin and vice versa.  The market cap of bitcoin will split over bitcoin and butcoin.  If butcoin's market cap remains over 51% of the total, miners will stay in the majority with butcoin.  We now have two coins: the original bitcoin, and an altcoin, butcoin.

However, if ever people sell off their butcoin for more bitcoin, and the market cap of butcoin drops, miners will move back to bitcoin (lower difficulty, higher market price).  If now, butcoin becomes minority, a major clusterfuck happens.

Indeed, more miners on bitcoin means that sooner or later, the total PoW in the bitcoin chain will overtake the total PoW in the butcoin branch.  From that moment on, butcoin miners will now accept the bitcoin chain as the valid chain, and mine on it (just to get orphaned each time by the majority: bitcoin's protocol being more severe, it is as if the former bitcoin protocol is a soft fork of butcoin so the majority wins).  It is only if they implement a full hard fork that makes the bitcoin chain incompatible with the butcoin chain that the butcoin chain will not undergo this major orphaning.

But with a BU split, the market cap will decide, and as long as the market cap of butcoin is larger than bitcoin's, it will stay alive next to bitcoin.  And if BU becomes fully incompatible with bitcoin, it will be an entirely different coin.

[lordoliver]

No, this is not true.  You only have no say with Segwit.  If 51% of the miners implement segwit, all of them have to follow or get orphaned, and there will be only one chain.  Whether your node supports segwit or not, only one single chain will be produced.  No choice.  Only miner majority.  Core is nice to not activate before 95%, but they don't have to.  51% is enough.

However, with BU, you have a choice.  BU being a HARD fork, (but backward compatible), it can make a NEW CHAIN forking off the normal bitcoin chain if it has 51% mining hash rate or more (if it has less, it will, being backward compatible, always be orphaned as it makes invalid blocks for the majority, but the standard chain is accepted by BU miners, which is the longest one).

If segwit will be used the other blocks would be orphaned with the software, thats true. But I bet at that time there will already be a software out, that fixes that and makes another altcoin with them like ETC did. So there is actually not really a difference. There will be 2 chains in either the way...
If people get crazy we maybe even have 3 chains. Classic, BU, and Segwit...

The main point is, it will always be a "double" amount of bitcoins with the fork. I suggest to think about different possibilities(https://bitcointalk.org/index.php?topic=1819160.msg18124177#msg18124177)

[dinofelis]

If segwit will be used the other blocks would be orphaned with the software, thats true. But I bet at that time there will already be a software out, that fixes that and makes another altcoin with them like ETC did.

No, because a segwit block is still a good normal block.   Because segwit is a soft fork.  Every segwit block is accepted by an old node.  Otherwise it is not a soft fork.  So as a non-segwit node, you cannot decide whether a given block is a non-segwit block.  Because segwit blocks are also OK.

You could, indeed, implement ANOTHER soft fork, a non-compatible with segwit, true.  Is that what you mean ?