Gold collapsing. Bitcoin UP.

AdrianX

Well-Known Member
Aug 28, 2015
2,097
5,797
bitco.in
Here are some interesting statistics regarding hash distribution on Slush - who have 5150 active users collectively supporting 265.4Ph/s 2.5% total hash:

https://slushpool.com/stats/?c=btc (scroll down)

Just 1.5% signaling for the NYX 2X agreement. - see Bitcoin blocksize voting results by Active Users and then check Active hash and distribution of has by user.

10% of "the pool decides" looks like a result of slush dropping BU support as a result of BIP141 rejecting blocks that don't signal Segwit.

at this time about 55% of slush will not follow the 2MB hard fork that's about 1.4% of total hashrate coming from slush is committed to 1MB blocks.
 
Last edited:

solex

Moderator
Staff member
Aug 22, 2015
1,558
4,693
nChain official statement following the 2-day workshop/meeting.
https://nchain.com/en/media/nchain-completes-workshop-bitcoin-unlimited-announces-support-bitcoin-scaling-initiatives/

PR newswire:
http://www.prnewswire.co.uk/news-releases/nchain-completes-workshop-with-bitcoin-unlimited-and-announces-support-for-bitcoin-scaling-initiatives-300494563.html

Thanks to @AdrianX @Mengerian @Peter R @Peter Tschipper for giving up their time and providing substantial input into the event.

We have two BUIPs for membership consideration which we will be raising soon.

The first is funding for servers and dedicated sysadmin for a high-powered Bitcoin testnet with a focus on onchain-scaling intended to attract participation from universities and ecosystem companies. It will enhance the current nolnet. nChain will provide matching funding and will also use this for testing.

The second is BU specific: a proposal for a New York agreement SegWit2x version of the BU client. This is for any miner or user who wants to be able to generate/validate/propagate SW witness-data extension blocks. Obviously, we expect most BU users to prefer our primary version which ignores witness-data blocks.
 

solex

Moderator
Staff member
Aug 22, 2015
1,558
4,693
No bearing. It became apparent during the discussions that it is a strategic error for BU not to participate in the NYA in good faith, especially as it looks like BS-Core won't.
To be clear: nChain are very anti-segwit and consider that it is an unnecessary kludge in Bitcoin.
 

albin

Active Member
Nov 8, 2015
931
4,008
@lunar

I'm starting to think that the Pi fetish is almost like a vanity thing, kind of like how back in the early days it felt really cool to explain to people that you can print something on a piece of paper, and it's a wallet that can receive money. The concept probably consists of useful idiots being culturally-manipulated to shape the p2p network into something conducive to the small-blocker narrative. (Even crazier I can't shake the suspicion that they're delaying as long as possible and encouraging Pi nodes to usher in a time where full nodes are so crippled that they easily run on smartphones).

Looking back, I find Jihan's tweets about just buying a cheap small form-factor PC instead to run a node a very logical counterargument. RaspPi's aren't even particularly cheap once you add on anything more than the Pi board, and realistically, you can get a crappy refurbished old regular PC for $100 or less, slap Ubuntu on it, console into it without running a display or input peripherals at all, and anybody can support way more massive blocks right now today.
 
Last edited:

torusJKL

Active Member
Nov 30, 2016
497
1,156
The second is BU specific: a proposal for a New York agreement SegWit2x version of the BU client. This is for any miner or user who wants to be able to generate/validate/propagate SW witness-data extension blocks. Obviously, we expect most BU users to prefer our primary version which ignores witness-data blocks.
@solex Will this go through a regular BUIP process and member voting?

No bearing. It became apparent during the discussions that it is a strategic error for BU not to participate in the NYA in good faith, especially as it looks like BS-Core won't.
To be clear: nChain are very anti-segwit and consider that it is an unnecessary kludge in Bitcoin.
Wouldn't validation and propagation be enough for good faith? Do we really want to be able to generate them even though we are against them?
 

AdrianX

Well-Known Member
Aug 28, 2015
2,097
5,797
bitco.in
@torusJKL
Wouldn't validation and propagation be enough for good faith?
I'd say yes, Segwit is a soft fork, and BU not being apart of the NY agreement won't be impeding the agreement by validation and propagation segwit blocks, BU is cooperating to allow it to succeed.

I don't feel we need to implement Segwit and enforce it and fores Segwit on the network**. Cooperation to allow Segwit to succeed on its own merit is my preferred approach.

Will this go through a regular BUIP process and member voting?
That was my understanding. The "our primary version" of BU is the V0.12 fork from Core, and we should work to preserve it, it's a different approach to the Core approach and I believe that gives the network more resilience.

In my mind the V0.12 fork is the base for BU, one option for the up coming BUIP is to take the btc1 implementation and add AD-EC to the implementation, this would be the version that would keep miners using AD-EC, and allow miners who want to enforce Segwit to do so, **unfortunately that is what most of our mining customers want at the moment, not all which makes me feel this is the best approach.

Another option is to take that further and add X-thin to that release, More work the risk of re-basing the 0.12 fork with 0.14.x, could would result in loss of lots of work done to date, over 2 years in fact.

From V0.12 to V0.14 Core could have introduced 0-Day bugs hidden in complexity @Justus Ranvier has called out BS/Core developers (i think Peter Todd) proposing this as a strategy and I expect it to be leveraged with brc1 given Samson and Louke-Jr have just spent a lot of effort (successfully) speeding the rummer in japan that Segwit is safe and Segwit2X is not in. I would say this could be part of their plan to get people to abandon the 2X fork and follow the 1MB chain.

I see no future for Segwit in an environment where there are no transaction limits, and I see no future for BU where there is Segwit and an enforced transaction limit. Segwit and an enforced limit facilitates cartel behavior where miners profit from cooperating to limit transaction capacity. The negative impact of this behavior is forcing future growth off chain in the pursuit of short term profit at the expense of long term survival and growth. Mike Hearn described this leading up to his rage quit.

Going back to the success formula that allowed BU to grow, we should do what is needed to allow the network to flourish and in time miners will adopt it (and abandon Segwit). BU has an industry partner who is capable of making that happen.

I would also be interested in what @theZerg and @Peter Tschipper the prominent BU developers think.
 

torusJKL

Active Member
Nov 30, 2016
497
1,156
Thanks @AdrianX for your thoughts.

I'm assuming that during the short time between the Bitcoin Cash hard fork in August and the planned 2x hard fork in November mostly idealistic big-block users will mine the BCC chain. (of course some economic driven pools as well)
During this time it should be easy to introduce new features and maybe even another hard fork if necessary.

Should the 2x hard fork not go through (which is very plausible) and BCC gain more attraction post November there will be more at stake and it will be harder to get new code accepted. (we will even get pro-Core miners back if it is economically more interesting to hash BCC)


In conclusion I think that our development effort should go into making BCC better and ready to accept the masses back in November and not work on a SegWit solution for a chain we do not believe in.

BU is still compatible with the legacy chain even with out SegWit ("thanks" to the soft fork).
 

AdrianX

Well-Known Member
Aug 28, 2015
2,097
5,797
bitco.in
I'm assuming that during the short time between the Bitcoin Cash hard fork in August and the planned 2x hard fork in November mostly idealistic big-block users will mine the BCC chain. (of course some economic driven pools as well)
I think only ideological miners will mine it to start. The miners with a significant investment will mine the most profitable chain, with the difficulty reset BCC could attract more economically motivated miners but it's going to be a small number of committed miners. There are also idealists that may attack it - at great expense :) , the BCC network would just have to wait it out.

Should the 2x hard fork not go through (which is very plausible) and BCC gain more attraction post November there will be more at stake and it will be harder to get new code accepted.
the 2MB fork is happening i cant see it being stopped, what may happen is BS/Core fundamentalists for some reason believe the 1MB is better and fork off (another ideological split) the 2MB chain may also have a Core Bug that has not been discovered yet. If that is the case the 2MB hard fork will be an Unlimited Hard fork, with a soft limit of 2MB. I think we could end up with 3 chains BCC, BTC unlimited with segwit scars, and BS/Core Coin. Whether or not BCC or Bitcoin unlimited is favored I cant project. I suspect industry will want to go with BTC unlimited, but the BCC chain need to be there as a viable option until the next 4MB hard fork, after which it may be just like LiteCoin without segwit.

In conclusion I think that our development effort should go into making BCC better and ready to accept the masses back in November and not work on a SegWit solution for a chain we do not believe in.
I some what agree, BU will defiantly have a foot in the door, and allow BCC its best chance of success, I'll be supporting the efforts (as well as holding). I think BTC unlimited has a good chance of success too. I'd like to see that as the main chain and a BS/Core fundamentalist split.

I've learned a lot of hard lessons in BTC, and I think many new comers are going to have to learn stuff the hard way. My best advice is hodl through this. Most people are emotionally invested and may be betting on the wrong outcome.
 

Norway

Well-Known Member
Sep 29, 2015
2,424
6,410
As a side note to @Peter R and CSW's interesting bet, I just want to point out something that most people get wrong:

The average time for a transaction to be included in a block (when they are not full) is 5 minutes, not 10 minutes.
 

Norway

Well-Known Member
Sep 29, 2015
2,424
6,410
Regarding the bet itself, I have a question about mining.

Do you increase the probability of finding a solution after mining for a while?

or

Is the probability of finding a solution almost* the same after mining for a while?

*) The probability is increasing but in a negligible amount.
 

Peter R

Well-Known Member
Aug 28, 2015
1,398
5,595
@Norway

Mining is a memoryless process. The next bitcoin block is aways expected 10 minutes from now. It's like a moving target. No matter how much you hash, you never get closer to the target, until -- just like seeing a shooting star -- you get lucky and solve the block.

Imagine that a block was just found. The next block is expected (as in the "expected value") 10 minutes from now. But then let's imagine 5 minutes have gone by without a solution. The block is still expected 10 minutes from now, which is 15 minutes from the starting point. Maybe another 5 minutes goes by, and still the next block is expected 10 minutes from now (20 minutes from the starting point).
[doublepost=1501241180][/doublepost]The reason the answer to the question is t = 15 minutes, is because it is stated that the selfish miner finds the solution to the next block (at height N) at t = 0. So this means then that the honest miners are still working on block N at this point. The haven't made any progress towards finding a solution because mining is memoryless. And so -- since they have 2/3rds the mining power -- the expect arrival time for a solution is t = 15 minutes.
 
Last edited:

Otaci

Member
Jul 26, 2017
74
384
Perhaps the problem is the word "expected". That's too strong a word - 10 minutes would be the average (mean?) amount of time it takes to find the solution over many, many, blocks. But it could be 1 minute, or it could be 99 minutes, for any particular block.
 

Peter R

Well-Known Member
Aug 28, 2015
1,398
5,595
Yes, the word "expected" is likely confusing for people not familiar with the terminology. The meaning of the "expected value" in this technical sense is the value you would calculate if you performed the experiment a million times, and each time measured when the honest miner finds his competing block, and then averaged all of those measurements together.