Gold collapsing. Bitcoin UP.

[Zarathustra]

i think the major distinction we're looking at is that there's only one set of BCH dedicated miners; Coingeek, SVPool, BMG pool. they're willing to go all out from the beginning of their existence and even now, losing money while doing so, to support the original Satoshi Vision.

Yes, but to follow Satoshi's vision in the future could also be an empty promise from Calvin. We can not know it. All we know is that today's ABC miners stabbed us in the back with empty promises and defended Segwit on tiny blocks against Bitcoin Unlimited. They voted with their CPU power, killed BU and created a minority chain and risked nothing, besides Satoshi's vision of one currency as universal sound money. That's the phyrric victory that the ABC fan boys now celebrate.

"We don't see that we don't see what we don't see." Heinz Von Foerster

 

[sickpig]

Not sure why I'm mixed into that discussion.

Because of this:

I don't have any data to provide you regarding SatoshiShotgun. Brendan Lee (@CoinstorageGuru on Twitter) can probably give you some answers.

In reply to a post of mine where I was explaining why in my opinion "ABC blacklisting the Shotgun txs" is not the reason why Nov 17th stress test failed, you chime in saying that Brendan could have some answers and facts to confirm or deny my theory.

I follow your advice and I looked Brendan tweets stream in search for an answer and I found it: according to Brendan the stress test failed because of "technical issues" on their side.

I thought it would have been a good idea to include my finding in my reply as a way to thank you to give me the right hint to where to find answers to my question.
[doublepost=1542703420,1542702341][/doublepost]

... and real big blocks. something no one thought possible.

Ok, this is not true.

Firstly, @theZerg, @Peter R, @Peter Tschipper and I set up the Gigablock Testnet, a geographically distributed actual network of nodes based on BU (8 miners and 12 transactions generators) and proved we can achieve 1GB block in size even if not in a sustained fashion.

We go through all the software bottleneck we found and remove it one by one. The most significant was the one that was limiting transactions admission to the mempool to a mere 100 transactions per second.

All that improvements have been ported to the BU code base, others are in the pipeline and we are testing it.

So we do think 64MB blocks are possible. We do think that 1GB are possible. More to the point we produced 1GB block. So saying that "no one thought" it was possible is plain wrong.

 

[Zarathustra]

All that improvements have been ported to the BU code base, others are in the pipeline and we are testing it.

So we do think 64MB blocks are possible.

Great job, @sickpig!

How did the 64MB BSV block behave on the BU nodes?

 

[freetrader]

And if you look at reddit ... they have become the new small blockers. What a pity.

Congratulations on the 64MB block for SV (and probably 128MB to come given the social media noises).

insert <look-at-me-i-am-big-blocker-now.jpg>

How soon you forget and/or shit upon the aims and achievements of Bitcoin Unlimited and the other clients that have been hard at work scaling Bitcoin Cash.

There must be something they want to win. Something big. nChain is clear about their intentions. Restoring Bitcoin 0.1, having thousands of patents. Take that like you want - both Satoshi and Faketoshi makes sense, and both would be an absolutely admirable act for the books of history. But Bitmain? What is their goal?

When I said "there will be no fork" I considered these known "Bitmain-Facts" and assumed a rational behaviour. It seems I missed something. Do they have a goal worth all this blood?

Is it this wormhole thing? I don't know much about it, and I have no clue why they need it so terribly that they are willing to burn hundreds of millions for it.

My guess is that they try a similar strategy as Blockstream: Demonstrate the ability to control a major blockchain protocol, and use this as an asset to get money from investors. This would make their bleeding a kind of investment. But it is still highly risky, while the precise goal remains unclear. As what can they sell their BCH-chain to the market?

A speculation would be that they promise to deliver the digital Yuan, powered by chinese Mining-Farms. Maybe even as a new world currency, to think big. Large mining facilities in inner China fuel the new world money.

Wormhole could be a perfect match for this venture: Bitmain holds BCH, and whenever the chinese Government, a chinese bank, or some foreign power, like a bankrupted nation, needs digital Yuan, they must buy a BCH from Bitmain, so that a new token can be created. That would be not a bad plan ... but not what I would support. But it is all pure speculation.

I also don't know that much about Wormhole, not having a great interest in it myself.
But I do know it's permissionless building on top of BCH, same as many other token solutions.
And no-one's forcing me to use it, or burn coins etc contrary to all the false narratives that have been put out by the SV side.
Anyone can fork Wormhole, anyone can apply it on BSV (well, until they soft-fork SV to oblivion w.r.t. OP_RETURN sizes etc).
Same as anyone could put a LN on top of BCH right now if they wanted.

*No-one* has shown how WHC gets any benefit from features in this BCH HF, even the OMNI devs have dismissed that (although you might argue that they could be giving a biased opinion - but then you'd need to bring some real arguments as to how it benefits).
So, I see a lot of speculation and fiction. FUD, mostly.
The conflicts between this one tokenization solution and the ones chaperoned by CoinGeek are clear.
Tokenized might be neat - I haven't had a thorough enough look at its lengthy whitepaper yet.
Some cleverer folks may tell in time if it contains IP that CoinGeek/nChain can use to prevent it being re-implemented by others. If so, it's not an open standard. I checked, and I didn't find any document licensing clauses either in the released version I was looking at. Which is a bit odd. This stuff should be right at the front of a document, or on every page. Maybe I missed it.

 

[Christoph Bergmann]

@freetrader

> How soon you forget and/or shit upon the aims and achievements of Bitcoin Unlimited and the other clients that have been hard at work scaling Bitcoin Cash.

I don't. I'm happy I got the block with two BU nodes, and I'm happy for it.

As I said, I don't know much about Wormhole, as well as I don't know if or how planned protocol changes help wormhole or not. It was not my intent to reproduce conspiracy theories, but to discuss the intentions of Bitmain ... after we discussed nChain's intention up and down, I thought this might be a good idea.

Unfortunately, your answer doesn't mention Bitmain once, but goes only about nChain, CoinGeek, FUD from nChain / CoinGeek (which I did not even mention), IP from nChain / CoinGeek (which I already mentioned) and so on. It's really sad.

 

[sickpig]

the is the sole rtoeason behind the accomplishments that are happening right now and has made it further along in this experiment than anyone like you thought he would get

@cypherdoc let me set it straight.

The 64MB (height = 557335) happened after 49 minutes from the previous block, which was 2.5MB in size, this mean that other than block propagation (like @Peter R reported) they have also problem in upping the throughput in terms of transactions propagating through the p2p network .

What I mean is that it took them 49 minutes to collect 330K txns. Which means around 110 tps, which is the same throughput achieved on the Nov 1st stress test (they reported 100 to 125 tps) when performed on the prefork network.

My question is: does it makes sense to have a 64MB block and wait for it for 50 minutes, rather than having 5 12.8 MB blocks, one every ten minutes?

I mean what really matter is not the size of the biggest block you found, but tps. if you're able to achieve a max block size of 128MB, but your tps remains capped at 100... well you didn't achieve much, imho.

 

[freetrader]

Unfortunately, your answer doesn't mention Bitmain once, but goes only about nChain, CoinGeek, FUD from nChain / CoinGeek (which I did not even mention), IP from nChain / CoinGeek (which I already mentioned) and so on. It's really sad.

The reason I don't go on and on about Bitmain is because in this conflict (even since the real small blockers) I haven't seen them behave dishonorably even once.

They make public commitments, and they stick to them.

Now on the CoinGeek, nChain side I definitely cannot say the same.

And the FUD that originates with them, gets repeated ad nauseum.

 

[Christoph Bergmann]

Without Bitmain BCH would be most likely dead. So it's better to care about them. But if you just want to focus on all the bad things nChain does - you are free to restrict your view.

 

[sickpig]

Great job, @sickpig!

How did the 64MB BSV block behave on the BU nodes?

Don't have any access to mining nodes only plain full nodes. All machine have 16GB of RAM, 320GB SSD disk, 4 cores Xeon, ntp time synced. This is what I got so far (time in UTC):

BU (in SV mode) London

2018-11-20 00:00:11.821717 UpdateTip: new best=000000000000000000d015adef111c81c84644a3a8635e6dec3ef6bd02f40e13 height=557335 bits=402847749 log2_work=87.729805 tx=276504268 date=2018-11-19 23:54:20 progress=0.999985 cache=37.2MiB(272468txo)

SV Amsterdam

2018-11-19 23:59:48.569611 UpdateTip: new best=000000000000000000d015adef111c81c84644a3a8635e6dec3ef6bd02f40e13 height=557335 version=0x20000000 log2_work=87.729805 tx=276504268 date='2018-11-19 23:54:20' progress=0.999996 cache=41.3MiB(272468txo) warning='2 of last 100 blocks have unexpected version'

SV New York

2018-11-20 00:09:31.517327 Timeout downloading block 000000000000000000d015adef111c81c84644a3a8635e6dec3ef6bd02f40e13 from peer=8492, disconnecting
2018-11-20 00:15:53.270622 UpdateTip: new best=000000000000000000d015adef111c81c84644a3a8635e6dec3ef6bd02f40e13 height=557335 version=0x20000000 log2_work=87.729805 tx=276504268 date='2018-11-19 23:54:20' progress=0.999985 cache=41.3MiB(272468txo) warning='2 of last 100 blocks have unexpected version'

As you can see the propagation time was pretty diverse.

This is the data reported by the SV explorer that I'm aware of (time displayed in your current timezone):

https://bchsvexplorer.com/block/0000000000000000004ec9fba8a5fa7139e2e4aeb094656bcbf5f7db3bcd79a8

https://www.bitfire.io/blocks/557335

 

[freetrader]

Without Bitmain BCH would be most likely dead. So it's better to care about them. But if you just want to focus on all the bad things nChain does - you are free to restrict your view.

I don't want to restrict my view.
Tell me about all the bad things Bitmain does.

That's a challenge not only to Christoph, but to all who support SV. Let's hear it, but back it up with verifiable FACTS. Not hearsay propagated by CSW or Calvin.

 

[Christoph Bergmann]

I don't say they are a evil company. But in your position I wouldn't bet on that they burn hundreds of millions just for doing something good. They are not a charity ...

 

[awemany]

Andrew's and Bitcoin Unlimited's CDS working for sound money (BCH):

https://old.reddit.com/r/btc/comments/9yrwk0/zero_conf_forfeits_zcf_on_electron_cash_pull/

 

[Norway]

Here is a polished graph of the sustained throughput on the BSV network during the stress tests. Despite the 64 MB and 38 MB blocks, sustained throughput was rarely above 10 MB per ten-minutes. The big blocks were made possible by producing them less frequently. 10 MB blocks every 10-minutes would have been more effective.

Nice data. I'm not suprised the sustained throughput is lower than the two largest blocks

It's important to understand that the stresstest is also a stresstest of the shotgun itself. You can't mine big blocks if you don't have enough transactions to mine.

 

[freetrader]

It's important to understand that the stresstest is also a stresstest of the shotgun itself.

Someone else made this analogy already elsewhere , but it's like saying "I'm going to test my shotgun by going to this mall and shooting people."

If you bring a shotgun to the mall for "testing", don't be surprised if a SWAT team takes your ass out.

 

[Norway]

@freetrader , you are seriously losing it. The stress test is nothing like that. There is no SWAT team.

 

[79b79aa8]

Tell me about all the bad things Bitmain does.

bitmain was instrumental to the creation of BCH, has supported its price by steady acquisition, and recently appears to have spent a lot of money dedicating hashpower to defend the chain.

but bitmain is looking out for bitmain, not for BCH. if BCH tanks, they'll have alternatives. if whoever is guiding their IPO -- about which the CPC will have a definitive say -- thinks their asset sheet is BCH-heavy, they will divest. when it is a public company, shareholders will apply pressure to obtain dividends, which will lead to decisions that may or may not be best for BCH as sound money. meanwhile, the BCH chain is under their benevolent control.

it is a sort of paradox. by attempting to protect something, you may end up destroying it. (that also goes for BSV.)

 

[cypherdoc]

Ok, this is not true.

Firstly, @theZerg, @Peter R, @Peter Tschipper and I set up the Gigablock Testnet, a geographically distributed actual network of nodes based on BU (8 miners and 12 transactions generators) and proved we can achieve 1GB block in size even if not in a sustained fashion.

We go through all the software bottleneck we found and remove it one by one. The most significant was the one that was limiting transactions admission to the mempool to a mere 100 transactions per second.

All that improvements have been ported to the BU code base, others are in the pipeline and we are testing it.

So we do think 64MB blocks are possible. We do think that 1GB are possible. More to the point we produced 1GB block. So saying that "no one thought" it was possible is plain wrong.

>We go through all the software bottleneck we found and remove it one by one. The most significant was the one that was limiting transactions admission to the mempool to a mere 100 transactions per second.

this is a mischaracterization. you missed this one. it only became apparent b/c of the stress test.

of course i'm provoking BU. i'm fully aware of your excellent accomplishments and work to date with the gigablock testnet. however, when the rubber met the road for you guys to step up to actually taking these big blocks and concepts live to mainnet, imo, you failed to step up. do you deny the back and forth arguments btwn the BU guys and the SV supporters have had in this thread over the last 6m or so of this debate about whether to fix the limit issue now vs later? @Peter R, @theZerg, and @solex have all come out in favor of keeping a limit in place right now in deference to the ABC/CTOR/DSV plan . and you haven't emphasized an adaptive solution either. several of you hate CSW and some have had direct conflicts with him. i'm not aware of all the gory details of those disagreements but i do think they have gotten in the way of permanently resolving the blocksize limit problem; which i contend is still the biggest problem in Bitcoin. like i keep saying, i don't care if CSW is a fraud, a cheat, a liar, a FakeSatoshi, whatever. all i care about is the current SV plan for 128MB and re-enabling opcodes. and then his further plans to remove the limit. that's good enough for me right now and i'm not afraid of forking away from his ass, if and when he decides to go rogue. the new pluses on the SV cake right now is the huge accomplishments he is proving are possible in terms of blocksizes w/o orphans.

I mean what really matter is not the size of the biggest block you found, but tps. if you're able to achieve a max block size of 128MB, but your tps remains capped at 100... well you didn't achieve much, imho.

i disagree with this. i see your solved BU tx throughput issue as a separate problem/step from blocksize propagation.

first point, i keep saying it's great how fast you guys solved the former in BU under duress. would you have ever realized it was a problem w/o the stress test? no. even Greg didn't realize he created this problem. this is what i mean when i say "necessity is the Mother of invention". all your voluntary testing and tweaking failed to find this problem; b/c you were never stressed in terms of high tx throughput. and b/c you guys volunteer and don't have the money or foresight to have tested this. and the only reason for the stress tests revealing this issue is b/c of the real big blockists wanting to probe and solve the blocksize limits. which is the biggest problem in Bitcoin.

second point, how do you know all the tx's in the 64MB were received into the mempool from the shotgun and network propagation? i've seen talk that they were self constructed. if true, the tx throuphput issue (TPS) is not the issue. furthermore, once miners have enough tx's in the mempool, or even if they don't, since they can self construct huge blocks, they push that huge block to the network and hope that it propagates w/o getting orphaned. the fact that SV was able to propagate these huge blocks successfully w/o getting them orphaned is still a huge accomplishment. if anything, i'd think the 40m delay may be from having to validate such a large block. but i'm not sure about that part.
[doublepost=1542729701][/doublepost]

@freetrader , you are seriously losing it. The stress test is nothing like that. There is no SWAT team.

if there's one thing i've noticed over the last several months is that @freetrader has stepped up his trolly hyperbole.
[doublepost=1542730320,1542729498][/doublepost]

I don't want to restrict my view.
Tell me about all the bad things Bitmain does.

That's a challenge not only to Christoph, but to all who support SV. Let's hear it, but back it up with verifiable FACTS. Not hearsay propagated by CSW or Calvin.

you said you don't know much about Wormhole. then read this and give us your thoughts. btw, i've always said i don't like the idea of burning BTC/BCH for altcoins. that hasn't changed:

https://medium.com/@craig_10243/vampire-securities-from-beyond-the-wormhole-8c4e691c809e

 

[sickpig]

@cypherdoc of all things you have said in your prev post I really want to address the following first.

We did not miss "this one". Maybe you are confusing ABC with BU.

We identified the txns throughput issue due to the tests we made in the context of the Gigablock Testnet Initiative (GTI).

Those results were presented by @Peter R and @theZerg at scaling bitcoin Standford on November 2017.

The tests took place during September and October 2017. Satoshi's Shotgun was not even an idea back then.

 

[freetrader]

@freetrader , you are seriously losing it. The stress test is nothing like that. There is no SWAT team.

Poor attempt at psych.

You're not going to get away from the fact that your "stress testers" felt the need to apply their "shotgun" (misnomer afaic) to the non-SV chain on which commercial activities were taking place after the split.
That is "the mall".

Want to play with your little gun?
Play on your testnet.
[doublepost=1542732342][/doublepost]

bitmain was instrumental to the creation of BCH, has supported its price by steady acquisition, and recently appears to have spent a lot of money dedicating hashpower to defend the chain.

but bitmain is looking out for bitmain, not for BCH. if BCH tanks, they'll have alternatives. if whoever is guiding their IPO -- about which the CPC will have a definitive say -- thinks their asset sheet is BCH-heavy, they will divest. when it is a public company, shareholders will apply pressure to obtain dividends, which will lead to decisions that may or may not be best for BCH as sound money. meanwhile, the BCH chain is under their benevolent control.

it is a sort of paradox. by attempting to protect something, you may end up destroying it. (that also goes for BSV.)

I asked about the bad things Bitmain does, not the good things it did, or the bad things it might do.

Still no-one able to give an example.
[doublepost=1542732472][/doublepost]@cypherdoc : maybe you want to supply an example instead of trolling the trolls?

 

[cypherdoc]

@cypherdoc of all things you have said in your prev post I really want to address the following first.

We did not miss "this one". Maybe you are confusing ABC with BU.

We identified the txns throughput issue due to the tests we made in the context of the Gigablock Testnet Initiative (GTI).

Those results were presented by @Peter R and @theZerg at scaling bitcoin Standford on November 2017.

The tests took place during September and October 2017. Satoshi's Shotgun was not even an idea back then.

ah yes, now i remember. the tx throughput was an issue only with ABC, is that it? well then, my apologies and great job.

still though, most of your team members are in favor of keeping the current limit right now. for me, i need to see continuous increases in the limit with each upgrade or a permanent solution to be fully onboard with a certain implementation.
[doublepost=1542732741][/doublepost]@freetrader

>You're not going to get away from the fact that your "stress testers" felt the need to apply their "shotgun" (misnomer afaic) to the non-SV chain on which commercial activities were taking place after the split.
That is "the mall".

look dude, the rest of us have already declared that all's fair in love and hashwars. i already conceded to stop harping on the checkpoint since @sickpig declared this is a war. doesn't mean i think it was a good idea. in fact, it was a terrible one considering the Solidcoin history.