Gold collapsing. Bitcoin UP.

[AdrianX]

I LOL,ed at this post and the underlying truth it unveiled. The Bitcoin ledger was segregated before BS/Core activated their Segwit code.

"The Bitcoin Legacy and Bitcoin Cash chains share the role of upholding the Bitcoin ledger in its now-bifurcated form (importantly: only bifurcated for those who voluntarily gamble on fork arbitrage trading).

BTC and BCH are two sides of the same coin, as it were."

 

[Peter R]

@Peter R
I always assumed Craig Wright was exaggerating a little bit, and the exaggerated numbers we're assuming nodes where all 20,000$ machines and his test consider only a single bottleneck like sigops/sec
he made his super computer do 50,000 sigops per second, and then he says "50,000 tx/s next year"

can you prove that no computer network in the world could be made to do 50,000 tx/s by next year?

Reading his recent paper, yeah, I get the feeling that his simulation consisted of sending 340 GB files between computers and processing 100,000 sigops per second via parallelization. That this is possible should hardly be surprising, so I don't really see the point of the experiment. What exactly did we learn?

Reaching this performance with a node that we could physically connect to the Gigablock Testnet and run through a ramp from 1 tx / sec to 50,000 tx / sec would be an important feat, but I have a feeling that he's nowhere near achieving this.

 

[albin]

@Peter R

100-500 tx/s is over-the-top fantastic for the near future!

Edit: just to be clear, by "fantastic" I mean as a generic superlative, not literally rooted in fantasy!

 

[Peter R]

@albin

I think so. And even with continued exponential growth, it will likely be a decade anyways before we have even that much demand (see chart below). A decade from now, I'm confident that the 50,000 tx/sec number being thrown around will have already been achieved on realistic test networks.

 

[freetrader]

@Peter R
can you prove that no computer network in the world could be made to do 50,000 tx/s by next year?

Reversing the burden of proof won't get us ahead in a discussion.

If you do the math using ~224 bytes for a standard transaction, you end up with 6.7GB blocks for that rate if the 10-minute confirmation time is kept.

Let's set aside the matter of lack of demand for such rates for now, and the speculation that maybe enough network participants would accept decreasing the target time to 2.5 minutes or thereabouts so one wouldn't have to transfer the data in ginormous blocks.

None of the publicly released Bitcoin Cash clients are able to handle > 1GB in production as part of their current make-up. And you can bet there's some major engineering work to be done system-wide to handle high rates.

So what Craig must be talking about is a piece of the puzzle that has not been seen by the world.

Now, it's fine talking about it if it exists and is planned for rollout next year. But it's not a particularly helpful way to promote Bitcoin Cash at this time. It would be better if such claims were accompanied by plausible system designs, studies on prototypes, public demos of the tech etc.

Like others have said, if we can get from where Bitcoin is now to an organic volume of 500 tx/s next year that would already be a massive accomplishment and extremely good for Bitcoin Cash. If we can grow the network faster, even better.

Bear in mind that generating artificial traffic of 50000 tx /sec wouldn't be that hard. If someone wanted to do it just to prove a point, they could. It would be much harder and more costly to have proper infrastructure ready and deployed to handle that as a matter of course.

 

[adamstgbit]

simulation consisted of sending 340 GB files between computers and processing 100,000 sigops per second via parallelization. That this is possible should hardly be surprising, so I don't really see the point of the experiment. What exactly did we learn?

we learn the current limits on the hardware side. i guess.

None of the publicly released Bitcoin Cash clients are able to handle > 1GB in production as part of their current make-up

i guess CW has alot of faith in you bitcoin cash devs, and believes you'll be able to fix that in about a year. I guess what needs to happen is removing any hardcoded limits and allowing the software to scale up to 7billion TX / sec IF its running on hardware that is actually capable of that.


500tx/sec sounds good enough for the next few years, maybe longer depending on bitcoin cash's success ( which can potentially be extraordinary >100,000$ BCH in <3 years IS possible IMO )

 

[AdrianX]

But it's not a particularly helpful way to promote Bitcoin Cash at this time.

While I don't like the level of confutation, if handheld respectfully in a world of trolls and the media was reporting over which was a more realistic limit to be achieved by the end of next year, is it the upper bound under x circumstances or the lower bound based on empirically validated date.

It would make for a nice Juxtaposition when looking at the ridiculous limit of 1MB.
[doublepost=1511220299][/doublepost]Worst case if we can't scale that fast we'll have a fee market and a free market building layer 2.

 

[79b79aa8]

Highly recomended interview with "the CEO" of the real bitcoin:
https://www.rt.com/shows/sophieco/410363-bitcoin-economy-pirate-party/

haha i love how he's keeping the CEO title, to teach a little lesson to those who did not get it the first time around. and if some newbs are confused or put off by the monicker, sorry, let that be part of their education.

but wait! that means . . . that means . . . i friggin' nominate forkius for Chairman of the Board! do i hear a second? i mean really, the prescience has been extraordinary . . .

which brings me to: i flipped my BTG today, for equal parts BCH and USD, using TREZOR, Changelly, and my US exchange. on average i got about 1% of BTC price; there is slippage. sorry to talk my book. i only mention it because in my estimation Gold will drop fairly quickly; if you share this view you should not drag your feet on performing fork arbitrage.

EDIT: i should note that it's more liquid to swap BTG for BTC, from there you can do anything you want (you do have to deal with the fees, but obviously those get dwarfed pretty quickly).

 

[Richy_T]

Why does Craig Wright continue to make ridiculous claims? There is no way that Bitcoin Cash will be able to handle 50,000 tx/sec by next year. Does he actually believe this, or does he think it's helpful to make magnificent claims that will not come true?

I believe he just sees the BCH for as a chance to ingratiate himself with disaffected members of the community. Let's never forget that he hanged Gavin out to dry.

 

[Norway]

If someone bought 500 super computers costing 20k USD each and gave them to 50 mining pools, it would cost 10 million USD. Would that be a good enough hardware infrastructure for the world?

(Blockstream is funded with 126 million USD, twelve times more. https://www.crunchbase.com/organization/blockstream )

 

[adamstgbit]

@Norway

if we have blocks that are >1GB, then it stands to reason we'd also have 1000's of bitcoin-companies making millions, and a one time cost of 20k USD a is a tiny operating cost, for each of these companies.

But i heard about some tests of 1GB block, and i'm pretty sure they were using fairly standard computers, even a bloody laptop was able to limp along, so i hear. The operating cost for a 1GB block limit, might of been exaggerated...

sometimes i wonder wtf core going to say when the newest version of Raspberry pie + external SSD is processing 1GB blocks. seriously...

 

[79b79aa8]

what if i told you . . . it is irrelevant what core says or does not say.

 

[Christoph Bergmann]

By far the most popular Core-supporter argument I see for running full nodes is that it's the only way to "really know" which chain you're on. As if it matters that much to most users whether you know this with 99.99999% certainty or 99.999%.

Try to discuss about "who decides about the rules" / "Who's king in the land of Bitcoin", and you will always run into "non mining nodes decide about rules."

If the argument for full nodes ends with "really knowing that you have been paid", arguments about bandwidth would be ridiculous, as you only need to recreive 1 block / ten minutes to do so. The argument about bandwidth only makes a minisculous sense, if you want your node to be a decider about rules, for what it needs to upload blocks to enough peers.

BTW: I am one of the technical illiterate who was and is impressed by CSW's small world story. I found it an interesting point of view, especially compared to mesh networks like Lightning. I think the story goes to "You fools misunderstood everything, Bitcoin is not a mesh, but a small world network, which is the only reason why it is sybil-resistent, which means, non mining nodes don't add anything to Bitcoins resiliency, and if we follow the vision of non mining node fans, we would even lose this attribute ..."

This said, I think it is possible that CSW is terribly wrong when he looks for sybil resistancy in network topology instead of Proof of Work (Bitcoin) or Proof of Stake (Lightning) ...
[doublepost=1511254163][/doublepost]

Why does Craig Wright continue to make ridiculous claims? There is no way that Bitcoin Cash will be able to handle 50,000 tx/sec by next year. Does he actually believe this, or does he think it's helpful to make magnificent claims that will not come true?

It's especially weird because nChain cooperated with you researching exactly this topic, right?

 

[Zarathustra]

Now, it's fine talking about it if it exists and is planned for rollout next year. But it's not a particularly helpful way to promote Bitcoin Cash at this time. It would be better if such claims were accompanied by plausible system designs, studies on prototypes, public demos of the tech etc.

Helpful or not? That's the question. Isn't the Elon Musk way of life (always promising a bit more than probably can be achieved) more successful on the marketplace than the way of understatement?

Who is getting hundred thousands of paid pre-orders? Tesla or the legacy production chains?

 

[Peter R]

 

[awemany]

@Peter R: Very interesting. I must say that this investigation reads like a raw debug log of some low level protocol parser or similar, though. So I think I need to do a couple of rereads to digest what is going on.

I now wonder about the censorship_notifier guy. It looks like they do quite the effort there, including lots of automated searches and correlation? Or is that a huge manual effort? Does anyone know anything?

EDIT: Ok, given that I have had quite a few exchanges with nullc, I was intrigued to look into this 4n4n4 guy's post history:

He's definitely interesting: He appears to very much like the 1:4 trade-off, that, as we all know, Greg pulled out of his behind during a Bitcoin dev IRC session: https://archive.fo/79sX4

Arrogance: https://archive.fo/YsMNt

Calls Luke-Jr just "Luke" as if he knows him personally: https://archive.fo/24DIe, https://archive.fo/veu0X

Has the expected opinion set on Ver: https://archive.fo/8AFP3, Jihan and Haiyang: https://archive.fo/2tbzw

Yes, I think this could quite well be an alt of nullc. I get a 'it is his style of writing' vibe.

 

[humanitee]

https://www.reddit.com/r/btc/comments/7eil12/evidence_that_the_mods_of_rbitcoin_may_have_been/

Did you guys see this shit!? HAHA

Wait--I meant to say--LOL

 

[awemany]

@humanitee: Yup. See above post / link-to-twitter by Peter. Quite convincing.

 

[humanitee]

Ah, Tweets don't load for me on here so I generally skip them. Sorry for the repost!

 

[adamstgbit]

@awemany
ya it does read like a debug log, its hard to say if this evidence is a smoking gun, or wild speculation based on limited information.

i wonder if there is a discussion about this post on /r/bitcoin...

edit: it is :https://www.reddit.com/r/Bitcoin/comments/7ej53t/more_information_on_the_bot_attack_of_rbitcoin/
a comment that will make you lol:

This seems a lot like the blockstream emails where it says a lot of words but it never reaches a solid conclusion