English
Chinese
Jiangzhuoer’s response to WhalePanda on leaked chatlog
in
Headlines
18
29
20933
Note: The article is translated from Jiangzhuoer’s forum post on 8btc , translation of which is released first via his twitter . The debate arise from WhalePanda’s disclosure of: Verified chatlogs: Why Jihan and Jiang want to block segwit at all cost.” – WhalePanda . The debate is heating up on 8btc forum with 67 comments so far.
I’d like to address the article and associated chat transcript published yesterday by WhalePanda, Why Jihan and Jiang want to block Segwit at all cost.
1. The source of the chat log
The chat transcripts included in WhalePanda’s article come from a Wechat group that is used by Chinese Litecoin pools and Litecoin developers to communicate with one another. There are 15 people in the group. WhalePanda selectively chose a screenshot to create the appearance that this is some kind of secret society.
2. Litecoin’s own Segwit debate
Our goals have always been explicitly clear: according to the Hong Kong agreement, everyone agreed to first activate Segwit, and one year after that to initiate a hard fork to a 2MB block size limit.
But Core did not uphold their end of the Hong Kong agreement. After it became clear that Segwit would not activate on Bitcoin, the scaling debate spilled over into the Litecoin community, and there was a push to first activate Segwit there as an example for the Bitcoin community.
As stated in my last article, Why I am Still Not Voting for Segwit:
While the cryptocurrency community’s attention was focused on the Bitcoin scaling debate, a mysterious new Litecoin developer, shaolinfry, appeared on the scene. Shaolinfry appears to be deeply familiar with Segwit, and in a short amount of time helped the rest of the LTC development team to finish writing their Segwit implementation. Once he had secured the title of “Litecoin developer,” he switched his focus to Bitcoin, proposing the “user activated soft fork” (UASF). After launching his campaign for UASF on Bitcoin, he did the same for Litecoin and piggybacked on the reputation of Charlie Lee to push for the UASF there, too.
Our wishes have always been simple and clear-cut: Segwit and a block size increase, as agreed to in the Hong Kong Roundtable agreement. Fortunately there was not the same opposition towards increasing the block size among the Litecoin community, and we were able to come to a consensus much more peacefully: Litecoin Global Roundtable Resolution. This agreement was closely modeled on Bitcoin’s Hong Kong agreement, with Segwit activation first followed by a block size increase. This fully demonstrates that the broader community is not opposed to the Hong Kong agreement, so I do not understand why Bitcoin Core has been consistently against the agreement that they already signed.
3. The “far greater cost” mentioned in the transcript
This question was answered in my very next message in the chat log:
This is going to have a direct impact on Bitcoin’s adoption of Segwit soft-fork. Very clear precedence […]
Both sides of the debate have been represented in the discussion of Segwit on Litecoin: Core dispatched shaolinfry, and the big block camp organized the “Litecoin Global Roundtable.” Fortunately, without Bitcoin Core’s stubborn opposition to a block size increase, the Litecoin community was able to easily come to an agreement on their own version of Bitcoin’s Hong Kong agreement.
WhalePanda’s article completely ignored my own explanation, and makes the claim (without any evidence to support it) that the “greater cost” I was referring to was ASICBoost. If this was the case, then the Litecoin version of the Hong Kong agreement that I have agreed to should have invalidated the “greater cost” argument. However, the “greater cost” I referred to was clearly about the implications for the larger scaling debate, and not about ASICBoost, as WhalePanda concluded.
4. The “Bitmain opposes Segwit because of ASICBoost” rumor
This rumor is easily refuted by asking: If Bitmain only opposes Segwit because of ASICBoost, then why did they agree in Hong Kong to activate Segwit before ever activating a block size increase?
I will address the technical arguments and the logic below.
ASICBoost only works with mining pool and miner collaboration.
It is easy to demonstrate that Bitmain’s miners do not have the ASICBoost feature enabled. Anyone can take any Antminer machine, point it to any pool of their choosing, or even on Core’s Segwit testnet, and see for themselves that the hashrate and power efficiency are the same, no matter where the machine is pointed. Bitmain has derived no sales benefit from ASICBoost.
ASICBoost’s theoretical and actual cost savings are not the same.
ASICBoost can theoretically deliver a 30% savings in energy consumption for the person using it. Practically, it would be more like 20%. Most people hearing this have significantly misunderstood the implications. They think that this would increase a miner’s total advantage by 20 to 30 percent. If you take the current bitcoin price, and assume a figure of 4 cents per kWh as electricity cost, then an S9 miner will cost 15.36% of its total revenue on electricity. Making the miner 20% more efficient will bring power costs down to 12.29%, delivering a total advantage of only 3%.
Do you feel deceived? Weren’t you told that Bitmain was getting a 30% advantage? How can it actually be only 3%?
Here we can dispel another myth: that Bitmain must be using ASICBoost on their own machines pointed to their own mining pool. Bitmain’s operations are not like those of a top-secret military project, they are like those of any other business enterprise. Anything the company does will involve dozens if not hundreds of employees. Every mining farm would have numerous employees who would certainly be able to figure out for themselves that the S9 machines they are running are only using 1kW instead of the expected 1.2kW. The employees would easily recognize that the 3 megawatt mining facility they work in should be able to handle 2500 S9 machines, but somehow they manage to have 3000 of them.
Bitmain obviously cannot pay their mining farm employees the millions of RMB in salary that would be required to keep quiet about such a controversial secret. How could they possibly keep such activity so well hidden from the public? Would so much trouble, secrecy, risk, and additional cost really be worth it for a mere 3% gain from their mining operations?
Bitmain supports bcoin’s EXTBLK proposal, which is also incompatible with ASICBoost
Even without ASICBoost, Bitmain’s hardware would still have the lowest power consumption and greatest sales volume of any other Bitcoin mining hardware manufacturer.
For years Bitmain has been vocally in favor of increasing the block size limit. They signed the Hong Kong agreement. They support the ToTheMoon extension block proposal. Nothing about their actions is inconsistent with this. In fact it is Bitcoin Core who have been vocally opposed to both of these things, going so far as to renege on the agreement they signed in Hong Kong. I sincerely wish that Core would give some kind of clear-cut explanation for why they do not support the Hong Kong agreement, even as the Litecoin community has shown how easy it was for everyone to come to an agreement under the exact same terms. Core providing their reasons for opposing the Hong Kong agreement would go a long way towards helping the community move forward together.
Of course as long as Core is unwilling to do so, then we are left with no choice but to look elsewhere. The EXTBLK proposal is enough to satisfy everyone’s requirements: It’s a block size increase, it activates Segwit (BIP141), it enables Lightning Network and Rootstock, and it deploys as a soft fork.
If you do not like the idea of a coin split caused by Core’s “UASF” (or DASF, Developer Activated Soft Fork), if you want more on-chain capacity, if you want Lightning Network and Rootstock, then please join me in supporting EXTBLK.
Operator of LTC1BTC.com, a highly profitable and GPU-based mining pool
COMMENTS(29)
Note: The article is translated from Jiangzhuoers forum post on 8btc , translation of which is released first via his twitter . The debate arise from WhalePandas disclosure of: Verified chatlogs: Why Jihan and Jiang want to block segwit at all cost. WhalePanda . The debate is heating up on 8btc forum with 67 comments so far.Id like to address the article and associated chat transcript published yesterday by WhalePanda, Why Jihan and Jiang want to block Segwit at all cost.1. The source of the chat logThe chat transcripts included in WhalePandas article come from a Wechat group that is used by Chinese Litecoin pools and Litecoin developers to communicate with one another. There are 15 people in the group. WhalePanda selectively chose a screenshot to create the appearance that this is some kind of secret society.2. Litecoins own Segwit debateOur goals have always been explicitly clear: according to the Hong Kong agreement, everyone agreed to first activate Segwit, and one year after that to initiate a hard fork to a 2MB block size limit.But Core did not uphold their end of the Hong Kong agreement. After it became clear that Segwit would not activate on Bitcoin, the scaling debate spilled over into the Litecoin community, and there was a push to first activate Segwit there as an example for the Bitcoin community.As stated in my last article, Why I am Still Not Voting for Segwit:http://news.8btc.com/jiangzhuoers-response-to-whalepanda-on-leaked-chatlog
Jiangzhuoers response to WhalePanda on leaked chatlog http://NEWS.8BTC.COM #Bitcoin http://news.8btc.com/jiangzhuoers-response-to-whalepanda-on-leaked-chatlog …pic.twitter.com/hcDh00ISW8
Here is the link to the original comment thread. Or you can comment here to start a discussion. Author: 8btccom
Astounding. Core released SegWit in summer of 2016. One year later would be this summer. How did Core fail to keep their agreement when jihad is the one who is not activating SegWit and the agreement was SegWit would activate first?
So Jiang agreed to segwit, then refused to signal for segwit, and now blames core for segwit not activating?
“He does not preach what he practices till he has practiced what he preaches.” Confucius
Este tipo es el administrador del tercer pool más grande en Litecoin.
I was reading exactly the same thing… I didn’t see a clear answer as to why he wouldn’t activate segwit now other than he didn’t agree with Core on the later scaling? That just seemed a very flawed argument.
La verdad es que ya no entiendo nada. Segun pone:
“Bitmain supports bcoin’s EXTBLK proposal, which is also incompatible with ASICBoost”
y
“The EXTBLK proposal is enough to satisfy everyone’s requirements: It’s a block size increase, it activates Segwit (BIP141), it enables Lightning Network and Rootstock”
Rootstock está participado por Bitmain por lo que les interesa que se solucione el tema de la malleabilidad para poder implementar una sidechain.
Esto ya parece juego de tronos.
Just UASF these scumbags off.
Core did release SegWit a year ago. According to jihad they would still have months to go before the 2mb block increase.
We will run a SegWit release in production by the time such a hard-fork is released in a version of Bitcoin Core.
https://medium.com/@bitcoinroundtable/bitcoin-roundtable-consensus-266d475a61ff
and
The Bitcoin Core contributors present at the Bitcoin Roundtable will have an implementation of such a hard-fork available as a recommendation to Bitcoin Core within three months after the release of SegWit… This hard-fork is expected to include features which are currently being discussed within technical communities, including an increase in the non-witness data to be around 2 MB
Putting aside for a second the question of whether F2Pool broke the agreement for everyone when they signalled a Classic block that one time, you can’t possibly think that the Core devs present at the meeting, and Adam Back, did their bit. There was virtually zero effort to work in public on a post SegWit HF, nor get code into a release of Core; the best they managed was Luke-Jr whipping up a last minute recommended block size decrease not at all in the spirit of the agreement.
I’m not saying Core should have included HF code in a release by now, but that was (seemingly) a condition for running SegWit code. Maybe there is some other reason 70% of the hash rate is not signalling SegWit, but you have to be honest and recognise that they (the miners) have been pretty consistent with their message — they wanted SegWit+2MHF.
Maybe the sentence is ambiguous, but it reads to me (and the miners have since said as much) that they want HF code in a release of Core before they run SegWit in production. Clearly we are a million miles away from HF code in a release of Core, so why would they signal/activate SegWit? They basically want S Lerner’s SegWit2M (or SegWit8M if you prefer).
See, rollock, they blame you and your presence out in the open, idiot …
Wow, they basically completely discredited ASICBOOST fud really easy.
If Bitmain really did use ASICBOOST, the miners would be using 1kW instead of 1.2kW of electricity. This would be easily noticed and practically impossible to hide.
Your wordpress “site” is shitting the bed. Pages do not load.
Desconocía el dato de Rootstock. Los ganglios del chino llegan a todas partes.
Still important. There’s only two options: Deeper down the rabbit hole or telling the truth.
Oh no wait… third option: BIP8 UASF
If Bitmain really did use ASICBOOST, the miners would be using 1kW instead of 1.2kW of electricity. This would be easily noticed and practically impossible to hide.
Alternatively, they’d still use 1.2kW after bringing 20% more devices online.
This would be easily noticed
By whom? Are you able to monitor their power usage? That would be… interesting.
each miners uses 1.2kW individually, that’s the spec. So it would be pretty easy for anyone to notice this difference, if the miners were only using 1kW, and they could report on it.
In that case, the entire point of all of this is that only Jihan would be using miners with the covert boosting function activated. Why would he report it?
Do you think Jihan literally runs 100,000’s of Antminers personally? He probably has hundred of employees, pretty impossible to keep that a secret.
I can only go by the publicly available estimation of his hashrate, so I have no idea how many individual devices he runs.
That said, it’s easy to run a datacenter with only a handful of employees; and, it’s even easier still to keep hidden advantages a secret using incentives, threats, and compartmentalization.
Source: My entire 20+ year career has been spent protecting extremely valuable data and secrets, and Insider Threat happens to be a specialty of mine.
ok, well Antpool and Bitmain personally probably has several separate data centers. I don’t know how many, but I’m guessing 5-10 large ones. Maybe they run the facilities like highly classified and top secret clearance type things, and employees are sworn to secrecy, to continue secret ASICBOOST. However, I also read recently that SegWit doesn’t prevent covert ASICBOOST. It just makes it less efficient, but can still be used, maybe 5-10% instead of 20%.
https://www.reddit.com/r/Bitcoin/comments/64qq5g/attempted_explanation_of_the_alleged_asicboost/dg4fspy/
Also, Antpool could still use covet asicboost with SegWit with full % efficiency, if they just do not include SegWit transactions. So, it could be empty blocks, or blocks with only non-SegWit transaction.
I’m not sure what that has to do with my original point, which was that their internal use of Asicboost for energy efficiency gains is still undetectable. (in refuting your original claim that the lower 1.0 kW would somehow be detectable)
You’ve somehow pivoted to a whistleblower argument of some sort, but that doesn’t make much sense.
by that argument the asic leak would have leaked earlier even if they are not using it given it was designed, implemented and tested on testnet
it would be noticeable if your entire server room was supposed to be running 1.2kw each, but instead they are only running 1kw each
This rumor is easily refuted by asking: If Bitmain only opposes Segwit because of ASICBoost, then why did they agree in Hong Kong to activate Segwit before ever activating a block size increase?
Maybe they weren’t aware that SegWit would block Asicboost, then they tested AsicBoost on testnet (as they already confirmed doing) and realized it didn’t work.
Again, who outside of Bitmain itself could notice and report such a thing if/when only Bitmain itself utilizes the hidden covert boost?
You’re not making sense; unless, of course, you’re still relying entirely on some sort of whistleblower working at the Bitmain mining facility?
The fact that they only really get 3% more profit certainly diminishes the the manufactured outrage about asicboost. Even if they were employing it in some fashion, it is not more of an advantage than plenty of other competitive aspects of mining.
Must have been one hell of an OMFG moment there…
Please sign in first