Overt AsicBoost Released today?

[itzazkrit]

That is exactly what I am saying.
As you said, doing public pre-sales sets a huge milestone that MUST be met regardless of how the technology co-operates. ref the fallout from BFL's Monarchs and Bitmine.ch/AMT A1 Coincraft miners. Unfortunately input from the Engineering side of many companies - not only those involved in crypto - is ignored once the Marketing group is unleashed. In the crypto mining business pre-sales translates to anything from 'we have this idea...' (ala the GMO vaporware) to Halong's 'we have some engineering test rigs built and running' (but still a work in progress and not ready for Prime Time).

Much to their credit Canaan and yes even the 'evil' Bitmain never did that. They stay quiet while developing and once factory production trials start only then announce upcoming products.

In the beginning of their thread Halong specifically stated that they were fully funded all the way through to at least pre-production trial runs and "we have no need of further funding via pre-orders". Then went on with their spiel never to this day saying *why* they did pre-orders.

Its been so long ago that I kind of remember them saying being fully funded but not sure.  But I almost guarantee that the only full funds they had were for the chips and first pre production run and maybe small batch, and rest of money was allotted for NRE.  If I had to write a conspiracy theory book about the whole ordeal I would guess that the principles had a good chunk of cash, had the connects to get relationship with samsung and contracted innosilicon to do R&D.  Pretty sure they had no choice to do pre-sale.  Behind the scenes of the business is rarely seen, and AB prevented much of anything, so pretty much nothing at all is what we got.  Not saying they were right, but saying in their shoes with limited resources as everyone has, it was probably the only way to get miners to the market.  So you either do it the way they did, or you never get off the ground because you ran out of cash.  I think positives of increasing technology to the market outweighs the negatives of forum speak.  Has anyone gotten hurt?  Did anyone lose their money?  I think we are all still better off today than if they never came along.....

[frodocooper]

... I think we are all still better off today than if they never came along.

I disagree. Halong's method of choice for patenting their implementation of AsicBoost — the Blockchain Defensive Patent License (BDPL) — actually sets the stage for further centralization of mining technology, and not the converse as advertised by Halong and the BDPL's website.

Under the first question of the BDPL website's FAQ page — "How do you benefit by using the BDPL?" — it is stated that in return for immunity from patent-related lawsuits from other BDPL users, each BDPL user receives "royalty-free licenses from every other user's patent portfolio." This looks great at first glance on its surface, because it seems to encourage technology-sharing and prevents any one BDPL user from being a patent troll.

However, upon closer inspection, the BDPL begins to look more and more like a very cleverly disguised weapon designed to obtain competitors' valuable intellectual property for free, and then snuff them out.

We know that Halong now has their implementation of AsicBoost patented under the BDPL. If then, for example, Canaan wants to use Halong's implementation of AsicBoost, Canaan would have to join the BDPL and license all of their patented intellectual property to Halong (and other BDPL users) for free. In effect, Canaan gets AsicBoost, and Halong gets access to all of Canaan's patented technology, on which Canaan had previously spent millions, if not billions, in prior research and development. The playing field between Halong and Canaan has now been leveled, but at Canaan's expense. (It may be argued here that Canaan also has access to Halong's portfolio of patented technology under the BDPL, but at this point it is reasonable to assume that Halong doesn't have anything else to offer that is of competitive value. The BDPL only works if the playing field was already level to begin with.)

Things get worse from here. If, in the above example, Canaan now decides to pull all of their patents from the BDPL, Halong still gets to keep licensing all of Canaan's patents for free. Halong, on the other hand, is now allowed under the BDPL to impose additional terms ("fair, reasonable, and non-discriminatory terms (FRAND)" according to the BDPL's website, whatever that means) on Canaan for Canaan's continued licensing of Halong's patents. In other words, Halong now gets to use Canaan's patented technology for free, while Canaan has to pay Halong to use Halong's patented technology, in addition to whatever terms that Halong may now impose on Canaan at Halong's discretion. Again, Halong wins at Canaan's expense. If Canaan hasn't already been forced to shutter their doors at this point, they may soon have to, due to them being placed at such a competitive disadvantage as a result of the BDPL's terms.

This is but one slice of how the BDPL is designed to unfairly benefit startups like Halong — startups who dangle one carrot in exchange for a thousand pieces of gold. A better and more equitable way of democratizing access to mining technology would be to release them under one of the many free and open-source software and hardware licenses.

So no, we are not better off today than if Halong and the BDPL never came along. We may actually have regressed far more than we realize.

[philipma1957]

... I think we are all still better off today than if they never came along.....

...

So no, we are not better off today than if Halong and the BDPL never came along. We've actually regressed far more than we may realize.

Fully polarized interesting.

One) the gear needs to show it can do 10-20 better then the s-9 or the 841.

So far more hash more power used and very close to 0.10 watts a gh

And limited to two pools.  At these numbers not good and not worth modding your pool or buying.

Pretty much a bust.

Now if it tweaks with firmware and goes to 16th and 1425 watts it will cause issues.

[NotFuzzyWarm]

Frodo, for that!

That is precisely how I feel about Halong's patent and it's being BDPL along with an excellent and on-point dissection of what a company joining the BDPL pool would mean to said company.

To the best of anyone's knowledge Halong's version of AB is all they have to offer. Their "you get to use our 1 patent (and any ancillary IP we may or may not have invented/refined) in return for you giving us (and all BDPL pool members) access to ALL of your IP. Oh btw: You also have to release all of your suppliers from any restrictions, NDA's, sales agreements, etc as well" is insanely 1-sided in their favor. Your line "one carrot in exchange for a thousand pieces of gold" is a spot-on description for what BDPL is to Halong.

As an Engineer and systems designer with a few Patents to my name I categorically state that it stinks to high heaven like the huge steaming it is.

edit: Interesting that the BDPL Users List so far has only Halong and 1 other company (Little Dragon Technology LLC) on it. Ta boot, The required BDPL Offering Page for Halong dated March 6, 2018 states "Licensor does not have any patent at this time".

edit edit: Today forum user Dr. Mann wrote in the Halong thread (color emphasis mine):

Halong wrote on its blog, "After Little Dragon Technology LLC acquired the patent from the original inventors, we negotiated a license to use AsicBoost in our miners on the understanding that AsicBoost would be opened up to everyone to use, under some form of defensive patent license, in the hopes it can help protect decentralization of Bitcoin mining. " (See: https://halongmining.com/blog/.)

I checked the Statement of Information on file with the California Secretary of State for the entity Little Dragon Technology LLC, and the filing does not note any association between Little Dragon Technology LLC and Halong Mining. To that extent, I am inclined to believe that Little Dragon Technology LLC is simply a shell company with no legal connection to Halong Mining, and that Halong Mining is using the AsicBoost license just as anyone else is authorized to do. The key point is that one cannot successfully sue Little Dragon Technology LLC for valid legal claims against Halong Mining.

So the gist of that is despite everyone calling it 'Halong's patent' - it isn't actually theirs unless it turns out that Halong Mining and Little Dragon Technology LLC are actually owned by the same (unknown) folks. That then leads back to Little Dragon and their BDPL AsicBoost actually being a means for Halong to gain IP they want/need and don't feel like developing themselves.

[Thetaj]

I disagree. Halong's method of choice for patenting their implementation of AsicBoost — the Blockchain Defensive Patent License (BDPL) — actually sets the stage for further centralization of mining technology, and not the converse as advertised by Halong and the BDPL's website.


So no, we are not better off today than if Halong and the BDPL never came along. We've actually regressed far more than we may realize.

You had me at "disagree"

[itzazkrit]

Technology advancement.  Thats what I'm for.  Thats the only way that blockchain will prevail.  However it gets there, as long as it does and nobody gets burned, I'm okay with it.  So yall are saying that you would rather have waited another 6-9 months for a 10nm asic boosted miner to be available?

[NotFuzzyWarm]

Technology advancement.  Thats what I'm for.  Thats the only way that blockchain will prevail.  However it gets there, as long as it does and nobody gets burned, I'm okay with it.  So yall are saying that you would rather have waited another 6-9 months for a 10nm asic boosted miner to be available?

Coming from a miner currently running >300TH/s-- YES.

1. The latest 16nm miners still have long legs left and depending on your power cost will be good to run at least another year or more. I easily see the latest A841's I've picked up being good for at least 2.

2. As expected by those who work in the semiconductor industries outside of this mad market, moving to lower nodes do NOT in-turn make dramatic improvements in mining chip performance. So far there very little to no advantage in power eff being seen. The Halong T1 miner uses 10nm chips and even with its AB is so far showing to be - meh. Ebang's 10nm miner actually has worse efficiency than their 16nm miner has.

3. As a miner wanting to get maximum life out of my equipment, the longer delay before truly faster/more efficient gear hits the market the better. Helps slow down the diff rise at least a little.

[Steamtyme]

I disagree. Halong's method of choice for patenting their implementation of AsicBoost — the Blockchain Defensive Patent License (BDPL) — actually sets the stage for further centralization of mining technology, and not the converse as advertised by Halong and the BDPL's website.

Under the first question of the BDPL website's FAQ page — "How do you benefit by using the BDPL?" —

I have 2 things I find funny here.
1) It definitely screws over the companies that have poured a massive amount of time and effort into their R&D at first. This is done by providing the new guy with all your secrets; this just isn't done anywhere.
So I don't know why BDPL, thinks it needs to be implemented in this environment specifically to "democratize" blockchain technology.

Really though at a certain point your trade secrets are stolen on the first models that ship to the world. It's tore down and reverse engineered and changed just enough to avoid a lawsuit.

2) In the end this up and coming company then proceeds to strive for continual improvement investing their funds into R&D. Now the older companies can effectively sit on their heels and let the new patents pour in. It's not a great long term strategy but if someone else is going to put in the work and you already have the "Brand power" to coast why wouldn't you.
This is the outcome I would expect from the given example of a company choosing to leave; as opposed to them paying to use Patents they didn't deem worthy on the off chance the next development is better.

Overall short term it's a no brainer for Halong to pledge their patents; access to all those secrets for the cost of only a few words.

Much like communism being great in theory; this is an idea that on paper seems great but is a legal and logistical nightmare that isn't worth the headache for larger established companies.
The best case scenario for this project would be the smaller ASIC producers joining, and thus helping themselves push advancements and creating their own "Brand Power" to chisel more of the market share in their favor.

Thanks for the new topics to read up on gave me something to do this evening.

[frodocooper]

[...]

2) In the end this up and coming company then proceeds to strive for continual improvement investing their funds into R&D. Now the older companies can effectively sit on their heels and let the new patents pour in. It's not a great long term strategy but if someone else is going to put in the work and you already have the "Brand power" to coast why wouldn't you.
This is the outcome I would expect from the given example of a company choosing to leave; as opposed to them paying to use Patents they didn't deem worthy on the off chance the next development is better.

[...]

The issue here isn't so much whether the incumbents would get to "coast" on the back of the newcomers' R&D, should the newcomers even decide to invest in their own R&D. In fact, the way the BDPL is currently set up actually incentivizes the opposite — the newcomers would be a lot better off simply leaching off the incumbents' patents and resources. And if the incumbents decide to simply stop developing because of the burden of spoon-feeding the other BDPL users, everyone loses, because the entire BDPL pool of developers and manufacturers would grind to a halt. They would all then very soon be overtaken by non-BDPL developers and manufacturers. In other words, it would be far better for the incumbents to have never joined the BDPL in the first place.

The issue here instead is that should any company choose to leave the BDPL, they remain obliged to license their patents to BDPL users for free, while the BDPL users would then be allowed to impose additional terms on the leaving companies for their continued use of BDPL patents. This isn't just unfair, it's extortion.

[Steamtyme]

*Snip*
The issue here instead is that should any company choose to leave the BDPL, they remain obliged to license their patents to BDPL users for free, while the BDPL users would then be allowed to impose additional terms on the leaving companies for their continued use of BDPL patents. This isn't just unfair, it's extortion.

I completely agree with what you are saying, was just speculating on ways things could play out. The overall contract implied here doesn't represent a balanced trade, nor do the implications of backing out of the agreement hold a fair "punishment" for exiting the agreement. These items would have to be addressed in detail before entering the agreement.

I can't imagine why any company would sign into this agreement anyways but if they were so inclined; there would need to be a level of openness prior to releasing all your patents. For example this Asicboost, it's still early but given what I've been reading it still isn't performing as expected (Though not enough samples are in the wild to confirm this). If it doesn't live up to expectation any other party should be able to walk away with their patent rights.

Prior to openly pledging support there should also be a set fee structure in place for what you would have to pay to use any of the competitors patents. At the same time each of these companies should be able to limit how long you have free access to their "original" patents should they walk away; as well as you retaining free access to all patents you brought to the table in the first place.

Overall it's a terrible system set up to create an oligarchy instead of using competition to drive innovation. These companies will dictate among themselves the marketshare for each and will pummel any small outside producer who try's to break into the system.

Can you imagine these companies locked into this for the long haul this early into the life of Bitcoin?

[picominer]

Could Innosilicon have created Halong to avoid sharing its patents with BDPL?

There are strong indications that they are related in some ways. The following post summarizes the key points:
https://bitcointalk.org/index.php?topic=2443327.msg33551682#msg33551682

How can a startup like Halong afford to create prototype device? A mask is a multi-million dollar investment. Is it possible they acquired their chips from another manufacturer or that Halong is actually Innosilicon? My bet is on the latter.

If Innosilicon is supplying chips to Halong, shouldn’t Innosilicon also be obliged to join BDPL if they manufacture chips with the AsicBoost design?

[HyperMega]

Could Innosilicon have created Halong to avoid sharing its patents with BDPL?

There are strong indications that they are related in some ways. The following post summarizes the key points:
https://bitcointalk.org/index.php?topic=2443327.msg33551682#msg33551682

How can a startup like Halong afford to create prototype device? A mask is a multi-million dollar investment. Is it possible they acquired their chips from another manufacturer or that Halong is actually Innosilicon? My bet is on the latter.

If Innosilicon is supplying chips to Halong, shouldn’t Innosilicon also be obliged to join BDPL if they manufacture chips with the AsicBoost design?

I’m pretty sure, that Halong didn’t design and implement a 10nm ASIC without help. Design teams capable of 10nm implementations are not coming out of nowhere.

In principle Innosilicon is a design service company. They are realizing any kind of microchips for any customer who is paying them. Normally that kind of service companies are “pure-play design service”, which means that they are not releasing products labelled with their company name. Innosilicon seems not to be so strict here, they have own products too.

However, I guess that Halong contracted/paid Innosilicon to implement their 10nm ASIC as a turnkey service. Halong covered the complete NRE costs, took all the risks and is the owner of the resulting ASIC design and mask-set. The engineering was done by Inno (maybe Halong developed the ASIC architecture and parts of the front-end design).
That is the way it goes, nothing really wrong here.

[Steamtyme]

I haven't really been following this to closely, as I don't own any asic boost hardware.

Bitmain tweeted that they have new firmware to activate this for the S9's

https://twitter.com/BITMAINtech/status/1054329450018435074

What sort of hash increase can people expect? Are th eefficiency numbers going to get better for the S9's that are currently available or is this just going to be more powerful?

[HagssFIN]

Hah, I knew it.

Did they figure it out like yesterday or have they been doing that with their own machines for a long time? 

[Longsnowsm]

That is exactly my question.  What is the effect on power consumption and hash rate for this.  Is it worth even bothering to do given the S9's mine at a loss.  Is it really going to change anything?  Would love to hear more and what you have to do to use this feature.

[sarcheer]

Apparently its a simple firmware upgrade available now for the S9 from the Bitmain support page. Some testing going on now with an S9 on Kano to see if there is any noticeable improvement in hash rate or a reduction in power consumption.

Given the new firmware was given a "LPM" designation, I'm leaning towards it reducing power consumption for the same hash rate.

[philipma1957]

That is exactly my question.  What is the effect on power consumption and hash rate for this.  Is it worth even bothering to do given the S9's mine at a loss.  Is it really going to change anything?  Would love to hear more and what you have to do to use this feature.

my guess would be   a 10 to 15% improvement in efficiency .

So if your s9i  was doing 94 watts a th it would do  80-85 watts a th

but  since the m10  is the same size chip and does 70 watts a th  I guess that would be possible.

[rifleman74]

Hah, I knew it.

Did they figure it out like yesterday or have they been doing that with their own machines for a long time?  

You would have seen more AB blocks solved by Antpoo and BTC and yadda yadda.

[philipma1957]

You would have seen more AB blocks solved by Antpoo and BTC and yadda yadda.

Most big businesses are thieves and practice illegal moves all the time.  Smart ones do it in ways they do not get caught.

So they  could simply mined to an unknown address
See below

[Steamtyme]

When it comes to Bitmain, who knows what goes on behind the curtain.

I want around for the "yay" Bitmain days just, the bad reputation and ill will part of their business so far.

Is there a way for the public to tell if a block was mined by an asic boost piece of equipment? I know the pool can tell but I sort of assume that's just because the hardware is connected, and you know what hardware runs asic boost.