Overt AsicBoost Released today?

[frodocooper]

[...]

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.

Yes, there is.

Non-AsicBoost blocks have a version number of 0x20000000. AsicBoost blocks may have different version numbers.

For example, block 546932 — mined by ViaBTC, which currently does not support AsicBoost — has a version number of 0x20000000. Block 544704, which is a known AsicBoost block, has a version number of 0x3fffe000.

[minefarmbuy]

If it benefits to mine with AB whether covert or overt, it's just bad business to not to on every piece of hardware. If this has been practice it's been a poor one.

[frodocooper]

If it benefits to mine with AB whether covert or overt, it's just bad business to not to on every piece of hardware. If this has been practice it's been a poor one.

Again, covert AsicBoost no longer works, because of segwit.

There is therefore no longer any miner that is able to use AsicBoost in secret. (This has been the case ever since block 481824, when segwit activated.) The non-standard version numbers of their blocks would immediately give them away.

[Steamtyme]

Yes, there is.

Non-AsicBoost blocks have a version number of 0x20000000. AsicBoost blocks may have different version numbers.

For example, block 546932 — mined by ViaBTC, which currently does not support AsicBoost — has a version number of 0x20000000. Block 544704, which is a known AsicBoost block, has a version number of 0x3fffe000.

Thanks for that.

I guess even with that information, you would still need another level of transparency to prove whether or not Bitmain had been using this firmware upgrade for any definitive amount of time. Just seeing blocks within their pools, or even the unknown pools can't be verified as the claim could be that it was a known asic boost miner.

With all this, I've been going back over the beginning of this thread and I admit to being lost.

IF Bitmain were to revise the physical logic inside the chips to use the Halongitosis's overt vs BM's covert AB protocols then yes they would be required to open all of their IP to the public. Don't see that happening.

How did this play into things, if the firmware upgrade works on all s9's does that mean the chips were already designed with this feature in mind?

Does this somehow violate the patent held by Halong?

[NotFuzzyWarm]

How did this play into things, if the firmware upgrade works on all s9's does that mean the chips were already designed with this feature in mind?

Does this somehow violate the patent held by Halong?

Yes the chips were designed with BM's version of AB in mind. However, after the kerfuffle over covert AB Bitmain seems to have decided to hold off turning on their overt AB until the noise over AB in general settled down AND until someone else got the pools to support it.

Why do so many folks still continue to wrongly say that Halong Mining owns the AB patent? As has been said time and again including being very clearly spelled out on the Halong site the patent is owned by Little Dragon LLC. Halong is just a licensee of it. Now as to who owns Little Dragon (and for that matter Halong) is anybody's guess as neither company wants anyone to know and both have gone to great lengths to prevent anyone from knowing. The only info on Little Dragon is that they are represented by a Chicago law firm (look on the Little Dragon LLC site) and all communications go through them.

As to violating the Patent - highly doubt it though of course I expect that to be contested. However BM made mention in their blog that the Patent has already been shot down in several jurisdictions. More to the point - you cannot Patent an end result (reuse of previous hash calcs) and there are always several ways to skin a cat. Case in point, Intel/AMD cpu's, the various makes of GPU's, etc. all do the same thing, just differently but producing the same results.

[Steamtyme]

Why do so many folks still continue to wrongly say that Halong Mining owns the AB patent?

What can I say lost in the details/lack thereof. I guess I did assume their identities as being under one umbrella, as that makes the most sense to me when I look at their overall corporate relationships. Regardless we'll see if they stick around or do anything in the future; or if they end it as an enigmatic puppet.

As to violating the Patent - highly doubt it though of course I expect that to be contested. However BM made mention in their blog that the Patent has already been shot down in several jurisdictions. More to the point - you cannot Patent an end result (reuse of previous hash calcs) and there are always several ways to skin a cat. Case in point, Intel/AMD cpu's, the various makes of GPU's, etc. all do the same thing, just differently but producing the same results.

This I find interesting and plays into my musings above. This "game changing" patent came out of nowhere and was being used to create a consortium of companies sharing patents. Now it is being shot down as non enforceable by a company that was easily able to adapt their machines to run in the same manner. In the end it looks like an improvement that was held back and used in an attempt at corporate espionage done out in the open.

I might finally have my chance to play around with AB gear as the T9+ is supposed to have some firmware drop, really can't make the machine any worse fingers crossed.

It's nice to know that there are manufacturers out there finding ways around needing this and getting great results. This is still where I plan on looking for new gear.

[NotFuzzyWarm]

This I find interesting and plays into my musings above. This "game changing" patent came out of nowhere and was being used to create a consortium of companies sharing patents.

Speaking of which, if you check the Little Dragon BDPL site for its list of licensees one company very obviously missing from the list is Innosilicon who are using AB in their T1+ and T2 series of miners. The companies that are listed have NO patents to their name to 'share' with others and aside from Halong nor do they produce/sell any miners. Are they shell companies owned by Inno or otherwise associated with them? Again, the BDPL covering AB is very obviously just an attempt to get access to other companies IP.

IMHO that rather points to Inno being behind Little Dragon and the AB patent...

[sarcheer]

So its been a week and a day since the S9 AB firmware was released... think they're working on the hashrate bug in the S9 firmware before released the AB firmware for the other BM1387 based machines?

[Steamtyme]

... one company very obviously missing from the list is Innosilicon who are using AB in their T1+ and T2 series of miners. The companies that are listed have NO patents to their name to 'share' with others and aside from Halong nor do they produce/sell any miners. Are they shell companies owned by Inno or otherwise associated with them? [...]

IMHO that rather points to Inno being behind Little Dragon and the AB patent...

Yeah I was thinking the same thing about Inno as I reviewed the hardware list I have yesterday. While it all seems to convenient I remember there being something in the Halong thread that there was some business relationship between Halong and Inno, just can't find/remember what it was. Regardless it shouldn't have been an open door to the tech unless there was something else going on behind closed doors.

Then I was also wondering about Bitfily they also produced miners that run AB, and I'm not sure who or how they fit into that list of licensees.

Regardless I guess it's safe to say any manufacturer who think s AB will improve their efficiency will be able to integrate it into their designs in the future, or at least test it and decide. That's assuming the patent continues to get tossed.