A Warning to ASIC Manufacturers.

[BobsSocks]

The company Quantum Blockchain Technologies was recently allowed a Patent on what they term ASIC Ulra Boost. The case file can be found here,

https://globaldossier.uspto.gov/details/US/18696073/A/129692

Marketing here,

https://quantumblockchaintechnologies.co.uk/patents-ip
https://quantumblockchaintechnologies.co.uk/images/QBT_ASIC_Ultra_Boost_Overview.pdf
https://quantumblockchaintechnologies.co.uk/images/ASIC_Ultra_Boost_White_Paper.pdf
https://quantumblockchaintechnologies.co.uk/images/AUB-Patent-Application-Abstract.pdf

Today they posted the following message to X,

https://x.com/QBTOfficial256/status/2093359908485730552

QBT is now planning to begin behavioural checks on commercial ASICs, screening for signatures consistent with ASIC Ultra Boost - to be carried out in our Milan lab. A black-box approach: no de-capping, just controlled inputs and observed response. A lead, not proof.

The graphic below the post includes the following words,

Planned not yet performed - any result would be circumstantial. This kind of analysis cannot see inside the silicon and any real conclusion would need further professional legal and technical review to mean anything.

This is a blatant threat to any ASIC manufacturer that should they find evidence that you are using the methods described in their allowed patent, it still needs to be granted which will happen 4 to 6 weeks after the allowance fee was paid, they will be coming after you.

In effect they seem to wish to become Patent Trolls.

Let me start by saying I am NOT a Patent Lawyer... However,

For any Patent Lawyers working for such ASIC manufacturers I would recommend that you follow the history of this application through to grant on the following pages,

https://www.search-for-intellectual-property.service.gov.uk/GB2113962.1
https://register.epo.org/application?number=EP22793447&lng=en&tab=doclist
https://globaldossier.uspto.gov/details/US/18696073/A/129692

In particular pay attention to the prosecution history for the US filing.

It is very important that you understand that as originally filed the applicants claimed many optimization methods that were ultimately disallowed by the US Patent Examiner, non final rejection and final rejection, as a result of prior art.

The US Patent Examiner insisted that the independent claims were too broad in that they would have protected this material and forced them to move the one piece of allowable material into the independent claims whilst allowing them to retain the disallowed material as dependent claims under this.

Hopefully this link will get you a copy of the Notice of Allowance that includes those claims and the wording insisted upon by the Examiner. The US website appears to use session cookies.

https://register.epo.org/ipApplication?documentId=MRWFN840X247X24&number=US.202218696073.A&patentScope=false

It is my belief that at least 50%, likely much more, of what is in the dependent claims can be implemented without infringing the restraints placed on the Independent claims and as such you will have a defensible case if you do so and also if you already have included these optimizations in your present designs.

It is my supposition that QBT have either been unable to generate any interest for their, now limited, IP amongst ASIC Manufacturers and this recent X post is PR show for prospective investors in the company.

If you are an ASIC Manufacturer then please give this message serious consideration and at least be prepared for this threat and better still work in collaboration with the community to publicize and act against it.

Give them the publicity they crave.

[NotFuzzyWarm]

QBT just refuses to give up don't they?
Up until last year they were trying to get the same thing patented in the UK & the EU and failed miserably mainly because their ideas are pure garbage. The only reason they succeeded in the US is because a US patent can be just an idea - it does not have to actually work, eg. Patents on perpetual motion and 'Free energy' ideas. (It was only fairly recently that the USPTO finally stopped accepting applications on those) The UK & EU denied their patents because there is no demonstrable evidence they could ever work.

What they have patented is nothing more than an Oracle - something that is analyzing previously mined blocks to try and find a pattern in them then apply its findings to create headers that have better chance of resulting in a found block. It is pure marketing fluff aimed at:
A. Generating more investor attention so they can scam more folks out of their money
B. The hope that miner mfgrs will capitulate and pay them to avoid the cost of (baseless) lawsuits

Problem is, THERE ARE NO PATTERNS IN MINING SHA256 COINS! Period. Unless there is a flaw in the software that generates the mining headers, it is an entirely random process. This has been cryptography proven countless times by many different researchers. That is why miners try every single possible hash when trying to solve a block.

Last year they tried to recruit both -ck and Kano who wrote cgminer to help them with this project because they could not figure out how to plug their Oracle results into cgminer to test their ideas. Both told them to fuck off because:
1. If they cannot figure out a well-documented program then they are idiots
2. There are NO PATTERNS that could be used

Because of this, no miner mfgr is using anything resembling an Oracle and never will. Ergo they cannot be successfully sued. That doesn't mean that QBT won't try though.