This application is claiming everything that is involved in Bitcoin mining or any other cryptocurrency mining using Asic chips where the block chain is segmented and hashed by multiple parallel hash units and then recombined. I would like to make sure open source cryptocurrency is not patented.
I know there should be something out there that teaches how to speed up mining with the use of GPU or Asic chips that is before July 11th, 2012.
- Publication number: US 2014/0019693 A1
- Assignee: Intel
- Prior Art Date: July 11, 2012 (but see caveats below)
- Open for Challenge at USPTO: ?
The three independent claims (1, 7, and 19) (as of June 27, 2015) are all essentially variations of claim 1, (substitute "computing device", "method", "machine-readable storage media"):
A ["computing device", "method", "machine-readable storage media"] for processing a data buffer, the ["computing device", "method", "machine-readable storage media"] comprising:
a data buffer processing module to: access an arbitrary-length data buffer having a buffer length and a plurality of data segments, each data segment having a segment length greater than zero and less than the buffer length;
directly read the data segments into a single data register, the single data register having a plurality of data paths, each data segment being read directly into a different data path of the single data register;
perform a serial data processing algorithm on each of the data paths of the single data register substantially in parallel to produce a result for each data path;
and combine the results produced for each of the data paths of the single register to form the output of the serial data processing algorithm.
The priority date for this application is July 11, 2012, and the assignee is Intel. However, there are six inventors listed, so any prior art from these individuals or Intel in general would avoid any issues regarding burden-of-proof for inventorship if they were earlier than July 11, 2011:
- Sean M. Gulley
- Wajdi K. Feghali
- Vinodh Gopal
- James D. Guilford
- Gilbert M. Wolrich
- Kirk S. Yap
Update (September 5, 2015): This patent has now undergone several office actions, and there has been no amendment to the claims. The inventors are repeating arguments with the examiner during each office action:
Claims 1-20 are rejected under pre-AIA 35 U.S.C. 103(a) as being unpatentable over Sprengers [GPU-based Password Cracking on the Security of Password Hashing Schemes regarding Advances in Graphics Processing Units] in view of Zalewski et al. [Case Study on FPGA Perfomrance of Parallel Hash Functions]. Sprengers teaches the use of GPU programmable ALUs for parallel processing of hashes. Zalewski teaches a proposed parallelizable hash function on an FPGA.
The examiner has determined the applicant's invention to be directed towards the parallel performance of cryptographic hashing schemes. The examiner knows that such improvements have been in use in the fields of password cracking and cryptographic currency mining such as Bitcoin mining. Therefore the examiner has determined the overall board aspect of the applicant's claims to be obvious given the teachings of programmable ALU's taught in GPU, FPGA, and ASICs. The examiner suggests the applicant further narrow what it is the applicant considers the novel aspect of the present application.
In short, the examiner is not budging on his position, but some additional (more specific) prior art would be extremely helpful in getting this patent rejected. Intel has shown no signs of abandoning this attempt to patent something that arose from the open source community. In the year since this question was posted, there has been no attempt to provide prior art. Where are the open source contributors?