During the 57th Ethereum Core Dev Meetings, one of the topics discussed was ProgPow, or Programmatic Proof-of-Work implementation. The protocol proposed by IfDefElse is essentially a “GPU-tuned extension of Ethash that minimizes the gap available to fixed-function hardware”.
Owing to previous meetings, ProgPow was up for an audit concentrating on two factors and there were also discussions about including it in the Istanbul hard fork. Hudson Jameson began the discussion on the topic by giving a brief-introduction of the audit.
He stated that there are two components to the ProgPow audit, Benchmarking and an efficiency check of ProgPow. He said, “looking at how efficient a ProgPow ASIC would be compared to a graphics card and what their efficiency gains would be.”
Hudson further stated,
“So the Cat herders are looking at both the proposals and we’re basically going to evaluate them and make a decision next week, which one we want to basically try to get funding for.”
This was followed by Martin Swende asking about the progress of the audit. To this, Hudson stated that “there has been no work done” on the audit. Furthermore, Hudson spoke about the questions the second audit would answer. He said that it would give an overview on the constraints of ProgPow in terms of its efficiency as there were claims that “it wouldn’t get that much of an increase and that would deter people from building a ProgPow ASISC in the first place”.
Hudson added,
“There would be an analysis on the proposed ASIC architecture, manufacturing assessment, which includes expert interviews. That’s something else that is in the deliverables for the first proposal we’ve received and potentially an economic analysis on the impact that ProgPow would have on economics of the Ethereum protocol.”
He added that they would not be able to fulfill all of these because of budget restraints. However, they were definitely looking forward to it, he said. This was followed by him speaking about benchmarking. He said,
“I’ve talked to Whiteblock and they are on the fence about what the next steps are because they haven’t secured funding for the audit. What might happen is we might not do the benchmarking part at all because that is the less important part of the two pieces of the audit. And we may only try to pursue funding for the second piece.”
Furthermore, it was said that ProgPow would go on Istanbul (update) unless there were problems that would eliminate the reasons why the team decided to implement it in the first place. Martin Swende said,
“Kind of consider ProgPow to be accepted, but then obviously, if the audit turns out to be some horribly wrong, we can remove it.”