Study: Detecting ASIC Miners in Zcash

Our announcement blog post: https://z.cash.foundation//blog/asic-equihash-study/

Biryukov and Feher concluded, “In the short period of time of this study (2 weeks) we could put an upper bound on the ASIC/FPGA potential presence in Equihash mining ecosystem as of end of May 2018 at 20 – 30% of total mining operations,” based on GPU miner’s dev-fee estimates and the fraction of above 8Ksol miners in the Equihash mining landscape. Dev fees are typically 2% voluntary fees that many of the GPU miners pay to the developers of the most optimized GPU mining software.

To be clear, that means that if ASIC operations were active in the Equihash ecosystem, they did not comprise more than 20 – 30% of total hash power.

Biryukov and Feher continued, “Most of the mining power (95%) is concentrated in pools so any large hidden ASIC/FPGA operation should be present in one or several of the pools, moreover four pools concentrate 87.5% of the total mining power.”

The study: https://cryptolux.org/images/e/ef/Zcash_Mining.pdf

7 Likes

Thank you for heads up sonya. Now we see some real commitment and willingness to go deeper into research of the problem.

2 Likes

Anyone care to wager what the percentage will be in December? I think it will be over 50%.

I have read the whole research and conclusion in the .pdf format.

I personally think it’s not really a perfect neither a very precise one. It relies on most part on the dev fee, not even bothering with the Mining Distribution Chart of Zcash which in my opinion is for hidden/secret ASIC mining operation the best indicator.

It was visible from the chart which Bitmains hidden pool is, and since some days we saw another way bigger unknown bigger pool, right followed by Innosilicons announcement of their 50k miner and i think someone is safe to put 1+1 together to get 2 and that is the miningpool Innosilicon is mining at the moment.

IMO in secret/hidden/private/stealth mining phases of the different Asic manufactures the mining pool chart of Zcash gives the very best indicators on what’s going on. The problem is after a given release when Asics get shipped because than the indeed other miningpools get the hashpower. It’s highly unlikely that any Asic Manufactor is using a known mining pool:

First of all, the miningpool owners would immediatly recognize the Asics and share info. It’s an easy task for a Mining Pool owner to from the hashrate per worker to see what’s going on there.

Second, fees and feedback. It makes 0 sense for whatever Manufactor to use a 3rd party miningpool, neither they can get inside feedback that way and have to pay fees as well. There is 0 logic behind to use a public mining pool at all for larger companies that can provide bigger hashrate power on a given network.

This said, i don’t think this research has provided any new info. Everything is just mentioned as a normal unbiased person can find out itself on whatching and analyzing the hashrate, price, difficulty, mining and other charts.

Even the Bitcoin Gold conclusion that it’s highly likely that it was done with Nicehash traffic and gpu traffic is just logical and common sense…

I personally have awaited a way more detailed, better researched paper/document to be honest and i hope the Foundation didn’t pay too much for it… Pretty sure their must be specialized organisations that can provide way more accurate researching, just my 2 mini-zcash…

1 Like

With that speed and hashrates all manufactores come up these days i’am betting on 75+% as a convervative bet. Having in mind that 6 months in cyberworld is like lightyears i even consider 90+% possible as well.

“Lightyears” is a measure of distance, not time.

The difficulty will be 8 to 12 times what it is now within 6 months.

@sonya - thank you for posting this very interesting report.

2 Likes

Just some observation on the Zpool Mining Pool Distribution:

It seems there is a new shifting taking place. There are now 3 unknown bigger pools visible ib the Zchain Miners Statistic 24h chart: https://explorer.zcha.in/statistics/miners#

Possibilities in my opinion:

  1. The supposed Innosilicon Mining Pool if true is shipping it’s miners.
  2. The supposed Innosilicon Mining Pool is splitting it’s hashpower to it doesn’t take that much attention
  3. High possibility that the Asicminer 40k by Asicminer is ready and “tested” in one of the new 2 pools.
  4. Additionally a new Asic producer is entering or Bitmain is testing their 40k Antminer.

These are the scenarios i can think about to whom the 3 unknown pools belong…

1 Like