Let’s talk about ASIC mining

Still whining, still no answeres. Still erasing people
Messages. This forum has led me to leave zcash. I have now pointed all my rigs towards etherium and zen cash. I have lost all faith in this coin, community and I’m selling the 150 coins I’ve mined asap. Was a good journey. But now I see the end for zcash. Core developers and foundation. This is your fault. Keep stringing your community along, that’s been here since day one. Bye Bye. Just lost another miner.


Wasn’t aware of anyone getting their post deleted. I did get editted by Daria, but in all fairness I got a little carried away on that post.

Can we stop pretending that for profit GPU miners only have the communities best interests at heart? Most miners are fickle anyways, they’re pointing their rigs at whatever coin is most profitable, ready to change allegiance at the drop of a hat. There are very few if any GPU miners here that would mine ZEC at or near a loss.

I’ve been mining ZEC since the genesis block, but even I don’t have all my GPU’s pointed at ZEC. I’ve always believed strongly in the project, but this is just simple diversification on my part. One can easily make the argument that someone who invests in ASIC’s is more committed to the network than their GPU counterparts, because they can’t just switch coins when profits start going down.

My small GPU farm will be profitable regardless of the direction Zcash decides to go. In the end, GPU and ASIC miners are motivated by the same thing, profit.


Here are a few historical references:

  • Solar Designer’s analysis

    “A conclusion is that while Equihash is not among the most ASIC-resistant known PoW schemes, its choice by Zcash may pose a reasonable tradeoff considering Zcash’s multiple goals and depending on which properties of the PoW turn out to be actually most important to users of Zcash. Another conclusion is that Zcash’s current n=200, k=9 Equihash parameters are suboptimal and may need to be changed.”

  • our blog post Why Equihash

    “Nevertheless, we can’t know for certain that Equihash is safe against these issues, and we may change the Proof-of-Work again, if we find some flaw in Equihash or if we find another Proof-of-Work algorithm which offers higher assurance.”

    By the way, in the context when we wrote that, it was during the Bitcoin Blocksize Debate, when a lot of people were arguing that it was immoral for a cryptocurrency community to ever change anything about a consensus algorithm. We didn’t want people (including miners and ASIC manufacturers) to think that the community would be incapable of changing the PoW again, or to think that there was a social contract that would be violated if we were to change the Proof-of-Work again. It never occurred to me that eventually the public conversation would shift to where some people thought there was a social contract that would be violated if we didn’t change the PoW! Remember back then the whole debate was about whether it was possible or morally acceptable to ever change a PoW. If it had occurred to me that the people (including some of my own employees!) would think the opposite social contract held, then I would have spelled it out and said something like:

    “Nevertheless, we can’t know for certain that Equihash is safe against these issues, and the community may collectively decide to change the Proof-of-Work again, if we find some flaw in Equihash or if we find another Proof-of-Work algorithm which offers higher assurance. Or the community may collectively decide not the change the Proof-of-Work again, even if some improvement becomes appealing. You can’t count on the community to not do it, and you can’t count on the community to do it. It will be up to them when the time comes.”

    If I get my hands on a time machine, I’m going to go back and change that post to say that, right after assassinating Hitler.

  • User Expectations at Sprout Pt. 1: Slow-Start Mining & Mining Ecosystem

    “One of the reasons we chose Equihash is for the ASIC-resistant properties which in theory make the network more accessible to “small” miners who are extremely valuable for providing a market balance to other miners with significantly more capabilities.

    While the lasting effect of ASIC resistance is, at least at this stage, unpredictable, the more people participating in mining, the stronger the network will be against any future growing disparity and centralization in mining power.”

  • The Zcash FAQ: Is Zcash proof-of-work? What mining algorithm do you use? Is it ASIC resistant?

    “Yes, since launch, Zcash has been based on proof-of-work. Maybe the community will choose to change it to proof-of-stake or something someday. We cannot predict what the community or communities will ultimately decide about such things but are very much open to improvement and evolution.

    We are currently using Equihash as the proof-of-work for block mining in Zcash. Equihash is a proof-of-work algorithm devised by Alex Biryukov and Dmitry Khovratovich. It is based on a computer science and cryptography concept called the Generalized Birthday Problem. Please read the Why Equihash blog post for more details.

    The algorithm is currently not economically implementable in ASIC. We’re still evaluating whether we think it will resist custom hardware (“ASIC”) implementation long-term.”

  • My answer to Shawn’s question “So then what hardware would be the most cost efficient for mining?”

    “To be honest, we don’t know the answer to this. Our goals are to make mining profitable for casual enthusiasts, using hardware that they already own, when it is idle (e.g. your smartphone while it is plugged in overnight). This would be a revival of the original Satoshi Nakamoto vision of a very decentralized network of miners.

    However, our actual process isn’t guaranteed to achieve that goal — it is just our current best attempt. Our process is:

    maybe we would consider tweaking the algorithm a little if it seems like there is a really valuable tweak to be made to give more advantages to laptops, tablets, and smartphones (in particular I’m thinking of Solar Designer’s MAXFORM tweak for Argon2d, and also his suggestion to make memory reads come in 1 KiB spans in order to reduce TLB misses for CPUs). But we probably won’t have time to mess with that before launch.

    Now, what hardware setup will be most cost-efficient after we do the above, and whether the above will succeed in our goal of rewarding enthusiasts who don’t put in a capital investment, your guess is as good as mine.”

Would everyone please, from now on when someone makes an assertion about my or the Company’s previous commitments, link them to this post? Thanks!

(Thanks to Simon for hunting down these historical references for me.)


Correct. Also zcash has lost a good amount of hashrate since the 24th, which is the date that the enlargement pill became available (50Mh/s with 1080 ti).

This is your personal opinion and not a fact.

Yep, Zcash lost as well about 20-25% hashrate. Notcied that as well …
So, proofs somehow one of my arguments, lot of “loyal” hoppers around … :slight_smile:

Opps, after we have Asics on the Zcash network than we even have maybe 25 - 50% gpu profit hoppers on the network …:thinking:

That’s hilariously over generalizing. 100 people in a room 20-25 decide to go next door because they have better music…guess everyone prefers the music next door…even though 75-80% stayed put.

20-25% of the hash power just went somewhere else. GPU miners are greedy and coin hopped. Neither of those hold true, your just bending the narrative to fit your perception.

Of that 20-25% how much of that was NiceHash? How much of that was another cloud mining service? How much of that was another auto flipping mining application? How much of that was a Large Mining operation who made the decision they are missing out on Ethereum right now, or maybe decided to go prospect some coins at RVN? How much of that were the ASICs getting powered down by Bitmain so they can blow the dirt out of them, and pack them up for sale??

That doesn’t even take into consideration the “normal” things that can happen. Power failures, network issues, pools having problems. How much of what you claim is disloyal miners just that??

1 Like

You obviously can not read a chart or you would have noticed that monero hashrate is down by ~25%. If that’s stable without fluctuation, so it be.

And in one post you called me a liar, an agent and compared to a Nazi Propanda Minister? Seriously???
Can it go any lower than this or did this hit the bottom yet?

1 Like

Agree, that was not called for in the least bit. You could have made your point without referring to him as a Nazi…gees.


OMG—everyone needs to read the entirety of David Vorick’s new blog post. There are many major truth bombs in there. This is going to force me to rethink a lot.


So much great info here, but one quote that seems especially pertinent would be:

The strategy of hardforking ASICs off of a network is going to lose potency the more it happens, because chip designers do have the ability to make chips that are flexible, anywhere from slightly flexible to highly flexible, with each piece of flexibility costing only a bit of performance. The Monero devs have committed to keeping the same general structure for the PoW algorithm, and because of that commitment we believe that you could make a Monero miner capable of surviving hard forks with less than a 5x hit to performance.

Equihash is an algorithm that has three parameters. Zcash mining happens with one particular choice for these parameters, and any naive hardfork from Zcash to drop ASICs would likely involve changing one or more of these parameters. We were able to come up with a basic architecture for equiahsh ASICs that would be able to successfully follow a hardfork that chose any set of parameters.

Hmm, have to read that post, but the conclusion is probably much alike in https://www.microsoft.com/en-us/research/publication/wheres-the-beef-why-fpgas-are-so-fast/ or How FPGAs work, and why you'll buy one before that – and why there’s https://www.microsoft.com/en-us/research/project/project-catapult/. Back in the uni the greatest pain was programming on the level of flip-flops (Verilog, VHDL or SystemC), but it was fast and one can always pick and choose various other parts (timings were an issue).

<edit: I know I refer to FPGAs, but Yossi’s post explains a bit of similarities.

And the abstract from the MS paper if someone is lazy :slight_smile:

Where do all the cycles go when microprocessor applications are implemented spatially as circuits on an FPGA? It is well established that certain sequential applications can be captured spatially and achieve breathtaking speedups when run on an FPGA, but why? Despite running at clock speeds orders of magnitude slower compared to their embedded processor equivalents, FPGA applications can “lose” enough cycles to create exceptionally fast spatially-oriented circuits. We profile and analyze three canonical applications amenable to FPGA speedup to quantify exactly where FPGAs gain that speedup. We compare the FPGA implementations to several idealized software platforms. The idealized software platforms give insight as to how FPGA implementations attain such dramatic speedups. We quantify the effects of parallelizing and pipelining instructions, streaming data, and eliminating the instruction fetch, showing exactly where the cycles are lost in an FPGA implementation. We also show how the memory interface to the FPGA will affect the performance. Our results show that custom memory interfaces are the most effective way at enabling much greater performance on the FPGA, and that memory interfaces traditional software use become a bottleneck when the FPGA uses the same interface. The results, though not surprising, provide a clearer and more intuitive understanding of the performance FPGAs can achieve, offering researchers and engineers alike a new angle to attack the task of parallelizing applications.

That article zooko posted was a great read. As I said 3 days ago, no idea why people consistently underestimate companies like bitmain.

Yeah that part is an eye opener, ASICs can be made to be hard fork resistant from just about anything but a major algo change. But I also give a nod to the first hand account of how underhanded and dirty Bitmain is with it’s customers and it’s competitors. They were warned about trying to manufacturer in China…and it bite them hard.

After any reasonable timeframe to reach out to another manufacturer, after-hours on a Friday night, our manufacturer reached out to us and said with little warning or reasonable explanation that they would be unable to manufacture for us. Just as we had been warned, our attempt to manufacture in China had fallen flat on its face. This setback is estimated to have cost us somewhere north of $2 million.

A lot of people think it’s an exaggeration when comments are made about Bitmain and dirty play…but it really isn’t. It’s not even considered to be bad business in China, it’s just another tactic to gain further efficiencies. Granted they couldn’t “prove” Bitmain had anything to do with it, and say so, but they were also warned repeatedly this would happen, and it did. This is how you stay on top, fair play, unfair play, all the same thing, which is Business.

People have been saying this for some time.
Bitmain is releasing these to public. They could have just kept mining away with them for who knows how long.
If miners keep spitting on Bitmain, maybe the next miner they make does not get sold to the public!!!
Maybe they decide to keep all the profit.
So much hate for Bitmain. But they have done a lot for crypto…yah some shady in there too…but I still think more good than bad.
The point is, any profitable coin will have some sort of asic built, it’s inevitable!!
So the public can go with it as natural progression, move gpu’s to the next coin.
Fight it, fork, and hope ASICS arnt built again, and secretly run, and then not sold to public because they are afraid of a fork.

The way I see it either way ASICS are on Equihash
They are here to stay
Whether we ALL get to buy one, or
They secretly run in the background for big corps like Bitmain. And we never see em.
They are here. Will be here. And won’t go away. And by all the info I read with the tech now. They can’t be stopped.

THEN why not just use master card…asic coins are more centralised then banks

1 Like

No. They are not magic. Just progression.
More sols for WAY LESS power.
Just technology. And progression

If we take the “The State of Cryptocurrency Mining” post at face value there are two issues that should be considered:

  1. Any conventional algorithm can be eventually mined via ASIC-s

  2. The problem about any major player, like Bitmain or Halong, is that they can compete in an uneven ground with the people who buy ASIC-s from them. Meaning, for the money we spend on their ASIC-s, they can make profit, as well as purchase more ASIC-s for themselves at a much lower production price. As well as many other issues along this one. That centralizes mining, and centralized mining centralizes crypto.

  3. The way I see it POS isn’t much different because it simply raises the stakes higher. And as soon as any POS currency gets recognized and it is strong enough to have the value/risk ratio profitable, the same issue will occur there as well.

Now, I like most of you guys am a believer in a free and open market, I don’t think it is optimal to think into the direction on how to dethrone Bitmain or Halong, since it is in their right to do what is best for them.

I do however believe that we should put our effort into a direction on providing the opportunity of mining or whatever system we use to keep crypto moving, into the hands of as many individuals as possible. I believe whoever gets this right and involves the most individuals in the process will create the most recognized and decentralized system. And the most trusted one too.

If you like the the thoughts I’ve posted here in 2015-2016 regarding involving more individuals in mining, consider Facebook as an option. I’ve read they are interested in gearing into crypto and their platform might offer us some solution :slight_smile:

1 Like

The problem here is you are a software engineer and think like an software engineer forget about what you know and go hire some ASIC engineers from Nvidia, AMD and Intel they will tell you to concentrate you POW on “flop/watt” no ASIC manufacturer can compete with Nvidia or AMD or for that matter Intel in this arena with generally available hardware CPU’s and GPU’s. The proof of work should be structured to run a load using the available computing resource here and that would be it for any custom ASIC, as the cost to develop new custom ASIC that would be comparable to generally available options (and Nvidia, AMD and Intel have access to the latest fabrication process nodes) and would level the playing field for the foreseeable future as Nvidia, AMD and Intel release new product which would offer increment performance increases for the POW loads. Have you not looked at the SPEC benchmark? look at the sample https://arxiv.org/ftp/arxiv/papers/1707/1707.04558.pdf