Future of Zcash mining (circa 2019 and 2020)


#23

I understand the goal of making ZEC accessible to more people. The post doesn’t make it clear how you intend to go about this specifically, which creates uncertainty. Uncertainty inhibits investment - maybe that is the aim here.

Generally I’m not convinced GPUs are more accessible. For example for many months earlier this year it was impossible to obtain GPUs because of global supply shortages.

There are also benefits to ASICs. They lock-in miners to the coin more-or-less whereas GPU miners can rotate through. They’re also far simpler to setup for someone that has never built a computer (most people).

Presumably no one is supporting GPUs for emotional reasons from growing up gaming, we’re all just wary of centralising hash. I’d be interested to here thoughts on approach taken by Grin community which is ultimately ASIC friendly but in way that is designed to foster competition. For example can there be investment in an open source hardware design?

Theoretically I understand performance can be improved with POS but there is a big engineering cost to implement POS plus various risks. If POS delivers towards privacy mission it’s a valid expenditure of resources but if POS is tangential to privacy the effort should go towards something else - there’s always an opportunity cost.


#24

smartest man in this thread. thank you almccann


#25

thanks zooko, good decision.


#26

Exactly correct that capital formation is centralising. This is true regardless of context - POW vs POS, ASICs vs FPGAs vs GPUs etc. We can never get away from this in crypto - there will always be a tension between decentralisation objectives and tendency towards centralisation overtime due to capital formation.

The question is what to do about this tension?

I disagree with the assumption that GPUs more accessible to more people. This is only true if:

  • you already own one (or at least there was a GPU shortage for a long time)
  • you know how to build a computer

Does POS promise resolution to tension? Not exactly, it just rewards existing holders, doesn’t necessarily get coin into more hands and you expend a lot of engineering resources to get to an unknown end state (as far as decentralisation). Granted you may get environmental benefits although I’m not sure that’s understood - POW may equally drive renewable resource investment and research and development.

Any protocol change relating to POW parameters is distorting market dynamics by picking a winner (GPU owners vs ASIC owners). I don’t like this from a conceptual standpoint. I’d rather attempts were made to make a level playing field for hardware manufacturing using principles of open source. However it could be made more understandable - I believe the community deserves a clear description of the research, consultation etc.

Why are we so sure GPU market is anymore decentralised than ASIC? I heard (happy to be corrected) that there are basically two GPU manufacturers globally. How is that any different from ASIC market? A lesson of crypto is we can rely on incentives to influence behaviour. All it takes is Nvidia to start mining on GPUs it manufactures instead of selling and Nvidia will own GPU coins. Please can we stop with anti-China and anti-Bitmain. All companies respond to market forces it’s naive to think Nvidia any different. Crypto is an unusual investment right now but that will change.


#27

This is a great idea and it really shows thoughtful consideration from @zooko and Zcash Co. All users on this forum want to see Zcash succeed and that means getting ZEC into the hands of as many people as possible. ASIC miners aren’t anti-GPU, despite what some users might believe. There are many positive aspects of GPU mining, the main one being accessibility for hobby miners. Allowing thousands of gamers to mine with their downtime could have tremendous network effects by increasing organic growth of the Zcash user base.

It’s my hope that we can stop with this ASIC vs. GPU nonsense that’s driving a wedge between the community. An ideal solution might be some kind of split of the mining rewards between Equihash and an ASIC resistant algorithm. It’s unclear whether multi-algorithm PoW is technically feasible, but it was brought up during the Mining Workshop so it’s definitely something worth exploring further.

Finally, one last comment: It’s PUBG Zooko, not PugBee!


#28

#29

@vgm I agree with this i cannot tell you how tired i am of getting on here and just reading such negativity between both parties… I started out as a GPU miner but then when ASICs were announced i decided to purchase some… Still have my GPU rigs just moved them over to a coin for a very long term hold which i have no idea if it’ll even have value years down the road… But main point being is there defiantly is a clear wedge in this community that needs to be solved in a fair manor to both parties.


#30

I can’t see how these points are valid at all:

  • IF they want to come up with a new algo than again engineering costs are involved.
  • Privacy shouldn’t change in POS to POW, from what i have readed so far it has 0 impact. Have in mind that all the ASIC/GPU resources are used to FIND a block, not to solve it at all. The solving alone is an more or less easy process. There is no reason why a POS consensus shouldn’t be able to keep the same privacy than with POW.
  • What are the other risks? Just saying there are risks without naming it doesn’t help at all…

#31

These funds also went towards hiring and paying developer salaries while the price was lower. I remember when Zcash was $20 for quite some time. A lot of this has been used effectively from what I’ve seen.


#32

So the assumption here is that the engineering cost of POS is considerably more than a POW change (see how much time it has taken Ethereum to develop POS - granted Ethereum is different to Zcash but I believe it is still a considerable effort relative to POW).

I’m not saying privacy is directly impacted by POW or POS. You’re right it isn’t. I’m saying that with finite engineering resources and a mission to promote privacy, we should only be considering a change from POW to POS if it is on the critical path to privacy adoption. This is because the opportunity cost of working on POS is that you’re not working on features that help privacy adoption - UX, keys management, more Sapling-like performance improvements, whatever.

There is a big engineering risk to POS because it is a lot more complicated than POW. There is also economic risk because it is not clear that POS coins will retain purchasing power relative to POW coins.


#33

You can get a Z9mini for the price of a GPU, anyone can participate already.

The miningrace finished this year, why start a new one? With a new GPU-algorithm it will just be a matter of time before there will be a new secret ASIC.

With a miningeconomy this big it’s pretty scary that decisions are made this lightly. 100’s of millions invested in hardware will be useless only because 14 year old kids should be able to participate.

How does Zcash consensus work, is not 51% hashingpower needed to fork the network?


#34

Find it hard that a lot of people have spent a lot of money for ASICS to support your coin and now your just going to throw them in the water.
Gpu miners are a problem for gameing pc builders and the miners put the price of gpus up and asic miners are a problem for gpu miners. If gpu miners can’t move to new machines then they need something else to do.

Plus asic will crack your new algo all they have to do is send a update to the machine then your back to square one :joy::joy::joy:


#35

Nearly daily we see projects going from Asic algos to POS, good luck with “cracking” these … :joy:


#36

Yeah if it goes to pos algo then we’re all done but don’t blame everything on asic miners. If every asic miner owner built the gpu rig then whats people going to blame it on then because things will be the same as what they are with a asic on the market.


#37

I really appreciate this feedback. Note that a change to mining isn’t the only thing we’re intending to do! (See https://z.cash/blog/zcash-company-proposed-roadmap/ .) But it is true that there is always opportunity cost and by doing this we’ll be forgoing something else. The reason I’m prioritising mining is the “security” and “adoption” parts of that draft roadmap.

We’ll work out a lot more detail about our future plans soon — give us a chance to catch our breath after the Sapling upgrade — I only mentioned this one now because I didn’t want everyone to assume we were settling into permanent status quo on mining, not because it is the most important or most exciting plan. :wink:


#38

Hi boxalex! Thank you very much for the thoughtful posts. I’ve always found your posts to be thoughtful and factual over the years. I liked a lot of your points, but the main two points that I want to reply to are: PoS, and Usability and Adoption.

As far as PoS, I would definitely support the idea of upgrading to it eventually, but the October 2019 Network Upgrade is still too early! The Zcash Network Upgrade Pipeline is about a year long. That is, if we’re going to activate a Network Upgrade in October 2019 then we need to have the software implementation of it merged to master and activated on testnet by around June 2019, to give the entire ecosystem time to test it, upgrade to it, and change their own infrastructure as necessary to accommodate its changes. That means we need to finish implementation by around April 2019 to have time for implementation security audits and “pre-testnet testnet”. That means we need to have the specification audited and frozen by around February, etc… It adds up to the fact that we basically need to commit to exactly what changes we’re going to implement in December 2018 in order to activate those changes in October 2019! PoS is still too experimental for us to be ready to commit to deploying it in October 2019.

The other thing you emphasized that I want to reply to is Usability and Adoption. We absolutely agree on the central importance of that, and we’re working on it. Fortunately a lot of those improvements can be implemented and deployed a lot quicker than a Network Upgrade can. Also fortunately there are numerous other parties than The Zcash Company who are working on this issue…


#39

Yeah, I really agree that everyone should be here because we share values and we have shared goals. Feuding doesn’t accomplish anything except to undermine our shared values. :heart:


#40

May I ask how can it be years? He joined May 2018!


#41

A 14 year old kid shouldn’t be able to mine anyway you have to be 18 to get a coinbase account to even sell your cryptocurrency. So why should they be able to mine if they can’t do anything with it :joy:


#42

Create a wallet and buy great things with crypto, there is no need to change it to FIAT actually e.g. https://www.kinguin.net/ / https://www.keys4coins.com/