Let’s talk about ASIC mining


#5302

Zcash foundation just awarded IfDefElse a grant to integrate ProgPoW into Zcash!


#5303

Interesting, shouldn’t there be a voting first or is this the result of the last voting?


#5304

I think voting is already done


#5305

ProgPOW for ASIC resistance is fully approved for funding by the Foundation and also Solardiz (hacker) to help evaluate and troubleshoot ProgPOW or another ASIC resistant PoW.

As I mentioned before


#5306

this war, at least for the private Asic and GPU miners is actually over in my opinion. Why?

  • From what i understand any changes are too apply soonest Mid-End 2020, about 2 years from now. Means meanwhile nothing will change or in other words, Asics will stay for that time.
  • Mostly all further ASIC batches won’t ROI for the private asic miner.
  • IF and maybe when ProgPOW comes into action in 2years ETH will have allready the constaninople and POS update, meaning all that gpu power gets shifted to the by than less and less remaining gpu coins. If someone really thinks the old good gpu mining times will return even than, sorry guys, won’t happen exactly to the reasons mentioned here. The huge ETH shifting hashrate will prevent this to happen.
  • Asics will face a similar problem, each coin from now forking away from current equihash puts more pressure on the profitability of the Zcash Asics.

I would call it even by now a psyeudo war as the 2 losers are allready known, private Asics and private GPU miners. A war makes only sense if there could be a winner at the end and in this one there isn’t a winner beside Bitmain/Innosilicon and of course Zcash themself as for them it doesn’t matter who generates the Fees at all.


#5307

Maybe of some interest for some members (not sure if it’s the correct forum topic, if not please move!)


#5308

The war is not over, but the entire army was lost in the battle.

Many GPU miners have found other coins to go and support. Why would we want to come back to ZCash, knowing ZCash might just let ASICs back on again because they are busy with more important things.

ZCash has shown ASIC resistance is not a importantant issue to them…


#5309

Zooko has already stated that the algorithm will be changing and economics are economics. If Zcash forks and all of a sudden Nvidia GPUs are the most profitable again miners will switch back or new miners will build new rigs. Just like many miners ran out to buy ASICs. Miners are always chasing profits and ROI, that’s the name of the game.

I have always thought that GPUs are better than ASICs for decentralization and fewer points of failure (like with Antbleed) along with wider distribution of ZEC to more Community members.

But ultimately I have no say in the matter, I’m just watching from the sidelines same as you.


#5310

I completely agree that miners who don’t care about Zcash shouldn’t be involved in mining it. This was one of the positives of ASIC mining. It forced people to commit to Zec while also giving us a higher hashrate.

If only there were a way to merge ASIC mining with widespread availability of miners and multiple hardware manufacturers.


#5311

This might have been true earlier this year, but in the current market noone will be running out to build a new rig to mine with. Spending $2,000 usd on a rig that will only make $1-2 a day is a foolish investment atm.

So unless GPUs start mining more ZCash on the new algo, than ASICs are right now, it still wouldnt make sense.

Only reason ASICs were selling was because the Z9Mini started at $30 USD a day. It is currently making around $3. You would be unwise to by any mining hardware currently.

Which leaves old GPU miners that were burned by the ZCash earlier decision. I speak for myself, unless Zcash becomes much more profitible to mine vs other coins, I doubt I would come back and support the coin if they changed the algo. To be forced off the network because “busy with other stuff”, then asked to came back later to help support it, when its more convenient, is kinda audacious.


#5312

Yet hashrates show people bouncing around Equihash coins using their ASICs. They are not forced to commit anything to Zec, this is completely false.

Just because the hashrate is higher now, does not mean there is more people commited to Zec, infact I would say there is far less people commited to Zec now compared to 6 months ago.


#5313

My gut says that it’s not on the same scale as miners switching to the many different GPU-minable coins.


#5314

I never mentioned scales or said its the same as GPUs, Just the fact ASICs do not force you to commit to Zec. Do GPUs have more freedom? Yes, yes they do. Atleast currently, if ASICs keep taking over coins however, GPUs will soon have nowhere to go.


#5315

You’re mischaracterizing users that buy equihash ASICs. They are buying them to mine Zec. If you buy a GPU to mine, you’re buying it with the intent to potentially mine many coins and then sell the card when you’re finished. There is a difference and the amount of commitment and risk are higher.


#5316

Yet there was many posts on here about when they get their Equihash ASIC they were going to mine Zen.

To say everyone that bought a equihash ASIC bought it FOR Zec only is mischaracterizing . They bought them to mine Equihash coins, what ever is most profitable.


#5317

That’s absolutely right, if nothing changes with the PoW/ hashrate ASICs will continue to dominate over GPUs making them not worth running on Zcash.

I’m reminded of the most recent Sia fork, few people had the Obilisk hardware ready when the fork happend which removed the Bitmain ASICs. I remember seeing the Obilisk hardware selling for less than $500 because they weren’t profitable. The Fork caused the profitablity of those Sia built ASICs to skyrocket to $90 a day: (today!) https://youtu.be/WsfnsChedDY

I could see a similar situation if Zcash forked and hardly anyone had thier GPU hardware set up to support it, the few that did would get a greater share. And of course over time everyone would learn about this and cause a shortage of the best mining hardware again. Long term the cycle of slowly decreasing profits would still continue as it always does.


#5318

They bought them to mine Equihash coins, what ever is most profitable.

You’re honestly saying that when the Z9 Mini was released miners here were immediately thinking, “Oh sweet. An ASIC to mine Zen.” :grin:


#5319

Yes they were, they needed the Zen to start up a super node, that is what alot of people wanted to mine Zen with a ASIC for


#5320

@shawn

I think this would be different because there are a lot of GPU miners that could pretty easily just switch over to Zec just as easily as they switched away.

The analogies to Siacoin make me cringe because instead of forking to GPUs or something that would theoretically increase fair mining, they decreased fairness by creating their own manufacturer monopoly. They took 99% of their miners, invalidated their hardware and then gave that hashrate to 1% who purchased miners from Obelisk. This is a totally unprecedented move. I actually have trouble understanding the logic of why they did what they did, but I think it was because those purchasers were threatening them with a lawsuit due to miner delivery delays.

I hope Zcash never does anything even remotely close to that. The result of what they did could very well kill Siacoin over the long term. I hope for their sake that it works out for them. They must have been out of options.


#5321

This is one reason they did not want to fork. They were afraid that it would leave them vulnerable with a low hashrate as people switched over to the new fork. Now for you to try and spin it around and say it would increase mining profits is kind of misleading.

Also if they were to fork, there would be months in advance notice, people wouldnt be caught off guard with unsupported hardware.

Sorry, im very skeptical about GPUs comming back, I just dont see it happening anytime soon, or without problems.

Zooko briefly entertained the idea when ASICs were first announced. These types of statements really made me cringe when he posted stuff like this, or two Zcash chains 1 ASIC 1 GPU. For him to even suggest such a ideas, instead of a strong no ASIC stance kinda shocked me.