Let’s talk about ASIC mining


#5322

True, if there is lots of advance notice (which Zooko mentioned that there would be)

then the chance of miners being caught “off guard” like with Sia would be unlikely.


#5323

Actually if we analyze the hashrate that is switching it’s not a big one and it’s very limited too, just to be fair here.

The other coins on equihash just can’t take a bigger hashrate while staying profitable. At the current situation it’s impossible that ZEN could take the 2.3GB from ZEC, even less the others ZCL, KMD, BTCP…

It seems that 20% switching forth and back is the maximum that the other coins can take for a short time until again ZEC gets more profitable for these switchers.

While i don’t want to support asics anymore, in this point it made sense to limit the switching possibility this low. While the equihash asics still can switch to some equihash coins it’s a very limited option. While in theory and as an example ETH had risen to new highs its nework could have taken the whole ZEC gpu miner network just in a sudden.

Just commenting the switching issue here, nothing more.


#5324

What is fair? You are trying to make it sound like my statement is incorrect. I did not make any GPU comparison, just the statement “ASICs are not forced to commit anything to Zec”.

ASICs are able to jump to different coins, this is not a comparison vs GPU, its just a fact. You are not FORCED to commit to Zec with a ASIC.

Do ASICs bounce around less than GPUs, Yes.
Do they have less choices than a GPU, Yes.
Are they FORCED to commit to Zec, No not at all, this is my point.


#5325

They are forced, there is no way the Asic ZEC hashrate can be shifted elsewhere staying profitable. This just means they are just that, forced to mine ZEC.


#5326

this is so absurd…

zcash should save asics because they have nowhere to go? well… you should known better when investing in EQUIHASH ASIC.

Besides you have some money made back.

I can’t believe that ZCASH TEAM is actually trying to protect ASICs.

ZCASH SUPPOSED TO BE ASIC RESISTANT!!!

BECAUSE ASICS ARE BAD FOR ANY COIN THEY COME TO!

Inflation, hiding, manipulating with prices - what asic is, and always will be.

There is huuuuge diffrence between GPU mining and ASIC mining.

ASIC are a hack for crypto. Hardware made as cheap as possible that is good for only one specific thing, hardware which has no any other use.

asics should be for sha256 only, and only because that algorythm can not be changed… While here we supposed to have a dev team which stands for its own promises, ffs!


#5327

IMHO your posts are way to emotional and they reflect always the same thing/story - it’s a repeating process in a never ending loop. Do yourself a favor and shift down a few gears, this could benefit your health.


#5328

pff… take some action.

absence of any decision on a never ending loop since may.

im trying to do zcash a favor, or this could take affect on its health lol


#5329

its very kind of you though. rare nowadays when someone cares so much about health of a complete stranger :)))


#5330

I see only care about yourself here.


#5331

lol bro, asic owners are the ones whos back parts are on fire, when we talk about change.
gpu miners surviving OK, on other projects.
and noone of you sudden supporters of passive position can explain the benefits of asic network.


#5332

Completely incorrect and false, they are not forced at all to stay commited to Zec.

Stop spreading lies please. Sure they might make less profits switching, does this mean they are forced to keep mining Zec, no not at all.

Stop saying people who bought a Equihash ASIC are forced to commit to Zec and contribute only to Zec. This is a complete lie. If a new Equihash coin came out and was more profitable than Zec, I know alot of people would switch to it. They are not forced to stay with Zec.

I never said it was more profitable, My point was buying a Equihash ASIC does not mean you are commited to Zec…

Do ASICs have less choices than a GPU, yes.
Can they switch to another coin, yes.
Are they forced to mine Zec, no.


#5333

It’s not a lie, i’am commenting the current situation. It’s reality based on current facts without if’s.

But to make it more enteraining i will use your argument IF a new gpu coin comes out that can take 5x times the hashrate of ETH and be 5x times that profitable than all GPU POW coin networks over all algos would just vanish in seconds … Just IF

If you like it or not, about 70-75% of the current equihash Asics are just forced to mine ZEC to stay profitable …


#5334

There are two ways to look at mining and it’s implications I will discuss my 2 cents of the two :

Economic Implications

Mining profitability is what brings people in the network and pushes for research and development spending . If we look at it from a biased perspective those that spend most on mining should collect most of the rewards alas this is biased because currently taking Bitmain as an example, it’s a billion dollar profitable company at this point Bitmain is able to spend on research for ASIC hardware more than the rest say a hobbyist . If all participants were equal the richer will always have a head start which turns us to the problem of rich growing richer .

Consensus Implications

I can’t stop thinking about Andrew Poelstra and his paper ref :


ASIC hardware is good for network security, centralization of ASIC usage is not bad either because there is really no incentive for Bitmain for example to attack the Network it will just incur a loss from them . Bitmain is a company who’s allegedly about to IPO, being a public company will introduce new people to it making Jihan Wu less relevant and Bitmain a conglomerate . By going public the vision of the company transforms from it’s founding team to shareholder happiness Bitmain’s goal will remain profitability BUT the this raises the question “Is the new participants in favor or not in favor of cryptocurrencies and Bitcoin” ?
So the question becomes then should we sacrifice a few life bars for a decentralized consensus ? In my opinion YES , because as long as consensus revolves around hash power the risks of centralization is present .

The Proof Of Stake Problem

Proof Of Stake makes networks self sufficient i.e the network doesn’t depend on an external source for security “electricity” but on it’s own, after the supply is bootstrapped “all coins are created” miners incentives becomes transaction and in this part miners hold the better cards because they can simply force users to pay higher fees, as we’ve seen it happen with Bitcoin . The second part is that the price will simply skyrocket to offset miners spending as the cost of mining is in theory constant miners will want price to go higher to make it profitable . This is great for all ZCash users but the first one is bad unless a new consensus rules that reject empty blocks is put in place .

Conclusion

I don’t know what’s good for ZCash but following on Monero steps by hardening Equihash parameters seems a logical first step as other solutions are just too risky and will make it problematic for the team to focus on other aspects of it’s roadmap .

I hope this was in point and constructive .

  • I look at Bitmain as it’s the standard mining group the one that represents a potential threat to what we are building

#5335

You gut say one thing but data show totally different thing …and that is that ASIC Zcash have much more hashrate oscillation then when it was when it was GPU coin …now hashrate oscillate even 30% daily …2 days ago it was even more then 50% …going from 1.9 to 2.9 and then 2.0 in matter off hours …on GPU it was never more then 10%.


#5336

My personal opinion is that this is no longer true (if it ever was at all).

  • While i agree that until now mining brought a lot of people into the network, it’s by now no longer true for several reasons:
    a. Bitcoin, the most and best known currency it’s impossible to start mining now at all. It’s not attracting NEW miners at all, even reverse, it’s driving out slowly but surely miners that can’t afford better hardware or have higher electricity costs for example.
    b. Nearly nobody is investing into new hardware, no matter if asics or gpu’s. Sure, here and there you find newcomers, but let’s be honest, it’s way less than it was a year or two ago.
    c. POW mining is coming slowly but surely to it’s end, most of us realize this, no matter if it takes another 6 months, 1 year or 2 years, the days are counted… Thus it’s not really attracking new people into mining.

  • About mining profitability and that these that spent most should collect the most rewards. Sounds good in theory but this would apply to POS only. You spent a given amount directly into the coin, you await about proportionally return/rewards. In POW and with mining it’s not really like this:
    a. First of all you spend the money out of cryptospace, it’s more or less lost money. Instead of investing it directly into a coin you gave it a 3rd party, means money is leaving crypto space.
    b. You pay again for electricity and have montly costs in FIAT, again money that goes to 3rd parties in no way related or supporting crypto.
    c. What’s the deal to pay (by now!) for hardware when the return never will reach the investment? It’s at best case a “get even” game. Pay for hardware, pay for electricty, mine coins and in some years you have reached the ROI, eventually with luck if nothing broke meanwhile, lol.
    d. What’s the deal if you pay montly (just an example here) ~50-150% for electricity, some other % to the mining pool, just to get a minor part as a reward? You feed your electricity provider daily way more than mostly is left for you in kind of rewards. It would be ok if it’s <25% that you spend and 75% of your rewards get into your pocket, but these times have past. Now with mining your electricity company gets the rewards, in FIAT.

Just to comment your Proof of Stake problem. In my opinion you are plain wrong here. First of all the high computing resources and of course the electricity used for it is not directly related to the security but on finding blocks. In POS you remove this “find a block” aspect which uses the most electricity resources (with asics, i didn’t dig further to see how it’s related with gpu to be honest). The security level stays the same, actually in POS you have improvements with security all over. The miners forcing users to pay higher fees is as well a POW problem and by my limited knowledge not possible in POS. Just my 2 cents on your 2 cents :grinning:


#5337

Yes it is a lie and its completely incorrect, saying Equihash ASIC miners have commited only to Zec is completely false.

Then you bring up a strawmans argument about ETHs hashrate, this has nothing to do with Zec. My hypothetical example was to show if a more profitable Equihash coin came out, people would switch to it, they are not commited to Zec and did not “invest in the network”.

Here you are talking about profits again. I never said it would be more profitable or even the same. My point is Equihash ASICs are not commited to Zec and did not "invest in the network".

And looking at WhatToMine, Zec isnt even the most profitable right now, so please tell me again how and why people are forced to commit to Zec?

EDIT: The way you try to word things, ASIC are commited to Zec in the same way GPUs were not forced to leave Zec. GPUs can still mine Zec if they wanted.


#5338

because this is zookos position, as simple as that…


#5339

You really have problems following common sense and logic, seriously.

Take the network hashrate of each coin on the what to mine list:

BTCP = 4.47 MH/s
ZEN = 215.5 MH/s
ZCL = 56.5 MH/s
ZEC = 2300 MH/s
Komodo = 18.5 MH/s
Nicehash = 212.5 MH/s

Now let’s say what happens IF all coins are more profitable than ZEC, what happens next? Logical first of all a huge portion of the Nicehash traffic gets switched. It needs only some funny 5 MH/s to half the profit of BTCP, some 215MH/s for ZEN, some 56 MH/s for ZCL, 18 MH/s for Kommodo and that’s it all over that can shift right now and you have ALL the other equihash coins at 1/2 profit immediatly.

~300 MH/s are needed to half the profit of ALL other coins. It’s basic mathematic that these coins can’t take the 2300 MH/s from ZEC.

Here a free joker for you: Notice the 7 days profitability tab? 103% inf case you have trouble finding it. Nicehash is leading, means these are the most that switch forth and back, hence the leading 103% over 7 days.

Now in your favour, IF ZEC wouldn’t have such huge equihash traffic (10X more than the next coin!), but way less than the majority of the Asics on equihash wouldn’t be forced to mine ZEC, but current reality is just like it is, they are forced if they want to stay profitable.

And just as a sidenote, this is mostly the only point i will argue in favour of equihash asics that in the current state they are forced to mine ZEC.

Just in case you didn’t get the logic behind it let me know, i will try to paint it in easy mode … :wink:


#5340

I think you misunderstood my point, moving to PoS for Zcash now isn’t a perfect solution as the supply is not totally created you still need a way to do that, unless of course you want to do a hybrid thing. PoW won’t die anytime soon it’s a natural process to create supply fairly or atleast the most fair way.

I also pointed that PoS makes the network self sufficient in terms of dependence on third-parties which is great,security wise if we are talking naive Proof Of Stake then no there is no security improvements in theory and in practice there are no metrics to judge that except maybe the fact that we’ve seen 51% attacks on PoW chains.
For Zcash I believe that hardening equihash parameters is a good temporary solution that fits the current roadmap and long-term a move to a PoS based consensus mechanism makes most sense.


#5341

Moving to POS now not only isn’t a perfect solution it’s impossible. It’s not a thing you do just with a snap. The question about POS isn’t actual anyway, it was stated that whatever changes will happen will happen soonest 2019-2020, at least that is what i got out of some official comments.

But to answer about your POS comments. The supply does not necessary to be mined until POS, it can be mined right with POS as well. Many people, for whatever reason, seem to think that POS can only calculate transactions, but that’s not the case.

About the security improvements and how to measure it. There are for some attacks ways how to measure it, the cost of an attack. In POW you actually could try an attack with way less than 51% it just lowers the probability of success. However, let’s take just 51%. In POW you need “just” to have the hardware and/or rent this hashrate, additionally a mining pool is a threat itself in case they get close to 50% or would several mining pools working as bad actors together. Compared to a 51% attack on POS it’s peanuts and just cheap.
In POS the miningpool attack vector doesn’t excist and for a successfull attack you would need to buy 51% of the coins. Needless to say that while the bad actor accumulates his 51% stake the prices would rise and rise and rise. Even without the additionally price rise it would be several times more expensive to carry out an attack.

Just as an interesting fact and to compare:
Right now for an 1 hour attack on Bitcoin it would cost 431,236 per hour renting hashrate, for a whole month you would need 310,489,920 compared to the 56,000,000,000 (without price increase!) to get a 51% stake if BTC was POS just … cheap. Too lazy to make another calculation how much hardware you could buy with 5.6B USD and if it would be enough to take over the BTC network, would still be only 10% of the amount needed to get a 51% stake.