Let’s talk about ASIC mining

I still think some companies are working on new FPGA equipment. If it could increase memory and be easily retasked to new algos, it could potentially be a way around any fork?

Yea but unlike asic’s, fpga’s have use beyond a single solitary functionality
Edit- I don’t know the extent of their reprogrammability or even their efficiency

As an FPGA developer, i can assure you, its not as easy as you think to make a “cheap FPGAs for X amounts of altcoins” with awesome memory on it (+ a cool “change the mining algorithm on the fly remotely” feature) that could also be developed in a couple of months on a large scale basis and sold at an affordable price.

  • This whole “being against asic is a lost battle” claim is pure fiction and just an excuse because one developer doesnt feel like doing something even though thats what the community wants.

  • The other argument posted here about “pool X can do 51% attack” is also nonesence because

  1. it is irrelevant to the discussion of being asic resistant since asic miners also mine to pools.
  2. no pool in their right mind would risk losing their 1% cut out of everyone mining for years to come, just to try to attempt a single double spend.
  • It has already been proven in previous comments here, that no business in its right might, would invest in creating asics/fpgas every couple of months when they realize the developers of those coins are clearly going to change POW or fork off to stop them, its a lost battle for Bitmain &CO and they are not stupid.

  • This discussion has gone way too far.
    Company promises ASIC resistance >> people trust the company and invests thousands >> one of the developers decides to change his mind and is very arrogantly telling the miners that are made his coin to F*** off and mine monero" ( although i have to say hats off to the other developer who posted here with smart arguments why ASIC resistance is a 100% MUST) >> The actions needed are clear:

  1. Raise memory requirement.
  2. If #1 is not enough, change POW.
  3. Official apology by zooko for his selfish behavior and total disregard for the community.

If any of the above points dont happen, the entire mining community should stop mining zcash and move to a different coin and let the ZCash coin vanish into oblivion because no coin that doesn’t stay true to its community or backs down on its original promises deserves to live. Let zooko become a bitmain slave all he wants, no reason for us to support that. As if the 20% cut wasn’t bad enough, now this is even up for debate? no thank you. I will take my megabytes else where and so should any sensible miner that wants their coin to have an actual decentralized system and a bright future.

Excuse any language mistakes, english is my fourth language, i am not an native speaker.

6 Likes

Thank you for your opinions. I will be moving to the new XMR in a few days. I was keeping some old cards on zcash, but had to shut them down as they are not that profitable. It’s kind of sad as I really like the premise of zcash. I will be watching closely to see if they take us smaller(10k sols)miners seriously, or do they just want to be another ASIC only coin like btc.

1 Like

i applaud you for having a sensible logic and proper way of looking at your investment. 10k is by no means a “small miner”. You are the middle class of the network i believe. Even if 100k miners like me or bigger are the majority, we are not stupid and we will not just stand aside and watch a single developer destroy our coin and go “oh well, i guess he loves ASIC miners, i might as well sell all my gpus and buy asics so i can keep supporting his coin”. We are talking over 30% loss of value here no matter the scale ( in addition to the fact of completely losing trust in this company).

1 Like

Hey folks! Just a note that I am paying attention and thinking about this. I’ve read about 1/3 of this thread so far. I’m not going to write my thoughts here right now — no time!? Plus I haven’t read everyone else’s comments yet — but even before reading the whole thread I’d like to challenge everyone to look for data that supports, or even better refutes, your hypotheses.

For example, suppose the hypothesis is that mining without ASICs is more decentralized than mining with ASICs. What sort of data could we find that could support or refute that belief? Here’s a stab at it: Monero mining, which doesn’t have ASICs. How centralized is it? I’ve been told that most Monero mining is done by botnets. Is there any way to verify that claim? And if it is done primarily by botnets, does that suggest that it is probably centralized?

Here’s a hypothesis that could be substantiated or refuted by the right data: non-ASIC mining as currently practiced in Monero has greater miner centralization than ASIC mining as currently practiced by Bitcoin. Anyone know how to get data on that?

6 Likes

Less centralized than any ASIC minable coin thats for sure. The same applies to the tons of GPU-mineable currencies right now, including our own ZCash.

Can you share with us who told you this? Do you have data to support this claim? why are we talking about Monero in the first place? it is completely irrelevant to GPU-only-mineable ZCash.

Nobody is asking to copy-paste Monero solution here nor is any mining farm that invested tens and hundreds of thousands of dollars into GPUS, will sell off their GPUS so that they can put that money back into some “botnets” that supposedly are mining majority of XMR according to what you heard.

Again, please provide data for your own hypothesis, specially when they make huge claims.

There is no way on earth that having 10000 GPU farms with 10 kSol/s average is going to be more centralized than having 100 ASIC farms with 1000 kSol/s average.

To be honest, this is starting to look a bit like a diversion tactic; The fact that you keep brining XMR to the discussion making unverifiable claims about it while avoiding to talk about the real issue which is completely going back on the promise of asic-resistance that was made by the ZCash network when it launched and completely going against what the community wants. Its making me have even less trust in this coin/team which is a shame but its a great coin.

Why not compare to ETH or other equihash coins? Their teams are clearly against ASIC and they are decentralized since the beginning. Just like ZCash was, and still is (unless we let it die at the hands of bitmain).

The community is clearly ASIC Resistant and so was the original promise of the Zcash network ( which is the reason i and majority of miners hopped on the board in the first place).
Please feel free to read the comments in the +20,000 miner gpu mining subreddit about this issue (which is not the only subreddit this is currently being discussed on).

The choices are clear:

  • Backing down from the orignal promise of the ZCash network = No more trust in this coin.
  • going against the clear will of the majority of supporters of the coin = no more miners support the coin.
    and this will sadly lead to this coin dying.

OR

  • taking a proper stand against ASIC as promised and staying true to the community = continued support of the shareholders of the coin, which are the miners.

Its not to late to take make the right choice, but it could be too later very soon…

7 Likes

Hey Zooko!

I’ve been supporting ZEC since ever since I started mining.
1.: Antminer E3=Ethash and Ethash!=Equihash
2.: Those who have long supported the community have all bought video cards.
3.: Unfortunately, I do not have data on centralization.
4.: I fully support ASIC resistance.
5.: If the Zcash Community accept ASIC miners break the initial promise and lost many people faith and the old community members faith.
6.: Zcash is not ETH or ETC or XMR or BTC, we have time to thinking and figure out good and better solutions.

2 Likes

More evidence would certainly be helpful. But Monero is a bit of a distraction here. There’s no evidence of any large-scale botnet involvement in mining of Zcash, or of other coins with GPU-friendly PoW algorithms like Ethereum and its forks, and that is probably because the difficulty is such that CPU mining isn’t practical even for botnets (where the botnet controller is not paying for power usage). The proposed change to the Equihash parameters would maintain that situation; the additional memory requirement would not exclude GPUs typically used for Zcash mining.

(There’s a separate question we should discuss about whether that parameter change is likely to be sufficient to resist ASICs, or more likely DRAM+ASIC multi-chip hardware. It probably wouldn’t be in the longer term, but that doesn’t necessarily matter if we are considering a move to PoS or PoW+PoS over that timescale.)

9 Likes

You said, “Monero mining, which doesn’t have ASICs. How centralized is it?”

You are about to see how WRONG you are in regards to Monero not having ASIC’s. On April 6, 2018, watch how much the network hash rate for Monero falls.

I’ve been mining with GPU’s and ASIC’s since 2014. I’ve owned almost every ASIC Bitmain has manufactured except the S2. I “know” Bitmain’s schemes. I have my suspicions Bitmain already has a ZEC ASIC mining now. I will provide more evidence to point to this later with screen shots.

As for your argument that ASIC’s are better for ZEC than GPU’s because of people switching their GPU’s to other coins. Can those with Equihash ASIC’s not switch to other Equihash coins using the same algorithm as ZEC? Certainly… and you KNOW that is truth. So, WHY such a lame statement as that? Is there something you already know in regards to Bitmain having an ASIC for Equihash that you’re not telling us?

1 Like

Existing empirical evidence is fine, but aren’t many arguments for optimal network security game theoretic? Where’s the dev game theoretic justification for colluding with bitmain? One of the reasons I liked zec is because I’m a professional statistician and liked the rigor behind the cryptography and algo.

So since the dev cur is 20% regardless of whether a bunch of ASICs mine your coin or not, where Is the incentive to resist? Faith to the community is a nice idea, but are you really going to sink opportunity cost into doing it when you get your money either way? You’re an altruistic exception to rational choice models? Then again, there may be non-trivial probabilities of farther-out unforeseen consequences than you think.

2 Likes

Your arguments are valid. Thanks for sharing them.

I too will join you in pointing my 36 rigs elsewhere if the developers we have supported have become lazy and indifferent. He’s got his millions now and travels the world. So, why should he give a shit now? At least this is the vibe I get from him now and his Twitter account.

1 Like

The fact that the developers change parameters to retain ASIC resistance will inhibit development of said ASIC’s. A company will not invest millions of dollars if they believe a coin will fork to retain ASIC resistance. They may try to mine in secret, but an increase in difficulty will expose them. The only near perfect solution would be to change the algorithm approximately every 6 months and increase the memory requirement every 2 years or so when new GPU’s are released and only allow the last 2 generations of video cards to mine. I think if that were done it would be a very strong solution.

2 Likes

rootMar 1
In the past few weeks we’ve seen the difficulty reach an all-time high of around 12 million. I wonder if this recent spike has anything to do with ASIC testing or people switching from ETH to Zec or just new systems coming online.

OopsMar 1
Global Sol/s have doubled in ~three months. Diff through the roof. There are a lot of new miners but I don’t think that accounts for such a tremendous increase. Spikes have to be coming from ASIC/mega-farms IMHO. My daily production has dropped by half and I have added GPU’s. Moving… :frowning:

I’m with both of these users…
It appears to me as well an ASIC has been mining Equihash for the last few months already AND increasing.

Your reflection has no flaw and is on point.

I really appreciate the fact that @daira and @zooko have different points of view on the subject, this is a very healthy sign IMO.

To follow along:

Since ASIC are accessable only for a minority I guess the question anwsers itself… if ASIC were highly available and distributed equally throughout nations no one would be so heart-hurted about this discussion.

Monero has ASIC (bitmain selling cryptonight related machine is proof) and when they’ll upgrade their algoryhtm(tommorow?) we can calculate the drop in hashrate. We must be grateful that Monero Dev Team have reacted so proactively, we will have some solid and valid data to discourage or embrace the unkwown impact of ASIC.

“most” is too vague. The question here is: why wouldn’t a “hacker” with 8000 i3 availble for 12 hours every night mine monero? People are developing games with miners, malware with miners and milions of webpages are reported having miners, this is a reality and personally I find this very good overall. In the end people will have much more awareness and care over processing power, and @ the dawn of AI development this is exactly what we need!

IN CONCLUSION

3 Likes

Please forgive if this is not the appropriate link to post in this chain. In a rush to head to the office :slight_smile:

Interesting in light of many of the comments I have read in this forum. Supports diversity.

1 Like

I’ll trust them more if they accept reality of ASICs and find a efficient way not change algorithm time to time.

how you got “shareholders = miners”?

I think the best way is PoS which Zooko already mentioned, very clever idea, good for community and also future of Foundation. for short term we can change some parameters of current algorithm, but for long view we need some basic change in rewarding system. Miners have to accept technology innovation in mining industry.

In light of the planning for Overwinter and Sapling (see end of discussion
in Issue: Plan how to change the proof-of-work that @Shawn linked to), a good way to expedite a post-Sapling change of Equihash parameters may be for current miner authors to add support, and advertise performance, for (144,5) parameters.

1 Like

oh so thats your argument? you called it a “reality”? wow, it must be true then. Pack it up people, lets all sell our GPUs and lose hundreds of thousands of dollars. lordfeo called it a reality even though it has been proven beyond the shadow of the doubt, several times in this thread, that its completely nonsensical to argue “fighting asic is a lost battle”. In reality its the other way around, but i don’t wanna repeat that again, read my previous comments as well as comments made by other miners here that clearly show how Bitmain&Co will not even attempt to attack ZCash again when they see the community/developers take a stand the first time ( as other coins are showing and leading by example).

Clearly you are not a miner, and judging by your utter disregard to the millions of dollars spent by the thousands of miners who are the soul and heart of the network (and ARE the shareholders of the company that can destroy it in one day if they all go mine a different coin), i don’t think i need to waste my time explaining things to some random “investor” who bought some ZCash coins because he thinks it will moon in 3 months and now is actively voicing his opinions on what miners and programmers should and should not do.

Sorry bro, don’t take this personally, but your opinion on this matter counts for nothing. You hold no power of any sort. Your couple of coins will be worth nothing soon without the support of the miners. Without hashing power, you have no say in this mining thread at all and should stop crating confusion and posting comments void of any real knowledge or experience.

2 Likes