RT @hasufl: Looks like 85% of all hashrate on Monero is ASICs, no surprise there. I lost a lot of faith in the project after they decided to "fork out ASICs". That only centralized ASIC production and made the network more vulnerable.
h/t @matt_odell for the article
RT @LucasNuzzi: "ASIC-resistance" is the one of the biggest lies that gets tossed around in this industry. Truth is, the majority of proactive measures to keep ASICs out *do not work.*
Instead, they further centralize mining around very few manufacturers that quietly developed a workaround. https://t.co/Gv9CH4WYiJ
Fuck Asics, and fuck you, you fucking corporate shill. With ASICs, one company produces them, then hoards them and mines them. They only sell them to the public once they've developed a better ASIC, so the public will always have obsolete hardware, and the corporations control the majority of the hashrate.
XMR needs the ability of individuals to mine without asics. The reason is that for privacy purposes it is better to convert your electricity and receive XMR than to go through an exchange. For the extremely privacy conscious, you need the ability to acquire XMR with no one else having the knowledge that you are doing so.
What a bad excuse. Buy with cash instead.
You can't really be ASIC resistant. All you'll do is prevent ASICs from revealing themselves (only the manufacturer will be mining with them), and favor big producers who can design and produce it fast enough.
How much another random individual will know of you is limited by physical features, really, and of course what you might end up telling them about you. The privacy-conscious will find ways to obscure their identity and physical features if they choose to buy XMR via cash instead of mining via GPU (which BTW requires action to safeguard one's anonymity as well, like setting up a VPN with a trustworthy provider(s)) as ways to avoid having the paper trail trace back to them (e.g. law enforcement using cameras in public settings to ID you in an in-person exchange)
...which will break everyone's mining rigs again.
I called this last year when XMR started doing their retarded ASIC-paranoid forks. Sure enough, now the ASICs are running in secret and they're not available for purchase. So their fork just consolidated all ASIC mining into one or two entities. And anyone who believed the "XMR-mining botnet" narrative in early 2017 is an idiot, I calculated that every PC in the world couldn't have produced even half of that Cryptonite hashrate . It was Bitmain "testing" their X series of miners.
The funny part is that most of the ASIC manfacturers would love to get their machines in as many hands as they can - that would mean more retail sales, and less big customers asking for discounts and special treatment.
OK so you like updating your mining code every 3 months? Seems silly when they likely fab a new ASIC within 2 weeks. And as I said, all of the ASIC makers are now mining in secret. Exactly the opposite of the intended outcome!
Large miner invests time and money to develop ASIC anyway. Takes 6 months. Costs a lot of capital.
Small miner cannot keep up, but uses the same capital to mine at small profit for six months.
small miners in profit, large miner isn't
large miner now starts raking in money. Still not profitable yet. Difficulty adjustment algorithm puts a cap on rate of income but they dominate mining... Small miners switch off, large miner takes 100% of possible income.
future is bright... Until
developers change PoW.
large miner was not yet in profit. Large miner's capital investment is wasted. Cannot possibly recover it. Custom ASICS are now scrap.
small miners (who had a profitable six months) switch their GPUs back on. Life returns to normal. Large investor has uncomfortable meeting with the bank and no profit. But at least they dominated the hash rate for a week, right?
Lesson: change PoW often. Punish those who try to control hashrate in this way by making them waste their time and money. They will stop doing that for selfish reasons.
> You're neglecting the "what happens next".
> ASIC resistant PoW algorithm change
> Large miner invests time and money to develop ASIC anyway. Takes 6 months. Costs a lot of capital.
> Small miner cannot keep up, but uses the same capital to mine at small profit for six months.
> small miners in profit, large miner isn't
> large miner now starts raking in money. Still not profitable yet. Difficulty adjustment algorithm puts a cap on rate of income but they dominate mining... Small miners switch off, large miner takes 100% of possible income.
> future is bright... Until
> developers change PoW.
> large miner was not yet in profit. Large miner's capital investment is wasted. Cannot possibly recover it. Custom ASICS are now scrap.
> small miners (who had a profitable six months) switch their GPUs back on. Life returns to normal. Large investor has uncomfortable meeting with the bank and no profit. But at least they dominated the hash rate for a week, right?
> Lesson: change PoW often. Punish those who try to control hashrate in this way by making them waste their time and money. They will stop doing that for selfish reasons.
This is unbelievably foolish.
There is no such thing as ASIC-resistancy.
If Monero ever came close to the popularity of BTC, then you know who would become the next ASIC manufacturers ?
AMD, Intel, NVIDIA, DELL, HP or even Amazon / Microsoft (cloud mining) - these kind of monsters.
They can produce and install universal mining machines (meaning PC-ATX computers) *en masse* cheaper and faster than any "small miner" ever could.
The only reason they didn't do it yet is because Monero is just too small for them to create significant profit.
Again: **There is no such thing as ASIC-resistancy.** Anybody who thinks otherwise is fooling himself.
PoW is the hash of the entire Blockchain in reverse order including the new block you're mining.
Doesn't matter how fast your hash engine is, you're limited by the speed of the bus. You can't overcome that with local RAM because the Blockchain is ever-increasing.
Ta-da. ASIC resistance.
What you described was nothing to do with ASIC. That was big guys having an economy of scale... So? That's not that big of an advantage and doesn't freeze anyone out. Corner shops still exist even if Wal mart is cheaper.
> Doesn't matter how fast your hash engine is, you're limited by the speed of the bus. You can't overcome that with local RAM because the Blockchain is ever-increasing.
Again, this is stupid.
1. You can get cheaper electricity, when you buy it in large quanities. Which means you can hash it cheaper.
2. "Speed of the bus" can also be optimized to be more efficient. Custom PCs can be built, which have 50% faster bus speed.
3. You have not changed anything. The fastest miner will be now SAMSUNG, producer of cheapest and fastest SSD doing 1000 TB/s in coalition with IBM/Intel, producer of fastest photonic RAM which can do 100000 TB/s throughput and reaction/switching time counted in single microseconds.
Smartest, largest, most invested miner will be ALWAYS more efficient and more profitable than small miner.
It is the law of the free market. The biggest & smartest who invests in the most efficient method of mining on the largest scale wins.
It's like you are trying to violate the laws of this universe. That's impossible.
**There is no such thing as ASIC resistance.**
None of that is anything to do with ASICs. Again, you're taking about economies of scale.
It's not about making sure there is no big player who can mine faster than anyone else. It's about making sure that mining isn't exclusionary. As long as general purpose hardware can be used competitively, then ASIC resistance is good enough. ASICs don't just give a little boost, as your statement about 50% bus speed asserts, ASICs are orders of magnitude faster per kilowatt. That makes it literally pointless to mine in your bedroom.
None of your objections have that drawback. That Samsung can make a fast SSD doesn't mean I can't buy a fast SSD, and have an incentive for that anyway. That a PC can be made with a faster bus doesn't mean I won't be willing to have that same PC for my own uses. I can buy them in ones and I don't have to order a production run of 10000 minimum.
ASICs are different, they are a barrier to entry. The complaint is not that they're used by rich guys in large numbers it's that they can only be used by rich guys in large numbers.
You have not understood anything I have said.
I do not have time to site here all day and explain you all of your logical mistakes.
Go on and believe whatever you want to believe, making ASIC-resistant coin is impossible and you cannot change the laws of this universe with your wishful thinking.
Well, disturbing complex (network) systems never is a good idea. Back then I felt like the change to ASIC resistance was rushed but a good move. See it as an experiment, learn from it, adapt and overcome™. This may have been the only way to find out that there is no such thing as ASIC resistance, or that we are not quite there yet.
Yeah that's why POW isn't the best long term solution. You're handing over control to whoever can make the most efficient ASICS, which they'll keep to themselves and sell off when they make better ones. These companies don't give a damn about the currency but they have strong control over it.
We need to figure out how to make POS work well.
KohTaeNaiRedditor for less than 60 days2 months ago
> You're handing over control to whoever can make the most efficient ASICS
That's just a definition of capitalism, it's not a bad thing..
In the free market, we "hand over control" of banana production to whoever can grow them most efficiently.
We "hand over control" over automobile production to whoever makes cars most efficiently.
There is nothing inherently wrong with the free market.
> There is nothing inherently wrong with the free market.
There is. It's when banana producers form a cartel and set prices of bananas on the "free market". When someone tries to intervene, the banana cartel buys them or kills them socially. Lastly when governments try to intervene, the banana cartel responds by infiltrating into it and bringing laws that help the cartel reinforce their iron grip on banana production and distribution.
KohTaeNaiRedditor for less than 60 days2 months ago
> It's when banana producers form a cartel and set prices of bananas on the "free market".
They need government intervention to form cartels. Meaning the government is somehow preventing others from creating ASICs. This is false.
A real life example would be cable television in the United States. It's an actual cartel because laws make it difficult for anyone to make their own cable company.
Anybody is free to create their own ASICs, it's just hard to do. You can't point to any law preventing people from doing so, like with Cable TV.
PoS is just as capitalistic / free-market, if not more. The previos poster is right, by definition in PoS the participants have more stake in the currency and therefore should care more about its success.
KohTaeNaiRedditor for less than 60 days2 months ago
Do you know how hard it is to create an ASIC?
You need to have a degree in Electrical Engineering and spend hundreds of thousands or millions in equipment to actually create the ASIC. Plus months or years spent designing the thing.
Contrast that with POS coins. Participants need to simply buy coins.
ASIC manufacturers have *way* more skin in the game than POS stakeholders. It's not even close.
Having to pay an investment is not having a stake. Not to mention the fact they can switch chains.
KohTaeNaiRedditor for less than 60 days2 months ago
> Having to pay an investment is not having a stake.
That's literally the definition of a stake.
> Not to mention the fact they can switch chains.
This doesn't make any sense. They can only mine the most profitable chain, or lose money.
They have incentive to make the chain they're mining as profitable as possible, they want more people using it, so it makes no sense to say they don't have a very strong incentive to keep their hardware useable.
If they ruin the chain, or do anything else bad for bitcoin, they're turning their investments into overpriced space heaters.
>ASIC manufacturers have way more skin in the game than POS stakeholders.
Let's not forget that PoS is, by definition, a **permissioned** system, as you cannot onboard into a PoS system without someone willing to sell at least part of their share. With PoW you just need hashpower and literally no-one can stop you from collecting rewards.
SiaCoin was mined by FGPA for months leading up to the release of the A3. Just after the ASICs hit the network there were people releasing code that showed they were mining at over 400x the profitability of a GPU.
How often does this happen, and how many "ASIC-resistant" cryptos are being mined by FGPA (or even ASIC) miners who just keep quiet? Good question.
PSA: [Really great interview with the CEO of whatever company develops SiaCoin](https://unchainedpodcast.com/why-asics-may-be-better-than-gpus-even-if-they-tend-toward-mining-centralization-ep-67/).
He discusses how SiaCoin's design enables ASIC development. More importantly he discusses the economics and theory behind why he believes ASIC's are actually the natural state of any of these networks, and thus a healthy sign of decentralization.
I own plenty of XMR, but those guys really hate it when I bring up these ideas.
Everyone argues ASICs benefit the incumbent producers. I actually think the opposite.
If you want to start producing a SHA256 miner, you can plan far in advance since you know the algorithm will be the same, and this brings in more competition. If you keep changing it, then the biggest and fastest producer will be able to keep up with it and keep producing a constant batch of new miners for the new hashing algorithm.
In the end, keeping it the same allows for more people to compete.
>In the end, keeping it the same allows for more people to compete.
That's a good summary.
Being anti-ASIC is a symptom of people who've been infected with the "everybody needs to be able to mine with Raspberry Pis and behind Tor" mantra.
Well rather than looking at the production, you should be thinking in terms of the stable high hashrates generated by having dedicated miners that can't jump ship across 16 other algos to find the best profit.
Asics create loyal miners and stable net hash, both are good for a coin's health, regardless who made them.
My approximation is 73.2% ASICs, using a different method.
Here's my work,
It’s not much of a secret anymore, is it? This was teased out of public blockchain data that could have been found all along. If anything it shows how ASICs have tried and failed to stay secret.
All it shows is that an occasional tweak is not enough to keep ASICs at bay, because all it takes is a tweak to the ASIC to bring it back.
Ethash itself shows how totally pointless it isn’t, because they managed to pull off solid ASIC resistance for 4ish years, and the best known ASICs aren’t orders of magnitude more powerful like they are with CN. All this shows is that not every solution to ASIC resistance is as good as the next.
CN V1 and V2 were weak upgrades to an algo with weak ASIC resistance.
ProgPoW is a strong upgrade to an algo with already strong (but flagging) ASIC resistance.
You’re comparing apples and oranges here.
It’s weak compared to your typical 30X+ ASIC but it’s strong enough to kick GPUs to the curb, especially when the fork drops the issuance. Difficulty not changing can be explained by the proportion of GPU miners dropping at nearly the same rate that ASICs are coming in. Its the pattern which you would expect to see when ASICs are 1.5-2X more powerful and not 30X+ more powerful - that doesn’t mean they’re not actually killing off GPU miners and wont eventually take over completely. They don’t need to stomp all over GPUs to do that, all it means is that it just takes longer and takes more of them to reach that saturation point.
ETH developers do not care one bit about ASICs. They never did. They only care about maximizing their wealth.
Unlike other blockchain projects, the people who control ETH’s future are wealth maximalists. Many of the ETH devs are the ones building the clandestine ASICs, since ASICs are a key part of ETH devs’ wealth maximalist strategies.
Except it doesn't mean or even imply that at all, because Monero isn't running ProgPOW. And the point of ProgPOW is that it makes any potential advantage of mining ASICs insignificant versus consumer GPUs.
This is what I have feared the last few weeks. I have redirected my remaining hashing power to XMR just to stabilize the network and fight the imminent centralization. What else can we do in this short span of time and before the next hard fork?
Things move fairly quickly in the crypto space. I doubt the monero community will be content with Asics while they dominate the network until 2022. A couple of things will happen before then, and hopefully for the good. Getting to the April fork is what's coming.
Monero's algorithm change was very minor and not at all designed or expected to have a lasting effect the way ProgPOW is. And the people working on ProgPOW for Ethereum have nothing to do with developing Serenity, so giving up on it would do zero to hasten POS and a lot to hasten mining centralization in the interim.
ProgPOW would result in a contentious fork, guaranteed, whereas the switch to Serenity is and has been expected by all from nearly the start. That is why I'm reluctant to get behind a POW switch at this stage, although my initial thoughts were in support of it.
I disagree that it would be contentious. ProgPOW would be better for everybody except ASIC miners. Those who think otherwise have either been swayed by unfounded FUD or are missing the forest for the trees.