Any asic-love and inflation-hate around?

If the 3k altcoins are anything to go by these opinions are extremely rare. But in my mind this tech is worth saving from the standard altcoin koolaid that gets passed around. So I’m throwing it out there early.

  1. funding botnets and the gpu shortage is rather bad, while sha256 is less than ideal as the race is ongoing, I don’t actually believe claims that new algorithms won’t restart the race, they are in a honeymoon period the race will begin again for many coins once there is enough money on the table to do so.

  2. The end state of the asic race, is a near perfect design being public-ish(handful of major companies) at the transistor limit of current tech. This is a perfectly acceptable state to be in as the race slows to the relative snails pace of moore’s law. Yes ideally you wouldn’t want any of it behind closed doors. But that isn’t in the cards right now; no one has produced a mathematically perfect asic algrothium pair and I don’t know how you would do something like that.

  3. Keeping a hardware requirement I believe slows down nation state intervention; a totalitarian can steal every chip, but if you need special chips and only special chips matter well your not generating enough money for the resistance when you steal cell phones from average people.

  1. I don’t think the speculation economy that’s spawning around altcoins is remotely healthy, cloud mining and stuff like ponzi coin are hyper degenerate. Real resources are being burned for “currencies” that are not being used. Yes rampant speculation will be a necessary step to cross the catch 22 of how to you create an economy with coins given away for free(or declaring you own it all). But it should be cut short how long it can survive bitcoins 100 year plan and eth style endless inflation should not be on the table with the 20/20 hindsight we have. The price doubles every 3 months, maybe try halving the coin reward that much as well? Or be on the safe side and slow to 6?

I don’t understand what you are saying - are you advocating for ASIC support? I know that GPUs keep getting more expensive and hard to find, but ASICs are still out of reach for most individuals. How can a person of modest means participate and earn GRIN without access to mining? Maybe there’s a way to make it ASIC and GPU resistant and only cpu mine? I am in favor of having maximum accessibility to mining however it works and for now that means I am in favor of ASIC resistance.

In terms of inflation, I am typically more in favor of a capped supply instead of a constantly growing one. I think having a capped supply makes the currency into more of a store of value than a means of exchange over time since the cost of a limited resource will rise. Even though I am an “inflation hater” I am intrigued by the linear emission rate of GRIN. I understand that the inflation rate drops over time and will never be as high as the inflation rate of most fiat currency. Some inflation can be a good thing because it makes the currency easier to use for more people as it becomes more widely adopted. By easier to use I mean that the price of the currency should stay lower as it is adopted, versus a non-inflationary coin (BTC) increasing in price over time due to limited supply. I think inflation is ok for a “means of exchange” currency.

I think its generally quite tricky to make a POW algorithm GPU resistant. Even if you could, mining would quickly get dominated by owners of bot-nets. Even GPU mined coins are being done in bot-net style, e.g. javascript Monero miners being executed without permission of the end user.

If you are a miner and you do not have to pay for mining costs, then it skews the incentives somewhat.

I am also not in the ASICs-are-evil camp. Especially given the end-game were ASICs can become just as “commodity-hardware-like” as the memory we depend on in POWs such as Cuckoo cycle. We arent there yet though.

I am of the position that speculation is actually quite important in order to bootstrap a currency. For the currency to be useful we need many people to be holding on to it. Getting the variables correct seems like a tricky problem and probably best left to economists, although I am skeptical that they know what they are doing either :stuck_out_tongue:

With an inflationary currency I might worry that people would do the rational thing, and simply exchange all their GRIN for a deflationary asset (e.g. BTC), as holding onto it for any length of time would be costly.

I do however predict that within a short duration of Grin’s launch there will be a number of competing MimbleWimble currencies which take different approaches, including the following:

  • Air-drops to BTC holders
  • Deflationary block-rewards
  • ASIC proof-of-work

The market will then decide what is preferable.

The constant emission rate is probably the most controversial decision about Grin… I wonder how that would play out. Probably someone would fork off Grin just to adopt a deflationary rate and it would be easy to market, because people are accustomed to and like deflationary emission schedule…

Air-drops to BTC holders

Is such a thing possible without being extremely ugly and privacy killing?

I’m aware of the one air drop that required a facebook login to claim vomit other than centralized databases making the genesis block I don’t see it

Probably someone would fork off Grin

I vote gir-coin

CPU-based mining protocols typically trade accessibility for security in the sense that although it is easier for anyone to mine (implies more decentralization), datacenters/large organizations won’t be securing the network with their hash power, so large outside attackers can rent a significant amount of computing power from Amazon AWS to attack the network. Specifically, an attacker that owns a significant amount of fiat could very easily launch an attack on a network reliant on CPU-based mining by renting Amazon AWS computing power. From my limited understanding, Grin mitigates this outside attacker risk by attaching range proofs and aggregating transactions in a verifiably secure way. This may already exist but it would be great if someone could compile a list of the attack vectors specific to Cuckoo-based PoW mining (in the context of Grin).

I don’t think this is correct. From what I have understood, the range-proofs are a necessity to prevent people from abusing the hidden transaction amounts, with negative values, or ones that can overflow, in an attempt to print money.

While the aggregation of transactions is a great feature for efficiency (saving space) and privacy gains.

While this is good & interesting stuff, it bears little relevance to the discussion of hash-power decentralisation.

It’s not just range proofs (I said “and aggregating transactions in a verifiably secure way”). The choice of using Schnorr signatures is what would eventually mitigate abuse of the protocol–

That’s not true. The two are very related. Hash-power decentralization implies less security if mass adoption is not immediate (and it’d be idealistic to assume it would be). So, we sacrifice security for decentralization, but a verifiably secure signature scheme bolsters security and solves this problem.

CPU-based mining protocols typically trade accessibility for security

Grin mitigates this outside attacker risk by attaching range proofs and aggregating transactions in a verifiably secure way

Your confusing two different security systems.

pow is about making sure the order of the transactions is agreed on, the transactions are about finite coins and ownership

attack vectors specific to Cuckoo-based

Asic races if someone puts L3 cashes on custom chips

1 Like

PoW is about ordering transactions but the reason we use the Cuckoo-based hashing algorithm is to prevent ASIC mining. This means ‘normal people’ are incentivized to mine on the network. If we take this to be true, then the hashrate is going to be lower than that of Bitcoin (or other alts that encourage ASIC mining). If the hashrate is lower, then the chain is more vulnerable to Amazon AWS attacks.

But, my whole point was that certifiable transactions (which Schnorr signatures would pave the way for) mitigate the risk of attacks in general because they allow for verifiably secure block validation, meaning that the Amazon AWS attack would quickly be detected or wouldn’t even work because you just can’t manipulate transactions in the same way that you would with Bitcoin…so double spend attacks would be easily detectable and maybe even impossible because transaction validity is verified by the encryption scheme.

I am not convinced that there will be many ‘normal people’ mining Cuckoo. I think it would still be more profitably done by large GPU mining farms and not home PCs or amazon AWS.

We also can’t compare hashrate across different POWs, but I would definitely agree that emerging blockchains such as GRIN will be far easier to attack with malicious hashpower than Bitcoin. Bitcoin would also have been pretty easy to attack in the early days.

That is interesting if hashpower attacks can be mitigated in this way, though I haven’t seen that claim being made much. Do you have some references for further reading?

certifiable transactions

Our Contribution: Certifiable Bitcoin Addresses. This paper describes
an extension of the Bitcoin protocol that preserves its decentralized nature, while
also enabling payers to optionally specify the involvement of a trusted authority
that attests to the identity of the payee,

Certified Bitcoin transactions: The figure shows all possible types of transactions
in a ledger with both standard and certified bitcoins

This work has been partially supported by the TENACE
PRIN Project (n. 20103P34XC) funded by the Italian Ministry of Education

Ew. Could we not?

Not all tech is morally neutral, and white lists are on the totalitarian side

1 Like

Lol yeah. That is certainly not a trade-off that I would be in favour of taking. Better to address the problem in hand and work to prevent hashpower based attacks.