So with GPU’s getting slowly left behind in Grin, I figured I would bring up the topic before the end.
As @lehnberg mentioned in his recent post about ASIC’s, the last GPU algo is going to be phased out in the next 5 months, as expected. Coming from a GPU miner standpoint is kinda sucks because so many networks keep phasing us out. Ethereum, Monero, Grin, etc. While I am aware some GPU’s can still mine C32 until they are outranked by future ASIC’s, GPU’s are again left to go somewhere else. While GPU’s may not be the best way to hash, we bring huge communities and passion with us when there is an opportunity.
Grin may consider a different sight of mining in the future. Why not support all the popular methods of mining. Heck, you could split ASIC’s GPU’s and CPU’s on the same network and fairly distribute grin. It is meant to be used by everyone so why not let it? While ASIC’s might be the most efficient, why constrain ourselves to a specific mining community? We honestly rely on miners to support the network and its future. In a way, other then the dev’s, GPU’s kept grin going and helped build its community. Without GPU’s who knows if Grin would still be around. Maybe there will be a new community excited about ASIC’s as well, who knows.
Anyways what sparked this is the phase-out of GPU mining, most likely will be forced somewhere else because of it.
This has been answered many times already, so let me be brief. An ASIC resistant PoW is only as good as its frequent tweaking. I think Grin has led the way in this department with 6-monthly C29 tweaks that no ASIC manufacturer could have conceivably prepared for. But these come with a centralizing force of frequent Hard Forks and trust in the PoW tweaker. As well as risks of leaving a weakness in a PoW that could not afford the time for extensive review (as was painfully obvious with Cuckaroom29). With Grin only having 4 preplanned HardForks and moving to a more decentralized and immutable state after 2 years, the time for such frequent PoW tweaks has run out. Leaving ASIC resistance infeasible…
I do appreciate the response tromp. I personally have nothing against ASIC’s and think they are awesome. But so far our biggest asset is GPU’s while ASICs have not been very interested if non-existant. We’ll have to see how ASIC’s pickup but maybe the network will have to adjust. Personally I agree that constant hard forks are not the answer but guaranteed there is a better one .
There are options out there if we wanted to explore them. Namely Progpow. It’s worked out wonderfully for Ravencoin so far. The intention isn’t to eliminate ASICS, that’s silly. It’s to put GPU miners and ASICS on similar playing fields.
It just sounds like we don’t want to explore that option.
Second that it sucks to be loyal to a coin then pushed out.
Progpow is terrible for Grin. Horribly complex, slow and memory-hungry to verify, and its ASIC resistance won’t last (expecting them to have to tweak it within a decade). About the worst possible choice for a minimal and lightweight coin.
You knew since before launch that GPU profitability would end.
@tromp You are absolute right from a technical point of view. However, I do also want to express my regret that GPU mining has to be phased out. The reason I regret it is not only since I am mining GRIN on my GPU, but because it conflicts with the ideal of Grin being FAIR in its distribution.
Pure ASIC mining will always give an advantage to larger, wealthier parties who have access to GRIN through mining for a lower price than what is asked on exchanges. ASIC’s are often hard to get for individuals who want to buy just a single machine. Personally I love that mining is an option to supply me with Grin without having to deal with KYC and exchanges.
Therefore, if somehow a feasible solution can be found to keep GPU mining in the game that would be great. Alternatively, Grin can set the example by making deals with ASIC producers to sell their equipment for fair prices to individuals. This could be insensitivised in many ways, for example by making a deal not to phase out the algorithm as long as miners are supplied to individuals for a certain price. Anyhow, just some thoughts on how to keep individuals interested in Grin and give them FAIR access to GRIN through mining.
That’s all I’d ever ask for is fairness in mining.
Once it goes ASIC though, that typically goes out the window.
I wish there was a way to incentivize the manufactures to distribute the machines fairly once they do come. Although that’s generally not the case as it’s easier to just sell one Oder to one person than 10,000 orders fo 10,000 people. Not to mention tarrifs make the ASICS a non starter for most Americans.
A second alternative to GPU mining would be to offer fair investment opportunities in Grin mining farms since in that way individuals can even benefit from low energy prices and economy of scales. However, privacy of the share holders, trustworthyness and managements of these farms is always a risk factor. Personally I would only trust such a fund if the core team is involved.
Changing the mining algorithm after it was very clearly defined and communicated before the launch would make this chain a joke in my eyes. Everyone knew the rules prior to launch. It would be unfair to change them now.
Yes, it would be unfair to ASIC developers who have already invested time and money. However, stimulating ASIC producers to offer small mining units for fair prices to individuals, woud be fair for both the producers and small time miners who will stay connected to the project. Additionally, fair prices and access to ASEC’s for small miners would improve the decentralisation which is good for security.
I agree, I would love that, but I don’t think that’s how business works. Asking them to have low prices is similar as going to Tesla and saying if they could lower the price of their cars so that people with less money could afford their cars. Would be a nice thing to do, but it hurts them so they are financially incentivized not to do it. I think the market will drive the price of ASICs just like for any other thing.
I am not saying the price needs to be low, just fair/stable. It could be as simple as setting a fixed price and a maximum for batch discounts. This would actually be beneficial for producers, larger scale for production drives the price down and greater insurance of ROI with a fixed (somewhat stable) price. Afterwards the marked takes over.
So no, there is no way to control the price, but fairness in price can be stimulated like in any other market. Even simply communicating with ASIC producers that there is a market for small mining machines or possibly lose mining cards, could already make a difference.
Sorry for digging old topic, just to have one question: Is GPU no longer able to mine Grin? Or the profit is too small comparing to ASIC? I need to update my blog as well.
.