@jamest, you’re glossing over the single most important part of that quote:
The Zcash team doesn’t believe that changing the Equihash parameters or switching to another PoW algorithm is either feasible, secure, or has better economic effects. Even if they believe that ASICs will benefit the Zcash ecosystem economically, that’s enough to justify not changing anything according to that statement. It seems like you’re just reading what you feel supports your position and ignoring everything else.
No, the part you have highlighted does not refer to the first sentence. It refers to the second sentence, which talks about more “drastic changes” such as switching to an entirely different system (which I assume means algorthim) or proof of stake.
changes are in the works, its called research and ot takes time
All of the other Forks happened immediately because they don’t seem to care as much about the value of research
I believe this is because the values surrounding their projects are not the same as Zcash
How can you honestly say that with a straight face? You can choose to interpret that quote in a way that suits your agenda, but it doesn’t make it true.
It’s EPIC how much you are forcing something that simply isn’t: mirror climber!
You can’t. It’s not the slightest bit ambiguous.
I find it more than strange that me, a foreigner and non native English speaker, has do give English lessons and look up English definitions, lol.
Both sentences are in a paragraph, it simply doesn’t matter if it’s written with 1 or 5 sentences:
paragraph, noun: a short part of a text, consisting of at least one sentence and beginning on a new line. It usually deals with a single event, description, idea, etc.
@jamest what hardware do you mine with? There are other projects you can mine for and exchange for Zcash so as to achieve your margins
Whether or not GPU mining for zcash will improve again I guess we’ll just have to wait and see
If i remember right jamest claims not to mine but only running 3 private mining pools.
How is this any different than large farms being able to buy 100 or 1000 GPU’s directly from the manufacturer?
I already answered this question:
Maybe retail GPUs were in short supply.
Maybe the buyer was able to get a slight discount if he bought in bulk.
Maybe the buyer was able to get a slight discount if the cards didn’t include DVI ports.
If we assume that the 125,000 GPUs were the equivalent of GTX 1060s and were sold at 80% of the retail cost, we can estimate that this resulted in roughly the following:
25 million spent
37.5 MH being added to the Zcash network
Now let’s look at the equivalent spending on ASICs:
25 million spent
550 MH being added to the Zcash network
Do you see the difference here? The huge bulk order of GPUs doesn’t drive up the diffiuclty in nearly the same way as the ASICs do. (For my estimates, I’m using $200 and 300 h/s for GPUs, and $500 and 11,000 h/s for ASICs.)
And again:
If they are 1080 Tis and the assumptions remain the same, then this is the effect:
25 million spent
30 MH added to the network
This is a good example of why GPU mining is so fair – most modern video cards are priced identically in terms of hashes per dollar. If anything, the budget cards are more efficient. This opens up mining to so many people, even those with an average PC.
By the way, you’d have to assume that GPUs were sold at less than 5% of their retail price to get the same effect on the network as ASICs.
GPU mining is far more fair and decentralized than ASIC mining.
The effect of highly efficient ASICs on the network is fundamentally different from that of adding more GPUs. It changes everything, and in a way that benefits only the people who manufacture ASICs. The entire Zcash block reward is now the revenue stream of Bitmain and Innosilicon.
It’s pretty sad to see all the hostility toward asic machines. I am running several GPU’s and my power bill is about 2400USD per month. I am tired of paying that much when there is a better way. Why would I want to use the wattage the GPU’s require when I can get a machine that can do 180,000 sols/hash at 2200 watts? I run several asics as well and I am quite happy with them. They withstand elevated temps better, are easier to ventilate, and the build out plan is much more simple. I am considering selling all of my GPU’s and just moving out of the stone age.
If you want to insist that your particular coin should continue to be able to utilize GPU’s for mining and want to keep throwing your profit to your power company then have a look at what the folks over at digibyte are doing. They recognize that people still want to be able to GPU mine their coin so that are making “some” of the algos that coin uses asic resistant while leaving others (sha256 for example) asic friendly.
Small time miners like myself who are just trying to earn a living instead of working a dead-end job don’t really deserve to be screwed just because people love to hate change. If there are to be changes, they should take the needs of all into account. Any solution that is fair would balance the needs of both camps GPU and ASIC alike.
The premise that sabotaging ASIC miners somehow encourages decentralization is a joke. Who do you think is buying the bulk of the GPU’s produced world wide? It’s not the average individual mining at home. It’s huge mining farms. They can buy 100’s or 1000’s directly from the manufacture at a much lower cost that I or anyone else I know will ever be able to get them. If I had it all to do over again I would never have wasted money on GPU’s, and instead would have bought ASIC’s. I was a no.0b then, and didn’t know any better.
Another thought is that the more a coin tries to prevent asic’s, the greater the chance that coin won’t be around in the future. This is the case for a number of reasons. To the same point, if we really want to graduate to real mining hardware and prevent centralization, then we can’t keep screwing the people making the machines. If the algos stop changing every time someone cries about asics, then other companies will start making good mining hardware as well. There are already quite a few companies making good hardware, and if they stopped getting screwed with algo changes then there would be even more. This would mean that no one would have a monopoly (as it were) on any one algo with their machine.
At any rate. I am probably fighting a loosing battle sadly, but change is inevitable and eventually all algos are just going to have to adopt or disappear.
Kwalker38,
Your analysis is flawed. You need to think about your hash rate relative to the network hash rate. That is it. That is the only thing that matters to a miner.
ASICs do not result in you getting a larger share of the network hash rate for a lower price. They do the exact opposite, because the ASIC manufacturers can build the ASICs at 1/10th the price they will charge you. The inevitable outcome is that networks with ASICs become mineable only by the manufacturers or their insider friends.
Thanks for the speedy and respectful response. Your point’s are valid, but how will there be any positive graduate change if it is continually prevented? You seem like an intelligent person good to have a conversation with Jamest. 
It’s a misconception that ASICs are a positive change.
I understand the superficial resemblance between, say, cars replacing carriages and ASICs replacing GPUs. You could argue that GPU miners’ attempts to resist ASICs taking over are just like carriage operators’ attempts to resist cars taking over. This is quite a tempting but specious argument.
The difference is that cars add value to society by enabling people to travel further and more quickly. There is a net gain in utility from the introduction of cars. It’s true that some individuals may be worse off (e.g., carriage operators or manufacturers), but the net change in society is positive because far more people derive enormous utility from cars than are hurt by them.
Similar analogies could be made between tractors replacing farmers or robots replacing factory workers. Some people are worse off (those who lose their jobs), but far more people benefit from more affordable and abundant food, as well as cheaper and better-produced cars.
When it comes to ASIC mining, the fundamental difference is that no utility is added to society when ASICs take over. Revenue from mining is entirely independent of what hardware is mining it. Moreover, we have observed an increase in power consumption from ASIC-based cryptocurrency networks – a greater percentage of the mining revenue is consumed by power costs compared to GPU networks. So you can conclude that not only do ASICs not add value to society, they subtract value!
In short ASICs are bad because they result in a net decrease in social utility and economic value in society.
How did you know i was thinking of carriages (modle T’s actually) and modern automobiles when I wrote that ?
RX 480 29.5 MH/s@135W(conservative downvoltage) mining ETH $0.84/day - $0.1/kwh electricity = $0.51/day
Antminer D3 16GH/s@700W(downvoltage) mining Dash $1.86/day - $0.1/kwh electricity = $0.18/day
Who pays bigger bills? Ethereum GPU miner or Dash ASIC miner?
P.S I know about efficient DaggerHashimoto ASICs from Inno but today Ethereum is mostly GPU coin and DASH is ASIC-only. And even though more efficient ASICs are in the ETH network GPU still show better cost and power efficiency than ASIC-only network.
Just saw this, and I want to point out that when I said “the memory requirements are currently set very low”, I meant it: at the time, the parameters were (96, 3). If you look at the linked issue (specifically this comment, and this subsequent PR), you can see that we bumped the parameters to the current (200,9), which at the time did take closer to 1 GB of memory; it was only after we froze the consensus rules (and thus were not going to alter the parameters again before launch) that mining software developers worked on optimised implementations that only required ~150 MB.
Then something is wrong with algorytm …memory requirements should be memory requirements …and not 1gb is memory requirements but still can run at full speed at just 144 mb.
ETHhash has memory requirements and they are real requirements in real life.
Bitmain Antminer E3 for ETH hash has 72 GB off ram …18 chips that run at around 10 mhs with 4gb ram each.
WHen you saw that it only need 144 mb off ram you should change parameters to much more very fast …you know that 144 mb was CALL for ASICs because they could just throw lot off chips with cheap ram and drow GPUs out which have 95% off memory unused on Zcash …and have extra profft …not to mention that memory speed on your parameters are totally irrelevant…on ETH Bitmain profit is probably very very low…on Zcash they profit is probably 100 times more then on ETH…which just make them stronger and stronger and 99.9% off others weaker.
I looked more up to this product.
Not only it promises boost in performance but also 30-40% less power draw from GPU’s.
If this is true it could be a very valid solution for GPU miners.
Just wondering if it can handle a 12/13/14 GPU rig…