1

I see all older questions [*]. But I read in reddit

Yes. But your efficiency is lowered. Use ethOS dagless mining.

My question is whether this is true and which software allows this? I see ethOS is mostly a collection of other programs. I don't want to use ethOS. Which exactly miner allows this?

Improve this question

1 Answer 1

Reset to default
1

The miner software in question is closed source, so I don't know exactly how it works. But from what I've read, IIRC, it generates the parts of the DAG as needed rather than precomputing it all up-front. I would guess it uses something akin to page faults to figure out what is needed. To the best of my understanding, nothing is ever evicted from VRAM (to avoid having to recompute parts of the DAG that have already been computed). Thus, I think it's more accurate to call those miners "non-pregenerated DAG miners" or maybe "JIT DAG miners".

In any case, memory for the entire DAG in such software is still set aside to work with a nice flat memory space, so there's no getting around the memory requirements with said software, AFAIK. In fact, the online help for command line arguments for this software explicitly state that larger buffers (to contain the entirety of the DAG for later epochs) can be preallocated to improve stability which implies that space for the entire DAG must be allocated at startup.

Improve this answer
7
  • thanks for the explanation, but where exactly can I find this miner?
    – akostadinov
    Commented Sep 24, 2017 at 17:04
  • You'll have to wait for someone else to give you a name. Giving a name is an almost implicit endorsement of the miner, but I feel like the percentage fee taken by this author exploits peoples' inability to do math and others' laziness in determining whether or not the miner is actually any better for them. Not that there's anything wrong with the software itself, AFAIK, but...
    – lungj
    Commented Sep 24, 2017 at 17:37
  • The fact that this person/people extracts (at my low-end estimate) of upwards of $3M/day from people on the ETH chain alone speaks to the economic rent being collected by this person/group (this entity also collects payment on the ETC chain and has miners for other chains). That is, this person simply pits one miner against another (elevating the miners' cost/ETH for elec+fee) in a zero-sum game and is currently pocketing 1.3%+ of all the ethers generated every day. This entity's economic incentives also don't align with miners as it seems to cash out to bitcoin regularly, lowering ETH prices.
    – lungj
    Commented Sep 24, 2017 at 17:37
  • 1
    That is, this entity is collecting money off the top of miners' revenue whereas miners only pocket the bottom (the profit) and, by cashing out, depresses the price of the cryptocurrency. Like a mutual fund, the entity makes money regardless of what happens to the people actually taking on risk. This puts the software author at odds with miners. While the author may not be predatory, human nature results in the author(s) collecting rent from individuals who would be better off without. So, sorry, I will not provide the name of this software. End rant.
    – lungj
    Commented Sep 24, 2017 at 17:46
  • 1
    Made an edit (it's closed-source). The second part is separate from your main question above. Mining is a zero-sum game: if miner A mines a block, it deprives B of the mining reward. It's essentially an arms race in a war: one can win more blocks through numbers or by a technological advantage. Here, the developer is war-profiteering: by giving someone a nuclear weapon, there is a strong incentive for everyone else to arm with nukes. In this case, the rewards are the same, but mining gets more expensive because everyone now wants "nukes" and the developer always takes a cut.
    – lungj
    Commented Sep 25, 2017 at 17:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.