[bseidensticker]

I managed a Bitcoin mining operation a long time ago. We made machines that made 50% of their BTC in the first month of their operation, and the remaining 50% over the next 5 months. After that they cost more to run than they produced.

The accountants wouldn't believe us when we told them that the machines were worthless after only 6 months.

[chucknthem]

How long ago was this? The rate of improvement on ASICS as well as new entrants to mining hasn't been this intense in a long time.

[bseidensticker]

2013

[chucknthem]

That makes sense, 2013 was when ASICs were getting popular and started dominating mining. The industry has slowed down significantly since. Presumably they've implemented most of the low hanging fruit optimizations. Antminers from 3+ years ago are still profitable in a lot of areas with <=$0.1/kWh electricity.

[Dmitriysz]

This makes sense. In the earlier days of bitcoin we’ve seen substantial improvements in mining every 3-6 months.

We’re talking about significant 100x-1,000x improvements in efficiency and profitability.

To stay competitive, these improvements in efficiency made it necessary for miners to upgrade mining equipment every 3-6 months.

Because of its closeness to silicon fabrication and cheap energy, China emerged as the de facto place to set up these mining farms. Those previous 100x-1,000x improvements are no longer possible.

The development of mining hardware caught up with Moore’s law.

This is the observation that transistors in a dense integrated circuit doubles about every two years.

It would be fair to assume ~2x improvements in mining every ~2 years.

This would mean that mining equipment would now have a longer shelf life, and no longer be obsolete in a few months.

[yuvadam]

Worthless? People still mine on S9s released in 2017.

[rtkwe]

You can mine on worse machines if you have lower power costs or are valuing the BTC higher than the electricity costs (ie you're long/bullish on BTC or you're doing it somewhat altruistically or alternatively not doing the math on how much it's costing.)

[redox99]

If you're bullish on BTC you will still get more BTC by buying them than mining them at a loss.

[SturgeonsLaw]

Miners might be happy to pay a premium to get non-KYC coins

[redox99]

True, there might be tax reasons too even if they don't care about KYC.

Although most miners will know people who they can trade face to face with if they don't care about taxes and don't want KYC.

[8note]

Sounds... sketchy? I'm not sure how that would assuage the risk adverse accountant

[q-big]

> I'm not sure how that would assuage the risk [averse] accountant

If the accountant's salary depends on being assuaged, this should hardly be a problem: the accountant will then just tell you that he does not believe this to be a good idea, and give reasons for his judgement.

[philihp]

That's an expensive premium. How much is a crisp benj worth compared to a gently used one, laced with trace amounts of cocaine?

[jimmydorry]

A crisp benj is worth about 21% extra (the typical tax rate of a business). :)

[threeseed]

Replace worthless with unprofitable.

And we don't know if it's unique to his situation or for any miner at a certain scale.

[pelasaco]

Well, I managed as well some hardware and in the long run, they paid themselves. I started to mine with FPGA and then with asicminers. I mined in total, 23 BTCs. Then I stopped and kept the BTCs, (sold some), sold my hardware via ebay (which helped to return almost all my investment). I then got the BTCs doubled when bitcoin cash was created, which I sold well enough too. So when I read this whole "it doesn't worth to mine, etc", i think that is relative. It worked well, even for me, running it in Germany.

[amelius]

Worthless? Did you use ASICs?

[VHRanger]

Necessarily so.

It's been over 7 years since mining on anything except ASICs is a guaranteeing a loss.

[jhugo]

, if you are paying typical prices for electricity.

Most people are, but many major mining operations are not.

[Thorrez]

Are you talking about mining on a GPU or CPU? You should mine a different cryptocurrency; it'll be way better.

[jhugo]

Eh? I'm not mining anything on anything, just pointing out what the commenter I replied to left out.

If power is free then any hashrate is profitable. With "typical" power prices you basically need ASICs. In the middle there are myriad possibilities, and many miners have access to very cheap power (eg during hydro gluts in monsoons).

[Thorrez]

I was asking what hardware you were talking about. "You should mine a different cryptocurrency" wasn't meant as something specifically for you the commenter, but as a recommendation to everyone in the world. Theoretically you'll get something by GPU mining bitcoin, but it'll be tiny. You shouldn't do it. Instead you should use that GPU to mine ethereum. That'll be much better.

As an example, the AMD Radeon HD 7870XT[1]. It gets 346 Mh/s for bitcoin, which is $0.00004308/day[2]. It gets 14.5 Mh/s for ethereum, which is $0.3508/day[3]. So you get 8000x more by mining ethereum.

[1] https://technical.city/en/video/Radeon-HD-7870-XT

[2] https://www.cryptocompare.com/mining/calculator/btc?HashingP...

[3] https://www.cryptocompare.com/mining/calculator/eth?HashingP...

[bseidensticker]

Yes, we made our own. Our company was third to market.

[andy81]

Bitcoin mining is almost entirely ASICs. GPUs aren't much good at the algorithm used.

[rcxdude]

GPUs are pretty good at the algorithm (certainly much more than CPUs), just nowhere near as good as ASICs. Mostly this was because the proof of work function wasn't really designed with any such target in mind, and the basic cryptographic primitives available tend to lend themselves to ASICs dominating everything else unless you explicitly design it otherwise (like ethereum's PoW which is intended to be memory bandwidth limited and thus best suited to GPUs with ASICs offering minimal benefit, or Monero's algorithm which is designed to be extremely branchy and thus best suited for general purpose CPUs).

[Snowworm]

Wow. I can't delete comments in hacker news. That's so stupid. Anyway, I was getting downvoted for a stupid opinion FYI.

[majewsky]

You cannot delete once someone has replied. And by editing away your opinion (thereby breaking the context of the replies) instead of owning up to it, you will be earning many more downvotes.

[jagger27]

GPUs haven't been viable for Bitcoin mining for over a decade.

[musicale]

> GPUs haven't been viable for Bitcoin mining for over a decade.

I guess that explains why GPUs are so cheap and plentiful.

Or perhaps they're mining another crypto currency.

[cinntaile]

Yes, they mine Ethereum. GPUs are no longer scarce though, the prices have gone down a lot as well.

[amelius]

Perhaps the ASICs should be designed so that they are a little more general-purpose then. Perhaps capable of mining for the next cryptocoin on the block?

[rcxdude]

ASICs only offer a benefit through significant specialisation. Specifically with bitcion the double-SHA256 hash lends itself to an extremely efficient hardware operation which beats any general-purpose system substantially.

In general what you want to do is design the algorithms differently. There's a few reasons to do this. Firstly your new coin probably doesn't want to run best on hardware which is already used by a bigger coin, especially bitcoin, because then there's a massive attack vector just looming over you as the existing miners could just turn a fraction of their processing power towards your network and destroy it easily. This means you generally want to come up with a novel proof of work function which is not easily computed by existing miners. The best form of this is using a completely different hardware resource.

Secondly (and apart from the few large coins which generally got to this idea first, competing with the first point), if you care about keeping your miners distributed in terms of ownership and control, it helps if said algorithm is tuned for some general purpose bit of hardware your users are already likely to have for some other purpose. Ethereum's PoW was expliticly designed to be optimal for GPUs (because computing it is memory-bandwidth limited), and Monero's was designed to be optimal for CPUs. Chia's system was optimised for storage space instead of compute power. Lower-effort clones (e.g. dogecoin) tend to at least change the hashing algorithm used at little to provide some protection, but to a lesser degree (e.g. Ethereum is probably safe from attacks by bitcoin miners, Monero is probably safe from ethereum miners, but while dogecoin is probably safe from attacks by bitcoin miners, it's not safe from the other larger coins, and especially bitcoin forks have actually been attacked by bitcoin miners).

[wyldfire]

But...they're application-specific, by design. If they were more generalized, they'd start to approach CPUs or DSPs.

If you devoted extra area to some other algorithm, it would be a big cost for no benefit. Put that logic on some other ASIC and buy that one instead.

[jagger27]

This is sort of what Nvidia did with the cards they sold directly to miners. At the end of the day they're still GPUs (albeit without display outputs) so they could in theory be repurposed for work that actually benefits society.

In practice that would never happen because by the time miners are done with the cards they'd be too out of date for machine learning or other general compute workloads, and useless for budget gamers because of the lack of display outputs. It's really a shame how much power and silicon is being wasted on cryptomining.

[RL_Quine]

That would make them less efficient for no reason.

[cortesoft]

Ummm a more general purpose version would just be a GPU. The point is that it has to be specific purpose to be efficient.

[greyhair]

It is really hard to design an efficient ASIC to predict the future compared to just designing an efficient ASIC after the future arrives.

The tools available today support such rapid design/test cycles.

[miked85]

You seem to be missing the point of what an ASIC is.