201–210 of 520 posts
Earlier quoted context omitted.
Heat = bad for silicon. Also electromigration. And probably a couple other effects I don't know about.
Heat is bad, but MTTF at usually achievable temperatures is hundreds or thousands of years. Electromigration matters on the order of 100 years.
Intel messed up their life expectancy calculations, and thus they died after 2~3 years. I think 2 are still alive.
Earlier quoted context omitted.
How long until gpgpu isn't cost effective for mining? I know a lot of fpgas went on ebay for cheap when the new xilinx line came out. The watt per hash has to be lower for fpgas already. Will the cost of a coin always reflect the current price to make a new one and thus gpgpu will always be a cost effective technique?
I think right now biggest earner for GPU mining is Ethereum. And (I think) it's aiming to move to Proof of Stake sometime in next 12-24 months, which kinda time limits the utility of both Eth-specific mining chips, and large investments in GPU mining (unless mining burns GPUs out really fast?). Regardless, most proof of work algorithms are already being optimized with ASICs (ala bitcoin)... Or, like Ethereum's ETHASH…
GPU mining is going to become much less attractive this year due to Ethereum moving to Proof of Stake, right? If this is the case (perhaps an expert can chime in) - it makes sense to limit the hash rate now, to prevent these cards being dumped later this year? And to redirect miners towards higher-margin products in the meantime, whilst ensuring better supply for gamers (who are the actual long-term customers).
The ETH2 'beacon chain' is a meta-chain that's supposed to checkpoint a bunch of 'sub chains' for scalability. But they have launched it with many big questions unanswered. What does a 'sub chain' actually look like to interact with? How do you coordinate with many of them? All these questions have answers in theory, but not answers in solid production ready interoperable code. Furthermore some of these problems like…
That could create a lot of momentum/adoption for ETH2, very quickly. And it wouldn’t take much: all these substrate projects are intentionally architected so that you can just develop for them as if you were developing for an ETH1 side-chain, and defer all the operational questions to network launch time. They’ve intentionally commoditized themselves!
So, as long as ETH2 is the best choice for a substrate when these projects go to mainnet, it’s what they’ll pick. And, for many reasons (that all mostly come down to “lifetime cost of bridging to either ETH1 and/or chain-foo-where-DEX-foo-lives”), ETH2 may be the best choice.
(Imagine if you developed your project against some Postgres-ish-DB-aaS cloud provider like Greenplum Cloud, but only used regular Postgres features; and then, when you went to production, you looked around and decided that Amazon RDS was good enough—and on top of that, required no re-engineering, since you weren’t doing anything fancy.)
Earlier quoted context omitted.
People invest in all sorts of things they don’t seriously expect a financial return from. Like sports teams.
Yeah but that's just spending money, not investing.
Earlier quoted context omitted.
How do you think gaming streaming services work?
So I need two machines to used this GPU? Thats even worse. I know some trickery with an onboard CPU maybe works and thought you meant that.
No? Why would they need to be in different machines? You can stream from one GPU to another, whether they're in the same machine or not. One GPU may be a cheap little one built into your motherboard. You probably already have it anyway!
And even if you did need two machines... again that's how gaming streaming services work. Cheap end-user machines or iPad or just TV, and then a gaming system streaming video output somewhere else.
Earlier quoted context omitted.
'One of the biggest contributors to climate change' - this is so far from being even remotely true
Perhaps overstated, but it is a noticeable percentage of global electricity use spent on what is at best an enabling technology for the somewhat quixotic pursuits of a tiny fraction of the population. You should of course make your own judgments, but many people do not consider that a good tradeoff.
Earlier quoted context omitted.
Gamers will generally try to avoid buying preowned graphics cards that have been used for mining. Running them 24/7 at full pelt tends to “age” they quicker than one that’s just been used for gaming.
Can electronics actually wear out?
Another less related to electronics, is simple heat cycles.
Earlier quoted context omitted.
This should be weighted against these mining cards becoming e-waste instead of entering the seccond hand market. The energy used to manufacturea GPU is a significant part of its overall impact.
I don’t know anybody who will willingly buy a second hand video card used for mining cryptocurrency. They’ve been pushed so hard, for so long, that their useful lifespan is pretty much used up. It’s a big part of the reason they’re being sold and not continued to be used.
Depending on the card some would have been run underclocked and even undervolted. I personally know friends who purchased mining cards that are still up and running.
For miners they're not being sold becuase they're not useful but because they don't have the cash on hand to continue the scale of operation when mining is no longer profitable. It's not that difficult to run the numbers yourself to verify this.
I have no sympathy for miners, but purely from software freedom perspective this is maximally evil: closed-source software that decides what you can use your hardware for. The CMP version is a straight-to-landfill product. As soon as it stops being profitable for mining it will have zero resale value, and become e-waste.
Imagine the PlayStation was able to be used to mine crypto. Game developers depend on gamers to actually play their games, what would you have Sony do? Encourage it or do nothing? Pretty sure their obligations to any shareholders would be to limit it so more gamers can play and buy games for it.