Nvidia might delay RTX 40 series on account of a 30 series market flood

[LAwLz]

1 hour ago, igormp said:

He'll just release tons of different videos, each one with a different 4000 launch month, so he'll be right at some point

 

/s

You're sarcastic but he has actually done that before. 

If he makes enough guesses surely one will turn out to be true. 

[LAwLz]

1 hour ago, HenrySalayne said:

You still cling to the fallacy of relative privation.

Except I am not... I am doing the EXACT OPPOSITE of clinging to the fallacy of relative privation. 

 

The people I am arguing with are the ones saying "we should ignore the envioumental impact of gaming because mining is worse". 

I am pointing out the hypocrisy of saying "mining is bad" while also playing games, because according to the estimates I have found and done, gaming uses a fairly comparable amount of energy as mining. I think that number will look even less favorable for gaming as crypto loses value (and thus less people mine) and gaming GPUs/CPUs keep on getting more and more power hungry. 

 

I am by no means saying "mining is fine because gaming also uses power". 

I am saying "if your argument is that mining is bad because it uses a lot of power, then it is hypocritical to not be critical of gaming for the same reason". 

 

The response to this is usually either complete denial, or this weird mental gymnastic where people go "gaming is okay because more people do it",which makes no sense. As I said earlier, the environment only cares for how much energy is used (or rather, how much emissions were released to generate said energy). If we are going to compare environmental impact then we compare the total harm, not "harm per user" or some other weird and irrelevant metric.

 

 

If you wanna talk about fallacies, I think you should look up "strawman", because that's what your entire post is. I have never, ever said "mining is okay because gaming uses more" or anything even remotely like it. Yet your entire post is based on countering that claim. 

What I said is that if you are against mining because "it uses a lot of power which is bad for the environment" then you should also be against gaming.

There are other reasons someone might be against mining and I am fine with some of those. 

What I am not fine with are these people who think they have a bullet proof argument against mining but then as soon as the same argument is used against gaming, which they like, they do a bunch of mental gymnastics and all of a sudden the environment isn't that important anymore. 

[Brooksie359]

9 hours ago, LAwLz said:

The entire argument of "gaming causes less damage so we shouldn't hate it" also hinges on the idea that gaming uses less power, which we don't know for sure. 

 

Sure, 1 GPU used for gaming for let's say 5 hours a day might not use as much power as 1 GPU used for mining 24 hours a day, but what if there are 5 times more gamers than miners? All of a sudden it's 5 GPUs running for 5 hours a day each for a total of 25 hours a day, VS one mining GPU running for 24 hours. 

 

The anti-mining gamer crowd likes to think that like 80% of GPUs sold are sold to miners, but all facts and professional estimates we got so far indicates that it's more like 10% of GPUs being sold to mining. 

 

What is most likely to use the most power, 1 GPU used for mining or 9 GPUs used for gaming? Assuming those numbers are accurate of course. 

Well that's a pretty crap argument as the whole premise is that because mining is less popular it uses less power as a result then it suddenly makes the hobby not as harmful? Obviously that is totally disregarding the fact that if mining is as popular as gaming now we suddenly see how mining is going to have a worse impact on the environment. The point is gaming doesn't use as much power per person than mining which is key if you want to look at the impact. Honestly though I mostly care about people who are buying a bunch of gpus and creating big mining facilities as those are what cause some of the biggest issues especially to power grids. Also a big issue with mining is it incentives to use as much power as you can possibly can with as many gpus as you can to make the most profit which is a very dangerous profit incentive and can cause huge problems. It's one of the reasons why Ethereum is going away from pow as this profit incentive is such a huge issue that the creators realized how harmful that could be for the environment. 

[tikker]

1 hour ago, Brooksie359 said:

The point is gaming doesn't use as much power per person than mining which is key if you want to look at the impact.

At the end of the day what matters is the total power consumed and from what source that power came. To first order it's not important whether that's 10 people consuming 10 TWh per year each or a 1000 people consuming 0.1 TWh per year. Beyond that you can start playing a numbers game to see what is "more justified", which is how we generally consider gaming to be more justified: at this point in time, for the same power budget, more people can enjoy some entertainment from gaming for say 2 hours per day than that people can mine 24/7.

 

I think/hope that most if not all of us know that mining is currently the bigger problem, and I do agree that it should take priority, but if this trend of power-hungry components continues (and even if it doesn't really) then we should not neglect the impact of gaming beyond that on our wallets and personal electricity bills as well.

 

California did a study, released in 2019, looking at gaming power consumption in their state in 2016: https://www.energy.ca.gov/resources/publications/energy-commission-publications?combine=CEC-500-2019-042&field_publication_classification_target_id=All&field_publication_program_target_id=All

Quote

Systems used for computer gaming in California consumed 4.1 terawatt-hours/year in 2016 or $700 million in energy bills, with emissions of 1.5 million tons carbon dioxide-equivalent allocated 66 percent to consoles, 31 percent to desktop personal computers, 3 percent to laptops, and less than 1 percent to emerging media streaming devices. Electricity use equated to 5 percent of overall statewide residential consumption among the investor-owned utilities, or the equivalent of about 10 million new refrigerators that use 400 kilowatt-hours per year.

A single state used enough power to supply entire small countries, for gaming. If that number would scale to the world's population, then (using Google's population estimates for both California and the world in 2016) you'd arrive at approximately 787 TWh spent on gaming. Now that's probably a substantial overestimate, but even if gaming would become so popular that 1 billion people would game we'd still be looking at something of the order of ~100 TWh per year, or something of the order of 0.05-0.1% of the world's energy consumption in 2016.

 

Steam reported 132 million active players per month in 2021. PlayStation and Xbox apparently also were around ~100 million in 2020. Scaling California's 4.1 TWh for 38 M inhabitants in 2016 to 300 M gamers would mean something like 30-40 TWh per year for 300 M gamers. For comparison, Bitcoin's annualised power consumption is currently 81 TWh per year (https://ccaf.io/cbeci/index). Now this is based on one estimated and I'd be interested to know if there are more precise or recent estimates of gaming-related power consumption, but considering this (and if my midnight math is correct), gaming might thus be merely half as bad as Bitcoin mining currently is. The rumours about 400-700 W GPUs don't sound encouraging for either camp.

[Brooksie359]

22 minutes ago, tikker said:

At the end of the day what matters is the total power consumed and from what source that power came. To first order it's not important whether that's 10 people consuming 10 TWh per year each or a 1000 people consuming 0.1 TWh per year. Beyond that you can start playing a numbers game to see what is "more justified", which is how we generally consider gaming to be more justified: at this point in time, for the same power budget, more people can enjoy some entertainment from gaming for say 2 hours per day than that people can mine 24/7.

 

I think/hope that most if not all of us know that mining is currently the bigger problem, and I do agree that it should take priority, but if this trend of power-hungry components continues (and even if it doesn't really) then we should not neglect the impact of gaming beyond that on our wallets and personal electricity bills as well.

 

California did a study, released in 2019, looking at gaming power consumption in their state in 2016: https://www.energy.ca.gov/resources/publications/energy-commission-publications?combine=CEC-500-2019-042&field_publication_classification_target_id=All&field_publication_program_target_id=All

A single state used enough power to supply entire small countries, for gaming. If that number would scale to the world's population, then (using Google's population estimates for both California and the world in 2016) you'd arrive at approximately 787 TWh spent on gaming. Now that's probably a substantial overestimate, but even if gaming would become so popular that 1 billion people would game we'd still be looking at something of the order of ~100 TWh per year, or something of the order of 0.05-0.1% of the world's energy consumption in 2016.

 

Steam reported 132 million active players per month in 2021. PlayStation and Xbox apparently also were around ~100 million in 2020. Scaling California's 4.1 TWh for 38 M inhabitants in 2016 to 300 M gamers would mean something like 30-40 TWh per year for 300 M gamers. For comparison, Bitcoin's annualised power consumption is currently 81 TWh per year (https://ccaf.io/cbeci/index). Now this is based on one estimated and I'd be interested to know if there are more precise or recent estimates of gaming-related power consumption, but considering this (and if my midnight math is correct), gaming might thus be merely half as bad as Bitcoin mining currently is. The rumours about 400-700 W GPUs don't sound encouraging for either camp.

You are joking right? Obviously the total isn't very important when comparing them as again you are comparing habits and how harmful it is and the most obvious way to do that is per person as anything else is just disingenuous. I mean if one guy got entertainment by dumping ton and tons of water in the middle of his yard that is obvious more wasteful than taking a shower even though if you looked at total consumption showering globally wastes you would see that it is higher obvious because way more people take showers. That doesn't change the fact that the guy is being way more wasteful. 

[Arika]

11 hours ago, HenrySalayne said:

You still cling to the fallacy of relative privation

No he's not, if that's what you think, then you have entirely misinterpreted what @LAwLz had been saying.  He's not been saying "mining isn't bad because gamers are worse" he's saying "miners are bad,  but so are gamers who are also hypocrites who refuse to acknowledge that their hobby is also destroying the environment to a similar degree that miners are" 

So if anything the fallacy that you keep referring to is more used by gamers as a defence rather than acknowledge that gaming is also bad for the environment. It's not to say that everyone should stop gaming, but don't stick your head in the sand with a sense of moral superiority because you think "miners are the ones destroying the planet, not me"

 

Also side note: it's good to be aware of logical fallacies, but if you're the kind of person who continuously calls them out by name in discussion as a way of refuting someone's point, that doesn't make you look smart, it makes you look like an asshole. You cited it 4 times in 2 posts. 

[Kisai]

8 hours ago, MageTank said:

Thank you for sharing this. Now where exactly does it have anything to do with what I have been saying? 

 

No power stations were ever spun back up because of the demand to play a game. Dirty power stations have been spun back up BY miners. By all accounts, mining is being done in more intensive, environmentally-negative ways then gaming is.

 

8 hours ago, MageTank said:

This is part of the problem I have with this rhetoric. Even when we remove the moral discussion from the equation and make it about ones personal views of what is or isn't entertainment, someone gives in ignoring the context of the conversation to make it a moral dilemma yet again.

 

What do you expect? Someone to go "oh gamers also die because they game for 24 hours in 40 degree C heat without hydrating at all." No, those gamers who die from marathon gaming sessions are a consequence of a mixture of bad factors. I'm sure every single one of them died because they smoked and/or didn't eat properly and/or didn't hydrate, and/or were using "energy drinks" to stay awake. All of those factors are within their control.

 

Industrial scale bitcoin mining is not within the control of the general population, their existence drives up energy costs, which leads to people not using air conditioning in 40+ degree heat, either by choice, or as a direct consequence of the utility shutting down more than just the farms that have voluntarily been shutdown. More people will directly die from crypto mining when the conditions (High energy costs and extremely high or extremely low outdoor temperatures) exist, and it won't be the voluntary shutdowns by industrial miners, it will be the "hobby" GPU miners directly responsible for that, who will continue to run their rigs regardless of conditions, causing the utility to have to shut down areas in rolling blackouts, making a critical problem problem worse.

 

Every time utilities have to resort to rolling blackouts, people in hospitals die, people at home but on medical equipment (such as oxygen) also have a risk of dying.

 

Do you really think that gamers are going to sit there and continue to game during rolling blackouts? No. I know of no people who go "I must have a gamer laptop, just in case my power utility shuts off my power", but when a miner has to shut down, they immediately scream bloody murder as they are now losing money.

 

You're free to do whatever you want with your own equipment, but saying "mining is a hobby" is like saying smoking is a hobby. It's objectively bad, but you can quit any time and cause no further damage. You can quit games, but you'll just replace it with an equally energy intensive source of entertainment like watching TV, or reading books. Or are people that naïve to believe that no entertainment is energy-free? 

 

Before Radio, Television, Film and Internet. People drove cars, or had to take care of large animals to tow their wagons. There is no such thing as a energy-free lifestyle today. Bringing everyone together has transportation costs, lighting costs. Arguably the Pandemic showed us exactly how much energy we waste in needless transportation commutes alone, where some fuel places had prices at levels last seen during the 1990's.

 

If mining was banned overnight, and something like the ATF investigated all energy theft, and large residential users of energy were investigated ( which, is easily justified https://pilotteacher.com/can-police-helicopters-detect-grow-operations/ ) for operating crypto mines, there would be no net loss to society.

 

[leadeater]

3 hours ago, Kisai said:

If mining was banned overnight, and something like the ATF investigated all energy theft, and large residential users of energy were investigated ( which, is easily justified https://pilotteacher.com/can-police-helicopters-detect-grow-operations/ ) for operating crypto mines, there would be no net loss to society.

Why not just eliminate all sources of income you personally do not agree with and doom people who are supporting their lives through that to either poverty or restrict their economic success. I mean according to you anything like that is "no net loss to society".

 

Removing economic success from societal impact and factors is quite the extreme take. 

 

Careful with voicing shallow thought out opinions in such ways as this, eliminating anything of any kind will never have no net impact (loss or gain) and often quantifying such things is near impossible because life just isn't that simple.

 

3 hours ago, Kisai said:

It's objectively bad

Having to sacrifice your only life on having to earn an income to live is objectively bad, therefore any work at all is objective bad. Earning any income at all in anyway is objectively bad 

[LAwLz]

12 hours ago, Brooksie359 said:

Well that's a pretty crap argument as the whole premise is that because mining is less popular it uses less power as a result then it suddenly makes the hobby not as harmful?

That is not the premise at all.

In fact, you pretty much got my entire post the other way around. Please read my posts again and try and calm down a bit. I am saying the opposite of what you think I am saying.

 

 

I am not the one saying "the less harmful hobby is totally fine". The anti-mining pro-gaming crowd is, and it is precisely that sentiment I am against.

If mining is bad because it is bad for the environment, then so is gaming.

 

 

12 hours ago, Brooksie359 said:

Obviously that is totally disregarding the fact that if mining is as popular as gaming now we suddenly see how mining is going to have a worse impact on the environment.

But mining isn't as popular, so I don't get why you bring that up.

How about we stick to discussing the reality we live in today and how harmful these things are to the environment based on what is actually happening today, not what might happen in X number of years?

You can't just say "X is fine but Y isn't because in my fantasy future Y is worse than it is today while X stays the same. So therefore I am justified in hating Y today but you're not allowed to say X is bad, even though today both are harmful to the environment".

 

 

 

12 hours ago, Brooksie359 said:

The point is gaming doesn't use as much power per person than mining which is key if you want to look at the impact.

See:

15 hours ago, LAwLz said:

I don't understand why you think it is more okay to use up 1GWh of power if it's multiple people doing it. It still uses up the same amount of natural resources to generate that energy. 

 

The ozone layer doesn't heal just because you tell it "don't worry, the hole causes by the burning of coal went to power 10 gaming rigs, not one mining rig". 

  

Again, the environment does not care about the "per capita" emissions. It only cares about the total emissions.

Being shot by 5 bullets hurts just as much regardless of whether it was 1 person who shot you 5 times, or if it was 5 people who shot you once each.

[tikker]

10 hours ago, Brooksie359 said:

You are joking right? Obviously the total isn't very important when comparing them as again you are comparing habits and how harmful it is and the most obvious way to do that is per person as anything else is just disingenuous. I mean if one guy got entertainment by dumping ton and tons of water in the middle of his yard that is obvious more wasteful than taking a shower even though if you looked at total consumption showering globally wastes you would see that it is higher obvious because way more people take showers. That doesn't change the fact that the guy is being way more wasteful. 

The impact is the same for a given amount used, be that electricity of water. Whether the individuals making up that total use more or less doesn't reduce the impact. I've nowhere disagreed that mining (in this context) is the bigger problem. I've literally said so. If gaming is around or within 50% of the same usage, however, then it is having a similar impact regardless of whether your low-power 3050 Ti system uses only 20% the power your neighbour's 3090 Ti quad SLI beast uses. Similarly to people, correctly, saying "if everyone started mining" it'd be much worse, the same holds for gaming and should be looked at with the same concern.

 

Yes the large consumers are the "more wasteful" ones, but there also is a certain danger in thinking the individual contributions don't matter. Since you bring up showering, there are absolutely constant efforts going on on telling people to be considerate and shower briefly to conserve water. Especially with temperatures increasing and droughts becoming more prevalent even my country is now asking people to help prevent a fresh water shortage by e.g. showering less and putting a rain barrel in the garden to water plants with. If 1 min less of showering saves 40 L that may not sound much to you, but multiply that by 17 million people in a country and all of a sudden you have half a billion additional litres of fresh water for potentially more important matters than showering.

[HenrySalayne]

14 hours ago, LAwLz said:

because according to the estimates I have found and done, gaming uses a fairly comparable amount of energy as mining. I think that number will look even less favorable for gaming as crypto loses value

Looking at the wall and just making up some numbers is not an estimate.

23 hours ago, LAwLz said:

1 GPU used for gaming for let's say 5 hours a day

The average time gamers spend gaming per day is roughly one hour (refer to my previous posts for sources). If you take all gamers agnostic of platform, PC gaming results in only 1.6 hours per week. Just to calculate the upper bound let's say it's two hours per day.

 

https://www.tomshardware.com/news/ethereum-miners-have-spent-15-billion-on-gpus

Miners with dedicated mining rigs (not including people mining at home with their single or dual gaming GPUs) bought approximately 10% (by numbers) of all dedicated GPUs manufactured between Q4 2020 and Q1 2022. Considering miners don't buy low-end SKUs the amount of compute performance and corresponding TDP is probably even higher. Not accounting for workstation use and rendering farms; 90% of the GPUs (by number) went to gamers.

So if we just include the GPUs sold between Q4 2020 and Q1 2022, the energy usage of the 10% used for mining is at least more than 30% higher than the energy usage of the remaining 90% used 2 hours per day. In a best case scenario.
 

And here we have another cool graph showing the mining craze:

 

Ethereum mining alone (on really efficient hardware with 0.5 MH/W) accounts for the same amount of energy per year as the entire country of Slovenia (~ 14 TWh)

 

 

[tikker]

18 minutes ago, HenrySalayne said:

And here we have another cool graph showing the mining craze:

 

Ethereum mining alone (on really efficient hardware with 0.5 MH/W) accounts for the same amount of energy per year as the entire country of Slovenia (~ 14 TWh)

Just as a side-note mining efficiency is measured in Joules per hash or hashes per Joule, not Watts, otherwise the units won't add up in the calculation. Nobody is denying that mining uses a large amount of power though. You already say you assume efficient hardware and there are far worse estimates out there: Digiconomist estimates 63 TWh.

 

I'll happily see mining go before gaming, but in my eyes this isn't a mere finger pointing of who is worse. It's also a wake-up call. Crypto has been excellent in showing us how much power GPUs and computer hardware in general can consume. I mined for a bit last year and even though a 200 W 1080 Ti doesn't sound like much, my monthly power usage pretty much doubled from having that single GPU run 24/7.

 

The key difference with gaming is that the majority doesn't value crypto and thus its power consumption is heavily scrutinised. The entertainment from gaming is valued, so we haven't paid much attention to it beside the immediate effect of needing a bigger power supply, but as linked, you can estimate that it may not be much better than crypto in terms of overall impact. Using your 1.5-2 hours per day figure, gaming is better for the environment or power consumption in that you can have 12-16 gamers for every miner. Those 12-16 gamers will still have the same impact as 1 miner, however.

11 minutes ago, HenrySalayne said:

Looking at the wall and just making up some numbers is not an estimate.

See above where California did a study in 2016 that estimated 4.1 TWh was used for gaming that year in California (across all PC, consoles and mobile devices).

  

If you extrapolate from that study using Steam's, Playstation's and Xbox's estimated active players from last year then you arrive at gaming consuming 35 TWh or almost the same estimated amount of energy as New Zealand at around 40 TWh. That is triple your Ethereum estimate and 2/3 that of Digiconomist's. It could even be an underestimate, since there I assumed all of California's 38 M inhabitants were the gamers contributing to that power draw and scaled that to the ~300 M active gamers last year across PC and consoles.

 

Yes the power consumption of mining is definitely not what we want right now and I would also consider it worse in the relative way based on the number of miners vs number of gamers, but reiterating on the 300+ power monsters that recent and coming GPUs seem to be evolving towards, we should probably also start looking at where the gaming market needs to go as well because it is probably not far from crypto-level power-draw.

 

 

[LAwLz]

3 hours ago, HenrySalayne said:

Looking at the wall and just making up some numbers is not an estimate.

Good thing that is not what I am doing.

Here is my old estimate for power consumption used by gaming.

[Falkentyne]

On 7/11/2022 at 7:19 AM, Rym said:

 

 

Summary

 Due to an oversupply of used RTX 30 series GPUs, as well as board partners and retailers having too many GPUs, Nvidia is considering delaying the release of the RTX 40 series GPUs.

 

Quotes

 

My thoughts

 Well, how do I put this gently? 

 

That is entirely NOT my problem, Nvidia knew they were selling everything they had to miners, I called out mining 2 years ago as being a short term thing and it will come crashing down while I ignored the naysayers. All those used GPUs would obviously find their way on the used market, driving the prices down. I thought about getting an RTX 4080-4090 but now I'm also hearing Nvidia trying to cause an artificial shortage to drive the prices of their new GPUs up. Honestly I will now very likely go for an AMD RX 7900 XT especially since it's looking to release earlier than Nvidia gpus. I'm also hearing people will not be buying an Nvidia GPU due to pure spite of what they did for 2 years to their tried and true customers. AMD on the other hand doesn't have this oversupply issue since they never made that many GPUs for miners to begin with and are looking to release their GPUs very soon.

 

Sources

 https://www.pcgamesn.com/nvidia/rtx-4000-gpu-launch-delay-geforce-3000-oversupply

Good.  More time for me to save money instead of buying $1500 video cards for the highest gaming FPS.  My shunt modded retail HSF liquid metal thermal compound RTX 3090 FE will serve me well.  I can put the extra money and time into some Steam Deck games.

[HenrySalayne]

4 hours ago, LAwLz said:

Good thing that is not what I am doing.

Where do I find this number which was the basis of your argument in this thread?

On 7/15/2022 at 1:19 PM, LAwLz said:

1 GPU used for gaming for let's say 5 hours a day

[Kisai]

21 hours ago, leadeater said:

Why not just eliminate all sources of income you personally do not agree with and doom people who are supporting their lives through that to either poverty or restrict their economic success. I mean according to you anything like that is "no net loss to society".

 

 

Look at both the wage disparity and income inequality. There are some people who are out there that do not give a single care about anyone but themselves. Bitcoin miners are in that group. They consume without giving a single thing back to society as a whole. They are the literal definition of "privatize the profits, subsidize the losses on everyone"

 

The losses here, being energy costs for everyone. The number one reason to ban cryptomining is that the country/state/city has dirty expensive energy costs. There are countries out there that have routine power cuts, as a daily problem. There are states, like Alaska, which have limited power production capacity because they have to import the fuel needed for generating power. There are states like Texas which have deregulated energy markets which results in absolutely everyone subsidizing energy generation for the biggest consumers of energy.

 

That energy price is not flat. I don't see bitcoin miners going "oh, I'm only going to mine off peak"

Crypto mining is what allows this kind of insanity. Spot prices of nearly 6000 on-peak? Yeah that wouldn't be happening if there wasn't a perfect storm of generation shortage, climate conditions, and "hobby" miners refusing to turn their rigs off because of the "I got mine" mindset.

[LAwLz]

13 hours ago, HenrySalayne said:

Where do I find this number which was the basis of your argument in this thread?

That was not the basis of my argument.

It was just an example to illustrate that the total power used might be more even if each individual user uses less.

 

The point of that thought experiment wasn't "all gamers play games for 5 hours a day".

The point was five people playing for 5 hours a day does more environmental damage than one person mining for a day. 

 

I think you got hung up on a small detail and lost sight of the bigger picture I tried to illustrate. 

The facts and estimates that I was referring to was the post I linked to, which contains sources for the numbers. It wasn't by throw away line of "5x5 > 24".

 

I think you should reread my whole post again and interpret it based on the context, as well as pay attention to my wording such as "let's say" and "a gamer". 

I never said "all gamers plays for 5 hours a day". 

 

 

 

Since you seem to be a fan of logical fallacies. The one you are doing right now is called "strawman argument". 

My argument was NOT "there are 5 times more gamers than miners and all of them play for 5 hours a day so they do more harm". 

 

My argument was "many people doing a bit of harm can end up doing more damage than a few people doing lots of harm each. For example 5x5 is more than 24. But we don't have that many numbers to go by so we shouldn't say we are sure". 

[leadeater]

7 hours ago, Kisai said:

Look at both the wage disparity and income inequality. There are some people who are out there that do not give a single care about anyone but themselves. Bitcoin miners are in that group. They consume without giving a single thing back to society as a whole. They are the literal definition of "privatize the profits, subsidize the losses on everyone"

There are far large groups of people and in total that are this that are not miners. This still isn't a good reason to say there is no net loss, because again there isn't no net loss.

 

7 hours ago, Kisai said:

The losses here, being energy costs for everyone

Your following Texas example has almost nothing to do with mining at all, why that came to be and like many other places has almost everything to do with not mining. Infrastructure investment is simply very expensive along with having to meet (only public perception wise) global emissions targets. Utility electricity costs and avoiding blackouts is entirely controllable, it's by choice these are not sufficiently done.

 

Also dirty energy = cheap not expensive like you said. That's why coal is still used, because it's the cheapest in consumable materials and cost of facility.

 

And the spikes you point at happen regardless of (crypto)mining, just ask Australia.

 

Here is some wholesale electricity costs for 16 days in New Zealand. There are no current factors putting pressure on electricity prices, hydro dams are full, wind is blowing etc etc

 

https://www1.electricityinfo.co.nz/prices#!#q.run_types_=F,A,I&q.market_types_=E&q.nodes_=BEN2201,HAY2201,OTA2201&q.main_date_or_rolling_selector=date&q.date_from=2022-07-01&q.tp_from=1&q.date_to=2022-07-16&q.tp_to=48&v.search_include=&v.search_exclude=

 

Go explore the data, 300x swings are "normal".

[Montana One-Six]

On 7/15/2022 at 2:20 AM, Kisai said:

That's a 71.9% change in performance if you went straight from a 760 to a 3060.

Very interesting post. But this line jumped at me because you used the percentages wrong. The jump from a 760 (with a score of 4768) to a 3060 (with a score of 16996) is far greater than just 71.9%. In fact it is more than 350% more performance if you went straight from the 760 to the 3060. The 71.9% is relative to the 3060 meaning going from the 3060 to the 760 you would lose 71.9% performance as the 760 only has 28.05% of the 3060s performance.

 

Haven't checked if you used the other percentages right or wrong in text so you may want to double check if you care about that.

 

Edit: Just looked through it again you often used the percentages the wrong way. Meaning you want to describe for example the gain from the 750 to the 3050 but used the percentages you would lose when going from the 3050 to the 750. Those are not the same. You wrote this:

Quote

That's a 73.5% change in performance if you went straight from the 750 to the 3050.

With the right percentage that comparison would sound like this: "That's a 377% change in performance if you went straight from the 750 to the 3050."

Again if you would have said going from the 3050 to the 750 the 73.5% (as performance loss) would be correct.

[05032-Mendicant-Bias]

23 hours ago, leadeater said:

Your following Texas example has almost nothing to do with mining at all

I think the power argument of PoW blockchain is indefensible, PoW has been around fourteen years, there have already been several nations was power grid failure was directly attributed to an exponential increase in mining operations, and the pattern has become common:

• China coal plant takes out a third of global BTC hash rate
• Malaysia destroys BTC mining rigs
• Kazakhstan unplugs BTC mining to avoud blackout

It makes sense, if your country has a 1% cheaper electricity, PoW miner will flood in and add tens of megawatts of baseline demands in a matter of weeks, and if the country makes investment to harden the grid, making the energy 1% more expensive, miners will just fly away leaving you with an unbalanced grid, meanwhile your country had to suffer rolling blackouts and more expensive energy. Mining demand is both reactionary and enormous, you don't see steel foundries relocating every month depending on energy prices.

I agree with you that the Texas grid would have been in trouble without miners (IMO), after all last winter Texas was eight minutes away from total collapse, and PoW miners had nothing to do with that. In this case I speculate PoW miners shut down their operations in Texas for profitability reasons, and advertised the shutdown to gather "good press". I also think PoW miners made things slightly worse by adding hundreds of megawatts of baseline load to a troubled grid.

23 hours ago, LAwLz said:

It was just an example to illustrate that the total power used might be more even if each individual user uses less.

Like stated earlier, I think it's plausible gamers energy demand and blockchain energy demand are comparable. Datacenters alone use probably four times more power than blockchain minings. I think the argument is that that expense of resources results in a net negative to society. PoS technology is worse for almost everyone given it's even more unsafe and unstable than PoW mining, but the power requirements are slashed by 100 folds. I won't mind as much about blockchain if all it does is transfer wealth from marks to fraudsters but and leaves our civilization's hardware and energy mostly untouched. Just like I don't care about MLM schemes or financial frauds. 

On 7/17/2022 at 3:06 AM, Kisai said:

Look at both the wage disparity and income inequality. There are some people who are out there that do not give a single care about anyone but themselves. Bitcoin miners are in that group.

I fully agree, this sums up my mental image of Blockchain technology and its use. I believe our civilization should actively ban PoW technology outright. Fraudsters don't need it, they can use PoS technology to achieve their wealth transfer anyway, and everyone that does not want to get involved in Blockchain at least is not damaged by it.
 

On 7/15/2022 at 3:12 PM, MageTank said:

I know a few miners that simply enjoy mining and trading as a hobby. It's fun for them, they enjoy it, so they devote their free time and excess resources to it. By your logic, since they are using it for entertainment, it is "justified", is it not?

Yup. There are believers that are in Blockchain because they believe it might be the future, or just enjoy interacting with it. As long as those people are not motivated by financial gains, or by defrauding, I believe society has no businness taking their toy away from them. I believe Satoshi themselves might have belonged to this category. I am confident in saying that this demografic is a small minority of people involved in Blockchain, I speculate BTC would be worth a few tens of USD$/BTC at most if the space only included non financially motivated believers, and that the blockchain technology would have evolved to become useful for some niche applications.

 

Line Goes Up has a great section about the PoW mining arms race. It's inevitable PoW mining grows until one of three conditions are met:

• The demand push the production price of tokens out of profitability
• Our civilization runs out of hardware, pricing hardware out of profitability
• Our civilization runs out of energy, pricing energy out of profitability

[leadeater]

1 hour ago, 05032-Mendicant-Bias said:

I think the power argument of PoW blockchain is indefensible, PoW has been around fourteen years, there have already been several nations was power grid failure was directly attributed to an exponential increase in mining operations, and the pattern has become common:

Correction, blame pointed towards with actually not so great reasoning and quantifiable calculations. Industrial manufacturing far, far outstrips mining and this is an absolute known.

 

1 hour ago, 05032-Mendicant-Bias said:

China coal plant takes out a third of global BTC hash rate

All this story is about is the issue of having such a large amount of bitcoin hashing done in a single geographical area. It's not a story about how that cryptomining caused an outage nor how much energy is being used.

 

1 hour ago, 05032-Mendicant-Bias said:

Malaysia destroys BTC mining rigs

Again doesn't at all back what you're saying. If you said a lot of cryptominers steal electricity then this would be a valid article to point to.

 

1 hour ago, 05032-Mendicant-Bias said:

Kazakhstan unplugs BTC mining to avoud blackout

One country does not prove a point and the cause of this is actually the result of what I pointed out, lack of infrastructure investment. Utility providers simply do not plan very well and rely almost entirely on co-operative projects so when for example a large manufacturing facility is built or expanded this is accounted for. General "unplanned" increases like distributed cryptomining simply aren't planned for. Many countries have had blackouts and problems before cryptomining was a thing and for the exact same reasons, poor planning and investment.

 

This of course ties in the argument that cryptocurrencies should be regulated rather than ignored, knowing what is happening goes a very long way to dealing with electricity demand.

 

1 hour ago, 05032-Mendicant-Bias said:

It makes sense, if your country has a 1% cheaper electricity, PoW miner will flood in and add tens of megawatts of baseline demands in a matter of weeks, and if the country makes investment to harden the grid, making the energy 1% more expensive, miners will just fly away leaving you with an unbalanced grid, meanwhile your country had to suffer rolling blackouts and more expensive energy. Mining demand is both reactionary and enormous, you don't see steel foundries relocating every month depending on energy prices.

It's not that simple and doesn't even play out like that. One of the biggest factors is land values and property costs. Also tens of megawatts is literally nothing at utility scale. 

 

Global electricity usage just for Aluminium smelting in 2021 was 901 TWh, or 2.47 TWh each day. Global electricity generation in 2021 was 27,520.52 TWh. Cryptomining, and gaming, is an insignificant tiny blip.

 

Now of course this isn't an argument to say PoW is efficient and scalable, but this being factually true does not then infer that it's actually having a huge impact on electricity utility grids. Large concentrated cryptomining operations cause local issues for sure, it's not however pulling down entire countries. Again regulation is the answer to this problem.

[Kisai]

On 7/17/2022 at 6:13 AM, Montana One-Six said:

Very interesting post. But this line jumped at me because you used the percentages wrong. The jump from a 760 (with a score of 4768) to a 3060 (with a score of 16996) is far greater than just 71.9%. In fact it is more than 350% more performance if you went straight from the 760 to the 3060. The 71.9% is relative to the 3060 meaning going from the 3060 to the 760 you would lose 71.9% performance as the 760 only has 28.05% of the 3060s performance.

 

Haven't checked if you used the other percentages right or wrong in text so you may want to double check if you care about that.

 

Edit: Just looked through it again you often used the percentages the wrong way. Meaning you want to describe for example the gain from the 750 to the 3050 but used the percentages you would lose when going from the 3050 to the 750. Those are not the same. You wrote this:

With the right percentage that comparison would sound like this: "That's a 377% change in performance if you went straight from the 750 to the 3050."

Again if you would have said going from the 3050 to the 750 the 73.5% (as performance loss) would be correct.

Nah, read the left column:

"G3D Rating(% diff. to max in group)"  

Which happened to be the left-most card in every example.

 

I then transcribed the other %'s by deleting the left-most columns. Losing 71.9% going from the 3060 to 760 is the inverse calculation. Yet it demonstrates exactly the same thing. Bigger numbers look better, but it doesn't change the fact that the differences in the tier of the card have pretty much been linear. No matter which 7xx card you upgrade from, the 30xx of the same tier was the same relative change in performance.

 

The point I was making was not "x card is 300% better", rather that sticking to the same tier of card yields the same level of performance gain generation over generation. So if you want to stick with exactly the same performance, you wouldn't pick the same tier, you'd pick the tier/half-tier below it. People do not make their purchase decisions that way though. If a "3060" is a baseline card, (which it should be) and nvidia starts short-changing performance tradeoffs by making DLSS permanently on, do you think people would notice? My guess is that if it actually worked that way, people would be able to game at 4K on any tier of card. But it doesn't. 

 

[Dracarris]

On 7/14/2022 at 3:14 PM, MageTank said:

"Evil is evil, Stregobor. Lesser, greater, middling, it's all the same. Proportions are negotiated, boundaries blurred" - White Haired Monster Hunter Man, 2019

Completely ignoring what I wrote. This quote is just horribly misplaced and unfitting here, well and in general quite close to complete bullocks and simply untrue. Ignoring the fact that "evil" in itself is a very bad fit for both gaming and mining.

 

Causing 10 - 100x more of whatever bad thing is not the same, period.

 

If I drive a compact car with 4l/100km of fuel consumption, only occasionally, and do short distances by foot or bike and my neighbor uses a truck with tenfold fuel consumption for every short way/distance daily, leaves it idling for hours in the driveway to have the AC running, he's doing significantly more harm to the environment. Period. They are not the same, not matter how often you or your mining friends here come up with ridiculous comparisons, quotes, and deluded views about quite simple mathematical facts. You are outright denying reality.

On 7/14/2022 at 5:59 PM, ewitte said:

Fully tuned I was pulling about 220W mining on my 3080 even that was enough to make the office uncomfortably warm.

And for how many % (on average) does your card actually operate in that mode? And how many cards do you use concurrently for gaming?

[Dracarris]

On 7/15/2022 at 10:26 PM, LAwLz said:

because according to the estimates I have found and done, gaming uses a fairly comparable amount of energy as mining.

No matter how often you repeat this utter, utter nonsense: It's not going to become true even in the slightest way.

I am also still waiting for the day where gamers bring down the electricity grid of smaller countries or cause a shortage of electricity and hence get gaming banned.

[05032-Mendicant-Bias]

27 minutes ago, Dracarris said:

I am also still waiting for the day where gamers bring down the electricity grid of smaller countries or cause a shortage of electricity and hence get gaming banned.

I found no reliable metrics on this, but it's is plausible that gaming overall consumes as much power as PoW mining. There are lots of gamers with few GPUs often turned off, there are few miners with a vast number of PoW rigs each always on. At the end of the day, PoW mining is the realm of a few tens of thousands of Blockchain enthusiasts, gamers are in the tens of millions.

 

PoW does vastly more visible damage because of two reasons:

1. PoW miners will change continent to chase a 5% cheaper energy price, the quick enormous increase in demand strains and rash the energy grid of the often developing nations that provide that energy (Texas is the exception as it's a developed nation with an unstable power grid).
2. PoW rigs are a baseline demands, they are on 24/7 to maximize profits. No tricks available to shape the demand curve, only more baseline power stations.

Unlike PoW mining, gaming rigs are all over the world, often in industrialized nation with a stable grid, and are only on when gaming, and gamers don't really care about energy prices, because gaming provide intangible benefits, gamers don't maximize fun per unit of energy unlike PoW miners.