This won't really do anything for availability or reducing scalper pricing. Miners are a sizeable proportion of sales but far from the majority. Gamers will be able to buy more of these GPUs but less of the non-LHR or AMD GPU's. GPU prices won't actually change.
The real reason Nvidia is doing this is that when mining profitability inevitably drops, either because of a crypto price crash or Ethereum moving to Proof of Stake, the miners will flood the used market and ruin Nvidia's profits for a few quarters. Nvidia is trying to get miners to buy mining-only cards, because those can't be sold to gamers down the line. They'd rather mining cards become e-waste in a few years than get resold.
People think Nvidia is doing this to try to help them get better prices now, but really Nvidia is trying to extract more money from gamers in the long term.
> The real reason Nvidia is doing this is that when mining profitability inevitably drops, either because of a crypto price crash or Ethereum moving to Proof of Stake, the miners will flood the used market and ruin Nvidia's profits for a few quarters. Nvidia is trying to get miners to buy mining-only cards, because those can't be sold to gamers down the line. They'd rather mining cards become e-waste in a few years than get resold.
Great commentary on the unsustainability of it all. Nvidia has no incentive to make sure the GPU market operates in a sustainable way by supporting a secondary market. Thus, they have every incentive to cut the legs out from under resellers. It would also be anywhere from hard to nearly impossible to give them an incentive to allow the secondary market to operate in such a way as to increase the sustainability of their product.
Edit: Aren't we a bit touchy today? (Turn showdead on and look a little down thread.)
> ”Nvidia has no incentive to make sure the GPU market operates in a sustainable way by supporting a secondary market.”
They do in the sense that buyers often consider future resale values when making primary purchase decisions. It’s much easier to justify purchasing a $2500 “enthusiast” graphics card if you’re confident you can get a decent chunk of that back when you sell it in 6-12 months.
Similarly, car makers are incentivised to try to ensure their vehicles have decent resale value. Expected
future residuals affect financing and purchase decisions today.
In my experience base level items have much better resale value then more expensive items. It's true for notebooks, graphic cards and cars too. For example my iPhone 11 Pro has fallen to half of the price in one year, but entry level iPhone 11 probably has still 70% of its value. Same for Macbook Air, entry level with 8GB and 256 SSD will hold value better then maxed out one. Same with cars, extra premium equipment has basically no value for people that are buying used cars. So doubt that premium graphic cards are exception. Their value falls sharply over time, so not sure people buying them are considering resale value as argument to buy them.
The entry level iPhone 11 and Macbook Air are still premium products. You're not reselling your €250 unbranded Walmart laptop for anything more than scrap.
Yep, it's premium to the point that's search function friendly. Apple machines will hold their value very well but it'll be weighted in favour of the basic version of each model.
I think when you get to PC parts it becomes quite a bit trickier as everyone who is willing to buy used PC parts is way more knowledgable on exactly what they're looking for. A lot of things do hold their value pretty well.
On the other side of things, you're taking such a risk selling computer parts on ebay (gotta be a lot of people frying components after buying them and not even realising it's their fault) that I can see why people couldn't be bothered taking the risk of reselling when they can just gift away the parts.
>Same with cars, extra premium equipment has basically no value for people that are buying used cars.
It mostly doesn't impact price as is, but the top equipped car will get much more traffic from potential buyers and therefore can get away with setting the price a bit higher than the current market.
> car makers are incentivised to try to ensure their vehicles have decent resale value
Is this actually true or merely a theoretical thought? Because it would seem that car makers are incentivised to ensure their vehicles annihilate a couple days after their warranty expires and/or after a couple of miles after their guaranteed mileage is driven, and comparing e.g. reputation of second-hand cars manufactured in 80es/90es/2000s/10s seems to support it.
One of the tactics Nissan ran into big trouble with was their extreme reliance on fleet sales. Fleet sales juiced the profits but killed the used market, thus pushing Nissan's cars down market thanks to reduced resale value. This was an active criticism by investors and journalists during the Nissan down turn.
Meanwhile Toyota is able to charge a meaningful premium because their cars retain value. So I do think it is safe to conclude in the case of cars the effect is real.
You came sooooo close to hitting the nail on the head. The first owners have a huge effect on a resale value of a vehicle.
Who buys Toyotas?
Why are those people's cars more valuable on the used market than corporate fleet's used cars?
Now see if your model explains the used value of a Dodge Journey, oilfield pickup and whatever car grandma owned.
Nissan spend the '00s selling to owners who's used cars are not held in high regard and this affected consumer perceptions of them. Toyota spent the 90s and '00s (and still is) selling to upscale people who treated their cars nicely and it affected consumer perception of them.
I'm not so sure. I bought my car mostly to minimize total-cost-of-ownership. I assume I'll resell my car when the TCO of maintenance exceeds the TCO of a new car.
I know a lot of other folks who do something similar to that too.
TCO includes my time, so if I need to take a car to the shop, or miss meetings because things break down, that raises TCO by that opportunity cost. Selling a car is a hassle, and by the time I'm ready to give it up, it's not worth too much (and the buyers will probably be poor, since TCO exceeds a new car) so I don't feel the need to maximize resale value.
While this is an overly rationalist way of phrasing things, but most people I know take a similarly pragmatic approach. Teachers. (Real, not software) engineers. Etc.
I don't buy used cars, since some people sell to upgrade, and some sell because there's some issue. The headache adds a lot to the TCO.
There's also material differences, there's a reason the Nissan CEO from that era is currently a fugitive and on trial for financial misconduct.
And not by some technicality either, he literally hired a paramilitary group to disguise themselves as musicians and smuggle him out of Japan while on bail.
Since 2001 Nissan has used their defective CVT transmission in the vast majority of their lineup from the compact Versa to the Pathfinder SUV (aside from the sports cars and pickup trucks).
The average time to failure depends slightly on the model, but it's often around 60-70k miles.
The cost to replace it is typically most if not more than the value of a typical Nissan.
Because they were selling defective cars, their buyers dried up, so they had to resort to buyers with sub-prime credit. At this point it's joked the Nissan Altima is the official car of bad credit or bad neighbourhoods.
While the poor upkeep by the sub-prime owners certainly doesn't help, the poor resale value is largely due to everyone knowing they are ticking time bombs.
I've shopped for many different used 10-15 year old cars, and through my research have found the used market is extremely accurate in pricing in expected upkeep or failures, even among different variants of the same model from the same brand.
When I wanted an W211 E-class Mercedes (2003-2009), I researched the common failures and the cost to repair. I found the whole range was actually priced pretty similarly once you factored that in.
There were major defects in the 2003-2004 models which would cost around $2500 to fix. The 2005+ models were worth about $2500 more than a 2004 (despite only being worth $5000 and $7500)
The E500 had an air suspension failure issue which would be around $2000 to remedy, the same model year E320 with normal suspension was worth $2000 more. And for 2003-2004 they were only worth $3000 and $5000 respectively.
After months of browsing craigslist and test drives, it was clear it didn't matter which model or year I bought, a Mercedes W211 E-class was going to cost me $7500.
At that point I figured I'd just keep saving and get a Lexus 2IS for $10,000 as it was much better value despite the difference in owner reputation.
tldr: The used market knows exactly what a car is worth regardless of owner reputation. Nissans are worthless because they're shit cars.
Until Masayoshi Son & friends face the music, or the convincion rates drop from a comical 98%, I'm inclined to believe this was entirely politically motivated.
>Is this actually true or merely a theoretical thought?
On one hand you have FCA (I'm sorry, Stellantis) which leans into its not so positive reputation and does just fine.
The other hand you have Toyota who spend the 90s and 00s digging a goodwill moat in the minds of premium consumers.
Both approaches work but for different demographics. You'll never see a 4Runner with a magnetic company sticker sitting in an industrial parking lot or in less well off part of town and you'll never see a Pacifica in an upscale suburban driveway.
> Both approaches work but for different demographics. You'll never see a 4Runner with a magnetic company sticker sitting in an industrial parking lot or in less well off part of town and you'll never see a Pacifica in an upscale suburban driveway.
Huh? Pacificas are everywhere, especially since they also have the only PHEV minivan in the US. In the PHEV Pacifica FB group I'm in (I do own one, so I am a bit biased here), many people have Pacificas alongside Teslas and other EVs. But even the gas Pacificas are everywhere, including "upscale suburban driveways".
Though your point does still stand about FCA/Stellantis' reputation: we had to get over their our perception of them when we got ours, and I would never consider any of their other vehicles. But our Pacifica has been a really great van. I'm just amused that you picked the one vehicle in their entire lineup that doesn't prove your point.
I wanted to say Dodge Journey but they don't make that anymore and it wasn't really a popular fleet vehicle. The nature of the comparison I had to make kind of forced my hand. The Pacifica (minivans in general, too much $$ new) doesn't scream negative equity nearly as loudly as the Journey.
You're right about the PHEV thing attracting a lot of people who otherwise would never own a Chrysler. Checking "just the right boxes" is something FCA is really good at with their more niche vehicles. The high horsepower Charger/Challenger/Cherokee and the Jeep Wrangler are amazingly good at creating a way for the married couple who can't agree on what vehicle want to both leave with enough of their boxes checked.
Pacificas are very common in upscale suburban driveways. Source: live in an upscale suburb with lots of Pacificas in the driveways. Honda Odysseys and Toyota Siennas are popular as well but probably not as much as the Pacifica.
I live in Palo Alto. I have a Pacifica (PHEV). I see many many Pacificas on the road and in driveways. The MSRP of a fully loaded hybrid is 54k. The MSRP of a fully loaded gasser is pretty close.
The reputation hasn't stopped FCA from selling to this market. Wranglers and Grand Cherokees are pretty common too, as well as the rare Hellcat.
Car dealers make their money from trade-ins, parts and service (new sales are often a loss leader to get you in for service work - everyone is driving for the best deal even if the dealer loses money), so the dealers at least care.
Car makers need to do well against the competition, and resale value is something that people look at. They want your car to last 10 years and then fall apart - that being about the time people start to be willing to call a dead car dead of old age. Though they do like having a few collectors cars that are a lot older than that around because collectors also feed into car culture.
I know that cynicism is popular, but in fact their incentives are for more than that. The executive who wants to last - and most do - needs to ensure there is a profit next quarter as well. Sure they will take profit today vs next week, but they also invest in r&d because they know if they don't eventually there will be no bonus at all next quarter.
No real data, but in my decades of gaming I didn't really hear of people buying the top card in hopes to resell it later and make it affordable.
In my experience people buy what they can afford. So mostly *60/*70 (Ti) Nvidia cards. Few people who I know bought the 1080 are still using it. The second life for gaming GPUs mostly is to serve friends/relatives who ask to help building a PC for them.
Some anecdata, I buy midrange cards and never try to sell them. They usually end up in the machines of friends or family.
The purchase price is what I am willing to spend on the device and it's usage, and a healthy dose of pragmatism about what is actually necessary.
I also overclock to stretch their usefulness out but I must admit the leaps in each generation recently, as well as some fundamentally new technology capabilities mean I am not keeping up as well.
I don't have data, but I do participate in an enthusiasts forum. It's a huge thing for a lot of them, because a top card costs quite a chunk of cash (even before the current shortage), maintaining resale value allows them to stay at the top for a relatively modest outlay every card generation.
Of course right now you can sell 4 year old cards for a profit, the market is crazy.
On average, I imagine it's low. However, anecdotally the gamers I know who are buying $1000+ cards, generally do so every 2 years with the intent to flip them and buy the newest hottest card.
In a similar train of thought - there might be better data on this question if one was to look at the photography equipment market. Or other used consumer markets with high re-sale value.
>car makers are incentivised to try to ensure their vehicles have decent resale value
Why? They are only incentivised to make more profit, and selling high-tech expensive cars that break after 10 years is great for that. The ones that will bear the cost of that will be either the financing companies (and it would be fun to see the market collapse as the current crop of shiny German cars hits the 2nd hand market and no-one is dumb enough to buy one), or those poor souls who insist on buying old-ish cars instead of leasing like everyone else.
Now days it is often the automakers themselves doing the financing/leasing, so if a model has poor residuals it directly affects their profits.
Even with third-party financing, the anticipated residuals influence the attractiveness of the lease offer. A cheaper lease offer means more sales for the automaker, and thus more profits.
Personally I don't think the market has yet caught up to the fact that today's new cars will make horrible purchases as the 3rd or so owner.
All the financiers have to do is make sure someone buys it for a good price after 3 years of being leased from new. That's not a bad proposition - the car will probably last a while longer with all of its electonic toys and gizmos working. It's the buyer after that (of which I am usually one) that will find out how expensive the car is to maintain.
Cisco and Sun were the go-to big name equipment suppliers to many startups around the dot-com days. A lot of it was stupid expensive (like 150k for a sun server with the performance of a 5k Linux box, or 20k for a basic Cisco router). Well, when the startups folded a huge amount of that equipment got sold for pennies on the dollar and Cisco basically almost went bankrupt - and Sun basically did.
the glory days of just bonkers sun servers! as a young pup consultant fresh in the USA, i was lucky enough to help setup a fully loaded sun enterprise e10000, gigantic (for the time) external disk arrays and a room-sized storagetek powderhorn[0]. it was in dallas for an insurance company ~1997. all in-house apps, crunching credit card risk stuff i think? an interesting quirk is that they were an almost all-X11 shop! they had a really sweet custom CDE setup. not what you'd expect at that type of firm. would love to know what that e10k eventually sold for.
I make $7-8 per day, after electrical costs, letting my 2080ti (normally used for flight sim, but when not in use…) run nicehash’s miner software. My electricity costs $0.0816 per kWh.
Not a ton of money but it adds up after a month and the GPU was a sunk cost since I got it for gaming. It literally is profitable, for my situation anyway.
If you're using nicehash you're not mining bitcoin. The payout is in BTC but it mines whatever it thinks is most profitable. On your 2080ti, it's not bitcoin.
I agree that the statement "you can't make a profit any more" is just incorrect because of the subjectivity of "you", but it's also worth noting that this is extremely variable - my electricity costs $0.38/kWh off-peak and $0.54/kWh on-peak.
Whether or not mining on existing hardware is profitable depends entirely on where you live, and since we're explicitly talking about the effects of an increased demand on GPUs due to mining, people are clearly not just using existing hardware.
So at around $2500 a pop, you'd be looking at a year just to amortize your costs, and that's assuming BTC stays at its current insane levels. (A mighty assumption, since it's dropped 30% in the last week.)
The volatility is the problem as it makes calculations unreliable but a year to cover your costs on a 2 year (if you upgrade to the latest graphics card every generation) or 4 year (if you go every second gen) life is a reasonable investment.
It's literally a ROI of 100-300%.
Or to put it another way it's similar to a risky stock with a compounding yearly return of 41% for four years.
Using a graphics card for mining that you have anyway is just gravy (apart from externalities which are another big problem).
He bought it for Flight Sim it sounds like, the mining is just a bonus.
And Flight Simulator 2020 needs crazy specs if you want the pretty graphics at decent frame rates. I was planning on doing the mining on the side just to reduce how much it cost, but I'm mainly getting it for the pretty Flight Sim graphics.
Or looked at another way, with only a year of mining you cover your costs, and everything after that is profit. I don't GPU mine, but I can certainly see why people do.
That's absolutely insane. I'm sure it's a nice card, but it's not worth three payments on my house.
Doesn't all this limit the game publishers as well? Surely someone who is currently developing a game has to consider whether or not to even care that something like a 2080 even exists, it might as well not.
And if games starts to be developed with older and slower cards in mind, to ensure that games can even play them, won't that hurt GPU sales in the future?
When even retailers have these prices, then those are the prices.
> I have bought around 4 3070s/3080s all for under 1000 at actual retail pricing.
How much time and effort did you spend to accomplish that?
I've signed up with multiple AIBs and Nvidia itself for months to get notified if anything I'm interested in gets in stock. Nvidia stopped selling directly in my region (Europe) a while ago, no AIB has notified me yet. Probably because I picked all the most affordable models, all of which are still more expensive than 3080 FE MSRP, while 3080 FE has not even an official retailer left in mainland Europe.
Sure, I could setup a whole range of alerts and spend all day hanging out in a bunch of discords to catch one of those super rare stock drops.
But tbh that's not the experience I'm looking for when I'm already willing to spend money: I don't want, and most people simply can't, having to organize the rest of their life around the possibility for a purchasing "opportunity" at any given moment.
Las vegas called, they want you to know they have the brightest and most energy consuming lightsource* in the world blasting at the sky, drawing fambling addicts like a moth to a flame.
*excluding laser research laboratories and weapons.
The luxor light was originally 39 Xenon bulbs running at 7 kW each. 243 kW wasn't that much energy to start, but they actually run only half the bulbs now on a rotating basis. So about 120 kW.
The light costs about $51 an hour. Arguably the smallest expense involved in running a casino hotel 24 hours a day. Less than a rounding error.
They cut back to half-brightness not because of money, or energy, but because they thought the light pollution was a little excessive.
The strip is mostly solar-powered so the Luxor is basically just taking sunlight and beaming it back up into the sky.
It's not like you could take that energy and ship it elsewhere, or do something more productive with it. The Strip's solar plant exists because of The Strip. If the Strip didn't exist then you'd just have sunlight hitting rocks in the middle of the desert like everywhere else in Nevada.
Oh, and moths to a flame? That's literal. There's an entire ecosystem built up around the moths who flock to the Luxor, the bats that eat those moths, and the owls that eat those bats.
I didn't encourage or discourage gaming. Also, mining means stressing the GPU 24-7, gaming is occasional. ... if you're gaming 24-7, then please stop doing that to, it's not good for your physical and emotional well-being :-(
In reality most mining happen with under-clocked GPU cores (and admittedly overclocked VRAM), and at very stable temperatures. The GPU that is used 24/7 for mining will be far less stressed than a GPU that is powered on and off, and cycles through a lot of thermal changes throughout the day
One of the issues people have mentioned in this thread is that cards sold by miners are often burnt out or nearly burnt out. In that case, it seems like cutting off miners from the secondary market would be an improvement for the secondary market.
I someone who hasn't bought a /new/ card since a 980ti, I find this talk of 'crypto burned gpus' to be a bunch of hooey. 1080ti x3, 1660ti x2, 2080ti x2... (We keep three gamers in cards around here, those aren't 'oh noes, this one fried, get another)... not one of them has had an issue.
Nor did the six P104s I picked up for an research project. And those cards smelled.. funny. I'm pretty sure they'd never seen aircon, and had no identifiable oem markings that I could find, but afaik, they're still chugging along 18 months after /whatever/ was done to them in their former life.
Honestly, I wonder if 'gpu burn' wasn't nvidia's first attempt to mitigate the cashflow drain of the crypto-cast-off-market.. before they decided it was more convenient to blame the need for drastic measures on a supply chain issue of their own design. -sigh-
Its an issue most people don't worry about because they're running exactly 1 GPU, which they wanted because they wanted to game, and if it dies, it dies.
But Google, Amazon et al. are very familiar with the concept of hardware lifetimes. Internally, calculations for some algorithms are done based on MTBF values for components - i.e. the cost of doing a thing is a certain number of CPU-hours, and you know a CPU will give you X-hours before it's liable to be hardware-dead.
A second hand gaming GPU has a very different power-on profile to a mining GPU - a gamer is going to struggle to on average put more then 4 hours a day or so through a card, most likely less. A miner is going to have that thing powered on the whole time, permanently. Even if the GPU is fine, the DDR RAM has been aging and that stuff fails faster.
The short answer is: people are going to be asking for a discount based on the expected lifetime of ex-miner hardware, once we know what that is. If the answer is "well it might last 6 months or it might last till you replace your machine" then the price will now reflect that expectation and risk profile. Not to mention the dilution factor which will be actually failing hardware being sold off by miners without properly marking it - after all, you might as well let someone gamble that bit flips in the GPU only lead to graphics aberrations rather then bad hashes.
While the cards are used heavily, thermal cycling arguably is worse than running a card 24/7. These cards do not generally experience any thermal fluctuations and were probably kept within adequate temps
There's three main ways GPUs wear out, silicon electromigration, electrolytic capacitor dryout, and fan bearing failure. And the first two happen much faster at higher temperatures. Temperatures are similar between gaming and mining, but miners run 24/7 compared to gaming 1-2 hrs/day, so the GPUs age like 10x faster. But still, GPU failures are pretty rare, I think the concern is overblown.
They also sometimes do GPU BIOS modifications, which is a fun thing to discover once the card ends up on the second-hand market and you happen to buy one. Luckily the fix is to reflash the BIOS using a matching one from TechPowerUp GPU database, but this assumes that you even know that this is something to pay attention to.
Source: happened with an RX 560 that I bought second-hand. Driver installation failed in Windows 10 due to the modified GPU BIOS. Was fixed with a reflash of a stock GPU BIOS using atiflash.
I didn't check it, but if I had to guess, it might have to do something with allowing the card to run using settings that make it more suitable for mining. Someone more familiar with GPU mining can correct me here.
Totally true- I went through the process of manually overclocking my 2080ti for max hash rate and the best settings involved setting a power limit that is less than half the card’s 100% limit.
Anedoctal but my two last GPUs burnt out after a few years. Bought new GPUs from reputable suppliers. Never mined or overclocked besides factory overclock (which for most of the time I had disabled). I was playing a lot more than 1-2 hrs/day. I guess it was the worse of both worlds. Heavy usage + cycling.
I’ve purchased graphics cards from the used market, and used them for years. Never any problems. One of my go to cards is 5 years old … that’s when I bought it - I’m guessing it’s much older.
Just normal wear, accelerated because it's always under load. The used cards are fine most of the time if the temperatures were stable under 80-90 degC.
The main cause of failure is the VRM, those components are the hottest, least reliable and least cooled.
Sadly, sometimes a blown transistor or capacitor can take out the whole GPU.
Flex is caused when the temp changes. Mining cards stay at the same temp and don't move around a lot while overclocked gaming cards are getting flexed multiple times a day
on contrary, miners tend to run gpus undervolted and underclocked for efficiency. gamers tend to run gpus overclocked (and sometimes overvolted) for performance.
Aside from a few top end models modern desktop GPUs are not designed for continuous operation at max load. When you combine than with higher temperature the lifespan of the cards is indeed reduced. The real question is how much this matters.
> The real reason Nvidia is doing this is that when mining profitability inevitably drops, either because of a crypto price crash or Ethereum moving to Proof of Stake, the miners will flood the used market and ruin Nvidia's profits for a few quarters.
Taking into consideration that the gamer segment is now well aware that miners and scalpers are ruining their supply, I really hope that when the flood of used cards happen, nobody with a grain of salt will buy devices that have run at 98% their thermal limit for years.
I really hope that market-ruining miners will find themselves with a ton of hardware nobody wants anymore.
Please take into consideration that this will be the same naive, impressionable "market segment" that repeatedly pre-orders new games, even though there is no real reason to do so, as the games are digital copies, so there is no scarcity. The same market segment also rewards shitty game companies, producing half-finished, barely working product, with their money again and again. All because of a combination of deviously good marketing and FOMO.
So yes, I fully expect those soon-to-break GPUs to sell like hotcakes, and a lot of crying soon thereafter.
It's not like pre-orders are objectively useless like some people may paint it.
> The same market segment also rewards shitty game companies, producing half-finished, barely working product, with their money again and again
Most people buying PC games aren't the ones buying top of the line dedicated GPU's for their gaming rigs. They also aren't ones who will scrutinize super technical details (nor even gameplay elements) as if they are Digital Foundry. Lotta weird generalizations here.
I'd rather buy a used card from a professional miner than from an enthusiast gamer. The former card will have run at very stable thermals, while the latter will probably have gone through a lot of thermal stress resulting from heating up and cooling down, and component stress from power on/offs.
Professional miners usually underclock and undervolt their cards. A constant load is also less stressful on the die and pcb than repeated heating and cooling cycles. The one thing you might have to replace on well kept GPUs are the fans.
2000 series was expensive and underwhelming, and raytracing wasn't really viable in real-time with other graphical settings cranked up.
3000 series is a huge performance boost, reasonably priced (at least at MSRP), and gives comfortable buffer on lots of scenarios like high-end VR, 1440p at very high FPS, and comfy 4K with all the good stuff turned up.
Meanwhile, the GTX 1080 was cheap and could basically play everything at 1080p. I think gamers just skipped over the 2000 because it was lame.
But hell, I hope you're right. I'm still rocking a 980, and I've not had any success trying to grab a 3070 to upgrade. Maybe these LHRs will actually free stuff up. (But to be honest, the 980 is still doing fine with most of what I play at 1080p.)
18 months ago I could get a 2070 Super for $550 and the 1080 Titan cards were $799 - 900 at launch and long after. The 2000 series was a lot of bang for the buck.
At the start I thought that miners weren't really a force to reckon with but when I saw that the scalpers are selling their received cards on the same day to miners I changed my opinion. I think people severely underestimate the proportion of miners at this point, overall they aren't a majority but during Q1 2021 they probably were. I'm part of a few botting groups, although I don't take part. The current price difference between the cards can also be partially explained by how much hashpower they have for mining ethereum.
Some back of the envelope math. The current hashrate is about 650000 Gh/s, which is almost 4x as much as at the start of the year. So an increase of about 50000000 Mh/s. A 3060 Ti produces around 60 Mh/s, that's about 8.3 million 3060 Ti's. Of course not all of the added hashpower comes from new cards so the numbers need to be taken with a grain of salt.
my 1080 ti sells for 30% more than i paid for it 3? 4? years ago. its price has fluctuated in mostly perfect tandem with crypto. just based on first hand experience, your claim (if its what you are actually claiming) that miners arent influencing the price is pretty wrong.
My point was more nuanced than that. Of course miners are influencing GPU prices. If miners stopped buying all GPUs, prices would probably go back to MSRP pretty quickly.
But stopping miners from buying one GPU model won't make it sell at MSRP. Miners switch to buying other cards, driving the price of those even higher, but the LHR cards are still going to be out of stock everywhere and scalped for 2x MSRP, because there are so many gamers buying GPUs.
I don't think this will help pricing or availability much either. LTT did a video on this topic somewhat recently and it turns out a lot of the reasons that you can't get a card right now is related to chip shortages and people (yes, regular gaming enthusiasts) buying any card they can find. Even cards that already have terrible mining efficiency are scarce.
Unless the miners are cutting off the supply at the silicon production level. What's to stop Antminer (or any other mining asic company) from paying 10x as much for a majority of TSMC capacity, RTX and Ryzen supply be damned.
Most of TSMC's capacity is allocated years ahead of time. They're contractually obligated to supply AMD wafers at a predetermined price and volume.
Also, GPUs are pretty good at mining Ethereum. ASICs are only about 2x as efficient, many miners would rather have the security of knowing they can always sell their GPUs to gamers if it all comes crashing down.
> Nvidia is trying to get miners to buy mining-only cards, because those can't be sold to gamers down the line. They'd rather mining cards become e-waste in a few years than get resold.
On other note, the "cloud" Vega that AMD sold to the very few cloud videogaming companies comes with exact same limitations, and a clause of non-resale, and non-refurbishment. This way buyers, can't simply solder display connector to those cards, and sell them on open market.
Imagine Intel would've sold their Xeons with the same limitations... but wait, they actually did this for "custom fused" Xeons they sell to big dotcoms.
Now, who said that silicon vendors can't "vendor lock" their dotcom clients? By buying direct, all those dotcoms have exposed themselves to an even bigger supplier risk of this type, and not reduced it.
> The real reason Nvidia is doing this is that when mining profitability inevitably drops, either because of a crypto price crash or Ethereum moving to Proof of Stake, the miners will flood the used market and ruin Nvidia's profits for a few quarters. Nvidia is trying to get miners to buy mining-only cards, because those can't be sold to gamers down the line. They'd rather mining cards become e-waste in a few years than get resold.
What they could have done is instead sell the video-capable cards in a way that discourages bulk buying (limits per customer, maybe a little bit bulkier and harder to correctly cool if several are in a rack) and then release a more bare-bone (no bulk, just a heatsink and optional fan) mining version that can be bulk-ordered and performs well in a rack mount.
Is it unsophisticated of me to assume that an ETH crash will just cause everyone with GPUs to switch to a different cryptocurrency? I'd expect miner demand to remain firm unless there's a broad crash.
tbf I wouldn't buy a cheaper used gaming card from a miner, because of potential ware. Electronics ware off, and a mining card would have been taxed much more than any card used for gaming.
I have and the truth seems to be that if the miner took good care of the card it will probably be fine. Thats why I said "potential". This is not an acceptable risk when Im spending big on gaming hardware.
edit: just to be clear, I dont have the expertise to check the wear, so it just seems like a bad gamble to me
With so many more of them out there, it'll be interesting to see if this becomes more of a thing. Alternatively, it's easy to imagine that depending on just how constrained these are, they might eventually find homes doing non-crypto headless DC workloads, such as deep learning stuff.
AMD isn't even going to limit performance on gaming cards, nor are they using a different node for their mining cards like NVIDIA. They are straight-up just going to shunt wafers over to mining cards, while letting miners have both the mining cards and the gaming cards.
Gamers won't be buying used 30xx cards from miners because by then nvidia will have released 40xx cards. Gamers are mostly kids running after shiny things.
The real reason Nvidia is doing this is that when mining profitability inevitably drops, either because of a crypto price crash or Ethereum moving to Proof of Stake, the miners will flood the used market and ruin Nvidia's profits for a few quarters. Nvidia is trying to get miners to buy mining-only cards, because those can't be sold to gamers down the line. They'd rather mining cards become e-waste in a few years than get resold.
People think Nvidia is doing this to try to help them get better prices now, but really Nvidia is trying to extract more money from gamers in the long term.