"If we didn’t have such a dire X I would be upset on principle, but I see this as a move that’s supposed to help ..."
What worth is a principle if you don't still have it when it get challenged? Then it's just a opinion, which is fine by itself and it's ok to change opinion, smart people do it all the time. But don't call it a principle.
Unlikely that mining will disappear because NVIDIA limits a specific algoritm/client software/however the limit is implemented. It will, like most "limit for your self-protection" systems, eventually be broken. Miners are already not afraid of picking apart the components they are using, so neither software nor hardware is safe.
I think your criticism is valid, but I don’t think it’s sensible to adhere to principles in a black and white way. For me, the minor restriction put in place by Nvidia is the lesser evil when you consider that they could just as well sell out to the miners and drastically increase their margins. In fact, I think it’s a very sympathetic move. I don’t think “it will be broken” is a good argument, because the intent displayed by Nvidia does matter.
Their intent is to limit what people can do with their hardware. That strikes me as almost morally wrong. I tell the computer what to do. It doesn't tell me what to do.
I think cryptocurrency is a scam and I'll be glad if/when it goes away and graphics cards become cheap again. However, I will never support hardware manufacturers infringing on the idea of general purpose computing for the sake of their business model.
If they want gamers to get more graphics cards then make so many graphics cards that everyone will be able to afford them. Don't try to artificially limit what the hardware can do.
>If they want gamers to get more graphics cards then make so many graphics cards that everyone will be able to afford them. Don't try to artificially limit what the hardware can do.
If there are poor people why don't they print more money? But seriously, I addressed this in my reply: there is an ongoing chip shortage in the industry (besides the fact that it may not make sense to scale a production line for a demand surge).
If it suddenly became wildly profitable to buy all of the bread and rice and any form of carbohydrate from food manufacturers and grocery stores and burn it in a field, would it still be morally wrong for food producers to figure out a way to cripple the utility of their food products by making them unburnable? Is there some capitalist principle or moral principle of the free market that overrides all other human needs and values?
What is the principle you are defending and at what point does it become less important than other principles?
If the choice was between restricting food - i.e making it less profitable to burn, to make it more available to the hungry and letting food be free and the hungry starve, then that's a pick between the best of two bad options. Clearly it's better to feed people than let food be unrestricted.
In this case though the choice is between restricting hardware to give gamers more affordable GPUs. The benefit doesn't match the cost.
The principle is something like freedom or ownership. Arbitrary restrictions are bad because freedom is good. By definition arbitrary restrictions aren't needed (arbitrary) and reduce freedom (restriction).
Freedom is only good until an incentive aligns that causes harm to the long-term stability of an industry or society. Right now, people who want to create value cannot do so because a paperclip maximizer has found a way to exploit the system for its own gain. Nvidia's move, futile as it may be, aims to maintain the stability of the GPU market for professionals and consumers, the actual value producers of society. Crypto mining does not produce true value commensurate to the burden it is currently placing on all other industries that use computing power.
The economist in me can't help but state the obvious: If those other industries were doing something so much more valuable, they'd aurely be willing to pay just as much for GPUs.
Of course, it's not really the big value-producing activities getting hit here, it's gaming.
You missed the (extremely good) point that istorical is making.
The argument in this subthread has been over whether we should be mad at Nvidia on principle. For example, as capableweb puts it:
> What worth is a principle if you don't still have it when it get challenged? Then it's just a opinion, which is fine by itself and it's ok to change opinion, smart people do it all the time. But don't call it a principle.
You seemingly agreed with a "principled" reading of the situation: "That strikes me as almost morally wrong. I tell the computer what to do. It doesn't tell me what to do."
istorical is arguing, by giving an extreme case, that we don't have a consistent principle against limiting the capabilities of a product when it will benefit a customer who matters more.
After all, if we've agreed that it's morally wrong to "cripple" a product, then what worth is that principle if we abandon it during a crisis? Let the people starve! You're rightly repulsed by that idea, and so you're retreating to the position that we're just going with a cost/benefit analysis. Preventing food from being burned has a cost and a benefit. Preventing GPUs from being used for crypto has a cost and a benefit.
The thing is, this gives up the supposed moral high ground that many in this thread have tried to promote as a reason to be mad at Nvidia. From the perspective you present, it's perfectly reasonable to look at the situation and say that preventing GPUs from being used for crypto will have more benefit than cost. (I certainly think so!) There's no inconsistency you can draw out of someone weighing the various benefits and costs differently than you do.
No, I think I addressed that point directly. It is morally wrong, or against principle, in my view to do things that cause people to starve to death and to impose arbitrary limits on things. When the only possible choice is between two wrongs, letting people starve or limiting the combustibility of our food, we should pick the least-wrong of the two options - limit food combustibility in order to feed more of the starving.
This is not abandoning a principle but rather picking a higher principle to follow when two or more principles conflict. The point I was making is that we are not currently in this situation. We aren't starving people by denying them access to the latest and greatest graphics cards for cheap.
> Not really. The intent is to get graphics cards in the hands of gamers and professionals, instead of in the hands of crypto miners.
Perhaps I'm missing something, but why doesn't nVidia just make the best cards they can, and sell them for the highest price they can? What does nVidia care if their customers are using their cards for gaming or mining?
> What does nVidia care if their customers are using their cards for gaming or mining?
Because it makes a lot of business sense to prioritize the long time gamer customer base, as opposed to the crypto market, which could disappear very quickly.
Prioritizing certain customers, that will be around in the future, makes a lot of sense.
It also helps build a gaming ecosystem. If gamers don't have graphics cards, then the gaming ecosystem as a whole suffers, and could hurt their future business.
> If you're selling out make more and/or raise prices.
What if you want to raise prices only to crypto miners, and not raise them for gamers/professions, in order to support your long term customers?
There are lots of very good business reasons to provide preferential treatment to a specific customer base, and to raise prices on a different customer base.
It is called price discrimination, and is very useful.
Do you not support companies making obvious business decisions like this, with their own company, that have large benefits to their existing customer base, as well as being perfectly rational from a business perspective?
Having crypto miners subsidize an existing customer base, makes a lot of business sense, and helps out a lot of people.
And it is all done, with people making voluntary decisions with the products that they choose to sell. Don't buy the product, if you don't like it.
Because what they want to do is ensure that the cards are not resold to miners, and instead stay in the hands of long term customers.
> You don't need to cripple the devices
First of all, they aren't crippled for the existing customers. They are only limited, for this other use case that the company doesn't care about.
And they do need to limit that functionality, if they want to prevent those cards from being resold to crypto miners.
> I don't support companies
Then don't purchase their product, if you don't like it. That is kind of the point of all of this. They don't want you as a customer. And it is their company, that they can do what they want with.
> Their intent is to limit what people can do with their hardware. That strikes me as almost morally wrong. I tell the computer what to do. It doesn't tell me what to do.
If you need to run millions of hashes per second then buy one of the non-LHR models. No one is forcing you to buy one of these chips and Nvidia is not taking anything away from you.
This is like saying that Intel limits what people can do with their hardware because they make both i3 and i9 chips. There is a spectrum of options to fit more people's needs.
This is different than just product segmentation or binning chips - this is hardware restricting software.
How would you feel towards Intel if they didn’t allow you to run Microsoft Office on an i3 chip because it’s a “consumer grade processor” - you have to upgrade to an i5 or higher to run “business programs”.
It's more like you wanting to buy an i3 that is up to spec for your games, but then Intel launches a new batch of those i3's with intentionally reduced gaming performance because it's not meant as a gaming CPU (in the eyes of Intel), please upgrade to an >= i5
> but I don’t think it’s sensible to adhere to principles in a black and white way
I agree with you, you should for sure listen to criticism against your principles, and challenge them yourself too. But if they change because "if we didn’t have such a dire X", I wouldn't call it a "principle" anymore. A principle (for me) would mean that you don't compromise on it "in emergency" but rather that it's something fundamental you believe in, without much reason in many cases.
> because the intent displayed by Nvidia does matter
Yeah, agree again that it matters, but as a gamer myself, I'm afraid it doesn't solve the problem for me. I want to be able to buy graphic cards myself too, and these "fixes" from NVIDIA looks (to me) like posturing without actually try to address the core problem (low supply). I'd love for them to come up with a real solution, that doesn't try to decide how people use their computers, even though I don't want to mine anything myself.
> What worth is a principle if you don't still have it when it get challenged?
Huh? To me, refusing to even acknowledge or entertain criticism for one’s principles is one of the worst things I can imagine. Principles ought to be routinely subjected to criticism, no different than scientific theories.
> refusing to even acknowledge or entertain criticism
I'm not advocating for you not listen to criticism aimed towards your principle. But if your principle gets put to the test and you abandon it, it's no longer something you fundamentally agree on, it was some "light belief" that you had.
If it were due to new facts and information, sure, but a lot of time the "new" principles appear to be self-interest biases manifesting.
Also, if your principles are likely to change quickly, it's probable that they haven't been thought through very thoroughly to begin with. It's not like you have to wait for situations to actually arise to consider them.
In the same way that a scientific theory was meaningless if it failed a test, a principle was meaningless if it needed to be adjusted when actually put to use.
Superseded scientific theories aren’t all “meaningless.” Newton’s laws of motion are superseded, but they aren’t meaningless, and you presumably wouldn’t say that scientists who updated their beliefs in the face of new evidence and explanations were doing anything wrong.
You’re right: “meaningless” was a heavy-handed attempt at continuing the parallel on my part. And sure, they weren’t doing anything wrong, but they fundamentally were wrong in the rules they followed previously. Similarly, if you adjust a principle, you admit that it didn’t hold for you in the first place.
As a tangent: while the parallel to scientific theories is good on the surface, it does lump in this idea that there’s a universal set of correct principles. Isn’t that just philosophy, then? Logical arguments about which principles are sound & universal. Maybe the lesson is to pick principles from the battle-tested ones rather than trying to develop your own.
This is nonsense. Principles should adapt to the context they're applied in. In principle I'm against big government intervention into our personal lives, but you bet your ass I'm in favour of lockdowns when thousands of lives are at stake. I'm not going to let thousands of my compatriots die out of principle.
This is just imprecise statement of principle. It's more like you're against the government intervening unless they're credibly doing it to save lives. If someone discovered how to turn a microwave into a nuclear bomb, presumably you'd support government intervention to collect all the microwaves, even if they had to intervene a lot in our lives. Likewise, you support government intervention to reduce covid deaths, etc.
Sure, you could just say that your “principles” are simply the entire exact sequence of actions you take in your life. That way you could by definition never change or violate your principles.
But usually people use “principle” to refer to a relatively concise statement that can be applied to a large variety of situations.
I just mean that it must be more concise than a lookup table from every possible situational input to the recommended output. The same is implied by the term “explanation.”
A principle is a moral heuristic so you don't have to scan to root of your moral argument tree on every moral question. As a heuristic, it attempts to compress the moral space down and will naturally have exceptions.
As a compressed form of your moral judgment, it is also useful for you to quickly communicate your position on the moral space, and it is up to participants in social interactions to decide whether or not someone is good at this moral judgment compression.
For instance, I have a principle not to lie in general. Most social participants know that this does not constrain me from lying in specific situations. After repeated interactions they can tell whether this principle of mine is held in a manner that's useful to them.
The same principle also allows me to quickly judge situations that require me to lie. Is it okay for me to tell this person that their house is painted poorly? Well, I have this anti-lie principle so that pulls me toward "yes". I might have other principles that pull me in other directions.
Of course you can treat principles as immutable moral rods. This presumably has some value to you. In general, people do not act in this manner, however.
I think your loose definition of “principle” does not meet many others’: it’s definitely intended to be stricter than a heuristic.
Of course even though you can have relaxed principles, e.g. “I usually don’t lie” vs. “I don’t lie,” I think the point here is that if you believe or preach a particular “principle,” but then adjust it for yourself when it’s convenient (for example a “just this once” scenario), it was never a principle to begin with.
Actually, observationally I believe my definition of principle to be far more accurate. People self-report principles as being far more immutable than they behave like.
I have not yet met someone who rigidly adheres to theirs. Perhaps if I had met the Dalai Lama but nearly everyone else will find that the moral space is so fraught that they cannot adequately compress it into immutable laws.
Usually, should one demand such adherence one would find one met with accusations of “looking at the world in black and white” and “lacking nuance”.
Of course I don’t really intend to change your mind, but I do intend to transmit what I have observed.
> What worth is a principle if you don't still have it when it get challenged?
The realities of life make most principles less finite than one would hope. If Nvidia doesn't want to rebrand as a crypto-hardware company, they're going to take action so their products and brand match _their principles_.
How big is each of your principles (in terms of information complexity) that they cover significant enough number of all possible conditions and circumstances?
I’m sure plenty of people would agree to an absolute “I don’t steal” principle—one that covers a pretty significant number of conditions & circumstances-and yet not go back to the register when they accidentally walked out of the grocery store with something, or compromise it more significantly under stress of hunger. In this case, one could argue “I don’t steal” was never a principle they actually held.
What worth is a principle if you don't still have it when it get challenged? Then it's just a opinion, which is fine by itself and it's ok to change opinion, smart people do it all the time. But don't call it a principle.
Unlikely that mining will disappear because NVIDIA limits a specific algoritm/client software/however the limit is implemented. It will, like most "limit for your self-protection" systems, eventually be broken. Miners are already not afraid of picking apart the components they are using, so neither software nor hardware is safe.