Aapje

What is reliable enough (or not) is opinion, but you are presenting your opinion as fact. Again, I have explained why I consider the available benchmarks good enough to have high certainty that the performance of the new cards is not worse. This is a very modest claim, much less so than people who throw around actual percentages. Is it 100% certain? No. But it is a core tenet of science that you cannot actually prove anything. So if you demand absolute proof to call something a fact, then facts do not actually exist. After all, there is a possibility that we are a computer simulation, and anything that we perceive to be a fact can be changed, no longer being a fact, or can be a delusion implanted in our brains. But it is silly to have to present everything as an opinion, so in reality, people just state things with certainty when it passed their subjective threshold. You do so too. This entire idea that your personal threshold of what is valid should be accepted by others, is authoritarian and arrogant. You deny others the right to their own opinion on the matter and you act as if you are perfect. Yet you do the exact same thing in the quote at the top of this comment. So again, you apply your double standards, where you get upset when I do the same thing that you do.

To be honest, I think that you are going to have a hard time finding a replacement that you will like with that mindset. Tweaking is a pretty common need and you didn't seem to have taken advantage of available resources, choosing to use trial and error instead. Having lots of those resources available is one of the benefits of the headset. The need for 3rd part accessories has benefits as well, because it means that you can tune things to your liking. I would argue that the price reflects this and you should just factor in some accessories into the price.

Yet you are making educated guesses as well, albeit with less education behind them. Why is it OK for you to do this, but not for me? Your double standards are very obvious. You're moving the goal posts from 'MSRP is fake' aka no one can get these cards for MSRP, to a claim that anyone should be able to get these cards from MSRP (on day one as well?). I've already explained that entire countries can have higher prices if there are tariffs or local laws that mean that stores will not have sufficient margins at MSRP. That doesn't mean that the MSRP is fake, because you can't expect companies to subsidize residents in countries that screw over their residents. But you seem to live in a fantasy world, where companies are supposed to pay tariffs out of their own pockets, factories can make many millions of cards in 1 second, etc. Which is the same thing that we heard for the previous generation, just before release. Ultimately, we lack the actual numbers, so we can do no better than an educated guess based on the vague statements being made. But again, you are perfectly fine using very imperfect evidence for your own educated guesses when it suits your argument, but you get upset when others do the exact same thing. I'm not sure why you are claiming victory, when you didn't claim that at least one AIBs would miss the release date with their products either, so you did no better on this front. Note that 'not being surprised' wasn't the issue at hand, but whether we can predict with some level of accuracy. But you seem to be moving the goal posts again when it suits you. I never claimed that I can predict things perfectly. What I do claim, is that I don't base my claims on market conditions no longer exist, like the mining boom, while you do base your claims on that. Note that the AIB statement is still quite vague about whether they will make up for the delays or not. We also have no statements by other AIBs, and the other statements are 'we've heard'-statements/rumors. These are not the most reliable. And we have no actual numbers. You also need to keep in mind that a delay doesn't have to mean that supply can't catch up later. For example, if the plan was to deliver 1 million units spread over a month (this number is just an example), but they have a delay where they deliver 1 million units during the second half of the month, then the supply situation is exactly the same at the end of the month. In both cases, they delivered the exact same number of products to market when you look at the entire month. Then the only impact is on the people who must have the card in the first half of the month. It really depends on what the cause of the delay is, which is only vaguely alluded to. If there is no hampering of production, then this should only be a temporary hiccup, where the issue is how fast they can bring the cards to market, and not how many cards they can deliver. But here also, I see no proper analysis from your side what possibilities exists, and to what extent we actually have evidence to support an educated guess. You just uncritically grasp onto evidence when it appears to support your position, yet are extremely critical when it doesn't. There is no objectivity where you are just as critical of all evidence, no matter what claim it supports. Yes, early on, which I already argued is a hard thing to pull off, and tends to require quickly grabbing a card from a very limited quantity. This is why I suggested that people resist the FOMO and realize that the supply situation will most likely improve over time, unless there is a shock event like a sudden mining boom, so they might want to resist the scarcity mindset, where they overspend based on the idea that they will lose out if they don't jump on the very first opportunity to buy a card. You don't seem to realize that this is not actually an argument in favor of your claims. If the cards are disappointing, then we can expect demand to be lower. This then means that there is less demand for the cards priced above MSRP, and AIBs will shift more production to MSRP cards, sooner. Furthermore, there will be less competition between buyers, so it will be easier to pick up an MSRP card. Note that if the cards are better than the previous generation, 'disappointing' is merely a subjective statement that is fully dependent on the extent to which one's subjective expectations are met. You'd still buy the 5080 when you are in the store and have a choice between a 4080 or 5080 for $999. There is a big difference between a product that is a bad purchase in general, or a product that offers a small improvement, so it makes sense for fewer people to upgrade, when they already have a decent card.

BTW, at least some AIBs are claiming that they cannot make the release date: https://videocardz.com/newz/retailers-aibs-and-system-integrators-warn-of-geforce-rtx-50-series-shortages-and-delays It's starting to look like Nvidia messed up with their release date. So perhaps we should just act like the real release date is half a month or a full month later.

My suggestion is to always test the cable using the speed tool, so you know that it is running at high speed. Both the cable and the USB port need to support the higher speed, and it is easy for one or the other to hamper the bandwidth and get horrible visuals. Furthermore, I would open the Oculus Debug tool and manually set the transfer rate to 500. I think that you had a bandwidth issue and that the text would be clear if you fix that. Yes, I would always budget for some upgrades with the Quest 3. The AMVR Facial interface fixes the light bleed at the nose and is more comfortable and hygienic. Do be aware that they fairly recently released the better Q3FC3 version, which you should get, not the old one. It's the one with the distinctive knobs for distance adjustment.

If you want to get one early for MSRP, you need to watch out for relatively small quantities getting dropped and snatch one up right away. For example, by following the GPU Drops or StockDrops discord, or whatever discords, telegrams or other trackers exist for your local market. This is true in particular if you want an FE model, which are produced in limited quantity and sold for a limited time, so you can't necessarily wait for supply to catch up to demand, because by that time the FE cards may no longer be sold new. Telling falsehoods that the MSRP is fake, when the reality is that MSRP supply is very low at first and obviously sells out faster than the more expensive cards, doesn't really help anyone. Note that scalpers worsen this issue, as they buy up stock and hang on to it a bit, creating fake demand early on, even if many of these cards never get sold on. Once the prices start to stabilize, the scalpers return the cards to the stores, so supply can suddenly shoot up at that time. We had reports of huge numbers of 4080's being returned to stores by scalpers during the last release cycle. Note that scalping is worse for stores or countries where there is a generous right to return, since scalpers can just speculate on high demand with almost no risk, as they can just get a refund if they cannot sell the card for the inflated price. However, given that scalping didn't work so well for the 4080, I suspect that there will be fewer scalpers this time around (although we might see plenty for the 5090). That was during the mining boom when demand shot up enormously, and supply was not even close to catching up. Back then most MSRPs were set just before the mining boom reignited again, so they were far too low for the supply/demand situation for the next few years, and thus it is logical that prices never stabilized until Ethereum going to proof of stake ended the mining boom. It is quite irrelevant for the current situation, where we are not in a mining boom. However, we are in an AI boom, but in contrast to the mining boom that impacted all cards, the AI boom primarily impacts the top cards, but the impact is still way lower than the mining boom. I consider it the most uncertain when the 5090 pricing will stabilize. For the 4090 it only took a few months, and the supply of professional AI cards is better than back then, so it's probably going to last a few months again, unless a country like China starts buying 5090s up en mass. But for the other cards, it makes most sense to expect a fairly standard pattern or better than that, where the MSRP-models will be in very low supply at first, and stores may ask inflated prices for the first months. Then again, the 5080 in particular seems to give a very poor increase in performance, so there is a good chance that the demand is particularly low. Rumor has it that the initial supply of cards is very low, and that is of course a major factor. You cannot buy cards that are not supplied to stores. In general the rule with new products is that demand is the highest and supply the lowest just after release. Different things happened in the past and you do not seem to understand what the actual causes were for various price changes and supply issues in the past, and because of that, not able to make a good educated guess based on the information that we have. Anyway, my advice in general is to either be willing to spend some effort to get a card for a good price early, or be willing to wait a bit for prices and supply to stabilize. If you don't give in to the FOMO and just wait for the pricing to stabilize, it is a lot easier. Especially for the 5080, I expect that it doesn't take that long.

People are working on a headset that shoots (very low power) lasers straight into your eyes, which in theory is the most efficient and least intrusive way to do it without surgery, if you can miniaturize the lasers and the related tech. However, it requires way more advanced tech than the relatively simple technology of shoving a screen in front of your face and it is thus very far from ready for the market. For example, current eye tracking is rather mediocre and doesn't track the retina, which is the part of the eye that you actually look with, but the laser solution requires much more accuracy than foveated rendering, where you can just render a larger part of the screen to account for inaccuracies in tracking the retina, and slow performance compared to how fast the eye can move.

I would say that the Quest 3 is by far the best value if you play both standalone and PCVR games. For only simming, a second hand Quest Pro and the Pimax Crystal Light are also in contention. The biggest issue with the latter is the QA issues and that support can be poor, so it exposes you to a fairly high level of risk. All the other stuff is not value for money, and sometimes even has severe downsides that make it questionable choices even if you are willing to spend the money. Are you Canadian or Australian? Because otherwise your prices are way off. Note that the official accessories are extremely overpriced and mediocre quality. You should instead get some good quality stuff by Chinese companies (of course, the official stuff comes out of China too).

Literally smoke and mirrors

Winwing claim to have a stronger motor providing 20+Nm, so there doesn't appear to be a motor limitation compared to the Moza/Rhino. They showed off their Cyber Taurus base in June last year at FSExpo, but it was running a fixed movement program, and could not be handled or be used to control a game. So I would say that this indicates that their software lacked even the most basic features. Potentially, what they showed was merely a prototype to test the ability for the computer to control the stick and for the device to handle the forces. That still leaves a lot of work to provide DirectFFB support, telemetry support, anti-cogging, and support for a working grip on top of the base. That prototype might also not have been production-ready, as it can take a lot of work to go from a hand-made prototype that more or less works after a lot of manual fudging, to a mass-producible device that is optimized for efficient manufacture, limited material costs, reliability, a low defect rate, etc. It's pretty much impossible to know the true state of the device and the all-important software for it, with no actual demonstrations since that very basic one half a year ago.

There are people for whom these prices are pocket money, or who put huge value on the extra performance, who earn so much money with the card that it pays for itself, etc. If bragging rights is really the only reason that you can come up with, then that speaks to an inability to understand how much diversity there is in circumstance, and in motivation.

There is a lot of income inequality in China, so one would expect there to be a decent group that could indeed afford a 4090. However, there seems to be an industry in China that converts 4090s to server cards: https://www.techpowerup.com/316066/special-chinese-factories-are-dismantling-nvidia-geforce-rtx-4090-graphics-cards-and-turning-them-into-ai-friendly-gpu-shape This seems to be a major source of the scam cards, which get sold without a GPU chip.

The time to finish a download/install is notoriously unreliable, since it is essentially an attempt to predict the future. So these times are often faked in some way. It might be the case that the time to finish is calculated based on the total download size, ignoring that files have already been downloaded. To test it properly, one cannot look at the displayed time, but you would have to time the actual download yourself.

I have not said that either won't happen, although rumor has it that Nvidia mandates that the companies bring out MSRP models. Then again, they also seem to squeeze the AIBs so the margins on MSRP-cards are very low. So the AIBs do seem to produce few of the MSRP models initially. However, this happens every generation, so it is not a valid reason to argue that this generation in particular is not going to provide a price/performance improvement. Again, you keep acting as if I wrote things that I never claimed. And the pattern with you is that you very aggressively oppose extremely reasonable statements, and take your own experiences as the only valid truth. If you don't want people to call you out, you might want to stop stating falsehoods. We have factual data, consisting of the published benchmarks by Nvidia. This is certainly not the best data, because they don't publish the exact numbers, and most of the graphs compare apples to oranges. However, the graphs have proven reasonably accurate for the (few) games that are presented that do not do false comparisons. Even if you apply generous margins or error, it seems quite unlikely that a 15-20% gain suddenly becomes zero or negative, aside from the fact that Nvidia has never gone backwards in the same tier.

AI is not a money-maker for most, but it is a field in which companies and states think that they can set themselves up for success in the future. And China definitely seems to have bought up lots of 4090s, given how there was a sudden shortage in the West when a ban on sales to China was announced, and almost all new stock seems to have been shipped to China before that ban. And even after the ban, it was worth creating a China-legal version.

That's not true. Bitcoin mining has transitioned to ASICs a long time ago, so they have not been a factor. But Ethereum mining was highly profitable, until Ethereum moved from proof of work to proof of stake. Basically, at first each Ethereum transaction required Ethereum mining to validate the transaction, but they switched over to computers having to put up a stake of quite a bit of money in Ethereum, and then these computers could validate transactions. At the point that they switched over, the mining boom went away. None of the later coins had anywhere near enough value and transaction volume to sustain the mining boom. The crypto miners had a few ways to evade the throttling, so it never was effective enough to stop the mining.

I've already explained that companies can't just give fake performance figures, due to false advertising laws. But the pattern with you in general is that you blow an enormous amount of hot air, and have a hard time actually listening to other people. I'm not going to keep repeating myself. If you want to keep reacting to things I didn't say, then go ahead.

My worry is that the F-35 simply worsens the issue that this game already has. Too much focus on individually cool things, but too little concern with what makes a good, coherent game.

https://www.facebook.com/VIRPILControls/posts/pfbid0EFtmuLszHQxEdi61tRJRPmM3e9L9VKA3Upo7CyVdForG3h9JhaoJVyYPgsBLhxpnl But I wouldn't expect it anytime soon. Last news is that they are still in the patent approval process. It's not just a kneejerk, since they showed off the hardware already, during the same event where Moza showed theirs. However, Winwing seemed to have been behind on the software side (perhaps because they didn't steal the Rhino code).

@kksnowbear If you refuse to accept the law and it's consequences, then there is nothing I can do to change your mind.

You need to understand that these companies can't just lie due to legal reasons, so they try to deceive people by comparing apples to oranges, or cherry picking, but they can't just make up stuff. So extrapolating in a smart way does get you into the right range. Besides, Nvidia is extremely unlikely to release a slower product in the same tier, so an price/performance is pretty much guaranteed if the price does not increase for a tier. But if you want we can agree that I'm apologize to you if price/performance of the 5080 & 5070 (Ti) gets worse or remains the same (using MSRP pricing), and vice versa.

If you were happy with the G2, you should be happy with the Quest 3. The Somnium is enormously more expensive. The value for money is not really there, IMO, but it is better quality.

That notification refers to the issue with the 72 update, which has nothing to do with a broken USB link connection.

Sounds like a shorted cable somewhere. The Boosted Media review already noted that the thing is not really made to be repairable or moddable.

There is objectively a gain in price/performance of rasterization, without factoring in the fake frames, for all but the 5090 (the jury is still out on that one, but the price/performance improvement is probably going to be very small at best). I don't particularly like the fake frames, or at least how much it gets pushed as the new default, but with slowing progress in process nodes, you won't get the kind of rasterization performance improvements as in the past, if they would get rid of the fake frames. Ultimately, they cannot keep increasing the number of fake frames, so this is something that they can't keep doing.