Aapje

AFAIK the general rule is that the product should perform well during the lifetime that is reasonable for the product. The difficulty is that the reasonable lifetime is dependent on the product category and the price class. Is it two years, three years, more? In any case, the easy part is that there is no malfunction in the actual headset that can be blamed on abuse by the user, but an end of the software support, so arguing that it will stop being usable (safely) at the end of 2025 is a slam dunk.

Same video ^

It's only going to get worse with hardware improvements getting harder and harder to come by. Also, you always had a choice to stick with older software, so I think that you don't actually mean that, because you could have chosen to forego the latest eye candy and other improvements, but you didn't make that choice. Your lamentation feels a bit like blaming the supermarket while you are stuffing your face with cake, shouting 'why are you making me eat this?!' While the reality is that they only offer the choice and it is up to you to take it or leave it.

You'll probably get upgraded to 24H2, so yes. I'd stick with Windows 10 and start planning to replace that G2.

I see that your disk is using the legacy MBR setup to store the boot data. This doesn't work with UEFI boot mode, so I would make sure that your BIOS boot mode is set to legacy. Secondly, you may have enabled soft raid in the BIOS in the past and perhaps turned it off by accident. If so, the system is only going to boot if soft-raid is enabled again (also in the BIOS).

No spring wobble and smoother use of the rudder. For heli flying, one can remove the spring and only use the damper, which means that you do have resistance, but no more return to center.

Apparently the ministick requires two connectors, while the MB1 spot on the old grip has only one connector.

Typically, some features don't work if you use a Virpil stick on a TM base, like the axis and some more advanced button options. But I don't really get why you would use that combo, since the Virpil bases seem superior.

Cinebench is more relevant to what are commonly called 'productivity applications,' and not so much to gaming. In games, the code that matters is nearly always the gameplay loop, which are the calculations that have to be done for every frame that is shown on the screen. This in large part uses the same data again and again, which is why the X3D-cache works so well for gaming, since the data that keeps getting reused is kept in this cache and so the CPU doesn't need to constantly retrieve it from RAM. The gameplay loop also pretty much never benefits from a huge amount of cores, since to get the calculation done quickly enough to get a high frame rate, you can't afford the latency costs of having a huge amount of cores work together. In gaming, the CPU also can't afford to work ahead too much, since each new frame needs to adapt to the user input, the behavior of enemies, etc. This means that lots of cores can't be used effectively, to work ahead. In 'productivity applications,' in many situations you have a lot of data that is not constantly reused for each calculation, so the X3D-cache doesn't help much. The data still needs to be retrieved from RAM, since mostly new data is needed for each calculation that is not present in the cache. Furthermore, these applications don't have to react to user input that much and can take a lot more time to do their work. For example, in Cinebench, making a single frame can take seconds. A game with a frame rate less than 1 FPS would be completely unplayable, so Cinebench is not representative of how games do things. So for productivity applications, you tend to want: lots of cores and lots of bandwidth to the RAM. The speed of the cores, a very big cache and low latency to the RAM are less important. For games, you tend to want: very fast cores, a big cache and low latency to RAM (although with a big enough cache this becomes much less important). The number of cores, and the bandwidth to the RAM are less important. So essentially, the strengths of the 9800X3D match the demands of gaming very well, but not so much the demands for productivity applications. The 285k is more tailored for productivity applications. If you want to have strong performance for both kinds of applications, then the 9950X3D is almost certainly going to be the best, although that is achieved by more or less turning it into a 9800X3D for games (where the chiplet without X3D is not used for the game).

Why are you looking at Cinebench when you actually want game performance? The reviews are clear. The 285k is beaten by both the 7800X3D and 9800X3D in games. If you are willing to spend a lot for relatively little gain, then I would wait for the stock situation of the 9800X3D to improve and to then buy that. Don't get the 9900X3D, because it's worse to have two 6 core chiplets, where one has the X3D-cache, rather than one single 8-core chiplet with X3D. You'll have more latency and there will be fewer cores on the X3D-chiplet. There is a chance that the 9950X3D will be slightly better than the 9800X3D, but probably not.

I think that a big issue is that they just picked popular Youtubers, even though many of them don't have the expertise or time to test it properly.

@Scott-S6 Moza obviously designed things with side-mounting as the goal, which you can also see in their desk mount. To be honest, I think that you are being a bit unreasonable in expecting near perfection from a company that is new to flight simming, and whose product is itself a very new product category (high Nm FFB bases). Right now I would characterize the AB9 as an early adopter product, but that a perfectly normal part of the development cycle of a new product category. Companies need to learn what consumers want, what the best technical solutions are, etc.

Using the MH16 is not a requirement, though. You can use the Gunfighter grip with the VPForce adapter. And if they fix the Virpil compatibility, then you could use that one.

Perhaps you should explain the joke. After all, jokes that need an explanation are the funniest ones.

It's fairly typical for up-and-coming nations to ignore IP rights, since they have little IP of their own, only for that to change once they become more innovative. The US used to violate European copyright constantly in the early days...

In your comment after this, you waxed poetic that with Virpil you get "a bunch of cams and springs for different feels and spring force can be easily adjusted in place from the top." But requiring a dismount and 'operation' to change these things is far inferior than just being able to tune them on them fly using software with a FFB base. It's their first flight sim product, so of course it was never going to be that great. They need to learn. And I don't agree that their sim racing hardware is mediocre and overpriced. Or at least, no more than the competition.

You will need a VPForce motor kit, a power supply and a grip, as well. Last I saw is a price of €399 for the DIY kit, and that should roughly double to get the full base. But that is without the grip, for which there are a variety of different options (at different prices).

@Nedum None of what you are saying actually disagrees with what I said.

How so? The entire reason why we have FFB bases now is because two dudes decided to take a gamble and make a FFB base despite not having a real company. They proved that there is a market for this product and further success of FFB bases almost certainly does not hinge on Moza. You can more justifiably argue that it's not in our interest to see people stop innovating in their shed because they are afraid that a big(ger) company will just steal what they made. Remember that Moza didn't innovate all that much, aside from making the base in a real factory, but a company like Winwing can do that too. And it looks like we will get bases from Winwing, Virpil and FliteSim, and the Rhino & FFBeast already exist. So it's not like we will have no options in the future if Moza pulls out. And it's certainly not a given that they will do so, rather than comply with the license or rewrite their software from scratch. Of course you are free to rationalize however you want, so you can square your morals with what you actually want. People do it often enough, and truly principled people are far and few between. It's really unfortunate to see all kinds of false allegations here. Again, I'm not going to sue Moza and I never said anyone should. It is the actions of Moza themselves, where they have most likely broken the law, that open them up to legal proceedings. And that in turn is a risk for consumers. Some people here seem to prefer that customers are misinformed. Now, I'm fine with people choosing to take these risks for themselves, but you cross a line with me if you want to prevent other consumers from knowing what they are getting into. You can just put a cover on it: https://qr4rigs.com/index.php?main_page=product_info&products_id=218

@Rosebud47 I won't sue them of course, because I am not their competitor. I also never said I will. I made it very clear that my argument is that there is a risk to customers that a competitor will sue and that this will impact customers.

Not just on a black screen. It's most obvious on any larger area of one color...like the sky.

End of 2025, although the X3D parts might come later. And note that the concept of putting the IO under the chip is speculation, not a leak. But AMD tends to move towards a goal gradually, and it makes perfect sense. AMD has also told us repeatedly that they can layer other things than just 3D-cache, which is a pretty strong hint. By then we might also see the CUDIMM-market mature, so we may be able to run faster memory as well.

@MAXsenna As Chinese companies have been generating more IP of themselves, they've started caring more about copyright and such. So one could sue them in China, but since they do business in the West, one could also sue them in the US and EU. They are extra vulnerable since their main business is racing gear, so they would obviously be far more likely to pull the AB9 from the market than to risk a total sales ban in the US/EU.

There are screens in the headset, that can have stuck or dead pixels. You should contact Pimax.

No, the issue is how they used it. There is nothing wrong with generating new software from scratch (other than the fact that AI tools can't actually write big programs on their own right now), but that is very different from creating a derived work by translating code to a different programming language. As I've explained before, doing this sort of stuff predates LLM AI's by a very long time and it has been deemed illegal to make derivative works. This has all been hashed out in the courts a long time ago. But I think that you have a fundamentally different point of view than me. You probably also were happy to see FC Technologies making an identical copy of a Virpil grip and wondering why people were upset at it. However, while you are entitled to your own morals, you don't get your own law. If someone sues Moza, the courts won't say: 'Rosebud47 disagrees with long established law, so we we will allow Moza to break the law'