Aapje

Just because it is written in a contract, doesn't mean that it actually happened. These kind of things are just a chore for everyone involved, and as long as there is no conflict, it is effort without a payoff. So it makes sense for ED's developers to have just ignored this, and the same for the Razbam developers. And once there was a conflict, it was of course no longer in Razbam's interest to share the source code, because it is leverage. I'm a programmer and I have to escrow some of our code, but for this we use a third party whose specific job it is to badger me to do it.

I do see a bunch of developers that make modules for DCS diversify into making MSFS modules as well. Whether that is due to lack of trust in ED or MSFS being a good opportunity we don't know.

Not sure why the discussion is about SLI. The video isn't about that. SLI is splitting the screen up in parts and then each card renders that part of the screen. The method from the video is frame generation. There one card renders the full frame, and then the other card generates the extra frames based of that.

It's the opposite. The yields of GB202 are so good that they don't have a lot of defective chips. Only the defective chips go into the 5090. The ones that are close to perfect go into the professional cards for way more money. When they used Samsung for the 30-series, the yields were so bad that they even sold extra cut down version of their top chip as the 3080. Of course, back then the miners bought a lot of them all up.

I agree with @Tensorial_Architect that if you are new to VR, you should weigh quality control, usability and how easy it is to get support very heavily. Also, if you are new you don't actually know what you really want, so it is better to start with something a bit more affordable, and to learn from that before you commit to something more expensive.

It just wants you to shake your booty, you see.

It's the first time they hand out the awards. It seems to be run by three European news sites (cruiselevel.de, FSElite and Threshold), but voting is by the community.

I think that pre-ordering gives you the opportunity to actually order/reserve early, but the pre-order is not the actual order.

Yes, but my point is that Russia never accepted this logic. The entire idea that both sides could agree on the value of targets is a rather ridiculous notion. In cultures where retaliation is used as a form of justice, you often just see escalation or cycles of violence, not an end to the conflict. It creates an even bigger incentive to never use them in the first place. And if there is no realistic way for Russia to keep up with the US in capability, then letting things rot away and spending the money on things that can actually be used makes a lot of sense. They do exist... I don't see why I should believe that the Western systems would be less effective than the Russian systems. At least the Western ones are being maintained, and get tested by people who don't just lie to their superiors that it works, when it doesn't. No, they thought they could win by making it too costly for the US, which is a big misjudgment of US mentality. You ignore that Hitler misjudged Chamberlain's appeachment as an unwillingness to wage war, when it actually was a ploy to buy time, thinking that he could take Poland, and they were thus drawn into the world war before they were ready. The declaration of war on the US was also a major misjudgment, with Germany thinking that the US would not scale up quickly, and that the subs could keep them from supplying the UK. Yet in reality, the US scaled up military production very quickly and by a huge amount. And although the submarine warfare was rather effective in the short term, it didn't do enough, and the US learned how to combat it. The treaty merely required them to declare war if Japan was attacked. Since Japan was the aggressor, they had no obligations. That is not how Russia interpreted it, at least according to their propaganda. Again, this brings it back to the infinite blackmail problem. If you leave an opening for blackmail, the other side can blackmail you with lies as well. And we called Russia's bluff of claiming that this is an attack by the West and hoping that we would call off the attack. The thing about bluffing about nuclear war, is that the consequences will be mutual. This just seems to be you agreeing that their bluff was getting called. Yes, so all we really need to do to prevent nuclear war with a somewhat rational entity is not to start a war of aggression with them, where we try to take their country, which is not something we want anyway. The real risks are irrational behavior and accidents. I think that due to an aging population with low fertility, the appetite for large scale war is very low, so realistically, we will only fight proxy wars, unless some dictator attacks us. You forgot Israel. And at the end of the day, it still doesn't protect against strikes, unless you are willing to respond to a non-existential attack by risking total destruction. So realistically, it merely raised the bar, and probably protects against an all-out attack on your country.

https://fselite.net/content/flight-sim-awards-2025-the-winners/

Except the goal is not to kill as many people as possible, but to hit specific threats first and foremost. And because many targets are very important to destroy, the plans would actually call for multiple nukes on the same target, to increase the chance of a kill. If the reliability is truly as low as that, and the Russians know it, they will have to use a lot of nukes per target, to boost the chance of successful destruction to high levels. This is actually one of the reasons why these arsenals are so big in the first place. Even with fairly high reliability, you need multiple nukes per target to have high certainty of a kill. And note that a potential 10% success rate certainly doesn't mean that you can expect 10% of the 4300 Russian nukes to actually end up destroying a target. You also have 'counter-battery' fire (nuke facilities would be a primary target in a nuclear war, for obvious reasons), as well as self-defense systems. Depending on the scenario, many of the nukes may never make it out of storage. Basically, any quick escalation would be likely to result in that outcome. Wrong. We already did when Russia threatened a nuclear exchange if their territory would be attacked, including the newly captured territories. Yet the West allowed Ukraine to conquer part of Kursk. We called their bluff. Ultimately, you cannot allow yourself to be blackmailed, so aggressive bluffs need to be called, or else you will lose everything anyway. Giving the enemy the ability to delude themselves into thinking that they have a shot is very dangerous. Japan went to war because of that delusion. Germany went to war in WW I based on that delusion. Germany went to war in WW 2 based on that delusion. The anti-nuclear treaties had openness at the core, in part to avoid misjudgments and misunderstandings. There were a bunch of theorists who came up with deterrence theory who were sure that the Russians would think the same way as them, but each time Westerners met with Soviet leaders and explained their deterrence theory, the Soviets made it clear that they didn't share these beliefs. Note that the deterrence beliefs include intentional weakness (not preparing for survivability). Basically, the population is offered up as hostages, without actually knowing whether that weakness reduces or increases the risk. After all, blackmail only works if there is a credible threat you can make. And blackmail is a very dangerous game, where both sides may end up with an outcome neither wants, because both sides feel that they need to defend their credibility by escalating. You should read 'The Great American Gamble: Deterrence Theory and Practice from the Cold War to the Twenty-First Century' if you want to learn about deterrence theory and its weaknesses. It also allows you to move past the kind of stupidity you read in the newspapers, because journalists tend to be rather dumb people who are also very uneducated, especially in matter of war. For example, I've seen journalists claim multiple times that the nuclear treaties led to disarmament, when it merely requires looking at a graph of the nuclear stockpiles over time, to see that (Soviet) disarmament began before the treaties were signed.

@Dragon1-1 All the major nations have tritium boosted nukes, which require a lot less fissile material for the same yield. This is why one avenue that anti-nuclear weapons activists are exploring is to try to ban tritium production, so the nukes become paperweights. Russian commanders are known for not doing maintenance if it is not easily discovered, and pocketing that money. And then you also have Russia funneling their military spending mostly on things that are visible and things that they can sell, to put up a good front and to preserve their status as a major weapons supplier. On both counts, the nukes are pretty much a perfect place to cut spending or pocket the money, since who is ever going to notice that most nukes won't work? If anyone ever tries pressing the big red button, the people who could persecute the military commanders, are probably more concerned with survival. So my guess is that only a few nukes get proper maintenance, and if Russia attempts to do a large scale launch, the dud rate will be enormous. I would not be surprised if less than 10% actually function. And I think that if the US gets serious about producing military goods and converts car plants and other factories into military factories, they will easily outproduce Russia. Don't forget that it was not just the West that deindustrialized, Russia has as well. Also on this front, China is by far the bigger threat, since China can actually produce in volume. @Rick50 Yeah, I also have my doubts to what extent the remaining Russian vehicles can still be refurbished without replacing a gazillion parts. But can you please refrain from making your messages bold?

The 9800X3D is currently at MSRP, and availability seems good now. The question is when the price will start dropping significantly. The issue is that Intel is not very competitive, so we might not see prices drop that fast.

Even back in WW 2, lots of weapons systems got produced for a long time, and got upgrades along the way. See the Bf 109, which was a 10-year old airplane design in 1945. And lots of tanks got more and more armor bolted on, and cannons replaced. Nowadays, there is relatively little progress in the actual flight stuff. So key for a modern airplane is the ability to keep upgrading the systems along the way. Only when that becomes hard to do, and you see stuff like sensor pods on the outside of the aircraft, I think that there is a really good reason to move to new different platform. Besides, the complexity of modern airplanes is so high that you can't even keep replacing them, because it takes many years to get everything (mostly) working correctly. And keep in mind that everything points to Russia doing a poor job at maintaining their stuff. So their big stocks seem to in large part just parts kits, that need a big overhaul to really be ready for war. In a conflict between Russia and the US, I see Russia struggling to overhaul their old stuff to even be somewhat competitive with what the US already has, and the US will then also outproduce Russia with much better weapon systems. And the Kuznetsov is just scrap metal. Even in a nuclear war, I have my doubts that Russia can actually deploy a meaningful amount of working nukes, given the apparent lack of tritium and other maintenance. But still, at least in theory this is the only real way that they can hit very hard. China is very different. And I agree that a mix of high and low is probably the best.

And the active armor sold and replaced by bricks. Even if they would put new T-90's in storage, I wonder how many parts would be stripped and sold, and how much maintenance left undone. They claim to develop cutting edge tech, but a lot of it seems to be very exaggerated or straight up lies. And like you say, they make it in such small numbers that most of their nicer stuff was off the battle field very quickly. And with such little actual usage, there is a lack of feedback, which is crucial to learn what actually works. I also think that the US is way, way, way, way better positioned than Russia to take advantage of the drone revolution. The US already invested heavily in systems integration. The obvious next step is that the drones will detect targets and local commanders can immediately verify the finding and use the appropriate weapons system to take out the target. Local decision making has historically been very advantageous with regard to reaction time, so you can actually take advantage of opportunities and not allow the enemy to prepare. The downside is a lack of coordination with other units. Strong systems integration can give you the best of both worlds. In contrast, Russia has a top-down culture, and a lack of trust in the chain of command, and between units. So aside from probably being unable to develop this technology at the same level, I don't think that they can use it effectively. And the US has Palmer Lucky, who has a combination of insight and force of will to make things happen. And the US is taking advantage of that with ABMS and whatever the project is called that they do for SOCOM. And with advanced systems integration, you actually need less munitions, because if the drone sends the exact coordinates directly to the artillery gun(s), it can be a one salvo or even one shot kill most of the time. It seems obvious to me that the US has written off Russia as a true conventional threat and only sees China as such a threat. That is why they try to starve China of high-end computing, which the US seems to consider crucial for next gen weaponry.

MRTV and his pal seem very happy with the current state of the Super:

I do think that in a full on war between the US and one of the potential adversaries, the US would benefit a lot from being able to take a lot of high quality planes, ships, etc out of the running. Even back in WW 2, it did take quite a long time to produce ships. Of the 19 Essex-class carriers ordered after the US declaration of war, only 2 saw active service. Having too few missiles for the planes you have, and aircraft-carriers that can't do a lot of sorties due to ammo shortages, is still a lot better than not having those planes and carriers at all. It's still a lot easier to produce ammunition than weapon systems. But I agree that the obsession with low cost and efficiency is not a good idea. Ideally, you have munitions factories that are too large for what they produce, and have too many machines (and spares for those machines), and those are constantly running at a level that keeps a solid workforce at work, who can then be used to train up people for the excess capacity relatively quickly if need be. Yeah, but the Soviet war machine has mostly stagnated since the cold war ended. They just kept the big old stocks, but don't seem to have been adding meaningfully to that anymore (especially if you factor in them selling stuff off to other nations), and develop relatively little new technology. Technology still matters, and the bigger the technology gap, the bigger the numbers have to be to make up for it. And satellite photo analysis suggests that they've been depleting a lot of their old Soviet stock of vehicles. The only way for them to have a big superiority in numbers is for them to keep production relatively high for quite a while, and to start stocking up a lot. But after the current conflict is over, is that likely to happen? I think not. PS. Don't forget about corruption either.

In that case it is indeed possible to find such a performance difference, but also utterly meaningless for gaming with a discrete GPU.

That's not what the other person said, nor is it true. I think that you are confused. The Bigscreen Beyond is currently a SteamVR headset, and does not natively support OpenXR. However, SteamVR is an OpenXR runtime, so the games can talk to SteamVR with OpenXR, but then SteamVR talks to the Beyond with their own protocol, not with OpenXR. You can compare it to buying something in China, where you (= the game in this analogy) speak only English and the vendor (= the headset) only speaks Chinese. Then you need an English - Chinese translator to be able to communicate. If you bring a translator, you can talk to the vendor, but that doesn't mean that the vendor speaks English natively. And similarly, you can't claim that the Beyond natively supports OpenXR, when SteamVR is translating the OpenXR API that the game is using, to the SteamVR API when communicating with the headset. I just saw that Bigscreen is claiming to develop an OpenXR integration, so they seem to be doing exactly what I am saying they should do: "Planned development Foveated rendering for improved VR performance and integrations with SteamVR and OpenXR are under development in 2025." This text comes from the homepage of the Bigscreen website. So clearly their engineers agree with me That's just your opinion. It took them quite a while to optimize SteamVR and for a long time people were extremely happy that they could use mbucchia's OpenVR to OpenXR translation software to greatly improve their performance, by getting rid of SteamVR. Even today you still gain some FPS if you don't use SteamVR. If those games were built for the OpenXR-standard in the first place, instead of the SteamVR-specific API, it would be far easier to switch runtime, less buggy due to not requiring a translation layer, and wouldn't be dependent on the very hard and unpaid work by mbucchia. But with the Bigscreen Beyond you are locked into SteamVR right now, so you simply don't have any other option. You are completely ignoring one of the main points I made, which is that a dependency on SteamVR means that you are at the mercy of whatever Valve does in the future. Even if you are happy with what Valve did in the past, that provides absolutely no guarantee for the future. Reverb G2 buyers also didn't imagine that the WMR-runtime would be removed with no alternative, yet it happened.

The benefit is that you are not at the mercy of support by an individual vendor. Look at what happened to the WMR headsets, once MS lost interest. Bye bye WMR. Who knows if Valve is going to stay interested? Perhaps the Valve Deckard will be a flop and they'll just give up on VR or do the very minimum to keep existing things going, but not improving things. That is the risk that you are taking by getting a headset that is linked to a company. It's pretty clear that the market is moving to the sensible choice: OpenXR. That way, the software and hardware is not dependent on a single company that may lose interest, or may want to abuse their power, but they base their API on the work of a neutral consortium that is not dependent on a single company. In the chain of OpenXR Headset - OpenXR Runtime - OpenXR Game Engine VR driver, you can swap out any of these elements without impacting the other ones. And this in turn keeps the companies honest, since they can't just get away with crappy quality because people are bound to their software. You say that you are happy with the performance of SteamVR today, but that wasn't the case in the past. What if it becomes bad again? What if SteamVR is slow to implement a feature that is present in another runtime? What if Valve gets into a conflict with Bigscreen and 'accidentally' breaks the integration with the Beyond? What if Valve decides that they want to be a console company, and restricts the API to their own hardware? In any case, it seems pretty clear to me that Valve lost a lot of money on VR and the only reason why they stick with it is because Gabe believes in it. But he's 62. That's an age where a lot of risks start to increase. The risk of him retiring. Health issues. Age related mental decline. Etc. Gabe is obviously a very good person, and someone who (currently) believes in VR, but what happens when someone else becomes in charge of Valve? Even a very decent successor who merely decides to cut the unprofitable parts of Valve may be the end of SteamVR.

Bigscreen needs to switch to OpenXR. That is the future and foveated rendering is part of that standard, so they can then implement it in a way that works for everything.

That's obviously jargon. Corporate "around the end of March" means 'April 31st' in normal language.

Why pick that picture when there is a car that came with a joystick from the factory? https://old.reddit.com/r/acecombat/comments/bbmjhp/flashback_to_the_time_saab_thougt_it_was_a_good/

That's not great, since with the shifter function, some people may want to switch between a joystick and shifter regularly.

Yeah, it looks like a perfect storm of immense hubris, failure to execute and AMD closing the gap (FSR 4 mainly). If one of these things wouldn't have been true, customers would probably not think of switching en mass.