MudMan

There are definitely ways to send backwards compatible data when required and separately support additional features in a new iteration of an API. This wouldn’t be in the top 10 backwards compatibility challenges MS has figured out.

But in any case, I don’t care if they call it XInput2 or Game Input. I just need it to support all controller features in all games. It’s a bit hard to tell whether Game Input will ever do that, but so far it seems more concerned with acting as a layer to explicitly support a bunch of different hardware, each with its own standards, than a XInput replacement for controllers. There doesn’t seem to be a concept for a “Game Input controller” there at all, actually, just supported controllers you can listen for regardless of what they’re sending through.

I guess over time if they stick with it and it does end up working as a Steam Input-style intermediary layer that just recognizes anything you’d just ship controllers that match whatever format with gyro support and Game Input-enabled games would just pick them up fine more or less universally, but that doesn’t seem to be what it does right now, or at least not something that either games or manufacturers are relying upon.

Anyway, this was interesting and informative, but I think I’m good now. I definitely don’t want to have a conversation formatted as an argument in which nobody is disagreeing with anybody else. Those are exhausting.

That’s interesting, but considering this note:

We recommend the GameInput API for all new code, regardless of the target platform, because it provides support across all Microsoft platforms (including earlier versions of Windows) and provides superior performance versus legacy APIs.

For games developed on the GDK for Xbox One, GameInput is the only input API

I’m really not sure this would do what we both want it to do. If everybody has had a GameInput version of their controller support since last-gen and we’re still getting limited to the XInput feature set I don’t think it sorts out gyro-on-Xinput at all. I am not familiar with the behind the scenes of how modern engine controller code is handled, but this sounds like maybe it’s how games with native PS controller support are doing that, but not necessarily a new standard that will allow the default XInput PC setting of new controllers to pass gyro input to games detecting them as an XInput device. I think it’s more like MS’s answer to Steam Input as an additional layer between the games and the hardware, regardless of what the hardware is using.

It does show that all the tools are in place. MS has control over all the involved APIs. They could expand the Xbox controller API feature set tomorrow, whether or not they add the hardware feature to their base controller model. They just… don’t. And Steam could deploy a Steam-independent Steam Input driver or software to just take over all controller support on a dedicated full-feature OS layer, but they also don’t (on either Windows or Linux, as far as I can tell).

Honestly, there are enough workarounds (add games as non-Steam games, use Switch modes and so on), I just bump against the edge cases of it often because I’m both a controller and handheld nerd, so I’m stuck with a GPD Win handheld that insists on injecting their internal gyro as mouse inputs, which confuses the hell out of half the games, along with a bunch of GameSir and Gullikit controllers that do weird things with gyro, like injecting it at the firmware level instead of passing it to the OS. And I mess around with enough emulators to also end up with “oh, this was on DI mode when I booted RetroArch, so now all my buttons are in the wrong places until I quit”. It’s only dumb for like ten of us… but man, is it dumb.

Yes, I’m aware, that’s why I’m calling out it’s weird that XInput doesn’t support gyro, because we’re a long way away of it being just based on Xbox controller support and a whole bunch of other controllers with a whole bunch of other features now go through it. If MS doesn’t want to add gyro that’s up to them, but Windows supporting it natively is way overdue. Of course at that point older controllers would probably need a firmware update, but hey, we’ll cross that bridge when we get to it.

In practice the situation we’re having is games are defaulting to Xinput and relying on Steam Input as an intermediate layer for additional features, so the end result is that gyro is… not NOT supported, but often not acknowledged at all, so you end up with a bunch of situations where you have to config gyro manually per game as a bit of a Steam-level hack, and then your controller is all wonky anywhere other than Steam because the way Switch/DI/PS input modes get picked up in non-Steam stuff can be weird.

And it gets worse in handhelds where you’re absolutely at the mercy of how the manufacturer decided to set up their controller and gyro support, and sometimes need to do a lot of weird stuff to pass it on outside of Steam.

It’s the jankiest part of controller set up left on PC gaming, and it’s all down to this weird “mom and dad aren’t talking” dance where MS keeps pretending PC controllers are fundamentally Xbox controllers at the XInput layer and Steam is the de facto curator of the controller support but has no interest (and to be frank no expectation or need) to have their controller layer work outside their launcher.

It’s hard to fault Steam’s controller layer, but I really wish they finally found a way to parse gyro data from third party controllers without having to run them on Switch mode. That goes for Microsoft and their own drivers. At this point it’s weird to keep pretending the Windows controller APIs are supposed to work on their first party Xbox controllers only.

I’m almost entirely sure that PS4 and XOne controllers did get upgrades at some points. Definitely Switch 1 ones, which matters or not depending on how you split the gens. There were definitely revisions in older controllers, though. Some were labeled and had obvious new features, some were quieter. And PC-side drivers got updates all the time, obviously.

Also, your current gen controller will also keep working indefinitely without an update. In this case Valve is annoyed about a particular dependency where THEY need the upgrade to happen for a feature compatibility thing, but the controller proper will work if you plug it in.

Well, I guess we’re only going to invent the microchip once.

In any case now that I’m… you know, not a kid and can look at 90s tech with some historical data, it wasn’t quite as straightforward as we remember. But then it also didn’t have memory prices spiking by 300% overnight because everybody knows a fake AI bubble is going to pop so they want to sell them a bunch of crap before all the money evaporates but not ramp up production because that money is clearly going to evaporate.

That’s 21st century through and through. 20th century tech would just keep slapping the same Z80 and Motorola 68000 chips on things for a decade. Z80 too slow as your CPU? Maybe it can be your sound chip now. Listen, we have warehouses of these, we’ll keep selling you new ones well into the 2000s, it’s barely a clump of wires held together with chewing gum.

Did you know the 68000 was 10 years old by the time it made it to the Genesis/Megadrive as a CPU? That’d be like the PS5 launching with a single core 2GHz Athlon from 2005. I didn’t know that as a kid.

Hah. Shocked that you didn’t just take the easy way out and wrote “salt to taste”.

…

Please do not use recipes from AI summaries without clicking through to the correct amounts.

I mean, hell, I won’t use a recipe I find on a recipe site without cross-referencing at least a couple other recipes from other sites, because some of those are janky as hell. But the best-guess of a summarization LLM on what all those numbers and ratios are meant to be? Yeah, no. Not by itself at all, unless you’re just trying to jog your memory on something you already know and can recognize the correct values if you see them.

Man, that’s a big lateral move. Honestly, my experience is spotty as well (using mostly Bazzite), and I’m sticking to dual booting for the foreseeable. I definitely would not change hardware for this reason. Especially because, having some installs on AMD-based hardware, my expectation that the issues would drop to zero is… very low.

If you can’t try before you buy I’d only swap to get a meaningful upgrade. Of course when a “meaningful upgrade” will be affordable or accessible in the future is anybody’s guess. Because isn’t tech in the 21st century fun?

Riiiiiiiight.

Man, social media is a trip.

What the hell just happened there? That was some weird breakdown.

I mean, I guess you didn’t catch the post of the person explaining why Diablo doesn’t work properly (Fedi threading issues, presumably?) but maaaaaan, that’s some off the rails unprompted weirdness.

The actual answer is below, but I’m constantly surprised about Lutris being proposed to GoG as opposed to Heroic.

I mean, Lutris is more flexible and you can build more stuff into it, but it’s extremely fiddly and not as well supported. As long as you only care about GoG or Epic, Heroic is just hands-off and Steam-like. I actually prefer it to Galaxy on Windows.

Yeah, but nobody would argue that GameStop was dying in 2002, which is seven years into GameFAQs existing and very much the heyday of Prima and other dedicated print guide writers. Seriously, it just doesn’t line up. GameFAQs and print guides were servicing the same need.

Again, I’m not saying it didn’t have an impact. I’m saying if Prima guides existed as standalone publications in dedicated gaming stores it’s partly because GameFAQs had killed monthly print magazines as a viable way to acquire strategy guides for games, so you instead had dedicated guide publishers working directly with devs and game publishers to have print guides ready to go at day one, sometimes shipping directly bundled with the game.

And then you had an army of crowdsourcer guide writers online that were catching up to those print products almost immediately but offering something very different (namely a searchable text-only lightweight doc different from the high quality art-heavy print guides).

Those were both an alternative to how this worked in the 90s, which was by print magazines with no online competition deciding which game to feature with a map, guide or tricks and every now and then publishing a garbage compilation on toilet paper pulp they could bundle with a mag. I still have some of those crappy early guides. GameFAQs and collectible print guides are both counters to that filling two solutions to the equation and they both share a similar curve in time, from the Internet getting big and killing mag cheats to the enshittified Internet replacing text guides with video walkthroughs and paid editorial digital guides made in bulk.

Well, I’d argue if there was no money to be made, then CNET wouldn’t have purchased GameFAQs. At the very least it served to bring people over to their media ecosystem, and I wanna say they did serve ads and affiliate links on the site proper (but adblocker is also old, so it’s hard to tell).

Video contributed, for sure. This is a process of many years, the whole thing was evolving at once. But the clean break idea that print guides existed and then GameFAQs came along and killed guides just doesn’t fit the timeline at all. It’s off by 5-10 years, at least. Guides weren’t residual in the 00s when GameFAQs was at its peak and being bought as a company, they were doing alright. It’d take 10 years longer for them to struggle and 15 for them to disappear. You’re two console gens off there. That’s a lot. If guide makers like Prima were pivoting to collectible high end books out of desperation you’d expect that process to have failed faster than that.

Instead they failed at the same time GameFAQs started to struggle and get superseded, so I’m more inclined to read that as them both being part of the same thing and the whole thing struggling together as we move towards video on media and digital on game publishing. That fits the timeline better, I think.

In any case, it was what it was, and it’s more enshittified now. I’ve been looking up a couple details on Blake Manor (which is good but buggy and flaky in pieces, so you may need some help even if you don’t want to spoil yourself) and all you get is Steam forums and a couple of hard to navigate pages. The print guide/GameFAQs era was harder to search but more convenient, for sure.

It’s not a “even if some existed” thing, Prima operated until 2018. I personally remember preorder bundles with Prima guides for 360 era games and beyond. They published incredibly elaborate collector’s hardbook guides (that honestly doubled as artbooks) for stuff like Twilight Princess and Halo 3, all the way to the PS4 gen.

Even granting that “booming” is probably a bit hyperbolic, if GameFAQs being free in 1995 was going to kill them, bleeding out would probably not have taken 23 years. The death of retail, print and physical games probably hurt print guides way more than GameFAQs ever did. You didn’t buy those because you were in a hurry to solve a puzzle or look up a special move. They were collectibles and art books first and foremost.

FWIW, guides going back to paid professionals wasn’t as much due to video. Video is still crowdsourced for that stuff. It was visual guides in html with a bunch of images and reference, I think. At least that’s what IGN was doing, and they’re the ones that went hard on that front first. Also for the record, that probably had something to do with IGN and GameFAQs being affiliated for a while. GameFAQs was bought off by CNET in '03, it was definitely part of the big online gaming press ecosystem. I can see how IGN thought they could do better.

I don’t know that the timeline works out there. GameFAQs is, as this post reminds us, pretty old. Even assuming that it didn’t break out until the very late 90s or early 00s as THE destination for guides, there was certainly a booming editoral market for highly produced guides all the way into the Xbox 360 era.

I’d say it was responsible for the press not focusing on guides as much and instead refocusing on news and reviews. And then news and reviews died out and the press that was left refocused on guides again because by that point the text-only crowdsourced output of GameFAQs was less interesting than the more fully produced, visually-driven guides in professional outlets. And now… well, who knows, it’s a mess now. Mostly Reddit, I suppose?

I mean… MK1 predates it by what? 3-4 years? Which in 90s tech time is an eternity.

MK fatality guides were mostly in print. Magazines were all over that type of stuff at the time. But it wouldn’t have been strange to get a familiarly formatted ASCII guide for them with, say, your pirated floppies of the DOS or Amiga versions.

Hm… I’m a bit mixed on that, because GameFAQs became relevant a bit later than that, but at the same time that type of format for ASCII game guides predates GameFAQs being the main place you went to get them, so… it evens out?

I probably didn’t start going to GameFAQs for this stuff until like 2000, but I certainly was using text guides for games in the 90s.

Eden/Azahar seem to be the current meta replacement for the newer Nintendo platforms. I wonder how much of that is a crowdfunded assessment of quality and how much is a result of modern emu communities being driven by preinstalled bundles in launchers, handhelds and install scripts (think ES-DE standards or Retrodeck/Emudeck defaults).

Either way, hey, they work fine, so… yay?

I mean, convenience is a factor.

And while Steam doesn’t typically sign exclusive stuff they are known to use store positioning as a bargaining chip for preferential treatment. You’d think Konami would be above needing that, but who knows.

Anyway, good game, whatever the reason for the delay. Someone who is on the fence about getting it on Steam go get it on GOG instead to make up for them tricking me.