Hacker News new | past | comments | ask | show | jobs | submit login

but not the best sellers, because devs have no incentive to release native linux builds

with that new console however it can change

valve has to make it compelling for them, by lowering their tax for example




Why on earth would you think trying to incentivise developers to release native Linux builds would be a more worthwhile investment than simply improving Proton? Linux desktop probably occupies about 2% of the steam market and that's being generous and assuming that their actual steam hardware surveys are underreporting adoption.

Even assuming wild success on this products part, that would only drive that number up a few percent since I don't see production for these running particularly high.

It might make more sense to encourage developers to provide proton support for anti cheat (like what Valve are actually doing leading up to this device's launch) rather than asking AAA developers to invest additional time on support that is arguably a waste of resources and certainly something of low priority.

Aside from that, my own experiences with Proton are far more positive than yours. I've hardly ever had to tweak things for it to work. Occasionally, I need to use an older proton version for a game but that's hardly a chore.


because that's the difference between:

- supporting a platform

and

- hopping everything is compatible

will they hook and reimplement EVERY WINDOWS APIs?, what about Windows 11? what about DX 12 and DX 12.1?

Oh, now imagine that windows 11 now requires X security features that are impossible to emulate with proton, how will that work?

native vs non-native

that's the difference

i'm not against Proton, i think it is a very nice idea to SUPPLEMENT my library

nothing can beat native games, built for the device and its constrains

and and an other WORSE thing

CONTROLLER SUPPORT in steam

valve never encouraged people to support controllers, WORSE, 99% of games expect you to use XBOX controller, as a result XBOX controller are displayed even if you use a DUALSHOCK one

valve could have provided APIs so they can feed games with proper icons etc etc, but nothing

so many wrong things, including the focus on proton as main selling point

that's part of the reason why you'll see majority of people install Windows instead, and "the year of linux on the desktop" will yet again be killed due to poor decisions

i'm kind of tired of trying to provide arguments, i reply to that kind of comment on a daily basis, it's painful to see linux people stuck with their WINE, it's WRONG, it's BAD, it's COUNTER PRODUCTIVE

microsoft helped a lot to drive linux adoption up with their crappy Vista/Win8/10 releases, but yeah, let's focus on emulating a bloated OS instead!

you don't come up with such a device over night, they had years to think about that, they focused on emulation

we will see who was right in the end, i think i am


"CONTROLLER SUPPORT in steam

valve never encouraged people to support controllers, WORSE, 99% of games expect you to use XBOX controller, as a result XBOX controller are displayed even if you use a DUALSHOCK one

valve could have provided APIs so they can feed games with proper icons etc etc, but nothin"

Out of all of your rather dubious points against Valve's plans, this is by far the most absurd. Have you heard of a little thing called Steam Input? Valve in fact have provided an API that does everything you've listed including feeding the "proper icons". It's up to the developers to use it. That would not change even if the game was 100% native because xinput is still the industry standard. Controller support on Steam is second to none.

"you don't come up with such a device over night, they had years to think about that, they focused on emulation

we will see who was right in the end, i think i am"

No, I think it's safe to conclude that you are wrong even from this date. You're asking Valve to compete with a platform that has all the games available. You're also assuming that native is always better than non-native but as I've already seen, a crappy native port is often beaten in both performance and feature set by Proton. Instead of asking developers to gamble away valuable time on the OS of a device that could be a flop, it makes a lot more sense to release the device with the games currently available in the market in mind and then give developers an incentive to add special support for a device that is actually out in the wild in the hands of paying customers.

You also really do not need to implement every Windows API, you only need to translate draw calls to Vulkan and then a small subset of API's that actually relate to games such as sound. Your example doesn't make a lot of sense, the only way a security API would matter for Proton is if the game is from the Windows store but most people do not in fact use the Windows Store even on Windows so games tend to get released on Steam as well - including even many Microsoft titles.

At the end of the day though, Valve's approach has made it so that I already do not miss Windows for gaming even with the current non-ideal situation with anti-cheat. If they adopted your approach, we would have likely had a small increase in the number of native games but most of the industry would have probably ignored such a call. It would also be very unlikely to magically make desktop Linux a real competitor. I think we need to stop chasing that dream because frankly, it just doesn't make a lot of sense to me. The only way you would be able to compete with Windows and Mac is by having a similar marketing and UX budget and by removing or downplaying many of the attributes of desktop Linux that likely drew us towards it in the first place - package management, hyper-extensibility, the terminal, GPL licensing, etc.

I do not think it's worth it frankly. There is nothing wrong with being "niche".




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: