Rich Geldreich On The State Of Linux Gaming

[smitty3268]

That's a bit far fetched, but he does seem to be a bit bitter about Valve.
It's probably that not everyone fits into Valves corporate cultue. They say they have a flat hierarchy, but that's just the formal hierarchy. The informal hierarchy is probably not flat, since humans are humans.

https://twitter.com/richgel999/statu...22992704577538

I think the "flat" heirarchy probably results in a lot of office politics, with people teaming up and having each other's backs.

That's probably something certain people really like and others really hate.

[madjr]

my projections: STEAM OS will be kept in dev indefinetly for anyone or any relative small company who wants to build their own Steam box(es)


However. SteamOS is still important.

The Alienware Alpha is $550, meaning that up to $100 may be from including windows...

With SteamOS you can slash that price down and compete with consoles. THIS IS THE WHOLE GOAL: TO KEEP *these pc component* CONSOLES FROM EATING THE WHOLE PIE and GETTING ALL THE EXCLUSIVES. The notion that PCs can also be used on the couch and get more gamepad oriented games.

Microsoft doesnt want to compete with it's own console, so I don't think they will slash the price of their OS too much.

SteamOS/linux-machines are not ready yet. So maybe in a couple of years when the "Steampad" is ready and prices has gone down and Linux has 1.5 to 2k games (has like 700 now).

Also Valve will need to create some "beta or time exclusives" (3 to 6 months) for those that invested in their SteamOS-machines.

Also STREAMING COULD BECOME BIG THING, if they make some type of "cheap Amazon fire-stick-tv" like Streamer. A sub $100 device that Streams the bigger 3d titles and plays natively the indie/older ones.

So yeah that's just my projections, I think it could happen if they're creative, competitive and stay interested in it.

[smitty3268]

So this comment on his blog was very intriguing

FWIW - while at Valve I seriously supported open source Linux GL drivers and I got seriously burned by doing so. There are some powerful concerns out there that are not yet on board with the idea of open source GPU drivers. It cost us quite some time.

Wonder who those "powerful concerns" are? NVidia and AMD? People at Valve? Game devs?

[johnc]

Wonder who those "powerful concerns" are? NVidia and AMD? People at Valve? Game devs?

Based on his other blog post it's almost certain he's referring to driver devs, most likely at nvidia. After all, that's their bread right there. They won't be too eager to help make their jobs redundant.

[tarceri]

You sound like some of the driver writers I've interacted with. The phrase "I don't care" came up a lot. There's a huge cultural mismatch between driver developers and game developers. Most game developers don't do this purely for the paycheck. We are not all drones that sit and peck away for mere dollars. If you don't engage developers at a deeper level, and the bottom line doesn't make any sense to the suits and business types with the spreadsheets, then the effort is in trouble.

Maybe you should subscribe to the Mesa mailing list or at least watch some presentations from the Intel open source devs [1] if you want to understand where the open source driver devs are coming from. Its pretty clear that most of them are not just doing this work for a paycheck, the fact is they have been playing catch up for years just trying to get Mesa up to par with features, its only recently that performance has been able to be really focused on. I've contributed to a number of open source projects and Mesa is by far one of the best communities I've been involved in.

[1] https://www.youtube.com/channel/UCXl...EhjzLFTUTm_U7g

[prodigy_]

Individual devs at Valve don't always choose the right things to work on, especially during bonus season.

Valve did enough just by releasing quality ports and thus proving it's both possible and feasible. Must they also fix drivers for AMD? I think it's something we cannot reasonably expect. AMD should have dropped Catalyst for Linux long ago because it never really worked. And they could more actively help the community with the open-source RadeonSI (I believe it's what they're going to do now but they sure took their time before making this decision).

[Michael_S]

Linux gaming has a long way to go but at the same time things are better than ever. To be honest five years ago nobody believed we'd be where we are now.

Thanks for pointing that out. I had been so pessimistic that I lost sight of it, and I do think it's true.

Valve does tend to take the long view for critical projects, and I wouldn't be surprised if the wizards over there figure out other ways to continue enticing devs over to Linux. Assuming GabeN still sees the thing as a critical project worthy of dumping funds into with no immediate/obvious returns. I'm hoping they keep the fire lit under MS personally.

I really like their move to make the complete set of Linux games on Steam free to all Debian developers. I thought it was a great move. I'm not sure if it will have an impact, but it was, at least in my view, creative.

I think it is important to remember that this is all part of being a Developer, though. A dev's job is to get the artists' creative ideas to display on as many devices as possible, with the best fidelity possible. Alternatively we'll just go back to the "everyone buy a PS4" hardware model.

In reality, no-one other than devs care about what devs have go through. To be fair - I don't care what the mechanic who fixed my car went through either!

I don't think anyone is advocating that. They're just pointing out that getting good graphics and 30 or 60 fps without crashing from a poor driver is tremendously difficult.

[richgel999]

Maybe you should subscribe to the Mesa mailing list or at least watch some presentations from the Intel open source devs [1] if you want to understand where the open source driver devs are coming from. Its pretty clear that most of them are not just doing this work for a paycheck, the fact is they have been playing catch up for years just trying to get Mesa up to par with features, its only recently that performance has been able to be really focused on. I've contributed to a number of open source projects and Mesa is by far one of the best communities I've been involved in.

[1] https://www.youtube.com/channel/UCXl...EhjzLFTUTm_U7g

You know, I should have totally qualified my statement about driver devs. The open source driver devs (from Intel) where a pleasure to work with. They were very open to feedback and it was obvious they were passionate about their driver. The closed source devs where a completely different bunch altogether.

[DMJC]

Honestly I think this is all a load of shit. Mainly because as I've stated before, the driver progression is screaming along.
http://nouveau.freedesktop.org/wiki/FeatureMatrix/ - Look at the feature matrix. Seriously. There are 4 main components of the Nouveau Driver remaining to be completed:
OpenCL (dead in the water right now), Power Management, SLI, and Tesselation support. Tesselation isn't even needed until you look at OpenGL 4.3+ We have about 4 OpenGL tasks left until OpenGL 4.2 drops. Realistically we will see 4.3 by the middle of 2015. Now look at the power management subpage from the feature matrix: http://nouveau.freedesktop.org/wiki/PowerManagement/ We see that for the current gen cards, 5/10 are MOSTLY, 2/10 are DONE, 1/10 WIP and 2/10 TODO. Every other card of consequence has more dones/mostly and maybe 1-2 WIP and 0 TODOs So realistically the Power Management status should be set to MOSTLY not WIP. Nouveau is under-reporting progress on Power management. I predict this: Linux 3.19-Linux 3.21. Nouveau will finish Power Management on Nvidia cards upto 700GTX models, possibly 900 GTX series. OpenGL 4.3 will land on Linux by mid 2015, and Nouveau will be a viable drop-in replacement for the Nvidia binaries for most gaming. AMD is still bumbling along with their crap-ass cards that aren't designed for GL and don't work well with it, but even they will see some gains coming from the code sharing with AMD/Intel/Nouveau. Intel is doing surprisingly well. OpenGL 3.3 will be working on Sandy Bridge by middle of 2015, and OpenGL 4.3 will be working on Intel hardware capable of supporting it by then. I don't think this guy is being realistic in his assessment of Linux and OpenGL and I think he's shilling for ATI/AMD. ATI/AMD are not necessary for Linux to have a strong gaming segment. What's needed is a lot more pro-active activism and coddling upto NVIDIA. We need to get NVIDIA to start cross promotion. We should be seeing a lot more works best on Intel/Nvidia/Nouveau/OpenRADEON signage/news/information. If AMD aren't going to step up and actually deliver a working product, then we should be promoting the alternatives. Why should gamers have to settle for inferior hardware on their platform of choice? If AMD is not delivering the performance but NVIDIA and Intel are delivering performance/features. They should be celebrated and rightly so. I'm sick of seeing these debates framed as if AMD is on a level field with NVIDIA. They aren't, they've had over 15 years to prove they were serious about Unix. They aren't. Nvidia works on FreeBSD/Solaris/Linux/MacOSX. AMD just tries to throw their own crappy APIs at people because they can't cut it in OpenGL. It's time the community got vocal about it and started to show people why AMD is a crap brand.

[DMJC]

On Steambox:

I reckon steambox hasn't come out because AMD's driver performance has been so rubbish. That Valve initially thought they could flood the market with cheap AMD gaming PCs in living room form factor. Now they're finding out oh crap, the drivers just can't do the job and they're either scrambling to get the drivers fixed, or they're trying to get their hands on low-cost nvidia parts. Nvidia is probably very reluctant to move back into the living room after they got burned so badly on the original Xbox, and are quite content to stick to their two segments of portable gaming and big desktop cards.