Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

SteamOS

store.steampowered.com

Post reply on HN
Print view View original

291–300 of 532 posts

Re: SteamOS

#291

Post by simias »

asdasf wrote: post #262
errnoh wrote: post #217

Earlier quoted context omitted.

1080p video is nowhere near 1458 bytes. For raw 720p RGBA you'd send: 1280 * 720 * 32 = 29,491,200 bits but obviously you're not going to do that. Suddenly you're not only sending data but also encoding / decoding it.

>1080p video is nowhere near 1458 bytes. It doesn't matter, that's the point. Your MTU is almost certainly 1500 bytes. You are sending 1500 byte packets at the most. That does not cause latency. If you want to argue that we're incapable of encoding or decoding video with acceptable latency go right ahead, but doing it in response to me correcting a misconception about network latency doesn't make much sense.

Having worked on the particular issue of streaming real-time video over wireless, I can say that the main issue is not link latency.

The main lag comes from encoding/decoding. If you do it naively you encode frame-per-frame (encoding slices is more difficult), and the encoder does not only outputs iframes: you get partial frames that depend on both previous and future frames. Also the decoder does not always output frames in order. So you have to expect something around ~10 frames of latency, maybe less if you optimize everything well enough. That still means easily more than 100ms of lag.

The network is really not the main issue here.

Re: SteamOS

#292

Post by kllrnohj »

redthrowaway wrote: post #46
Glyptodon wrote: post #22

The streaming of other OS games from your other OS systems seems like a pretty compelling and surprising feature. I wonder if it will work with all games.

I can't see how they'd manage to get twitch games like online FPSes to be playable with the addition of another layer of latency between the player and the game. A 100ms delay between input and response is death for games like that.

Input latency is going to be Audio & Video, on the other hand, is more interesting. Let's assume everything is running at 60fps. Your encoder is therefore not adding any more than 16ms (if it was more, it wouldn't be keeping up and therefore not doing 60fps - it could definitely be less, though). Blu-ray is 48mbit/s, but for 24fps (youtube 's 1080p is ~8mbit for comparison). Let's take that same 48mbit/s, but adjust for 60fps instead. 1 frame would therefore be 0.8 mbit, or 100KB. If your network can do 48mbit/s, your video latency would therefore be 16ms. If your network is faster, though, the latency drops proportionally. Even a cell phone can do 100mbit/s on wifi, which would be 8ms of latency. A gigabit network would transfer that 0.8mbit needed in about a millisecond.

So your total latency of input, encoding, and transmission of the frame would be 32ms at most, and if you have a fast network & computer it could actually be down in the 5ms or less range.

Re: SteamOS

#293

Post by gavanwoolery »

rkalla wrote: post #26

This is effectively the software-only equivalent of a console launch. Is anyone else skeptical that they can motivate publishers to spend time/money porting their games to the Linux platform? Valve certainly has a better chance than most at pulling this off (and likely enough user/market data to make this seem like a valid investment) I am still super skeptical that these publishers are going to spend the time portin…

Ultimately, what other choice does Valve have? Every single other company is coming out with its own "App Store" - few are exclusive at this point, but the tide might turn that way (already I believe Microsoft and Apple have taken steps that have made it more difficult for Steam).

That said, OpenGL is an increasingly relevant player due to WebGL, mobile devices, Apple's growing market share, etc. It is also possible to simultaneously target OpenGL and DirectX with things like ANGLE. If your game is written in OpenGL, it is relatively easy to port to different operating systems, assuming you don't rely heavily on platform-specific libraries (and if you already support Mac, it is typically even easier to port to other *nix-based platforms). If the XBox One is OGL capable (via its Windows layer), porting could get even easier.

To top it off, I'm pretty sure it will get support from at least of the few big engines: ID Tech, Unreal, Unity, and CryEngine all seem to be potential candidates. 90+ percent of games use one of these engines.

Re: SteamOS

#294

Post by sefy98 »

rkalla wrote: post #188
jsnell wrote: post #65

Earlier quoted context omitted.

Well, there is some teasing along those lines there: > Watch for announcements in the coming weeks about all the AAA titles coming natively to SteamOS in 2014. But more importantly, control of Steam gives Valve a lot of leverage in this. It's easy to see the reasons for Linux support in indie games. But harder to see why mid-tier developers like Paradox or Eugen started to port their latest games to Linux. It's a goo…

Juho, great call-out with the "Steam-related carrot", I hadn't really thought down that path but you are absolutely right. Valve could easily offer big discounts in their split with the publishers for the first few years for Linux sales. My (possibly very incorrect) assumption was that the profit from the Linux sales were so low as to not matter, but maybe they can provide a catalog-wide carrot... something more sign…

This was my thought also. Offer $100 in games with the console, or release hl3 only on the console.

I'm really glad you guys pointed out this was a console launch because I tried to point it out on reddit and got shut down...

http://www.reddit.com/r/gaming/comments/1mywq0/steamos_annou...

Re: SteamOS

#295

Post by jon-wood »

rkalla wrote: post #26

This is effectively the software-only equivalent of a console launch. Is anyone else skeptical that they can motivate publishers to spend time/money porting their games to the Linux platform? Valve certainly has a better chance than most at pulling this off (and likely enough user/market data to make this seem like a valid investment) I am still super skeptical that these publishers are going to spend the time portin…

"In-home Streaming"

This one is going to be the killer I think - it means two things to me.

1. Not everything has to be ported to Linux straight out of the gate.

2. Its going straight on my living room machine! I've been waiting for the ability to use my beafy gaming PC (which is at a desk, with a keyboard and mouse) to drive games on the TV for a long time.

Re: SteamOS

#296

Post by giovannibajo1 »

Karunamon wrote: post #227
MrMember wrote: post #62

Earlier quoted context omitted.

It seems to work like the nVidia Shield's streaming. Another computer on your home network handles the heavy lifting and streams the video output to your SteamOS box.

This is genius. All you need now is a micro box running windows with appropriate video card that you can stick in the corner next to your router and never have to look at again. Assuming the streaming and UX works half as well as they're playing at, they've just managed to commoditize hardware compatibility. I don't think the importance of this can be understated.

You could do this already for years with Apple TV and Airplay mirror. The real problem is that of controls, mouse and keyboard don't match well with couch

Re: SteamOS

#297

Post by mrschwabe »

Karunamon wrote: post #227
MrMember wrote: post #62

Earlier quoted context omitted.

It seems to work like the nVidia Shield's streaming. Another computer on your home network handles the heavy lifting and streams the video output to your SteamOS box.

This is genius. All you need now is a micro box running windows with appropriate video card that you can stick in the corner next to your router and never have to look at again. Assuming the streaming and UX works half as well as they're playing at, they've just managed to commoditize hardware compatibility. I don't think the importance of this can be understated.

My living room is perfect use case for this. Has an older PC just for simple games. Now it could stream my entire Windows/Steam library of games, powered by the more beefy hardware in my home office rig.

As you say, if the streaming feature works well, this is a huge win. At the very least, it should save me from running 60+ feet of HDMI/USB cables which was my original plan.

Re: SteamOS

#298

Post by gcb0 »

makmanalp wrote: post #108
rkalla wrote: post #26

This is effectively the software-only equivalent of a console launch. Is anyone else skeptical that they can motivate publishers to spend time/money porting their games to the Linux platform? Valve certainly has a better chance than most at pulling this off (and likely enough user/market data to make this seem like a valid investment) I am still super skeptical that these publishers are going to spend the time portin…

The interesting thing is that convincing AAA publishers to port merely requires them to think it'd be profitable. 20 years ago, if someone asked to raise a quarter of a billion into a video game, they'd be laughed out of the building. Now it happens often. Over the years, increasingly big-budget games got funded and created. So, all they need to do is to replicate a similar process. First you move all the indie games…

Not so simple nowadays.

Games are not fire and forget anymore. Now it's all about downloadable content and In-Game-Currency (which is basically a way to allow gambling legally)

Just being profitable initially doesn't mean jack. I think everyone learned that from the emails from Jobs and the Printing industry (and what i'd give to read the ones with the music industry...)

Re: SteamOS

#299

Post by GauntletWizard »

speeder wrote: post #245
wonderzombie wrote: post #203

Earlier quoted context omitted.

If you're using an off-the-shelf engine (Unreal engine, say), it's probably somewhat easier, as some of the work would be one-off, at least in theory. But I don't honestly know how that breaks down.

Most off-the-shelf engine support Linux. This include all Quake engine variations (including Source, and GoldSource) and Unreal variations.

Don't forget Unity, and C#/XNA support via Mono/Monogame.
Post reply on HN