Platform Discussion Thread

+

Which API do you think CP 2077 will use?


  • Total voters
    135
Yeah, I believe in might well do that. Up to a point, at least.

My opinion is just that "computing power" hasn't really been a real issue for a long while anymore, but it's rather made to be that by excessively focusing on everything superficial in games that makes them more resource intensive, which in turn - I would guess as a layman - creates a snowball effect with everything else the game has to offer and roll on the screen. So in that sense I don't see much point with this kind of thing.
Yeah, computing power often seems to translate to "prettier graphics" rather than "better gameplay systems."
 
If worst comes to worst, eventually, the only hardware-based gaming that will still be left is PCs, and who knows - perhaps developers will decide that our audience just isn't worth catering to anymore. I don't know if things will go that far, but it will certainly mean big changes for the PC industry one way or another.

Believe me, i'm a PC gamer, i don't own a console, i spent a good portion of my savings to buy a high end PC, i get it.

IEven the most hardcore of PC players will have to give themselves a pretty damn good reason not to spend, say, $15 a month to play all the games they could ever want at any FPS and resolution they want instead of shelling out hundreds of dollars upgrading their PCs.

My opinion is just that "computing power" hasn't really been a real issue for a long while anymore, but it's rather made to be that by excessively focusing on everything superficial in games that makes them more resource intensive, which in turn - I would guess as a layman - creates a snowball effect with everything else the game has to offer and roll on the screen. So in that sense I don't see much point with this kind of thing.
But, let me ask you this.. IF you could choose between a game that benefits from the computing power of multiple high-end pc's and one that dosen't, one that is confined to what you can shove in a box, console or otherwise, what would you pick? And when i say computing power, thinking "graphics" is undestandable, but.. narrow. Physics, AI, graphics, everything in a game makes use of it, the more complex and intertwining with the rest of it's systems, the more it uses. It's the next logical step, it's our Oasis. Not Stadia, not yet.. but it's one way of getting there.

That's my take on it. Oasis or not, it can help game development with how much it could do for the developers, for their capabilities of translating their vision as close as possible to the game, whithout having to trade-off one thing or another by the platform limitations, what it could do for the time it takes for a game to go from concept to playable with the tools Google can place at their disposal (Deep Mind). It could be SO much more than just another PS or XboX. Or not and i'm making a fool out of myself but hey, a man can dream right?
 
Believe me, i'm a PC gamer, i don't own a console, i spent a good portion of my savings to buy a high end PC, i get it.




But, let me ask you this.. IF you could choose between a game that benefits from the computing power of multiple high-end pc's and one that dosen't, one that is confined to what you can shove in a box, console or otherwise, what would you pick? And when i say computing power, thinking "graphics" is undestandable, but.. narrow. Physics, AI, graphics, everything in a game makes use of it, the more complex and intertwining with the rest of it's systems, the more it uses. It's the next logical step, it's our Oasis. Not Stadia, not yet.. but it's one way of getting there.

That's my take on it. Oasis or not, it can help game development with how much it could do for the developers, for their capabilities of translating their vision as close as possible to the game, whithout having to trade-off one thing or another by the platform limitations, what it could do for the time it takes for a game to go from concept to playable with the tools Google can place at their disposal (Deep Mind). It could be SO much more than just another PS or XboX. Or not and i'm making a fool out of myself but hey, a man can dream right?
Okay, fair enough. From a developer's perspective, I can see the potential.

I still think developers will focus way more on graphics than they need to, but you make some good points.

From a player perspective, my concerns still stand, but thanks for broadening my view a bit.
 
it all depends on the internet infrastructure way too much, you get a dodgy bit of intermediate hardware you input lag suddeny hits 500ms and the game becomes unplayable. and you will hit dodgy intermediate hardware, you can't not, is why the performance of MMOs which are asking a lot less can go totally to shit for no apparent reason.

Streaming anything interactive isn't the future, unless they are about to rebuild the internet.
 
only limited by bandwidth ...
... what that tells me is it's ONLINE gaming. Online gaming, for me, personally, is a non-starter. What ruins a game for me is OTHER PEOPLE. I'm firmly in the offline, single-player, local machine installed and run gaming camp.

One reason I'm in this camp is that I'm rather fond of modding. I'm never really completely satisfied with any studio's "finished" work, and the customization, as well as additional features and changes that modding allows opens up a whole new level feature rich content that I'm always happy to see whether take advantage of and use what's on offer, or not.

Online games, on the other hand, require an online connection, and are slaves to the control of whoever's hosting them, plus, there's usually 100 dozen ritalin soaked 12 year olds running around, screaming into their headsets and I'm like all sorts of NOPE on that.

Maybe I'm a dinosaur suck on the old paradigm of single player, local machine, offline play, but, I suspect i'm not the only one.
 
Personally, since I'm not fond at all to DRM and exclusively-online services could be considered by default DRM (a passive layer of DRM one could say), not to mention the unreliability of streaming services a la Netflix (games licenses could expire for any reason and, thus, be suddenly removed from the platform for good), Stadia has no appeal to me.
 
To me what will make or break it is how much this all ends up costing. First there's the bandwith. A lot of people have a data cap that if you go over you get hit with fines. So that's bad right off the bat. The second thing.. will we be paying for a service that allows you to stream any game or will we still need to purchase the games? What rights will we have to those games if any? (can we mod them, can we cheat in single player games?) And how will game publishers be making their money and what's there to motivate them to put their games on the platform?

Like any of these types of things I'm a wait and see guy. I am at least looking forward to see how it all works.
 
I see it as as having positive side effects for Linux gaming, since Stadia will prompt more developers port their games to Linux and use Vulkan (if you followed the presentation and read the technical details, Stadia uses Debian Linux for their backend and Vulkan as their primary graphics API, however they require their own specialized SDK for handling input and such).

I'm not planning to use Stadia personally, so for me, the most interesting question is, how hard would it be for developers to make proper desktop Linux releases after releasing for Stadia. Something like adding support for SDL would be required.

Also, would be cool if CDPR could release CP2077 on Stadia, following proper desktop Linux release once they run it on Linux already.
 
I see it as as having positive side effects for Linux gaming, since Stadia will prompt more developers port their games to Linux and use Vulkan (if you followed the presentation and read the technical details, Stadia uses Debian Linux for their backend and Vulkan as their primary graphics API).

I'm not planning to use Satdia personally, so for me, the most interesting question is, how hard would it be for developers to make proper Linux releases after releasing for Stadia. Something like adding support for SDL would be required.

Also, would be cool if CDPR would relese CP2077 on Stadia, following proper desktop Linux release once they run it on Linux already.
Gilrond, I love you. You never stop preaching the good word of Linux, no matter what.

I need to go back and find your argument with Sard from years ago... I should bookmark that...

EDIT: Ahhh, this is where it all began... https://forums.cdprojektred.com/index.php?threads/platform-discussion-thread.110083/page-2 It was, of course, a different thread at the time and I remember begging Sard not to merge it purely to preserve this discussion. But oh well.

EDIT 2: I'm tagging @Sardukhar for nostalgia's sake. Sorry, buddy, please don't hate me.
 
Last edited:
Gilrond, I love you. You never stop preaching the good word of Linux, no matter what.

I need to go back and find your argument with Sard from years ago... I should bookmark that...

Heh. I think Stadia can be quite a big deal for Linux gaming. That is, if Google are going to sustain it, and it won't end up like many other Google projects, abandoned after some time, even though they had some success. I wouldn't count that out, knowing Google's history. They launch stuff with fanfare, and after some time just silently shut things down. If this succeeds however, this will have a real incentive for developers to make Linux games (with the catch that it won't be exactly desktop Linux versions, but the major parts like Vulkan renderer will be there).

I.e. the amazing thing about that is, that it jumps over the desktop market share issue, and simply says: release for Linux, and anyone with the browser will be able to use it. If that works, amount of studios using Vulkan and Linux will baloon. So it's likely some of that will spill out to regular desktop Linux gaming, so on one hand I don't want this to become the expected norm and the only venue of release, but on the other hand I wish their project would be significant enough to push developers to release for Linux.
 
Heh. I think Stadia can be quite a big deal for Linux gaming. That is, if Google are going to sustain it, and it won't end up like many other Google projects, abandoned after some time, even though they had some success. I wouldn't count that out, knowing Google's history.
Yep! Google does this all the time. Hangouts, Google+, etc.

Google is a bit like an easily-distracted puppy - always chasing a new noise, or the next shiny thing.

However, this is apparently going to be a much more open development environment, which could mean it doesn't even need Google's avid support at all to thrive on its own - and, at the very least, if this proves successful, we will see competition; possibly competition that cares more about its services than Google does.

I'll admit I'm not well educated on the subject of Vulkan, Wine, or Linux development in general - all I know is what you've wrote here over the years, which is a lot, but obviously not quite a college-level run-down of this stuff. So, I'll take your word for it that its good for Linux gaming.

Forgive me if this is an obvious question, but lets say CDPR is developing their game without Vulkan, and they later want to release it on Stradia.

Is the port over just automatic? Do they have to do some funky behind-the-scenes work to make it function with Vulcan? Does Google do the heavy lifting?

Also, what about Vulkan makes it ideal for Linux development, and what is the alternative that ISN'T good for linux development (you've explained this particular point before, but my memory is poor)?
 
From some additional information including from Google insiders, it seems that so far Google are focusing on native Linux games, so not on Wine. You can also notice that CodeWeavers (who are the primary sponsors and backers of Wine development) aren't listed as partners for Stadia project. So I think it's enough of an indicator of where it's going, at least for now. I.e. to release for Stadia, Google will pitch developers to release for Linux natively and to use Vulkan (that's big I'd say, if Google can convince many developers to do that).

If some game / engine has no Vulkan support yet, adding it is quite far from automatic. Depending on the engine, it can be even quite difficult and it clearly requires good level of graphics programming expertise. So if let's say CDPR never had that yet, they'll have to find (hire?) experts. Whether Google provide assistance - no idea. I know some companies like AMD do it (AMD are partners with Stadia). It's quite likely that they did that exactly for this project by the way. Here is one example:


AMD engineer Adam Sawicki worked with some big gaming company (not named) to port their engine and game to Vulkan. I wouldn't be surprised, if AMD and that company did that having Stadia in mind, since Google could reach out to many companies in advance about this, which would make a lot of sense.

Regarding why Vulkan is beneficial. It's a bit hard to explain without going into much detail, but let's just say that Vulkan gives programmers more control over usage of the GPU, unlike higher level APIs like OpenGL. I.e. they can tailor their game in a way that saturates GPU efficiently utilizing parallelism and memory management approach of their own making. To put it differently, if they have good experts, they can make very efficient games that use hardware to its full potential, thus providing very well performing games, which fully scale according to existing resources (like CPU cores / GPU pipelines and amount of RAM). Without Vulkan that's just not possible.
Post automatically merged:

Also, interesting to note that Google deliberately went with using AMD and the fully open graphics stack (from the Linux kernel to amdvlk - AMD's own open Vulkan driver). That avoids the need to pay any licenses to Microsoft or Nvidia (since Nvidia would most likely require paying them for using their graphics blob driver on Linux for commercial purposes).
 
Last edited:
From some additional information including from Google insiders, it seems that so far Google are focusing on native Linux games, so not on Wine. You can also notice that CodeWeavers (who are the primary sponsors and backers of Wine development) aren't listed as partners for Stadia project. So I think it's enough of an indicator of where it's going, at least for now. I.e. to release for Stadia, Google will pitch developers to release for Linux natively and to use Vulkan (that's big I'd say, if Google can convince many developers to do that).

If some game / engine has no Vulkan support yet, adding it is quite far from automatic. Depending on the engine, it can be even quite difficult and it clearly requires good level of graphics programming expertise. So if let's say CDPR never had that yet, they'll have to find (hire?) experts. Whether Google provide assistance - no idea. I know some companies like AMD do it (AMD are partners with Stadia). It's quite likely that they did that exactly for this project by the way. Here is one example:


AMD engineer Adam Sawicki worked with some big gaming company (not named) to port their engine and game to Vulkan. I wouldn't be surprised, if AMD and that company did that having Stadia in mind, since Google could reach out to many companies in advance about this, which would make a lot of sense.

Regarding why Vulkan is beneficial. It's a bit hard to explain without going into much detail, but let's just say that Vulkan gives programmers more control over usage of the GPU, unlike higher level APIs like OpenGL. I.e. they can tailor their game in a way that saturates GPU efficiently utilizing parallelism and memory management approach of their own making. To put it differently, if they have good experts, they can make very efficient games that use hardware to its full potential, thus providing very well performing games, which fully scale according to existing resources (like CPU cores / GPU pipelines and amount of RAM). Without Vulkan that's just not possible.
Post automatically merged:

Also, interesting to note that Google deliberately went with using AMD and the fully open graphics stack (from the Linux kernel to amdvlk - AMD's own open Vulkan driver). That avoids the need to pay any licenses to Microsoft or Nvidia (since Nvidia would most likely require paying them for using their graphics blob driver on Linux for commercial purposes).
Well explained! Thanks for the information. Had no idea Vulkan was actually better for performance, just assumed its main purpose was for Linux compatibility.
 
Well, Vulkan doesn't mean automatically better performance, but it gives game programmers enough control, to make the game perform as much as it can on given hardware, which they can't do using higher level APIs. I.e. the final performance depends on how well the game code is designed when using Vulkan.
 
it all depends on the internet infrastructure way too much, you get a dodgy bit of intermediate hardware you input lag suddeny hits 500ms and the game becomes unplayable. and you will hit dodgy intermediate hardware, you can't not, is why the performance of MMOs which are asking a lot less can go totally to shit for no apparent reason.

Streaming anything interactive isn't the future, unless they are about to rebuild the internet.

Indeed.

They do touch a bit on that at around the 39 and 49 min mark, i'm no expert but they do seem to adress something very close to what you wrote.
Post automatically merged:

only limited by bandwidth ...
... what that tells me is it's ONLINE gaming. Online gaming, for me, personally, is a non-starter. What ruins a game for me is OTHER PEOPLE. I'm firmly in the offline, single-player, local machine installed and run gaming camp.

Nope, Stadia has nothing to do with ONLINE gaming, not in the 12 year olds screaming in the background, nope, thank god for that.

What it means it that it utilizes Google's vast network of nodes and data centres to deliver a streaming quality and computing power to it's users unattainable by it's competition or users in general.

One reason I'm in this camp is that I'm rather fond of modding. I'm never really completely satisfied with any studio's "finished" work, and the customization, as well as additional features and changes that modding allows opens up a whole new level feature rich content that I'm always happy to see whether take advantage of and use what's on offer, or not.

The platform dosen't automatically exclude modding, if the amount of flexibility promised to content creators on platforms such as youtube is an indication, if anything it facilitates modding through the tools they will allegedly provide. How will that compare to current PC modding, i can only guess.

Online games, on the other hand, require an online connection

Well, yes, in a way, no more that 99% of games today. Even retail "bought" games NEED an online platform to "activate" even run them. But nowadays who dosen't have an internet connection? It's mandatory, especially in the way games are released these days, more like pushed out the door, gamebreaking bugs? sure, poor optimisation? why not? cut and add later? it must be so.. release first, finish later. Don't have an internet connection? don't use our platform? well have fun playing then.
 
Last edited:
Indeed.

They do touch a bit on that at around the 39 and 49 min mark, i'm no expert but they do seem to adress something very close to what you wrote.
Post automatically merged:



Nope, Stadia has nothing to do with ONLINE gaming, not in the 12 year olds screaming in the background, nope, thank god for that.

What it means it that it utilizes Google's vast network of nodes and data centres to deliver a streaming quality and computing power to it's users unattainable by it's competition or users in general.



The platform dosen't automatically exclude modding, if the amount of flexibility promised to content creators on platforms such as youtube is an indication, if anything it facilitates modding through the tools they will allegedly provide. How will that compare to current PC modding, i can only guess.



Well, yes, in a way, no more that 99% of games today. Even retail "bought" games NEED an online platform to "activate" even run them. But nowadays who dosen't have an internet connection? It's mandatory, especially in the way games are released these days, more like pushed out the door, gamebreaking bugs? sure, poor optimisation? why not? cut and add later? it must be so.. release first, finish later. Don't have an internet connection? don't use our platform? well have fun playing then.
I can play all of my (singleplayer) Steam games without an internet connection, and I have over 270 at this point.

Let's not normalize always-online connections for singleplayer games. It's (one of) the reason(s) SimCity was so widely criticized.
 
One issue this is clearly missing. Not everyone wants to rent games. We should be able to buy them too. Not only preserving them is a lot more probable when they are DRM-free, when we run games on our own hardware, we can decide what performance and settings to run them with. And not some service decides for us.
 
Stadia is Nexon on Steroids? Time couldnt be any better as EA is making changes with Origin. WoW things are getting interesting especially Amazon is enterting into the ring too.

Holy Cow, Google’s Stadia will let you jump into a game in seconds straight from YouTube
 
Last edited:
I can play all of my (singleplayer) Steam games without an internet connection, and I have over 270 at this point.

Let's not normalize always-online connections for singleplayer games. It's (one of) the reason(s) SimCity was so widely criticized.

You still don't own them, you are lincensing them, you (we) have been for quite a few years now. And to activate that licence you need an internet connection. Granted not all the time, but if internet is all you NEED? I mean i'm fine spending 2000 bucks every few years, deal with all sorts of hardware - software incompatibilities AND deal with the whole platform-exclusive-day one day two day three day #@! patch while HOPING games will somehow magically go back to the model from years back. It's not going to happen. GPU's are becoming more and more expensive, their time-performance efficiency becoming more and more reduced. It's fine for a hobby but don't fool yourself thinking it's the best option (or will be for much more) when talking games for anything except indies (but more are making their way to the consoles).

One issue this is clearly missing. Not everyone wants to rent games. We should be able to buy them too. Not only preserving them is a lot more probable when they are DRM-free, when we run games on our own hardware, we can decide what performance and settings to run them with. And not some service decides for us.

I don't think ANYONE would want to rent games for the same price you can buy others, and yet, we see it more and more. But thinking of the online part of the platform to just "Always online" like a limitation, is understandable, since up till now most games that did that did it for dubious reasons, but it could be a lot more, like the youtube integration example from the presentation. It's really up to the developers. I'm sure there will be corporate greed, shady practices, and everything else the industry is plagued by. But then there's studios like CDPR or Techland or 4A that just might make something truly special that would otherwise not be possible.

I'm pretty sure choosing how the game looks is not exactly removed from the table. The way i see it you can have even greater freedom in doing so with the platform since, like i said, you're not confined to what you can "shove in a box".

Just to be clear, i don't think, nor desire Stadia to replace anything. All i'm saying is to approach this concept as open minded as possible, see it for what it is, what it could be. If it's the thing that's going to shake the PC industry to it's core, fine! It's about time something does, maybe it's going to turn out for the better, maybe it's the final blow. Out with the old in with the new.
 
Last edited:
Top Bottom