Are Valve And AMD About To Ruin PC Gaming?

Yes they do, and now they have Mantle, a more efficient (in theory) low level api specifically for the hardware... read what I referenced. Slow dow before you respond please.
 
Kenjiwing said:
Dude its the real world.. the pc market is incredibly small now. Its a be happy with what you can get kind of scenario now.
Click to expand...

We might have to redefine "incredibly small" ... EA in FY13 made a little over 1.1 Billion on Xbox, around 1.1 Billion on PS3 and a little over 900 million on PC ... mobile and handhelds only made them around 400 million ... That was only one gaming company and there are still plenty of companies that make PC exclusive titles (especially independents) and Steam (a PC specific platform) is estimated to have somewhere between 30 and 50 million user accounts ... and PCs remain as ubiquitous as microwave ovens and TVs (much more so than consoles) ... PCs may be suffering a little from lack of the "killer app" or a victim of their own productivity but the total potential market for a good game remains in the billions (hardly incredibly small) ;)
 
Kai said:
Yes they do, and now they have Mantle, a more efficient (in theory) low level api specifically for the hardware... read what I referenced. Slow dow before you respond please.
Click to expand...

*sigh* read the article yourself. Mantle Benefits: "Leveraging optimizations from the next gen consoles to pc's". Please stop..it's a PC API.
 
It's been suggested by media outlets that the X1 supports a Mantle-like API, but no details have been confirmed.
 
I can't put it together for you. What kind of hardware do the new consoles use? Pc compatible hardware. What are these console games being developed on? PC's. The new consoles have Graphics Core Next. The new consoles are x86. Mantle is specifically for Graphics Core Next. Put two and two together. Holy shit they cam use Mantle or a version (already used for xbone) to make games? Giving better access to the specific hardware? Mantle (or derivitives) happens to be the dev kit given out by AMD...
 
They dont need to be confirmed. It makes logical sense to provide the tools for the hardware to create the best games possible. Mantle.
 
Holy Christ, somebody else want to take this...its like talking to an uninformed wall.
 
Read perspective 2 in the link provided. It requires a little reading comprehension and critical thinking. But I think its spelled out rather nicely. Its not explicit, but read it. It akes good sense.
 
pcjunkie said:
Holy Christ, somebody else want to take this...its like talking to an uninformed wall.
Click to expand...

um no its cross platform as much as i hate the idea
it IS a way to leverage the console in AMDs favor on the PC which is why i dont like it
it makes it very tempting to make mantle only games
 
It's hard to say how much of this Mantle will make it into the consoles since right now it's just an EA exclusive and AMD won't be making the SDK available for awhile (2014-2015), and I assume other devs are already making games for XB1 and PS4. So I think the AnandTech speculation might be off a bit.

And it's nice to hear that AMD is "looking" to make their OpenGL drivers better, but what kind of resource allocation are they actually doing over there? I suspect they have far, far more DirectX developers over there than OpenGL.
 
Considering how many PC Gamers/Geeks I know who say they would have switched to Linux years ago If it wasn't for games I think that MS is the only big looser here..
 
Someone mentioned here the other day that AMD had a driver development staff of three. I don't know if that's accurate or not. To be fair to AMD, though, driver development is not an area that's particularly human-scalable. Throwing more people on such projects doesn't yield tremendous payoffs. You need exceptionally proficient people, not large numbers of them.

What both AMD and NVIDIA should be doing is making larger financial and human investments in OpenGL-related tooling and educational resources. Drivers are important, but it's more important that developers have the resources they need to build robust and efficient OpenGL applications. OpenGL is currently uncompetitive with D3D in this respect. Extensions and themselves can wait.
 
Ok, I'll lay it out for both you. AMD has dev optimizations for both Xbone and PS4. They took some of the same optimizations from consoles (that were closer to hardware by bypassing some higher level DX functions, not sure about OpenGL?) to decrease draw calls amongst other things to the pc side. There is no "Mantle" developer kit or api for consoles. The pc api of these optimizations is called "Mantle". Hope that helps...
 
kbrickley said:
We might have to redefine "incredibly small" ... EA in FY13 made a little over 1.1 Billion on Xbox, around 1.1 Billion on PS3 and a little over 900 million on PC ... mobile and handhelds only made them around 400 million ... That was only one gaming company and there are still plenty of companies that make PC exclusive titles (especially independents) and Steam (a PC specific platform) is estimated to have somewhere between 30 and 50 million user accounts ... and PCs remain as ubiquitous as microwave ovens and TVs (much more so than consoles) ... PCs may be suffering a little from lack of the "killer app" or a victim of their own productivity but the total potential market for a good game remains in the billions (hardly incredibly small) ;)
Click to expand...

Yeah, PC Gaming is bigger than it has ever been.

True, it's a smaller slice of the overall gaming pie, but in absolute figures, PC gaming has never been bigger.
 
pcjunkie said:
Holy Christ, somebody else want to take this...its like talking to an uninformed wall.
Click to expand...

So you're using an opinion piece full of pc gaming FUD for your reference on what Mantle is?

How about reading up a little?

From HardOCP:
Mantle allows "bare metal access" to the GPU. This means that Mantle is able to allow developers to work directly to core features on the GPU at levels that are not possible under DirectX. This is how games are already developed on console systems. Developers can access core GPU functions at a much lower level. This is why consoles can perform as good as those can or look as good as those can with hardware that is light-years behind the PC. Now that AMD has been developing this low-level support on its silicon on the console systems, AMD can bring this to PC GPUs.
Click to expand...

http://www.hardocp.com/article/2013/10/07/amd_radeon_r9_r7_series_rebranding_introduction/2


From Anadtech:
At the same time, when it comes to writing APIs we also have to briefly mention the fact that unlike the PC world, the use of both high level and low level APIs are a common occurrence in console software. High level APIs are still easier to use, but when you’re working with a fixed platform with a long shelf life, low level APIs not only become practical, they become essential to extracting the maximum performance out of a piece of hardware. As good as a memory manager or a state manager is, if you know your code inside and out then there are numerous shortcuts and optimizations that are opened up by going low level, and these are matters that hardcore console developers will chase in full. So when we talk about AMD writing APIs for the new consoles, we’re really talking about AMD writing two APIs for the new consoles: a high level API, equivalent to the likes of Direct3D and OpenGL, and a low level API suitable for banging on the hardware directly for maximum performance.
Click to expand...

http://www.anandtech.com/show/7371/understanding-amds-mantle-a-lowlevel-graphics-api-for-gcn
 
wonderfield said:
Someone mentioned here the other day that AMD had a driver development staff of three. I don't know if that's accurate or not. To be fair to AMD, though, driver development is not an area that's particularly human-scalable. Throwing more people on such projects doesn't yield tremendous payoffs. You need exceptionally proficient people, not large numbers of them..
Click to expand...

You can't make a baby in one month with 9 women... :p

Not everything scales well. Software development tends not to, in many cases.

There's a good book on the suject:
http://en.wikipedia.org/wiki/The_Mythical_Man-Month
 
kbrickley said:
We might have to redefine "incredibly small" ... EA in FY13 made a little over 1.1 Billion on Xbox, around 1.1 Billion on PS3 and a little over 900 million on PC ...
Click to expand...

And Valve's annual revenue is in the billion+ to multi-billion range and they've been reportedly experiencing near 100% growth year over year for the past several. Nothing to sneeze at.
 
22_crunkaintdead_lg.jpg
 
Two things:

1. Mantle is only a HUGE deal if Direct3D is woefully inefficient. If it is doing a reasonably good job, and insulating the developer from a lot of low-level stuff that is time consuming and possibly not stable across hardware revisions, then it's worth a _modest_ performance hit. It seems there is some worry that Microsoft has not been as aggressive as it should have been on optimizing D3D. In that case, Mantle could be a kind of wake up call. If MS's API is not good enough, the industry will make a better one, etc...

Notice - this isn't a threat to OpenGL. It's used in consoles, tablets, phones, Windows, Mac, and Linux. It is pretty well optimized, with OGL versions of some games running faster than their D3D counterparts. OpenGL has a bigger reach than Mantle can since it runs on such diverse hardware.

2. SteamOS is only a HUGE deal if Windows becomes gamer-unfriendly. If MS decides it wants to lock out their distribution competitors (since now that it has a built-in store, MS is now competing with Valve and EA), they have nowhere to go. If Windows9 cripples gaming to drive more customers to the Xbone, too bad for everyone. SteamOS is more of a long-term play. It's a way to standardize a Linux gaming experience and give the vendors a push to get/keep their Linux drivers in good shape. People who wish to game on a Ubuntu box will just run Steam for Linux and have access to those games anyway, but this way the game developers don't have to worry about supporting all the fragmentation. They get a stable target from Valve, and let the fragmentation sort itself out on the margins.

Both of these technologies are about keeping Microsoft hungry for their PC gaming customers. AMD has to tread lightly since it is a major hardware vendor for the Xbone. But Valve doesn't need MS's (or anyone's) permission to push for gaming-targeted improvements to Linux.

Bottom line in my eyes: Mantle is important for a year or two. The industry responds one of two ways. 1) It sees how far D3D performance has fallen relative to the hardware, and MS responds by optimizing it. Or 2) If SteamOS takes off and MS doesn't make D3D better, PC gaming now has a place to fall - Linux/OpenGL for the enthusiasts, Windows/OpenGL for more titles but at a performance cost, and Windows+Mantle for some exclusive console ports.
 
One of the developer criticisms of D3D is that it's tied to OS version. So for awhile there you had 50% of people on Windows XP and thus stuck with D3D9. I think that caused some developer frustration.
 
I don't think Mantle or SteamOS will get very far. Mantle will get a few Frostbite engine game optimizations but apart from a handful of games, it will be a Gaming Evolved API where AMD has to pay the developer off to use it. That means NVIDIA could easily develop their own TWIMTBP API and pay developers off to do the same. And between NVIDIA and AMD, NVIDIA is clearly better off financially to do so.

As for SteamOS, streaming Windows games has always been a failure, just ask all the operators of those streaming services that have come and gone. PC gamers want to minimize input lag, not raise it. Plus with a compressed stream, you lose image quality which is unacceptable. SteamOS is completely pointless and will fizzle away into obscurity.
 
jwcalla said:
One of the developer criticisms of D3D is that it's tied to OS version. So for awhile there you had 50% of people on Windows XP and thus stuck with D3D9. I think that caused some developer frustration.
Click to expand...

DX being held hostage by Microsoft behind the Windows version paywall is ridiculous and the practice continues. It hurts everyone and holds graphics back because creating an artificial barrier, fragmenting the renderer among different versions of Windows and in turn discouraging most developers from embracing the latest/greatest (DICE is kind of an exception) who instead shoot lower for the lowest common denominator to ensure the largest pool of people can buy their game.

A case in point, Battlefield 4: the developer does its part embracing the latest/greatest tech in DX11.1 but most Windows users (meaning Windows 7, which outnumbers Windows 8 users by more than 3-to-1) end up getting crapped on and can't take advantage of the improvements.

This absurd practice is one of the main reasons I hope SteamOS takes off and we take graphic improvements out of the hands of one corporation that's more focused on tablets and phones these days anyway.
 
5150Joker said:
I don't think Mantle or SteamOS will get very far. Mantle will get a few Frostbite engine game optimizations but apart from a handful of games, it will be a Gaming Evolved API where AMD has to pay the developer off to use it. That means NVIDIA could easily develop their own TWIMTBP API and pay developers off to do the same. And between NVIDIA and AMD, NVIDIA is clearly better off financially to do so.

As for SteamOS, streaming Windows games has always been a failure, just ask all the operators of those streaming services that have come and gone. PC gamers want to minimize input lag, not raise it. Plus with a compressed stream, you lose image quality which is unacceptable. SteamOS is completely pointless and will fizzle away into obscurity.
Click to expand...

SteamOS isn't just for streaming. That seems to be a way to break into the living room more effectively. It will run its own native games - the ones on Steam Linux. The streaming feature allows people who have Windows games to use their cool living-room interface while leaving their Windows box at the desk.

I think you're mostly right about Mantle having a limited audience. If you want to make a cross-platform game from scratch, you'll likely be coding to OpenGL anyway. But if you're a middleware or engine developer (Frostbite, idTech, Unreal, Unigine, etc...) then you'll support that rendering path since you basically have to support ALL render paths all the time. But since you'll already be doing so for the consoles, it won't be a big deal to add it in to the PC versions of the engine.
 
5150Joker said:
As for SteamOS, streaming Windows games has always been a failure, just ask all the operators of those streaming services that have come and gone. PC gamers want to minimize input lag, not raise it. Plus with a compressed stream, you lose image quality which is unacceptable. SteamOS is completely pointless and will fizzle away into obscurity.
Click to expand...

No offense but you might want to do some actual reading on SteamOS and what it aims to do because you come off as a little clueless here citing online streaming services.
 
Morphish said:
It seems there is some worry that Microsoft has not been as aggressive as it should have been on optimizing D3D. In that case, Mantle could be a kind of wake up call. If MS's API is not good enough, the industry will make a better one, etc...
Click to expand...
It's not the case that D3D isn't well-optimized (they've reduced draw call overhead itself by an order of magnitude since D3D9), but it's a relatively thick abstraction. Creating resources, for example, happens at an extremely high level: you create an API resource; the resource is allocated by the driver through the API call through the kernel; the driver hands back a pointer to the API; the API gives the user a handle to it. Mantle probably just gives you some sort of opaque pointer type to memory addresses and lets users fill them with data arbitrarily, which would let you iterate over containers of resources with pointers and access resources through a simple dereferencing operation. Comparatively, that's low-level but fast, and possible only because Mantle would understand GCN and its memory model.

Direct3D is really just bound by a lack of architectural commonality between different vendors and by Microsoft's release cadence. There's no common 'instruction set', tile and swizzle formats or memory model between vendors, so D3D has to abstract implementation details away at a high level. OpenGL is in a bit of a better position due to its open extension system, whereby vendors can provide arbitrary functionality through their own sets of function pointers which can be invoked (almost) directly, but vendor-specific extensions are seldom used to their full effect as they entail writing multiple paths through a renderer to provide fallback functionality. D3D provides a set of functionality that's guaranteed on GPUs that claim that level of support, which inhibits how quickly they can iterate on its functionality: if they introduce a feature, they're essentially stating that everyone who wants to play has to support it.

I actually believe D3D's model is the better model for promoting usage of new features.
 
5150Joker said:
I don't think Mantle or SteamOS will get very far. Mantle will get a few Frostbite engine game optimizations but apart from a handful of games, it will be a Gaming Evolved API where AMD has to pay the developer off to use it. That means NVIDIA could easily develop their own TWIMTBP API and pay developers off to do the same. And between NVIDIA and AMD, NVIDIA is clearly better off financially to do so.

As for SteamOS, streaming Windows games has always been a failure, just ask all the operators of those streaming services that have come and gone. PC gamers want to minimize input lag, not raise it. Plus with a compressed stream, you lose image quality which is unacceptable. SteamOS is completely pointless and will fizzle away into obscurity.
Click to expand...

and if they did would put it past NV to not try to make a NV ONLY game
this is why its a bad idea from ether team
 
I guess it should not surprise me that none of you have done any reading on what Mantle is. :rolleyes: The game engine is what supports it which eliminates most of the support needed to implement the Mantle API. Also, the Mantle API is a cross platform API, meaning it is supported on PS4, Xbox One, and PC.

Also, from what I remember reading, the Mantle is platform agnostic. It will predominately support the AMD hardware of course since AMD is producing it and they hardware is all that is in the new consoles. However, it is or will be available to use for Intel and Nvidia as well. It really is a good thing that none of you guys are CEO's because you would all make Hector Ruiz look like the greatest CEO ever in the history of the world.

I am looking forward to it very, very much. Also, if SteamOS helps drive competition, I am all for that. (Also, NOONE is NOT a word at all! It is NO ONE. <-------Notice the space.)
 
Thank you gathagan for pointing out what so many others have totally missed. :D I made that comment above before seeing what you wrote so, at least there is some sensibleness in a sea of uninformed opinions.
 
The message seems clear.
Kill the golden goose at all costs.
Its the only way to control distribution and maximize dwindling profits.
Eventually the only need for a PC might be to develop for specialized software and or hardware.
If the need for a directory service is ever matched or removed then we call the empire of the PC dead.
 
DPI said:
No offense but you might want to do some actual reading on SteamOS and what it aims to do because you come off as a little clueless here citing online streaming services.
Click to expand...

I guess Anand must be clueless too:

SteamOS. Similar to Google’s Android and Chrome OS, SteamOS will be based on Linux, and obviously there will be a lot of tuning to make SteamOS work well as a living room OS. Valve specifically mentions support for in-home streaming; music, TV, and movie services; and family options to allow the sharing of games between Steam profiles.

Valve is also promising the ability to play all of the current ~3000 titles available on Steam, but unfortunatley it appears Windows/OS X titles will only be available via streaming (e.g. similar to NVIDIA's SHIELD streaming games from your PC to your SHIELD device).

...

Sure, we’re getting a new gamepad at some point, and another Linux-based operating system, but if you already have a Windows PC connected to your HDTV and running Steam, this hasn’t really changed the equation much. And depending on how many games need to be streamed from a Windows/OS X system to your Steam Machine, it's even less of a big deal.

Valve has the potential to change the equation; with an OS focused much more on gaming, performance in SteamOS could be competitive or even better than what we see under Windows. Of course, for non-Linux titles that are streamed from your Windows/Mac system, performace will be no better and you'll incur a small penalty in latency. Even if SteamOS were to get something similar to WINE, we’re talking about adding additional overhead to DirectX/OpenGL, at least initially. It’s a pretty big stretch to expect better performance from SteamOS when it initially launches in 2014.
Click to expand...

http://www.anandtech.com/show/7373/...m-machines-and-steam-controller-announcements


Basically it's gonna fizzle out.
 
ManofGod said:
I guess it should not surprise me that none of you have done any reading on what Mantle is. :rolleyes: The game engine is what supports it which eliminates most of the support needed to implement the Mantle API. Also, the Mantle API is a cross platform API, meaning it is supported on PS4, Xbox One, and PC.

Also, from what I remember reading, the Mantle is platform agnostic. It will predominately support the AMD hardware of course since AMD is producing it and they hardware is all that is in the new consoles. However, it is or will be available to use for Intel and Nvidia as well. It really is a good thing that none of you guys are CEO's because you would all make Hector Ruiz look like the greatest CEO ever in the history of the world.

I am looking forward to it very, very much. Also, if SteamOS helps drive competition, I am all for that. (Also, NOONE is NOT a word at all! It is NO ONE. <-------Notice the space.)
Click to expand...

This is odd because the actual truth is 100% opposite of everything you wrote here.
 
All this means is more crappy console ports for the PC.

Steam Shit Box is a death sentence to NON-casual players.

- Console Ports... check
- Controller support as primary input device... check
- Low system requirements plateauing the advancement of gaming technology... check
- More casual n00bs to have a say in pc gaming development.... check
- More ragers, more script kiddies.... check
 
Ill be upfront: i dont like the concept of mantle. I think its moving backwards not forwards. That being said i'm HOPING that it will make microsoft pull its head out of its ass with both developing direct x and not limiting releases to one version of windows
 
I am interested to see a game developer for DX and Mantle played on the same hardware, if Mantle show a fairly large improvement, then it will definitely be a good thing for PC gaming. If developers can get just 20% better performance out of games using Mantle then that could really help give a boost to the PC market since games will be playable at higher settings on lower end hardware, or played at higher FPS.
 
You must log in or register to reply here.
Back
Top