• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

AMD porting Mantle to Linux/SteamOS

Looks like a fully adopted API to me, good job AMD.

Doesn't make it fully adopted in my book. They need to open their source and let intel and nvidia implement it into their chips (if they so desire) before it'll be fully adopted. Tbh, it shouldn't be adopted unless implementation is atleast offered to the others in the graphics sector

That being said better AMD support for linux is always a good thing.
 
Yup, this must be what AMD was hinting at with their last post. "With DirectX 12 games still over 18 months away and no alternatives in sight for Linux gamers, Mantle’s future looks bright."

Good stuff!

Actually Steam Boxes with an APU and Mantle... reminds me of something... ;)

50 Mantle games, wow didn't expect that much support so quickly. That's a good number of developers learning and working with Mantle in less than a year since Mantle was introduced.

Once Mantle for Linux is functional, I wonder how long it will take to port Mantle games between Windows & Linux. If it's less than 4-6 weeks, it means devs saving money, and they like that. And then the other consoles...

In any case, this is looking mighty interesting!
 
Nice, about 50 Mantle games in development or being released soon.
Important question: How many of those 50 games support Mantle and are getting a Linux port?

Supporting Mantle on Linux doesn't really help anyone if none of the games that use Mantle actually have a Linux version...

Also, what if you don't have an AMD card? Are you then totally unable to run these games on Linux (aka, horrible hardware lock-in)?
 
50 Mantle games, wow didn't expect that much support so quickly.

Pretty much every single game using Frostbite and Cryengine. Both of these engines have (will have in the case of Cryengine) built in support for Mantle. Add to those every game developed on Oxide's Nitrous engine and Firaxis' Civ V engine plus I'm sure there are a bunch more unannounced titles in development.
 
Important question: How many of those 50 games support Mantle and are getting a Linux port?

Supporting Mantle on Linux doesn't really help anyone if none of the games that use Mantle actually have a Linux version...

Also, what if you don't have an AMD card? Are you then totally unable to run these games on Linux (aka, horrible hardware lock-in)?

Read the article. They answer your concerns.
 
Pretty much every single game using Frostbite and Cryengine. Both of these engines have (will have in the case of Cryengine) built in support for Mantle. Add to those every game developed on Oxide's Nitrous engine and Firaxis' Civ V engine plus I'm sure there are a bunch more unannounced titles in development.

Agreed, it's looking like a lot of development time has been invested, and is increasing every day, going into Mantle. Looking at it this way it only makes sense these developers wanting as much ROI as possible on their investment and having a "window" into easier Linux/SteamOS porting in addition to making their own games faster and look better makes a lot of sense.
 
This sounds awesome on paper at least. Hopefully they will be able to get it running on Linux before DX12 launches. I also would like to know which games are thinking of porting to Linux. Will be interesting to see if Intel onboard graphics and Nvidia adopt Mantle or develop their own solutions.

Thx for the post Lorien!
 
Doesn't make it fully adopted in my book. They need to open their source and let intel and nvidia implement it into their chips (if they so desire) before it'll be fully adopted. Tbh, it shouldn't be adopted unless implementation is atleast offered to the others in the graphics sector

That being said better AMD support for linux is always a good thing.

Mantle will be an Open API per AMD. They haven't released it as such yet, but that's the plan.
 
Looking at it this way it only makes sense these developers wanting as much ROI as possible on their investment and having a "window" into easier Linux/SteamOS porting in addition to making their own games faster and look better makes a lot of sense.
Mantle only (potentially, but not necessarily practically) gives developers an easier path to SteamOS if only Mantle is supported by the renderer. In cases where developers wish to target users who run Intel and NVIDIA hardware — which comprises somewhere north of 70% of gamers — they need to provide an additional rendering path (that obviously being OpenGL).

It's a false presumption, also, that any game built on technology that supports Mantle as an option will also offer Mantle support (in that Mantle is an available path). That there are fifty games in development on these engines is not an indicator that there are fifty games in development that will offer Mantle support. Some will; some may; some will not.

Well we need something to replace OpenGL, its a pretty shit API.
For what reasons, specifically?
 
This sounds awesome on paper at least. Hopefully they will be able to get it running on Linux before DX12 launches. I also would like to know which games are thinking of porting to Linux. Will be interesting to see if Intel onboard graphics and Nvidia adopt Mantle or develop their own solutions.

Thx for the post Lorien!
It would be logical for Nvidia and Intel to adopt Mantle for their own GPUs. AMD has said already that Mantle can easily be ported directly to DirectX 12 with minimal effort. If that's the case and I was either of the two companies, if I wanted to be ready for DX 12 in 2015 and stay competitive, I'd adopt it if and when AMD makes it available. Very likely this will happen after it exits beta.

AMD still wants this to be available to as many different architectures and platforms. Only thing we have to realize is time-- how long it's going to get there. It's still beta so I'm thinking they want it in a more finalized form before trying to get Nvidia and Intel on board.

Imagine if a hardware review site like [H] or Anandtech test Intel's latest Iris GPU in Broadwell compared to the latest AMD integrated GPU and see how fast both perform in Mantle. It may be an interesting comparison, to say the least, to see this happen someday. Or, see who performs better at close-to-metal performance outside of DirectX 11 between Nvidia and AMD.

Having Mantle available for Linux is going to be icing on the cake honestly. Since so many games are still on OpenGL 3.x and 2.x, having Mantle means Linux games may actually be using features on par with DirectX 11. I would like to see more games on OpenGL 4.x but I have yet to see one, honestly. Mantle might be the push needed to get better games on LInux since porting them from Windows would be easier to do. No more translating DX calls through some other software or pretend-Wine, games will probably perform as good as they are on Windows. If Nvidia jumps on board with Mantle, they'll have a competitive product inside Linux and adoption of the OS may increase.
 
The open-source and Linux community have historically not been kind to AMD and rival Nvidia, particularly criticizing slow progress by the companies to bring graphics cards drivers to Linux. Linus Torvalds famously showed the middle finger to Nvidia in 2012, saying it was the “single worst company we’ve ever dealt with.”

I love Linux as well as the open-source software movement, but Linus is quite possibly a bigger douche than Steve Jobs was. Removing Android from the equation, what is the actual market penetration of Linux in the countries that spend money on PC games? I know Linux is very popular in: countries where the cost of a Windows license is prohibitive; government facilities looking to control licensing costs; and of course it is a major underpinning of much of the web. These markets do not care about gaming performance because it is not their focus. Giving nVidia the finger because they are not supporting a minute fraction of the PC user base is just plain stupid, unless you're an arrogant asshole bent on pushing open-source on everyone.

I am glad to see that AMD is taking steps to make the API accessible to Linux game developers and I hope that the SteamOS/SteamBox is able to break into the PC gaming market in a big way, but up until recently, there was a reason that the market segment has been overlooked: no one was going to make any money targeting an audience of a fraction of a percent of PC users. How many games are ported to Mac OS X?

If Linux picks up major market share due to the proliferation of SteamOS/SteamBox and nVidia/AMD continue to thumb their nose at Linux, then I could see Linus' point being justified. Until then, it makes business sense to focus the majority of their efforts on the Windows platform. It's not like they both don't have enough bugs to work out in their Windows drivers, though AMD admittedly has a bit more ground to cover in my experience.
 
I love Linux as well as the open-source software movement, but Linus is quite possibly a bigger douche than Steve Jobs was. Removing Android from the equation, what is the actual market penetration of Linux in the countries that spend money on PC games? I know Linux is very popular in: countries where the cost of a Windows license is prohibitive; government facilities looking to control licensing costs; and of course it is a major underpinning of much of the web. These markets do not care about gaming performance because it is not their focus. Giving nVidia the finger because they are not supporting a minute fraction of the PC user base is just plain stupid, unless you're an arrogant asshole bent on pushing open-source on everyone.

I am glad to see that AMD is taking steps to make the API accessible to Linux game developers and I hope that the SteamOS/SteamBox is able to break into the PC gaming market in a big way, but up until recently, there was a reason that the market segment has been overlooked: no one was going to make any money targeting an audience of a fraction of a percent of PC users. How many games are ported to Mac OS X?

If Linux picks up major market share due to the proliferation of SteamOS/SteamBox and nVidia/AMD continue to thumb their nose at Linux, then I could see Linus' point being justified. Until then, it makes business sense to focus the majority of their efforts on the Windows platform. It's not like they both don't have enough bugs to work out in their Windows drivers, though AMD admittedly has a bit more ground to cover in my experience.


Well Linux doesn't need Nvidia.... But Nvidia does need Linux, does that make any sense?
 
The only thing I want to know is - will all the CPU overhead benefits still work with Nvidia and Intel graphics when it eventually goes "open?" I still don't know how hardware-level API's are going to be open unless Nvidia and AMD make a joint-effort architecture together, or if developers omit hardware features for certain devices for compatibility's sake (thereby eliminating the advantage of Mantle - unless it improves CPU performance for EVERYONE). I don't know. As a gamer, I want choice, and Mantle doesn't provide that. Hell, as a CRT user, my only option is Nvidia right now, unless I can get a good DAC that can handle high resolutions like 2304x1440 @ 75hz.
 
the whole point of mantle is lower cpu overhead.... how would adapting it for nvidia remove the whole advantage of it ?answer itt won't. They all have unified shaders and draw calls work the same.

dx11 is already being split in half with proprietary .dlls , while mantle is just made to work on amd right now. When the game don't work, just say no to gameworks!
 
the whole point of mantle is lower cpu overhead.... how would adapting it for nvidia remove the whole advantage of it ?answer itt won't. They all have unified shaders and draw calls work the same.

I'm not sure thats true. The draw calls work the same because they're going through DX, but the close to the metal differences are real. If the instructions are written in a way that they specifically fit nicely into GCN architectures, then you need to define what "adapting it for nvidia" means. Does it mean re-writing all the code so that it targets specific nvidia architectures or does it mean adding an abstraction layer to translate GCN specific code into something that'll run on nvidia? The former might very well mean a full re-write (not gonna happen) while the latter would defeat the entire purpose.
 
There's more chance of mantle going on linux than nvidia doing that. If physx is being implemented via gameworks for linux, chances are even lower now. I'd see intel doing it tho.
 
dx11 is already being split in half with proprietary .dlls , while mantle is just made to work on amd right now. When the game don't work, just say no to gameworks!

Difference between Mantle and Gameworks - you can still play a Gameworks game on AMD. Mantle - can't happen.
 
Read the article. They answer your concerns.
I did read the article, it answers none of my concerns.

Important question: How many of those 50 games support Mantle and are getting a Linux port?

Supporting Mantle on Linux doesn't really help anyone if none of the games that use Mantle actually have a Linux version...

Also, what if you don't have an AMD card? Are you then totally unable to run these games on Linux (aka, horrible hardware lock-in)?

The article does not mention which of those 50 mantled-powered games will have a Linux port.
The article does not mention what mantle-powered Linux games will do when an AMD GPU is unavailable.
 
The article does not mention what mantle-powered Linux games will do when an AMD GPU is unavailable.

From the article:

Most games today are being written for Windows using DirectX 11, and the games are then ported to OpenGL for Linux and other operating systems. The port from DirectX 11 to OpenGL takes about four-to-six weeks, and it’ll take about that long to port games to Mantle, Huddy said.

It's a PC World article, but I think it's safe to say that there will be the default OpenGL path available for Linux games as always.
 
Difference between Mantle and Gameworks - you can still play a Gameworks game on AMD. Mantle - can't happen.

With UBISOFT, it was almost like neither AMD or Nvidia could play Watch Dogs without stuttering. ;) I don't know if UBISOFT's programmers are Gameworks skill deficient, or they give less than a shit about the PC. At any rate you can still play Mantle games using the default DirectX path. With Mantle in Linux there should still be the obligatory OpenGL path available.
 
From the article:

Most games today are being written for Windows using DirectX 11, and the games are then ported to OpenGL for Linux and other operating systems. The port from DirectX 11 to OpenGL takes about four-to-six weeks, and it’ll take about that long to port games to Mantle, Huddy said.

It's a PC World article, but I think it's safe to say that there will be the default OpenGL path available for Linux games as always.

Why would any company spend the extra time and money to port the game to OpenGL and Mantle? Unless AMD plans to fund this like they did for BF4, it's probably just marketing BS.
 
Why would any company spend the extra time and money to port the game to OpenGL and Mantle? Unless AMD plans to fund this like they did for BF4, it's probably just marketing BS.

They would port to mantle for the same reason people use gameworks.

If the games engine is already in mantle, there in no reason to port.
 
Why would any company spend the extra time and money to port the game to OpenGL and Mantle? Unless AMD plans to fund this like they did for BF4, it's probably just marketing BS.

50 games being developed have chosen to support it. It's free will for them to do so. AMD doesn't have the resources to pay 50 different companies to do this so obviously there is a reason for them to do so. More than likely I would speculate that after you've done it within your engine once; that porting to Mantle for any other games using that same engine is a lot faster than 4 - 6 weeks.

Plus one extra month of development time isn't going to run you out of business if you can sell your game to people running SteamBox computers that look like this. They come in Intel and AMD variants.

http://www.gigabyte.us/products/product-page.aspx?pid=4930#ov

9663_big.jpg
 
Nvidia will have to be dragged to Mantle kicking and screaming. If Mantle isn't in a majority of the major releases, then I only see Nvidia pushing DX12.
 
Nvidia will have to be dragged to Mantle kicking and screaming. If Mantle isn't in a majority of the major releases, then I only see Nvidia pushing DX12.

DX12 doesn't work in Linux. It's proprietary for Windows. Nvidia already has working OpenGL drivers so you're probably right in that regard that they won't touch Mantle. What Nvidia could be banking on is that OpenGL 4.0 titles start showing up. I don't keep track of the Linux scene, so I'm uninformed about how many OpenGl 4.0 games are in the marketplace currently. I've never heard of one personally. OpenGL 4.0 is supposed to be the DirectX 12 equivalent for Linux from what I understand. If I'm wrong feel free to crucify me.
 
The thing with mantle is you can still play games in dx11 *rimshot*. I'm here every Wednesday!
:p

That's if a developer decides to write to DX (same thing with OpenGL). Whatever... No rimshot here - I think we're not understanding one another. I think that the possibilities of Mantle are pretty sweet, but I don't think it's anywhere near being a long-term option. It looks like multiplatform console development is what it looks like - you got your Mantle version, your DX version, blah blah. I'd rather have one version and have it so that all PC hardware could run it.
 
With UBISOFT, it was almost like neither AMD or Nvidia could play Watch Dogs without stuttering. ;) I don't know if UBISOFT's programmers are Gameworks skill deficient, or they give less than a shit about the PC. At any rate you can still play Mantle games using the default DirectX path. With Mantle in Linux there should still be the obligatory OpenGL path available.

Yeah, let's not count Ubisoft. :D I think the second option of what you said is probably more in tune with reality (but hey, never underestimate ye olde incompetence!). I understand that you can play the OTHER path if you don't have a GCN card, but that's not really the point. Mantle is bad because of its hardware lock. I don't care if it will allow other cards to use it in the future (until it does, that aspect is vaporware), the point is that I just don't trust game developers these days with multiple code path options. Hell - they cannot even manage it on current consoles and PC's. They always seem to develop for the lowest common denominator and then add on "extras" to the more powerful platforms. There have been some EXCELLENT multiplatform releases, but those tend to be the exception and not the rule.

Why oh why would we expect it to be any better when we give developers even more code paths to write for?

Maybe the long term solution is to dump Windows and Linux as we know it and create an OS that handles the hardware differently? I'm just coming up with stuff off the cuff now, because I don't develop games and can't speak from that angle.
 
50 games being developed have chosen to support it. It's free will for them to do so. AMD doesn't have the resources to pay 50 different companies to do this so obviously there is a reason for them to do so.

It's not 50 different companies. Most of it is DICE which AMD paid something like $8 Million. So far though only 1 or 2 games have made it to the market nearly a year after it was announced.
 
It's not 50 different companies. Most of it is DICE which AMD paid something like $8 Million. So far though only 1 or 2 games have made it to the market nearly a year after it was announced.

I personally own:
Lichdom
BF4
Thief

Day one purchase for me that support Mantle includes:
BF Hardline
Dragon Age Inquisition
Civilization
Crysis
Bunch more, but my buddy is instant messaging me. :)

I mean we just got decent Mantle Beta drivers on May 27, 2014. It's new tech. Rome wasn't built in a day. AMD even went so far as to release the 14.6 Linux driver.
 
From the article:
Most games today are being written for Windows using DirectX 11, and the games are then ported to OpenGL for Linux and other operating systems.
That explains the current workflow we see today, involving only DirectX and OpenGL. It doesn't speak to any sort of fallback renderer for Mantle.

The port from DirectX 11 to OpenGL takes about four-to-six weeks, and it’ll take about that long to port games to Mantle, Huddy said.
Right, so they're saying here that the port from DirectX to Mantle takes about four-to-six weeks. There's still no mention of any kind of fallback renderer for Mantle games on Linux.

It's a PC World article, but I think it's safe to say that there will be the default OpenGL path available for Linux games as always.
So they'll spend four-to-six weeks to port to OpenGL (for everyone) and an additional four-to-six weeks to also port to Mantle (for users with AMD cards, who could already run their OpenGL port)?

The article doesn't actually state that, but even if it did, that seems like an odd division of effort.

If the games engine is already in mantle, there in no reason to port.
They're still stuck porting to OpenGL if they want to support more than just a small sub-set of Linux users with Mantle-capable AMD graphics cards.
 
Last edited:
I'd rather have one version and have it so that all PC hardware could run it.

Wasn't direct X supposed to be just that ? Look now ,the industry is being parted on dx11 already, Nvidia is pushing gameworks to make newer effects for their cards only.While dictating how developers input dx11 code.

AMD's received builds of games a month before release and optimizing to have black box .dlls changed just before release. Game developers are contractually obligated to implement any dll change nvidia provides,so you can't blame the developers for that. It's a war on resources and we all know amd has less resource. It's brilliant :p Win win .Just not good for gamers. (See watchdog release ran like crap on both lol)

You don't get any of that with mantle..,tho they can... as there was talks bout implementing additional rendering techniques.

Future looks bleak ,divided either way ,even more when developers give different visual experience per gpu brand. Maybe intel can come up with their own effects, some type of realistic BloodFX, Bloodinside....Blood+ . Runs best on intel hardware.

Looking forward to lower cpu overheard in dx12 , mantle , on windows or linux either way.
 
@Unknown-One.

You really think a company is going to spend hundreds of millions of dollars developing a game for WIndows, spend a few more bucks porting it to Mantle for Linux, and then completely skip porting it to OpenGL for Linux so that they can lock out Intel notebook users and Nvidia graphics card owners just to be spiteful?

Come on man.
 
You really think a company is going to spend hundreds of millions of dollars developing a game for Windows, spend a few more bucks porting it to Mantle for Linux, and then completely skip porting it to OpenGL for Linux so that they can lock out Intel notebook users and Nvidia graphics card owners just to be spiteful?
Uh, no, I don't think that. I never said anything like that :confused:

I'm pointing out the elephant in the room. If they already have to port from DX11 to another API, and it takes the same amount of time to port to either OpenGL or Mantle, why bother porting to both Mantle and OpenGL when they can just port to OpenGL (which already runs on everything)?

If they try to cut costs by doing only a Mantle port, they lose a pile of customers and lock out most hardware (doesn't make sense, which you and I both noticed).
If they try to cut costs by doing only an OpenGL port... there's no real down-side (aside from missing out on some performance on a specific sub-set of users machines).

See? I was pointing towards the exact OPPOSITE of skipping an OpenGL port.
 
Last edited:
Uh, no, I don't think that. I never said anything like that :confused:

I'm pointing out the elephant in the room. If they already have to port from DX11 to another API, and it takes the same amount of time to port to either OpenGL or Mantle, why bother porting to both Mantle and OpenGL when they can just port to OpenGL (which already runs on everything)?

If they try to cut costs by doing only a Mantle port, they lose a pile of customers and lock out most hardware (doesn't make sense, which you and I both noticed).
If they try to cut costs by doing only an OpenGL port... there's no real down-side (aside from missing out on some performance on a specific sub-set of users machines).

See? I was pointing towards the exact OPPOSITE of skipping an OpenGL port.

I don't think mantle would be a full port (doesn't replace DX / OGL )

Anyhoo if the engine supports OGL already, then it shouldn't be the elephant in the room, there are many other things that need to be changed / ported going from Windows -> Linux.
 
Uh, no, I don't think that. I never said anything like that :confused:

I'm pointing out the elephant in the room. If they already have to port from DX11, why bother porting to both Mantle and OpenGL when they can just port to OpenGL?

Because they can bring triple A titles to devices such as the Gigabyte Brix. The little AMD APUs can run BF4 without a dedicated video card on Mantle. If Intel decides to use Mantle then whatever they have in the same form factor and price range might be able to run something like BF4 without a video card also. If something like the Brix drops in price to "disposable device" prices, imagine how many of those devices Intel could sell? I could even see it shrinking enough to be built into future Smart TV's.

But what you're asking is why bring Mantle over to Linux when there is OpenGL there already. If I already ported my engine to Mantle on for DX11 then why not stick with it for Linux? If OpenGL was so fabulous then why aren't there tons of triple A titles using it? That's my retort. :) But seriously, last thing I remember using OpenGL was ID Software for their Doom titles. Quite sure there are other titles, but those are the ones that I remember.

I did look at the OpenGL Wiki page and I saw a bunch of Indie titles that have Linux ports. I did see the obligatory Valve titles like Dota 2 and HL2. Blizzard disabled the OpenGL path for the Windows version of StarCraft 2. In the end Linux gaming is a new emerging market and it looks like consumers are going to have plenty of choices in the future.
 
Back
Top