Much has been made over the advent of low-level graphics APIs over the last year, with APIs based on this concept having sprouted up on a number of platforms in a very short period of time. For game developers this has changed the API landscape dramatically in the last couple of years, and it’s no surprise that as a result API news has been centered on the annual Game Developers Conference. With the 2015 conference taking place this week, we’re going to hear a lot more about it in the run-up to the release of DirectX 12 and other APIs.

Kicking things off this week is AMD, who is going first with an update on Mantle, their in-house low-level API. The first announced of the low-level APIs and so far limited to AMD’s GCN’s architecture, there has been quite a bit of pondering over the future of the API in light of the more recent developments of DirectX 12 and glNext. AMD in turn is seeking to answer these questions first, before Microsoft and Khronos take the stage later this week for their own announcements.

In a news post on AMD’s gaming website, AMD has announced that due to the progress on DX12 and glNext, the company is changing direction on the API. The API will be sticking around, but AMD’s earlier plans have partially changed. As originally planned, AMD is transitioning Mantle application development from a closed beta to a (quasi) released product – via the release of a programming guide and API reference this month – however AMD’s broader plans to also release a Mantle SDK to allow full access, particularly allowing iit to be implemented on other hardware, has been shelved. In place of that AMD is refocusing Mantle on being a “graphics innovation platform” to develop new technologies.

As far as “Mantle 1.0” is concerned, AMD is acknowledging at this point that Mantle’s greatest benefits – reduced CPU usage due to low-level command buffer submission – is something that DX12 and glNext can do just as well, negating the need for Mantle in this context.  For AMD this is still something of a win because it has led to Microsoft and Khronos implementing the core ideas of Mantle in the first place, but it also means that Mantle would be relegated to a third wheel. As a result AMD is shifting focus, and advising developers looking to tap Mantle for its draw call benefits (and other features also found in DX12/glNext) to just use those forthcoming APIs instead.

Mantle’s new focus in turn is going to be a testbed for future graphics API development.  Along with releasing the specifications for “Mantle 1.0”, AMD will essentially keep the closed beta program open for the continued development of Mantle, building it in conjunction with a limited number of partners in a fashion similar to how Mantle has been developed so far.

Thie biggest change here is that any plans to make Mantle open have been put on hold for the moment with the cancelation of the Mantle SDK. With Mantle going back into development and made redundant by DX12/glNext, AMD has canned what was from the start the hardest to develop/least likely to occur API feature, keeping it proprietary (at least for now) for future development. Which is not to say that AMD has given up on their “open” ideals entirely though, as the company is promising to deliver more information on their long-term plans for the API on the 5th, including their future plans for openness.


Mantle Pipeline States

As for what happens from here, we will have to see what AMD announces later this week. AMD’s announcement is essentially in two parts: today’s disclosure on the status of Mantle, and a further announcement on the 5th. It’s quite likely that AMD already has their future Mantle features in mind, and will want to discuss those after the DX12 and glNext disclosures.

Finally, from a consumer perspective Mantle won’t be going anywhere. Mantle remains in AMD’s drivers and Mantle applications continue to work, and for that matter there are still more Mantle enabled games to come (pretty much anything Frostbite, for a start). How many more games beyond 2015 though – basically anything post-DX12 – remains to be seen, as developers capable of targeting Mantle will almost certainly want to target DX12 as well as soon as it’s ready.

Update 03/03: To add some further context to AMD's announcement, we have the announcement of Vulkan (aka glNext). In short Mantle is being used as a building block for Vulkan, making Vulkan a derivative of Mantle. So although Mantle proper goes back under wraps at AMD, "Mantle 1.0" continues on in an evolved form as Vulkan.

Source: AMD

POST A COMMENT

95 Comments

View All Comments

  • ppi - Wednesday, March 4, 2015 - link

    While we would be on disagreement on not speeding up DX12/glNext, Mantle was certainly great implementation beta-test for both standards.

    By that I mean, the roughest edges in actually shipping a game with it were found, and so they may be smoothened for DX12/glNext.

    In practice, Mantle helps most owners of AMD CPUs, which the reviewers (including Anand) still do not seem to grasp.
    Reply
  • klagermkii - Tuesday, March 3, 2015 - link

    "In most important ways OGL 4 was just as good as Mantle"

    Yes, technically that's true. The most important thing of a graphics API is to be able to get 3D graphics on the screen and they DID both achieve that.

    But developers seems pretty excited about it, and the existence of DirectX 12 and whatever glNext will be called are both trumpeting as their main feature Mantle-like improvements.

    I think if you're not living in some crazy conspiracy world it's more likely that they wanted to see how DirectX 12 turned out in reality before ditching Mantle. I see Mantle as being really important in proof-of-concept-ing this kind of API, getting it running in real games, and helping DX12 and glNext get it right. Kudos to AMD.
    Reply
  • FlushedBubblyJock - Friday, March 27, 2015 - link

    That would be wonderful if the amd fans hadn't subjected all of us to the lofty fairly tales their master and dominator told them, while those with common sense and a clear head saw all this coming and said so.

    Very "open" of the holy manifesto for gamers card company, huh.
    Reply
  • kyuu - Tuesday, March 3, 2015 - link

    Never ceases to boggle the mind how emotionally invested people can become in hating a brand, as though the brand in question had kicked their dog and raped their grandmother.

    What is your proposed solution for this, by the way? Should AMD proceed to release the SDK and spend resources supporting it when it's simply not likely to be used? Just to prove people like you wrong? Because that's obviously the best way for AMD to use their limited resources.

    If you knew anything about the game development community, you'd know that developers have had an issue with the draw call limitations imposed by DX and OpenGL for a long time now. And we have no idea if Microsoft or Khronos were looking at improving the situation before Mantle came along, but either way, Mantle certainly didn't hurt and certainly did help bring the problems with DX and OpenGL into the spotlight.
    Reply
  • FlushedBubblyJock - Friday, March 27, 2015 - link

    Hate the lies, not the brand. Lot's of lies, and parrots. Reply
  • TheJian - Tuesday, March 3, 2015 - link

    Exactly. Carmack etc said as much on stage (even dice guy was there and couldn't say much...LOL), that OpenGL had it all with extensions. Cass Everitt did a speech at Steam dev days a while back demonstrating how to speed up draw calls massively on OpenGL.

    https://www.youtube.com/watch?v=-bCeNzgiJ8I&ht...
    Reducing driver overhead etc. An hour on the topic. Devs just didn't know how to use what was ALREADY there. He mentions a few lines of code gave a 4x improvement for a guy and that you could hit 8x+ with more in depth look. Note the room was full, and they were surprised considering the VR speech next door and even admitted they'd be in that room if they weren't giving the OpenGL speech...LOL. Clearly most devs had no idea. I'm guessing Carmack already knew all this ;) Funny Cass now works at Oculus :) I guess he meant what he said in the video.
    Reply
  • HalloweenJack - Tuesday, March 3, 2015 - link

    OGL has the same overheads as DX11 so it cant be `as good`.

    anyway - wait for Vulkan , big rumours about it using mantle as part of it , which would fully explain AMD`s cryptic PR
    Reply
  • Beany2013 - Tuesday, March 3, 2015 - link

    Vulkan is a derivation of Mantle, with the cruft and non-cross-platform stuff purged, and some more useful bits added.

    So the idea that Mantle was a 'waste of time' is a load of shit, frankly - and by being part of Vulkan, that means large chunks of Mantle *are* now open, by the nature of being part of Vulkan itself - which is open.
    Reply
  • TheJian - Saturday, March 7, 2015 - link

    It was a waste of time and money AMD doesn't have. You have no proof any actual direct code will be used by the time it ships (it isn't even done, a long way from it in fact). Were some ideas used initially? Probably. Direct Code that would affect performance of NV cards with NV being in the driver’s seat (IE NV's mobile ecosystem's guy is the president of Khronos, and he came up with OpenGL ES)? Doubtful.

    "For AMD this is still something of a win because it has led to Microsoft and Khronos implementing the core ideas of Mantle in the first place"

    I call that a loss, and just IDEAS, not anything that will affect perf. The major feature AMD touted was HLSL and that is totally cut out now.

    http://www.pcper.com/reviews/General-Tech/GDC-15-W...
    “When AMD made Mantle, one of their selling points was that the Mantle shading language was just HLSL.”

    “Khronos is doing something else entirely.”
    Note unlike the rosy Anandtech “it all came from mantle” stuff, PCPER tells a totally different story and barely mentions mantle in the context of it being a building block on which Vulkan was built. They changed the cpu and gpu side, and completely tossed out AMD’s major selling point, HLSL.

    Further from Anandtech, showing Mantle is dead and not even open.
    "Thie biggest change here is that any plans to make Mantle open have been put on hold for the moment with the cancelation of the Mantle SDK."

    From the Vulkan Article here:
    "Next Generation OpenGL Initiative (glNext). Designed around similar goals as Mantle, DirectX 12, and Metal,"

    Goals and actual code use are two totally different things in the end. A few lines that are not going to vault AMD into the sky won’t matter and this isn’t a finished product anyway.

    "In fact Khronos has confirmed that AMD has contributed Mantle towards the development of Vulkan, and though we need to be clear that Vulkan is not Mantle, Mantle was used to bootstrap the process and speed its development, making Vulkan a derivation of sorts of Mantle"

    What else could they do when it was clear they couldn’t win the API war? NOT MANTLE, just a bootstrap to get things going and as noted the largest most important part tossed. It pretty much sounds like they took some ideas. The standard isn't even done, so no idea how anyone can say what will be in the finished product when it might be a year or two away at best (is any game even using OpenGL 4.4 yet fully?). For any line of code to count, AMD has to be faster in it right? If you don’t end up with some sort of advantage over the enemy, what was the point? They already seem behind in their own “MADE FOR AMD” Star Swarm DirectX12 contest. That doesn’t bode well for OpenGL either and shows they shouldn’t have been doing mantle, but instead should have been concentrating on DX12/OpenGL (vulkan).

    You pointed to two places that mirror the same post. If a few lines of code make you feel better about AMD's management screwing them yet again, so be it. Personally, I'd rather have had them spend all that effort on DX12/Opengl so perf so far wouldn't look like all the benchmarks here:
    http://anandtech.com/show/8962/the-directx-12-perf...
    Killed in everything DX 11 or 12 vs. NV and this is AMD's sponsored showcase for Mantle. Since they are so far behind in DX12 (or 11, which anandtech dismisses as, well no reason to optimize DX11, forgetting it affects every game out on windows pretty much), I don't see how Vulkan will be much different. I mean, AMD is in the XBOX1/PS4, yet Nvidia was chosen for GDC 2014 demo of an XBOX1 exclusive (forza) running DX12 on Titan Black. So MS wasn't confident enough in AMD's drivers to choose them to demo a game that is exclusive to AMD/MS/Xbox1...Ok.
    http://blogs.nvidia.com/blog/2014/03/20/directx-12...
    Nvidia running Forza, hand in hand with MS since at least March 2013, and he mentions 4yrs before the demo. Not sure when we'll see the first Vulkan benchmarks, but I don't see how NV will be behind when they are the president of Khronos and already winning DX12 in Star Swarm & picked for Forza demo (AMD's game on xbox1). Odd or just a company spending money on optimizing where they should? IE, drivers for the two most critical upcoming API's on the planet! AMD instead spent money to help their lagging cpus, and that was a mistake. They should have made a MONSTER IPC cpu at worst (no gpu, just straight cpu monster for some cpu pricing power vs. Intel!), and definitely should have spent on dx12/Vulkan.
    Reply
  • FlushedBubblyJock - Friday, March 27, 2015 - link

    How come all the amd fans blabber on about competition and how amd makes everyone else be on their game - yet 2 seconds later they are bloviating about a single standard... which is no competition - and thus no innovation !

    Someone explain it to me, how the babbling communists "love competition", then demand monopoly.
    Reply

Log in

Don't have an account? Sign up now