Even if this is "the death of Mantle" - it has served its purpose. Other companies are emulating it, and DirectX12 should be similar.
Where DirectX12 will succeed, Mantle will follow, and vice-versa. DirectX11 was stagnation, and now Mantle will probably exist as a project to ensure the likelihood of that not happening again.
Either way, this has been a win for AMD. Now their "weaker cores" are not so terrible anymore, and people have invested more into their products. They gained a bunch of credibility in the process.
@Calculatron Agreed, and if there is something that needs to be addressed hopefully AMD has an answer for that in the future. if no one else will address it.
it proved beyond a shadow of a doubt that there was nothing wrong with AMD hardware (cpu's mainly) and the fault (gaming wise) for poor performance lay directly at dx11's feet.
I haven't read too much of the articles yet so I may be missing some finer points.
I would imagine as a business unit amd would only cancel Mantle IF dx12 served all of its needs.. if it didnt then they would have just carried on with it.
Seriously.. once bare metal becomes the norm we will all be gaming quite happily on near potato level hardware at higher settings :D
When I saw that Mantle granted me 15-20 FPS benefit when running Thief, I could imagine some of the AMD engineers/programmers silently chuckling to themselves, smug with knowing that their "good enough" performance they've used in defense has been right all along. I realize that everyone's mileage has varied, but Mantle isn't even a finished product yet and look at the change it has caused.
Anyway, I don't think AMD will eliminate Mantle for a while, even if DirectX12 fills all the gaps. (Although, no doubt, if DirectX12 is successful enough, people will bash AMD over Mantle.)
I was hoping there would be more tonit than this but it seems the talk of it being a catalyst (sorry could not resis), to get DirectX and OpenGL moving at speed was true.
Unfortunately again OpenGL just did not get it together again and we are once again stuck with the tyranny of Microsoft control over our games.
I wonder what this means for SteamOS as OpenGL and Mantle were going to be the underpinnings of it graphics capability. and at this time OpenGL still lacks the full feature set of DX11 and that will only widen now.
Although because they are retiring Mantle the pressure is off Microsoft to push DirectX again. So after 12 it may well stagnate for another 2 or 3 generations.
In all this is bad news for everyone involved. Less choice and only one that makes sense. All the while locking the games market to a failing and confused OS (don't quote numbers please I know Windows dominates home PC still they are just starting to slip).
Sorry I was channeling Zoltan a bit. But I am really quite annoyed at this, even if I never used mantle, it just seems like a back step.
Oh well wheel out AMD proprietary software it is and stop the GPU competition again for a while. Should keep the fanbois fueled with GPU feature fights for a while.
The only reason there would be for this is if they are working on something better and locked down. The fact that they killed it before getting the open source back end up and running is saying something. Kill it, stop the support, kill the SDK and dump it on the community to sort out, good job.
I like AMD and will conitune to support them as the alternative is nVidia.
I am more interested in the Khronos Group's APIs. They are releasing Vulkan and GL.Next. Vulkan is slated to come out this year and from the looks of it will make DX 12 look bloated.
By announcing the discontinuation of Mantle 1.0 all that has happened is that AMD dont have to follow through with making it an 'open' technology.
Plenty of games in 2015 will be launching with mantle as its baked into quite a few game engines now.
so in the short run (up until dx12 arrives and then its a level playing field) amd will have an edge which no nvidia tech can match.
DX12 does everything amd want so it just makes sense to put it to pasture, anything else is wasted resources for a company that doesnt have the resource to spare.
Mantle was destined to follow suit, and it does so today as **we proudly announce that the 450-page programming guide and API reference for Mantle will be available this month (March, 2015) at www.amd.com/mantle.** This documentation will provide developers with a detailed look at the capabilities we’ve implemented and the design decisions we made, and we hope it will stimulate more discussion that leads to even better graphics API standards in the months and years ahead
The newest verions of Mantle API's are being posted in March. There is nothing dead about Mantle.
You misunderstand, Mantle as an API is NOT going to exist anymore. The features are not going away. They are being implemented into DirectX12 and OpenGL-Next. And it clearly states that in the blogpost. It states developers who wish to use Mantle 1.0, invest into DirectX12 or OpenGL-Next. AMD's intentions for mantle was not to compete with DirectX or OpenGL, it was meant to show what a low-level API is capable of. If Mantle wasn't going away, AMD wouldn't tell developers to go check out DirectX12 or OpenGL. However developers who have already have a game that uses Mantle, it will still have support for AMD.