AMD Will Release Mantle Programming Guide, API Reference This Month

[Guest]

The exact opposite. Mantle is here and now, and has been for over 12 months evolving. Vulkan/DirectX 12 are in their infancy and will not be touched for years, while OpenGL 4.5 adoption evolves. Gaming will use Mantle. Microsoft will dump a ton of resources into Directx 12 but with AMD able to control the console hardware they don't have to jump when MS says jump.

That's the point that industry slowly realizes, whereas decade back it was all about releasing whole product on single market (specific console or DX+PC), nowadays companies put so much money into single game that they don't wanna be crippled by platform restrictions. While DX used to have market coverage (PC+XBOX), it's pretty clear (or should be) that open standard (not necessarily opengl) is a way to go if you aim to support whole market.
Recent XBOX sales didn't help much (their decision to sell console and make direct profit of it, Sony however went their tried and true way of selling console for cost to manufacture it and earn back profit on games/goods).
This mistake might have costed MS console wars and resulted in open standard proliferation (I'm speculating, please no angry comments picking on two words out of context).

[justmy2cents]

They even suggest Vulkan/DirectX12 themselves.

Mayble they see Mantle as an unstable branch of Vulkan/DirectX where they can test and refine new API ideas.

nah, they just pulled the plug from something that would prove costly in the long run. and even though i never liked AMD approach to Mantle, the one thing that needs to be acknowledged is that AMD kicked everyone in their arse and that Mantle is probably nr.1 reason for rapid improvement

i'd even go as far as saying that not releasing any spec until too late was just for this. they probably just got what they wanted and got it sooner than expected. normally, they would need to prove competition, this time suddenly everyone just jumped on the train

[bridgman]

Mantle is all separate from those two, it is special "right to the metal and custom" API, thus is not generic design like DirectX or OpenGL.

Not really. It was a bit lower level than previous DX/GL and required a modern GPU, but it wasn't really specific to AMD GPUs.

[dungeon]

Not really. It was a bit lower level than previous DX/GL and required a modern GPU, but it wasn't really specific to AMD GPUs.

I can believe you that it wasn't specific to AMD GPUs, but hard fact it is/was only available for AMD GCN GPUs... and thus that is same said to me .

[Kano]

@bridgman

If AMD wanted to support Mantle for LONG time then

A) Full specs would have be provided at release
B) Linux binaries would exist (since the beginning)

It was 100 % designed as technolgy with obsolescence.

[justmy2cents]

Not really. It was a bit lower level than previous DX/GL and required a modern GPU, but it wasn't really specific to AMD GPUs.

sorry to butt on this one, especially since you probably know better than me.

that was statement in second last announcement, last one specified "most of functionality does not require GCN" which would say at least something is GCN reliant

please feel to correct me

[dungeon]

that was statement in second last announcement, last one specified "most of functionality does not require GCN" which would say at least something is GCN reliant

That sounds to me like "most of the world does not use dollar, but those who don't... well... they can adopting it"

[smitty3268]

sorry to butt on this one, especially since you probably know better than me.

that was statement in second last announcement, last one specified "most of functionality does not require GCN" which would say at least something is GCN reliant

please feel to correct me

That's like AMD saying DX12 requires GCN. Doesn't mean nvidia cards won't too.

[Kano]

Well Nvidia does not use the term GCN, but i think the current WDDM 2.0 is working with Kepler+Maxwell, most likely Fermi will do as well (all DX11) later.

[clementl]

Really? I don't see all 3 consoles being a failure. More to the point, Mantle will most certainly be adopted by vendors developing across PS4/XBox and Nintendo WII replacement.

Next Gen OpenGL won't be mature for another 24-36 months. Directx 12 will take another 12-18 months. Meanwhile, vendors are wise to use Mantle.

Any one who thinks R9 300 series GPGPUs won't fly off the shelves knowing vendors will be leveraging the crap out of them are very short-sighted in their reasoning.

PC Gaming and Mantle will only grow, right along-side Consoles. Nvidia's dominance will dwindle as more games get far more performance out of AMD GPGPUs and switch their backends to leverage it. This means more used Nvidia cards on EBay.

I'm kinda counting on this being a troll but I'll bite anyways.

Mantle will most certainly be adopted by vendors developing across PS4/XBox and Nintendo WII replacement

Well, both Microsoft and Sony stated not supporting Mantle of their consoles.
And how many cross-platform Wii U games are there really? Most developers walked away since it has too little horsepower, even with standard console optimizations. The Wii U doesn't even have a GCN GPU, which AMD stated to be Mantle's requirement.

Next Gen OpenGL won't be mature for another 24-36 months. Directx 12 will take another 12-18 months.

This is where I think you're mistaken. In the past API's still had to be designed. And most time consuming of all: there had to be a lot of communication with developers.
But in this case (Vulkan and DX12) we're talking about a mostly implemented standard, which a bunch of dev's already have experience with. Depending on how similar it is, AMD possibly get away with a 1:1 translation of their Mantle drivers to support either.

Any one who thinks R9 300 series GPGPUs won't fly off the shelves

I think even if AMD will produce the far better GPU's, they won't be flying off the shelves. The market for dedicated GPU's has changed. APU's sell like hot cakes, Intel is getting a larger market share all the time.

Nvidia's dominance will dwindle

While indeed being the largest, the Nvidia/AMD GPU ratio's are pretty good compared Intel/AMD or Windows/anything else.


Also keep in mind that those optimizations have shown to be almost exclusively beneficial when used with slow CPU/fast GPU combo's. The CPU is where any API forms the bottleneck.
I don't see any developer build a new back-end for such a small market (one vendor + one architecture + one OS), so there won't be any benefits for end-users so soon anyways. The biggest draw of Vulkan is supposed to be the cross-platform support (PC's and mobile devices alike), which is a theme recurring in all of the more successful engines (e.g. UE4, Unity). Cross-compatibility has become too important.