Mantle is Over (for pc gaming)

[WiseWilson]

So mantle is done before it gets going as dx12 supplents a need for it. Think mantle pushed dx12 to have the close to metal features or not? They are releasing the documents for the api and keeping the SDK private.

AMD Gaming: On APIs and the future of Mantle | AMD Blogs

Key quote
The Mantle SDK also remains available to partners who register in this co-development and evaluation program. However, if you are a developer interested in Mantle "1.0" functionality, we suggest that you focus your attention on DirectX® 12 or GLnext.

 

[SKYMTL]

Correction. Mantle is more or less evolving into Khronos' Vulkan API.

 

[Unxetas]

Makes sense to me.. Looks like they at the very least helped kickstart DX12, which I bet must have been their goal to begin with? They've made their point and now they can focus on something else and let Microsoft do their thing.

 

[WiseWilson]

Correction. Mantle is more or less evolving into Khronos' Vulkan API.

I stand corrected wonder if this will lead to better linux performance in the future, Vulkan is the next version of opengl correct?

 

[IRQ Conflict]

DX 12 is gonna really take advantage of multiple core CPU's as well is it not? That would be HUGE for physics.

 

[SKYMTL]

DX 12 is gonna really take advantage of multiple core CPU's as well is it not? That would be HUGE for physics.

DX12 is all about optimizing pipeline operations.

More here: http://www.hardwarecanucks.com/foru...-12-detailed-dx11-gpu-compatibility-more.html

 

[Dzzope]

At the end of the day it's down to each dev team whether or not they implement certain functionality.. DX12 titles does not automatically mean they will be better for xyz feature of DX12

As for mantle pushing dx12 to be closer to the hardware or such.. I highly doubt it, usually the lead time of these projects are quite long.. That said unless we hear from the horses mouth(Microsoft DX team) we will never really know.

 

[SKYMTL]

I think it is a chicken versus egg scenario with DX12 and Mantle. DX12 has been in development for years, as has Mantle. I think BOTH MS and AMD talked to developers and received the same feedback: give us more control over resource allocation.

 

[Fragman]

"Mantle must take on new capabilities and evolve beyond mastery of the draw call. It will continue to serve AMD as a graphics innovation platform available to select partners with custom needs.

1. The Mantle SDK also remains available to partners who register in this co-development and evaluation program. However, if you are a developer interested in Mantle "1.0" functionality, we suggest that you focus your attention on DirectX® 12 or GLnext.

As AMD have told we will hear more tomorrow. I am guessing there will be some details on Mantle 2.0. AMD wants to use the API to showcase the best of their GPUs. Mantle needs to be something to differentiate AMD's product just as Nvidia has PhysX or Gameworks. Mantle can be even more closer to metal than DX12 as AMD designs both the hardware and the software (Mantle drivers). AMD has to get the major engine developers like DICE (Frostbite), Crytek (Cryengine) and few of the other big guys to keep supporting Mantle's latest version and features. Epic is the big challenge as they are firmly in Nvidia's court with Gameworks and Physx. To get them onboard is AMD's biggest challenge.

Unless you are a studio like DICE, Epic or Crytek whose game engines are widely licensed or a studio who has their own internal game engine used widely across multiple game franchises it does not make much sense to invest resources in Mantle. The smaller studios are best served by just sticking with DX12 and/or Vulkan.