Dual GPU cards may finally be able to combine their VRAM.

[Thracks]

Would all of these features work on the R9 290X? Because I might be replacing my faulty GTX 970 with one after the RMA is sorted.

 

Would work on any GCN-based product.

 

It is very important to note that Mantle is like any other API, though, and a developer has to choose to support the feature. Here and now, developers are much more interested in simply learning the API and cutting down software overhead. Getting better multi-threading, reducing overhead and having an API that can fully saturate the GPU resources are mission critical for the next 10 years in gaming. Collectively, the development community just started on that with Mantle, so it will take time to explore the more exotic features.

 

//EDIT: I'm reminded that the initial applications with any API are relatively rudimentary. Things like contact-hardening shadows, global illumination, compute depth of field were unthinkable in the earliest DX11 titles. But here we are, global illumination can be done with DX11 now, and was shipped in a real game in 2012(ish). So, yeah, this is normal for the industry.

[Dabombinable]

Would work on any GCN-based product.

 

It is very important to note that Mantle is like any other API, though, and a developer has to choose to support the feature. Here and now, developers are much more interested in simply learning the API and cutting down software overhead. Getting better multi-threading, reducing overhead and having an API that can fully saturate the GPU resources are mission critical for the next 10 years in gaming. Collectively, the development community just started on that with Mantle, so it will take time to explore the more exotic features.

I don't get why this wasn't done years ago. I've noticed that a lot of games offload visual aspects such as physics to the CPU when everything graphical should be handled by the GPU. It will be interesting to see the performance gains I'd get on my Old Core 2 and Xeon, maybe for once they wouldn't bottleneck graphics cards.

[flibberdipper]

Finally.

[Frankz]

AMD, loving the innovation these days . 

I think its gonna be a great year for them

Yea I really hope they knock it out of the park with their upcoming GPUs. I might have a GTX780 but I would go AMD instantly if they deliver. Started to dislike Nvidia more & more over the past couple months, even before the 970 fiasco.

[Humbug]

Actually if they make Mantle completely open and then bake the tech into it...that would force nvidia to take a closer look at building something on mantle. Or they'll just build there own system in response. Honestly the tech will come to both sides of the fence eventually, i'd just prefer not to segment the market more than it already is over arbitrary in game features. I just think if they baked it into something of their own, even if 100% open it'll make nvidia down create thar version just because pure ridiculous pride wouldnt let them use anything built by AMD.

I think now there is not much point in opening up mantle itself. Instead we need to all get behind a related truly cross-platform API

http://linustechtips.com/main/topic/303393-next-generation-opengl-to-be-shown-at-gdc/#entry4125491

[silberdrachi]

I think now there is not much point in opening up mantle itself. Instead we need to all get behind a related truly cross-platform API

http://linustechtips.com/main/topic/303393-next-generation-opengl-to-be-shown-at-gdc/#entry4125491

 

Mantle served its purpose, to force the industry to move forward and improve APIs and drivers that have stagnated and in some cases hardly functioned for the new demands that games have. Opening up mantle would be nice, but at this point it doesnt matter as much as it wouldve if they had when they first announced it.

 

That being said, OpenGL is going to have a very hard time gaining traction without having something amazing to really kickstart it forward.

[Humbug]

Mantle served its purpose, to force the industry to move forward and improve APIs and drivers that have stagnated and in some cases hardly functioned for the new demands that games have. Opening up mantle would be nice, but at this point it doesnt matter as much as it wouldve if they had when they first announced it.

Agreed

 

That being said, OpenGL is going to have a very hard time gaining traction without having something amazing to really kickstart it forward.

If the rumours are true then writing to Direct X 12 and writing to OpenGL Next are both quite similar techniques to Mantle. The design goals are certainly almost exactly the same.  

 

So it's the age-old OpenGL vs DirectX thing again. If we consider this as a kinda fresh start then the fact that in the last few years previous iterations of DirectX have won over previous iterations of OpenGL doesn't really matter any more. With a ground up redesign of openGL we can be sure that the technical issues of the previous version and the legacy stuff will not be carried over.

 

The advantage that OpenGL has is that it can be truly vendor neutral and platform indepdant. Andriod, IOS, Windows, MacOSX, Linux, consoles... whatever. A good implementation has the potential to simplify our world and you can see why it would be more appealing to devs than DirectX.

 

However people who are more knowledgable about this subject than me tell me that DirectX with Microsoft's backing has for the last few years had better documentation and better tools which makes it more appealing. Obviously if the new OpenGL is to make a challenge that is something that has to change.

 

GDC should be interesting.

[marldorthegreat]

It technically does not "double" the vram. It just allows ALL the vram to be used

[Will_Man_Dude]

Finally the autobahn hammer will have 24gb of Vram (more than all the ram in my own system and school laptop)