AMD Mantle Support?

AMD and DICE have just announced a new graphics API called Mantle it aims to reduce the CPU overhead and improve graphics performance with 9x draw calls per frame reported.

http://hothardware.com/News/AMD-Brings-LowLevel-HighPerformance-Graphics-To-PCs-With-Mantle-API/

ED Mantle support?

Very interesting yes. If FD were able to incorporate this into their COBRA engine that would instantly ease up development for next gen consoles AND PC's. I wonder if the API will be open enough to also be incorporated by NVIDIA but I'm affraid that is a little bit too much to ask.

EDIT: Ninja'd by Pecisk
 
Perhaps Frontier will update the game gradually, when new technologies graphics are created. Surely it takes time to adapt the game code or embed it into Cobra
 
Love the enthusiasm people, but wild speculation aside this is a flawed way to look it this.

It is a great idea and along with their unified memory architecture, it is something that I will have great interest in as it develops; but you have to face the facts here.

It is vendor specific, nVidia are never going to use this as it only works with AMD GPUs. Personally I hope that nVidia implement a similar system so they can be compared and competitively improved.

Unfortunately I would not want to have to write an additional code path in my renderer for each vendor implementation and I assume most other developers would feel the same.
 
Unfortunately I would not want to have to write an additional code path in my renderer for each vendor implementation and I assume most other developers would feel the same.

They are already doing it with DX for most of graphics intensive games. This is just embracing harsh truth.
 
Render paths.

I suppose, but most engines these days are exclusively DX, GL or both.
This takes us back to the days of Glide, powerVR etc. where do you draw the line?

Layers of abstraction are there for a good reason, they make implementing your code easier and faster. Yes you get better performance by working at a lower level, but you have to weigh up the pros and cons or each choice.

Like I said, I find the idea very interesting and I can't wait to see more, but I don't believe it is worth while for FD to uproot their engine to add this. Much in the same way that Oculus Rift support just does not make sense for them just yet.
 
Sounds really interesting. I think for the sake of ED and its future, the more rendering technologies that are available and supported, the better. DirectX and OpenGL might not be around in the future and we don't want to have to run ED in 20 years time in a sort of DOS Box environment do we?
 
Sounds really interesting. I think for the sake of ED and its future, the more rendering technologies that are available and supported, the better. DirectX and OpenGL might not be around in the future and we don't want to have to run ED in 20 years time in a sort of DOS Box environment do we?

I think OpenGL will be around for a very long time to come, and maybe DirectX too, but I'm not so sure about a proprietary API that ATI are offering. I too see this as a Glide equivalent. But that really didn't take off either (pun intended).

If it was an open API that all graphics cards manufacturers could support in their drivers, then I think it might have a chance. But without that, they are just serving up another PhysX.
 
I think OpenGL will be around for a very long time to come, and maybe DirectX too, but I'm not so sure about a proprietary API that ATI are offering. I too see this as a Glide equivalent. But that really didn't take off either (pun intended).

If it was an open API that all graphics cards manufacturers could support in their drivers, then I think it might have a chance. But without that, they are just serving up another PhysX.

I agree. Still they could always code it all directly in ASM. No chance of x86 getting thrown away now is there.... :p
 
Think of it from the game engine providers perspective, if you adopt this rendering API and take advantage of it's features your games run faster with lower latency and you can add more umph to your game.

As AMD are providing the GPU/CPU solutions to both consoles, Mantle is probably going to be a go to API for AAA developers.

The API is Open, so there is nothing stopping NVIDIA from adopting it.

It sounds to me that due to legacy and complexity issues the potential performance you can get from a multicore CPU and modern GPU is being limited by DirectX and OpenGL.

Maybe Mantle will be limited to AMD / XBox One / PS [n?].

But hopefully it will kickstart DX and OpenGL to provide similar features and adapt so they are not getting in the way of performance.

End result we should be getting faster better looking games. :D
 
...
The API is Open, so there is nothing stopping NVIDIA from adopting it.
...

You mean apart from Mantle being a software interface to proprietary AMD hardware? Sure if nVidia are going to ditch their GPU business over night?
 
Last edited:
You mean apart from Mantle being a software interface to proprietary AMD hardware? Sure if nVidia are going to ditch their GPU business over night?

Well if you take open to be like OpenGL where both vendors NVidia and AMD provide their own bespoke API features that over time are rolled into the standard version.

The main features of Mantle are the way in which it allows multiple CPU's to work with multi-core GPU's. Which is similar to the Compute/'Stream'/CUDA interfaces Nvidia and AMD have been providing to allow parallel processing across the CPU and GPU.
 
Excellent article, thank you for finding that.

You have to remember that API's are just layers so I still can't agree about nVidia ever using Mantle however, it's analogous to Microsoft building Windows on top of a Linux kernel.

Indeed it would be a sort of a WINE. Maybe feasible, but with a performance penalty. Not that nVidia deserves anything else though.

But I disagree that Mantle is Glidev2 - it has the potential to become much more.

Glide has some similarities to what is known of Mantle - somewhat "close to the metal" and basically only run on a proprietary HW. But while Glide came to be inferior to open standards, namely what became Direct X, Mantle is promoted as overcoming the crud that accumulated on the high level open standards and give more performance and power to the developers. All it needs to entice them is a market.

And with the consoles they have that market.

Anyway, IF Mantle is that more efficient and good dev tools are available, it could become the primary development target API. That would not mean that PC games would be developed solely for Mantle enabled cards, but *IF* "Mantle code" is very easily converted into DX/Open GL compatible language, then we could see plenty of devs targeting Mantle and also creating a DX/OGL path for non Mantle cards. All Console devs would probably go down that path.

It will depend on the tools available, whether Mantle is included in game engines, and the performance advantage of Mantle. But it can become a very good thing for a mostly stagnated market in terms of revolutionary stuff.
 
Back
Top Bottom