Combine and Ati and an Nvidia GPU?

[Lyonheart]

Not sure why you would want to do this.

Maybe i misunderstood it.. but it sounds like DX12 harnesses more of the power of multi GPUs, and ALL of the combined ram, than currently.. The combining of multiple types of GPUs is a side thing to me.

[Ughmahedhurtz]

Maybe i misunderstood it.. but it sounds like DX12 harnesses more of the power of multi GPUs, and ALL of the combined ram, than currently.. The combining of multiple types of GPUs is a side thing to me.

I suppose it depends on how they plan to optimize things. How would you put two GPUs in the same system where they supported different features? Least common denominator? How would they get the timing right so you got smooth framerates? I've been out of the video driver business for some years but unless they're turning GPUs into GPGPUs, I don't see how this benefits a video card manufacturer in terms of standing out from the crowd. NVidia and ATI have their own "special sauce" e.g. PhysX, EyeFinity, CUDA, etc.

It'll be interesting to see what the point is once we have some better clues, and what NVidia and ATI have to say about it.

[MiRai]

Maybe i misunderstood it.. but it sounds like DX12 harnesses more of the power of multi GPUs, and ALL of the combined ram, than currently.. The combining of multiple types of GPUs is a side thing to me.

Assuming it's not just marketing hype BS, the developer is going to have to use their own time to implement these things and it's not something that's going to "just work" when a game comes standard with DX12 (which is likely going to be years from now anyway) -- We're talking both multi-GPU between different vendors and adding the VRAM together. Having glanced at a few threads about all of this, I read, perhaps, the perfect quote:

You make it sound like having a multi-GPU setup's performance limited to whichever vendor's card is slowest for a given scene, featureset limited to the intersection of card features both could support reliably, on drivers never meant to work in concert with their competitors, with a PC installed with two proprietary driver packages with separate update methods and schedules, drivers that frequently have problems with conflicts with older versions of themselves, with different .NET or other environmental requirements, on hardware that was not designed/developed/tested in the presence of a competitor that frequently has problems with differing hardware from the same vendor, with low-level differences in execution behavior, little history of cooperative implementations, multiple render paths in an engine that were never designed/coded/tested to run simultaneously, in a platform that has at best problematically supported switching between one GPU or the other in mobile, between vendors who have every interest and an ongoing history of getting in each other's way could lead to undesirable outcomes.

[Ughmahedhurtz]

I was trying to not come off just straight negative, but that post pretty much sums up what I was thinking.

[MiRai]

I was trying to not come off just straight negative, but that post pretty much sums up what I was thinking.

Haha, I didn't mean to sound like a dick, but I find it so odd that something like this is even being marketed with DX12, and it's almost as if Microsoft was just pulling ideas from a hat labeled "things that just don't make any sense and nobody's asking for."

"I really wish I could buy an AMD GPU to run in SLI with my current GPU," said no nVidia GPU user ever. (the reverse is also 100% true for any AMD GPU user).