I will not even call this a thinly-veiled rant. Linus admits it. To make a point, he assembled a $5000 PC running a pair of NVIDIA GeForce 780 Ti GPUs and another pair of AMD Radeon R9 290X graphics cards. While Bitcoin mining would likely utilize all four video cards well enough, games will not. Of course, he did not even mention the former application (thankfully).
No, his complaint was about vendor-specific features.
Honestly, he's right. One of the reasons why I am excited about OpenCL (and its WebCL companion) is that it simply does not care about devices. Your host code manages the application but, when the jobs get dirty, it enlists help from an available accelerator by telling it to perform a kernel (think of it like function) and share the resulting chunk of memory.
This can be an AMD GPU. This can be an NVIDIA GPU. This can be an x86 CPU. This can be an FPGA. If the host has multiple, independent tasks, it can be several of the above (and in any combination). OpenCL really does not care.
The only limitation is whether tasks can effectively utilized all accelerators present in a machine. This might be the future we are heading for. This is the future I envisioned when I started designing my GPU-accelerated software rendering engine. In that case, I also envisioned the host code being abstracted into Javascript – because when you jump into platform agnosticism, jump in!
Obviously, to be fair, AMD is very receptive to open platforms. NVIDIA is less-so, and they are honest about that, but they conform to standards when it benefits their users more than their proprietary ones. I know that point can be taken multiple ways, and several will be hotly debated, but I really cannot find the words to properly narrow it.
Despite the fragmentation in features, there is one thing to be proud of as a PC gamer. You may have different experiences depending on the components you purchase.
But, at least you will always have an experience.
Sooo, this means i can run
Sooo, this means i can run Nvidia Quadro (or AMD’s Firepro) for work, and then power up the secondary consumer GPU such as Radeon (or Gforce) for gaming, and all I need is two PSI 8x slots?
in short, YES.
in short, YES.
Fully agreed.
As a experiment
Fully agreed.
As a experiment right now I have a 680GTX and 290X.
the 680 is for rendering and any Cuda task, the 290x is for gaming/mining :P.
Still, for Borderlands 2.. I can not get Physix to work in any shape. I really don’t like the idea of having two cables to my screen. But would like to see how different the experience is.
In part, I would build a system that was mentioned here, or at least SLI/Crossfire combo. too bad they couldn’t be used to perform a single task at once. even for rendering… unless I render on the Nvidia and play on the AMD at the same time 😉
Nivida just offered initial
Nivida just offered initial open source support for K1 graphics, I hope that Gabe is evaluating any Tegra K1 based tablet platforms built around the K1 for a Steam OS based tablet, it would be a shame if the K1 were relegated to mostly Android, when the Tegra K1 will come with the full desktop versions of OpenGL, OpenCL, and other drivers that were before only available on PCs/Laptops! Some of the demos at CES with streaming to a tablet from a PC or server, home or from online, makes me want to build a powerfull headless gaming server for streaming to a tablet/s for highend gaming and other home based cloud inside the server closet uses. Maybe even having the home server system and connected tablets become a home based distributed computing platform!
there are cheap x86 tablets
there are cheap x86 tablets available you could try loading steamOS on one of those
Not with Kepler graphics!
Not with Kepler graphics! Screw Intel’s x86 with its overpriced graphics, and crappy driver support! I want a STEAM OS based tablet, and the Tegra K1 will have plenty of GPU power, and much better graphics driver support. I also want a tablet built for, and Branded as Steam, for all its open Debian based goodness, and no WINTEL strings attatched.