One of the biggest issues that exists when it comes to PC gaming performance has to do with the fact that GPUs inside different graphics cards from all brands don’t speak the same language and using a real-time translator is required. , which prevents you from taking full advantage of it. Why don’t they all speak the same language if they are going to run the same games and programs, at least in the visual part?
Why is a universal ISA GPU needed?
When it comes to running programs, it is normal that all processors compatible with an ISA can run programs designed for it without problems, unless they are extensions to all of the registers and instructions thought well after launching said program. In other words, the binaries work perfectly and there is no need to compile them again for each new chip and architecture that appears on the market.
Unfortunately, that’s not the case with GPUs, where Intel, AMD, and NVIDIA not only use a different set of registers and instructions not only between themselves, but also between generations of graphics cards. This means that while the game is running, the various shader programs need to be compiled to save them to a file on the hard drive which will be retrieved later. This causes early games when hitting certain areas of a game for the first time to have huge frame drops on PC, although that’s not the only issue.
Having to store so much code written in high-level shaders, whether in HLSL, GLSL or any other such language, requires a greater amount of video memory. All this makes us wonder why NVIDIA and AMD, and if we also have the rest of the GPU manufacturers and designers, do not want to create a universal ISA GPU, which will allow shaders to be compiled on any platform for better performance, avoiding the shader compilation issues that exist on PC.
The Steam Deck is the perfect example.
Valve’s console offers a series of games with a profile designed for it which, among other things, includes shaders already compiled for the ISA GPU it uses, that is, RDNA 2, which, curiously, is the same as in consoles such as Xbox series and Play Station 5. So there is no shader pre-compilation process and all its associated problems. However, on PC this is impossible, given the large number of graphics card models that exist and the differences between them.
Why was this solution not applied?
One of the historical curiosities happened today with the first Xbox where the GPU vendor was envious and took advantage of their advantage to price Microsoft well above but could afford which led Redmond to seek reverse engineered or rather to use a chip from its rival ATI for its console, but given the ISA exclusivity of each of the chips, this was not achieved, although it did help ATI Technologies to become the supplier of the Xbox 360 graphics chip.
Currently ATI Technologies is part of AMD, it’s the current Radeon Technology Group, and of the three big brands, only Lisa Su’s has the ability to deliver a processor and graphics card in a single chip with the necessary power to a console. NVIDIA is completely lacking in x86 licensing and Intel is no one yet when it comes to graphics cards. So a universal ISA GPU wouldn’t be a problem, and consoles aren’t the issue.
Rather, it is the ecosystem that NVIDIA has created through CUDA today and with which it has absolute dominance in various markets. If a universal ISA were designed for GPUs, the green brand would lose its edge because envy-created applications and libraries for its hardware would run flawlessly over the competition, even if it temporarily had a first-mover advantage.