I feel like even though full vector architecture might perform a lot better, the use cases may be much narrower than SIMD, especially on typical desktop or server (web applications etc, not scientific computing, deep learning or image processing -- many of which are already vectorised on GPUs) workloads. As others have mentioned, SIMD allows you to do a little bit of vectorisation in an otherwise non-vector workload, or use it for the wider registers or whatever. I don't know enough about it personally to be able to judge either way, though. I just know that I've attempted to vectorise some hobby game code for fun a few times and typically found it much harder to achieve than it first seemed, even though the data seemed trivially vectorisable at first. Perhaps that's just lack of experience.
Comments
I feel like even though full vector architecture might perform a lot better, the use cases may be much narrower than SIMD, especially on typical desktop or server (web applications etc, not scientific computing, deep learning or image processing -- many of which are already vectorised on GPUs) workloads. As others have mentioned, SIMD allows you to do a little bit of vectorisation in an otherwise non-vector workload, or use it for the wider registers or whatever. I don't know enough about it personally to be able to judge either way, though. I just know that I've attempted to vectorise some hobby game code for fun a few times and typically found it much harder to achieve than it first seemed, even though the data seemed trivially vectorisable at first. Perhaps that's just lack of experience.