It looks like you're new here. If you want to get involved, click one of these buttons!
Technically speaking, yes ECC (Error Correction) uses redundancy (Duplicating bits and sending them multiple times through error correction circuits.) This extra encode/decode step will naturally cause more latency and propagation delay.Torval said:Does ECC ram come with a performance hit in latency or something such as that? I can't imagine error correction comes for free so I would think so. But I've always wondered why the tech manufacturers have kept this as an enterprise only feature for so long.Ridelynn said:If you are doing very detailed design work, such as complicated CAD or other 3D modeling, doing professional grade production rendering (either proofing or post production), medical imaging, rocket science, working on advanced AI or simulation - just some examples.albers said:So, what compute video cards are good at? How it will be useful for me to have giga, tera, mega flops of memory bandwitch ?
Gamers definitely don't need ECC VRAM or Gigateramegaflops of memory bandwidth. Most of that is for very high precision and/or low fault tolerance work. There is a market for it, and you would know if you needed it.
I'm comparing nvidia's performance with graphics libraries and api's. They lack on DirectX 12, and OpenGL 4.3+ but beat AMD hands down (New cards.) in backwards compatibility: DirectX 9/10 / OpenGL 2/3. In relation to new cards running older software.Cleffy said:Right now, nVidia's driver support for obsolete hardware is pretty bad compared to AMD. There is a reason why AMD GPUs have had a fine wine effect. They are only supporting 2 architectures for consumer GPUs. To cover the same amount of generations, nVidia would need to offer driver support for 6 architectures. It also translates to better backwards compatibility with older games for AMD as their drivers are built upon the previous generation. Right now AMDs driver support is much better than nVidias with more frequent updates and more stable drivers.
I think what you meant is the advantage nVidia has with certain DirectX versions. The cards are just more efficient at it now. Since DirectX 10, nVidia has been lagging at implementation of the standard. As a result AMD typically played newer DirectX versions better until nVidia eventually caught up and surpassed AMD. I expect the same to happen with Volta. With Vulkan, the API is based on Mantle, so AMD will have an advantage here for some time.
what about players that have bought the original SW ? Do we get something more compared to the new players of SWL ?