To cater to ever increasing processing needs by large dataset applications, AMD has announced its latest Radeon Pro Solid State Graphics (SSG) – a technological leap that will provide more than an order of magnitude greater memory capacity compared to existing GPU memory implementations.
With increase in 4K video consumption, we aren’t far from the times when 8K will become the norm. Higher resolution videos mean greater memory and graphics demands as is the case with the next wave of demanding use-case scenarios including high-resolution rendering, VR content creation, oil and gas exploration, computational engineering, medical imaging and life sciences.
AMD wants to take the challenge head on and for the purpose is developing the new Radeon Pro solution. The company has already demonstrated the solution wherein 8K raw video timeline scrubbing was accelerated from 17 frames per second to a stunning 90+ frames per second. AMD says that the quality of life, productivity and efficiency of the 8K video post-production will increase thanks to the new Radeon solution.
Pointing out the current scenario, AMD says that current limitations require slices of data to be processed individually and later merged by software, and this causes significant latency for fetching additional data from system memory. Because of these issues, developers in these domains often refrain from leveraging the advantages of the GPU.
AMD addresses this with its Radeon Pro SSG solution. The new solution is poised to speed-up processing for large dataset applications with and will also facilitate new application experiences by utilizing data persistence of non-volatile memory. Increased amount of local memory on the graphics card is known to speed-up processing for many applications – specifically those with large datasets – and the new technology will allow increased adoption of GPUs for many graphics and compute applications.
The new GPU technology enables one terabyte of extended frame buffer thereby allowing much larger datasets to be loaded locally, connected over a dedicated PCIe® bus. This means that the GPU will be offloading fewer requests to the CPU and instead it will look first in the local frame buffer and only needs to involve the CPU if the data is not already in the extended frame buffer.
AMD is currently accepting applications for developer kits, which are available for $9,999. Full availability is planned for 2017.