Even though Advanced Micro Devices openly demonstrates its next generation graphics processing unit, it has never confirmed specifications of the product. Nonetheless, multiple leaks have already revealed the vast majority of technical details about the upcoming AMD Radeon Fury graphics card and recently specs of the device were confirmed once again.
According to database of CompuBench benchmark, a test suit used to compare parallel compute performance of CPUs, GPUs and accelerators, AMD’s code-named “Fiji” graphics processor has 64 compute units, reports VideoCardz. Since every GCN compute unit has 64 stream processors, the total amount of SPs in the new GPU is 4096, as reported a number of times before. The chip is expected to feature 256 texture mapping units.
AMD’s Radeon Fury will be the first graphics card to use SK Hynix’s stacked high bandwidth memory (HBM). The new type of DRAM is expected to provide unprecedented memory bandwidth of 640GB/s thanks to 4096-bit memory bus and 1.25GT/s transfer rate. The graphics boards will be equipped with 4GB of HBM memory.
AMD’s upcoming flagship graphics card will use hybrid liquid cooling solution and will be physically shorter than today’s high-end graphics adapters.
Pricing of AMD’s Radeon Fury X remains unknown. It is expected to be formally introduced on the 16th of June, 2016.
AMD did not comment on the news-story.
Discuss on our Facebook page, HERE.
KitGuru Says: The main intrigue about AMD’s Radeon Fury X remains its actual performance. The graphics card is expected to demonstrate excellent results in high resolutions, but it is completely unclear how it will compare against Nvidia Corp.’s GeForce GTX Titan X. According to CompuBench, the new graphics processor from AMD is considerably faster than the Titan X. However, actual performance in games depends not only on pure compute performance.