With just over a month until the expected launch, more information on AMD’s Radeon R9 380X have surfaced. Last time around, we got a glimpse of the XFX Double Dissipation model and today we’re treated to the full specifications
Unlike the R9 285/380, the 380X will feature the full Tonda die. Tonga was originally launched last year cut down to 1792 shader units, 112 TMUs and 32 ROPs over a 256bit GGDR5. Keeping the same 256bit bus, the 380X will feature 2048 shader units, 128 TMUs and 32 ROPs, a hardware parity with the aged 280X. Given the architectural improvements GCN 1.2 introduced starting with Tonga, the 380X should place at least 10% faster than its predecessor. Clock speeds also get a boost up to between 1000~1100Mhz while GDDR5 speeds will stay about the at around 5500Mhz~6000Mhz.
With better performance and DX12 support, among other advantages compared to the GTX 770, AMD has a good chance to dominate the $150 gap between Nvidia’s GTX 960 and 970. The 380X may also come standard with 4GB of VRAM as 2GB is probably a bit too low for this tier of performance. If the 380X does well in the market, it will be interesting to see if Nvidia will respond with an even more cut down GM204 in the form of a GTX 960Ti, cut prices on the 970 or simply just wait it out till Pascal.
Thank you HWBattle for providing us with this information