RTX 2080 Super has arrived on the market today, and with this launch a somewhat unexpected detail has been discovered. Apparently, the performance of your Samsung GDDR6 memory is lower than it could be. These are memory modules that can reach 16 Gbps, but Nvidia has limited them to 15.5 Gbps. If you reach the potential figure, the memory of the RTX 2080 Super could reach a performance of 512 GB / s of performance, which is only a 3% improvement in bandwidth, but the question is obvious: Why why Nvidia has decided to reduce this figure in its new 2080 Super?
According to Anandtech, the green has wanted to manufacture the RTX Super series in the simplest way, with the same Turing chips that already existed, as well as the Same heatsinks and PCBs. The original line has 14 Gbps GDDR6 memory. The PCBs that Nvidia uses for it are designed for that bandwidth, however, to be able to place memories at 16 Gbps, the company should have redesigned the PCB. Being this Super series a slight renewal of the product, why design a new PCB only for one of the models?
The good thing about this is that most of the RTX 2080 Super can reach 16 Gbps through overclocks with ease, since the memory is designed for it. On the other hand, with this release it is confirmed that Samsung is perfectly capable of manufacturing large-scale 16 Gbps GDDR6 memory modules. This is good news for future generations of Radeon and GeForce graphics, especially after the Korean has announced its 18 Gbps GDDR6 memories. All in all, the memory of this 2080 Super is still 10% faster than that of the 2080 model.