NVIDIA's GeForce RTX 20 Super series launch in two phases across July is topped off by the new GeForce RTX 2080 Super. Designed to displace the RTX 2080 from the company's product stack at $700, the RTX 2080 Super not just maxes out the 12 nm "TU104" Turing silicon, but also comes with higher GPU clock speeds and faster 15.5 Gbps GDDR6 memory. NVIDIA's launch of the RTX 2080 Super may have been necessitated by, of all things, the $399 Radeon RX 5700 XT.
AMD's Radeon RX 5700-series "Navi" launch destabilized NVIDIA's lineup. The RX 5700 XT outperforms the RTX 2070, and the RX 5700 beats the RTX 2060, which forced NVIDIA to launch the RTX 2060 Super and RTX 2070 Super to recapture the $399 and $499 price points. This had a ripple effect on the RTX 2080 because the RTX 2070 Super is carved out from the same "TU104" silicon and the original RTX 2070 maxes out the "TU106". The RTX 2070 Super performs within 6-8 percent of the RTX 2080 while being $200 cheaper. Custom-design, factory-overclocked RTX 2070 Super cards narrow the performance gap to as little as 3-4 percent. NVIDIA hence had to refresh the RTX 2080 lest it gets cannibalized by the RTX 2070 Super. We thus have the new RTX 2080 Super.
NVIDIA's objectives with the RTX 2080 Super are to increase the performance gap with the RTX 2070 Super, so it can justify selling it at a $200 premium, at $700, and avoid tapping into the larger "TU102" silicon, which would increase costs. The company's approach hence is to supercharge the "TU104". For starters, all 3,072 CUDA cores physically present on the chip are enabled. The original RTX 2080 only features 2,944 of them. Secondly, the GPU Boost frequency is dialed up by a further 100 MHz, to 1815 MHz (up 6 percent). Lastly, the company increased the memory bandwidth by 11 percent by dialing memory clock speeds up to 15.5 Gbps using fast 16 Gbps-rated memory chips.
NVIDIA is also allowing its add-in card (AIC) partners to innovate custom-design RTX 2080 Super graphics cards with generous factory-overclocked speeds and meaty cooling solutions. With prices starting at $700, NVIDIA partners have $200 to mark up prices of their premium offerings because beyond $900, you should instead aim for the RTX 2080 Ti.
In this review, we have with us the EVGA GeForce RTX 2080 Super FTW3 Ultra. The card uses a strong 12+2 phase VRM and a large triple-slot, triple-fan cooling solution. On top of that, the card features EVGA's iCX sensor technology, which adds nine additional sensors at strategic location throughout the PCB. EVGA's FTW3 is clocked at 1845 MHz rated boost, while the memory is left untouched at 15.5 Gbps. EVGA is pricing the card at $790, which is a $90 premium over the $700 MSRP for the RTX 2080 Super.
|RX Vega 64||$400||4096||64||1247 MHz||1546 MHz||953 MHz||Vega 10||12500M||8 GB, HBM2, 2048-bit|
|GTX 1080 Ti||$700||3584||88||1481 MHz||1582 MHz||1376 MHz||GP102||12000M||11 GB, GDDR5X, 352-bit|
|RX 5700 XT||$400||2560||64||1605 MHz||1755 MHz||1750 MHz||Navi 10||10300M||8 GB, GDDR6, 256-bit|
|RTX 2070||$440||2304||64||1410 MHz||1620 MHz||1750 MHz||TU106||10800M||8 GB, GDDR6, 256-bit|
|RTX 2070 Super||$500||2560||64||1605 MHz||1770 MHz||1750 MHz||TU104||13600M||8 GB, GDDR6, 256-bit|
|Radeon VII||$680||3840||64||1802 MHz||N/A||1000 MHz||Vega 20||13230M||16 GB, HBM2, 4096-bit|
|RTX 2080||$630||2944||64||1515 MHz||1710 MHz||1750 MHz||TU104||13600M||8 GB, GDDR6, 256-bit|
|RTX 2080 Super||$700||3072||64||1650 MHz||1815 MHz||1940 MHz||TU104||13600M||8 GB, GDDR6, 256-bit|
|EVGA RTX 2080 Super|
|$790||3072||64||1650 MHz||1845 MHz||1940 MHz||TU104||13600M||8 GB, GDDR6, 256-bit|
|RTX 2080 Ti||$1100||4352||64||1350 MHz||1545 MHz||1750 MHz||TU102||18600M||11 GB, GDDR6, 352-bit|