NVIDIA GeForce GTX 1660 SUPER Clock Rate Confirmed, Will Beat The AMD Radeon RX 5500 XT

Igor over at Igor's Lab has managed to confirm the GTX 1660 SUPER's clock rate as well as the memory standard it will be shipping with - allowing us to get a very good idea of how the theoretical performance of this upcoming NVIDIA GPU is going to stack up against the likes of the AMD RX 5500 XT. Since I don't really like leaving someone hanging, here's the overview: the NVIDIA GTX 1660 SUPER is going to beat the AMD Radeon RX 5500 XT but not a possible RX 5600 XT (this card is only rumored at this point).

NVIDIA GeForce GTX 1660 SUPER will be clocked at 1785 MHz and feature GDDR6, performance just under the AMD Radeon RX 5600 XT

The problem, as many of our readers would remember, is that NVIDIA's standard GTX 1660 variant supports only GDDR5. Considering AMD's Radeon RX 5500 is launching soon with GDDR6 and clocked at roughly 1.7GHz, this would mean NVIDIA's vanilla 1660 stands no chance against the RDNA powered competitor in red. The answer, it turns out, is to equip the card with GDDR6 memory as well. With memory clock rising to 14 Gbps from a mere 8 Gbps and a boost clock that is much higher than the succeeding GTX 1660 TI, the GTX 1660 SUPER offers 94 to 96% of the performance at a much lower price point.

Related StoryHassan Mujtaba
Colorful Unveils GeForce RTX 3060 iGame Bilibili Edition While Colorfire Launches GeForce GTX 16 Series Pink-Colored Graphics Cards

In terms of specifications, the NVIDIA GeForce GTX 1660 SUPER would utilize the TU116-300 die, same as the existing GeForce GTX 1660. It would feature the same core config of 1408 CUDA cores, 80 TMUs, and 48 ROPs.

This also means that the sales of the Ti are going to get cannibalized since the performance per dollar value of this card is going to be superior to the vanilla Ti. Using the core count, clock rate and memory Igor calculated the expected performance of not only the NVIDIA GTX 1660 SUPER but the AMD RX 5500 XT as well. He also included a possible RX 5600 XT. Why? Well, the difference between the RX 5700 XT and RX 5500 XT is huge, which means AMD will probably roll out a card to fill this gap in the near future (this is how yield theory works after all).

Performance estimation based on architectural calculations using core count, clock speed and memory.

According to Igor's calculations, the GTX 1660 SUPER will be 2.8% faster than the AMD RX 500 XT and 4.3% slower than the GTX 1660 Ti. The rumored AMD RX 5600 XT will be 11% faster than the GTX 1660 SUPER, meaning it is probably going to face off against the next rank and file in NVIDIA's lineup rather than the GTX 1660 SUPER. Pricing of the GTX 1660 SUPER is not confirmed at this point but considering the original retailed for $219, you are probably looking at a price point of anywhere from $200 to $279 to account for the more expensive GDDR6 memory vs. strategic price cuts. This memory is faster than the GeForce GTX 1660 Ti which features 12 Gbps dies. The memory would be featured along with a 192-bit bus interface and deliver a total bandwidth of 336 GB/s.

It is worth noting that Igor has confirmed the existence of the GTX 1660 SUPER - so this card no longer has rumored status in our books. According to Videocardz, the GTX 1660 SUPER is going to be launching near the end of the month on October 29, with availability next month. Things are starting to get heated in the mainstream segment where the majority of GPUs are shipped and gamers are soon going to be able to purchase more fps for the same dollars.

NVIDIA GeForce 16 Series Specifications

 GeForce RTX 2060 FEGeForce GTX 1660 Ti 6GB
GeForce GTX 1660 SUPER 6GB
GeForce GTX 1660 6GB
GeForce GTX 1650 SUPER 4 GBGeForce GTX 1650 4 GBGeForce GTX 1630 4 GB
Architecture (GPU)
CUDA Cores
19201536 140814081280896512
Tensor Cores
240N / AN / AN / AN / AN / AN / A
RT cores
30N / AN / AN / AN / AN / AN / A
Texture Units
120 96 8888805632
Base Clock
1365 MHz1500 MHz1530 MHz1530 MHzTBD1485 MHzTBD
GPU Boost
1680 MHz1770 MHz1785 MHz1785 MHzTBD1665 MHz1800 MHz
Memory Bus
Memory Clocks
14 Gbps12 Gbps14 Gbps8 Gbps12 Gbps8 Gbps12
L2 cache
3 MB1.5 MB1.5 MB1.5 MB1.5 MB1.5 MB1.5 MB
160 W120W125W120W100W75W>75W
10.8 billion6.6 billion6.6 billion6.6 billion6.6 billion4.7 billion4.7 billion
Die Size
445 mm²284mm2284mm2284mm2284mm2200mm2200mm2
Which NVIDIA GeForce 16 series card is more well-positioned against AMD's entry-level Navi?
WccfTech Tv
Filter videos by