NVIDIA GeForce GTX 1660 Ti Max-Q vs NVIDIA GeForce RTX 3050 6GB Laptop GPU

The Nvidia GeForce GTX 1660 Ti with Max-Q Design is a mobile graphics card for laptops that is based on the Turing architecture (TU116 chip). It is designed for thin and light laptops and about 10-15% slower than a regular GTX 1660 Ti for laptops (depending on the cooling capabilities). According to the specifications, the Max-Q variant clocks 22% slower for the base speed and 16% for the boost speed.

Compared to the faster RTX 2000 GPUs (e.g. RTX 2060), the 1660Ti integrates no Raytracing or Tensor cores. The performance is similar to the old GTX 1070 (Max-Q) but at a reduced TGP of 60 Watt (versus 80 of the laptop version 115 Watt of the desktop version). 

In April 2020 Nvidia refreshed the GTX 1660 Ti with the new codename N18E-G0-A1 with similar specs and pin compatible to the refreshed RTX chips (e.g., RTX 2070 Super Mobile).

The Turing generation did not only introduce raytracing for the RTX cards, but also optimized the architecture of the cores and caches. According to Nvidia the CUDA cores offer now a concurrent execution of floating point and integer operations for increased performance in compute-heavy workloads of modern games. Furthermore, the caches were reworked (new unified memory architecture with twice the cache compared to Pascal). This leads to 50% more instructions per clock and a 40% more power efficient usage compared to Pascal.

Nvidia specifies a power consumption of 60 Watt TGP (Total Graphics Power) and therefore 20 Watts lower than a normal (Max-P) 1660Ti for laptops. Therefore, the Max-Q variant is well suited for thin and light gaming laptops. The TU116 chip is manufactured in 12nm FFN at TSMC.

The Nvidia GeForce RTX 3050 6GB Laptop GPU (mobile, GN20-P0-R, 2023 Refresh) is the refresh of the RTX 3050 4GB Laptop GPU (slowest RTX 3000 mobile card) and also based on the GA107 Ampere chip. The 6GB variant offers more CUDA cores (2,560, +25%) but a cut down memory bus to 96 Bit. The clock speed depends on the TGP variant and can range from 713 - 1530 (base) and 1058 - 1740 (boost) for the TGP variants of 35 to 80 Watt (see table below).

The performance is between the old 4GB RTX 3050 and the RTX 3050 Ti and therefore best suited for full HD gaming (1920 x 1080) in medium to high graphic settings. The performance is not sufficient to enable Raytracing in most games, but the Tensor cores can be used for DLSS in some games (and get performance boost with slight quality reduction). Beware, that the low TGP variants will offer a significantly lower performance.

The GA107 chip offers 3,072 FP32 ALUs of which half can also execute INT32 instructions (i.e. 1,536 INT32 ALUs). The RTX 3050 however maybe won't use all ALUs on the chip. With Turing all shaders could still execute FP32 or INT32 instructions. The raytracing and tensor cores on the chip were also improved according to Nvidia. The Ampere chips also include an improved 5th generation video encoder (NVENC for H.264 and H.265) and a 7th generation decoder (for various formats now including AV1).

The GA107 chip is manufactured by Samsung in 8nm (8N), which is not quite able to keep up with the 7nm node at TSMC (e.g. used by AMD and also for the professional GA100 Ampere chip).