NVIDIA GeForce RTX 3050 4GB Laptop GPU vs NVIDIA GeForce RTX 2080 Super (Desktop) vs NVIDIA GeForce RTX 2070 Max-Q

The Nvidia GeForce RTX 3050 Laptop GPU (mobile, NVIDIA_DEV.2583, GN20-P0) is the slowest RTX 3000 mobile card and based on the GA107 Ampere chip. It offers 2048 CUDA, 16 Ray Tracing and 64 Tensor cores. The memory bus is limited to 128 Bit and supports 4 GB of GDDR6 dedicated graphics memory.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).

In early 2023 Nvidia did refresh the RTX 3050 (GN20-P0-R) with a bigger VRAM of 6GB but reduced memory bandwidth. The 3050 6GB Laptop GPU also offers 2,560 CUDA cores as the 3050 Ti mobile.

The performance should be between the old GTX 1650 Ti and the 1660 Ti Max-Q depending on the used TGP variant. Therefore, it is best suited for full HD gaming (1920 x 1080) with high detail settings in demanding games. 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 faster RTX 3050 Ti offers 25% more cores at similar clock speeds and TGP settings. 

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).

The NVIDIA GeForce RTX 2080 Super or 2080S sits behind the GeForce RTX 2080 Ti as the second fastest desktop GPU in NVIDIA's current Turing line-up (mid 2019). It is the refresh of the one year old RTX 2080 (Desktop) with slightly higher performance. The GPU offers 3,072 cores (up from 2,944) and is based on the TU104 chip. The VRAM is still 8 GB of GDDR6 but now higher clocked at 15.5 GHz (496 GB/s vs. 14 GHz and 448 GB/s).

Features

NVIDIA manufacturers the TU104 chip on a 12 nm FinFET process and includes features like Deep Learning Super Sampling (DLSS) and Real-Time Ray Tracing (RTRT), which should combine to create more realistic lighting effects than older GPUs based on the company's Pascal architecture (if the games supports it). The RTX 2080 is also DisplayPort 1.4 ready, while there is also support for HDMI 2.0b, HDR, Simultaneous Multi-Projection (SMP) and H.265 video en/decoding (PlayReady 3.0).

Performance

The RTX 2080 Super slightly outperforms the older Nvidia Titan XP and of course the RTX 2080 which it succeeds. Therefore, the RTX2080S is located in the absolute high end and well suited for 4k gaming with high to maximum detail settings (depending on the game). Using raytracing features, will reduce the playable resolution however.

Power Requirements

NVIDIA estimates that the RTX 2080 Super should consume a maximum of 250 W, which is 35 W higher than the company rates its GeForce GTX 2080.

The NVIDIA GeForce RTX 2070 with Max-Q design is the power saving variant of the mobile RTX 2070 graphics card with reduced clock speeds and power consumption. Therefore, it is suitable for thin gaming laptops. It is based on the same TU106 chip with 2,304 shaders and 8 GB GDDR6 memory connected by a 256 bit interface (12 GHz).

Currently it looks like there will be two variants, a low power 80 Watt variant with 885 - 1185 MHz core clock and a faster 90 Watt variant with 1080 - 1305 MHz core clock). Both clock the memory at 6 GHz versus 7 GHz of the normal RTX 2070 (115 Watt TDP).

The clock is not the only difference compared to the regular GTX 1080 though. The drivers for the Max-Q version were optimized for efficiency (and not performance; only for Max-Q models), there are optimized voltage converters for 1V operation, high-end cooling methods, and a 40 dB limit for the fan noise (with clock adjustments to ensure this at all times).

Features

NVIDIA manufacturers the TU106 chip on a 12 nm FinFET process and includes features like Deep Learning Super Sampling (DLSS) and Real-Time Ray Tracing (RTRT), which should combine to create more realistic lighting effects than older GPUs based on the company's Pascal architecture (if the games supports it). The RTX 20870 is also DisplayPort 1.4 ready, while there is also support for HDMI 2.0b, HDR, Simultaneous Multi-Projection (SMP) and H.265 video en/decoding (PlayReady 3.0).

Performance

The Max-Q version of the RTX 2070 should perform a bit slower than the mobile RTX 2070. Therefore, it should be able to run all demanding games in 2560x1440 with maximum details. Raytracing features however may reduce framerates a lot.