NVIDIA GeForce RTX 4070 Ti vs NVIDIA GeForce RTX 3080 vs NVIDIA GeForce RTX 3050

NVIDIA GeForce RTX 4070 Ti

The Nvidia GeForce RTX 4070 Ti is a high-end desktop graphics card based on the AD104 chip using the Ada Lovelace architecture. The 4070Ti offers 7,680 cores and 12 GB GDDR6X graphics memory with a 192 bit memory bus and clocked at 22.4 Gbits. The card offers a good raytracing performance thanks to the 60 dedicated raytracing cores. For AI calculations (and DLSS 3 in games), the RTX 4070 Ti has 240 Tensor cores.

The gaming performance is well suited for 4k60 gaming in most games, but for very demanding games you might have to reduce to QHD resolution. With Raytracing, lower resolutions and DLSS are crucial for playable frame rates. The performance on average is similar to the old RTX 3090 / 3090 Ti.

The AD104 chip is manufactured in the modern 5nm process at TSMC (4N with EUV) and specified with a TGP (Total Graphics Power) of 285 Watt (compared to the 450 Watt of the RTX 4090).

NVIDIA GeForce RTX 3080

The Nvidia GeForce RTX 3080 is high-end desktop graphics card based on the Ampere generation. It uses the big GA102 chip and offers 8,704 shaders and 10 GB GDDR6X graphics memory.

The gaming performance in 4k is clearly better than the old GeForce RTX 2080 Ti (approximately 15% faster). Therefore, even demanding games should run with maximum detail settings in 4k. Enabling raytracing will however reduce the framerate and therefore maximum details won't be playable. 

The RTX3080 is produced in 8nm at Samsung and rated at 320 Watt but the power efficiency is still quite good.

NVIDIA GeForce RTX 3050

The Nvidia GeForce RTX 3050 is a mid-range desktop graphics card. It is based on the GA106 Ampere chip and offers 8 GB GDDR6 graphics memory with a 128 Bit memory bus. It offers 2,560 cores and supports PCIe 4.0.

The performance is similar to the older GeForce RTX 2060 and therefore best suited for 1080p gaming of modern and demanding games in high settings.

The GA106 chip offers 3,840 FP32 ALUs of which half can also execute INT32 instructions (i.e. 1,920 INT32 ALUs). 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 GA106 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 TDP of the RTX 3050 is specified at 130 Watt and therefore better than the previous RTX 2060 (160 Watt).