At CES 2025, NVIDIA announced its new RTX 50 series graphics cards with DLSS 4 tech, which include GeForce RTX 5090, RTX 5080, RTX 5070 Ti, and RTX 5070. These cards are based on the latest Blackwell architecture and are reportedly faster than their predecessors. If you are a gamer or need serious graphics power for video editing, these cards might just be what you’re looking for.
RTX 50 series pricing and availability
Now, you might think that these graphics cards will be very expensive compared to its predecessor. Thankfully, not all RTX 50 series come at a higher price. For instance, the RTX 5080 is priced at $999 which is the the same price as the RTX 4080 Super. It is also less than the RTX 4080’s launch price, which was $1199.
The RTX 5070 Ti costs $749, which is $50 less than the launch prices of the RTX 4070 Ti and the RTX 4070 Ti Super. Speaking of the RTX 5070’s price, it costs $549; here also, we get $50 less compared to the RTX 4070 Super and the same as the RTX 4070.
However, we see a price hike with a high-end RTX 5090 card that comes at $1999, which is $400 more than the RTX 4090 price. Both the RTX 5090 and RTX 5080 will launch on January 30. The RTX 5070 Ti and RTX 5070 will launch sometime in February.
RTX 50 series comparison
The new RTX 5080 is twice as fast as the old RTX 4080, and it costs less, too. The RTX 5070 Ti is 2x faster than the old RTX 4070 Ti. Also, the RTX 5070 should be twice as fast as the RTX 4070. Nvidia claims that the RTX 5070 GPU will offer “4090 performance at $549”. The RTX 4090’s launch price was $1,599. The secret behind the claim could be DLSS 4, which is Nvidia’s newest AI-powered upscaling tech that uses artificial intelligence to generate additional frames.
The RTX 5090 has 21,760 CUDA cores, 3,352 TOPS Tensor cores, and 318 TFLOPS ray-tracing. It comes with a boost clock speed of 2.41 GHz and is equipped with 32 GB of GDDR7 memory on a 512-bit bus, demanding a power draw of 575W. In comparison, the RTX 4090 comes with 16,384 CUDA cores, 1,321 TOPS Tensor cores, 191 TFLOPS ray tracing, and a 2.52 GHz boost clock, with 24 GB of GDDR6X memory on a 384-bit bus at 450W.
The RTX 5080 has 10,752 CUDA cores, 1,801 TOPS Tensor cores, and 171 TFLOPS ray tracing at a 2.62 GHz boost clock. It is equipped with 16 GB of GDDR7 memory across a 256-bit bus and needs 360W. In comparison, the RTX 4080 comes with 9,728 CUDA cores, 780 TOPS Tensor cores, 113 TFLOPS ray tracing, and a 2.51 GHz boost clock. The card shares the same 16 GB GDDR6X memory and 256-bit bus but with slightly reduced power consumption at 320W.
The RTX 5070 Ti sports 8,960 CUDA cores, 1,406 TOPS Tensor cores, 133 TFLOPS ray tracing, and a 2.45 GHz boost clock, plus 16 GB of GDDR6X memory on a 256-bit bus at 300W. In comparison, the RTX 4070 Ti comes with 7,680 CUDA cores, 641 TOPS Tensor cores, 93 TFLOPS ray tracing, and a higher 2.61 GHz boost clock but uses 12 GB of GDDR6X memory on a 192-bit bus, consuming 285W power.
Lastly, the RTX 5070 comes with 6,144 CUDA cores, 988 TOPS Tensor cores, 94 TFLOPS ray tracing, and a 2.51 GHz boost clock, supported by 12 GB of GDDR7 memory on a 192-bit bus with a 250W power draw. In comparison, the RTX 4070 comes with 5,888 CUDA cores, 466 TOPS Tensor cores, 67 TFLOPS ray tracing, and a 2.48 GHz boost clock, with 12 GB of GDDR6 or GDDR6X memory on a similar 192-bit bus, all while consuming just 200W.
