Nvidia has officially launched its highly anticipated GeForce RTX 5000 series at CES 2025, introducing four new graphics cards based on the cutting-edge Blackwell architecture. The series includes the RTX 5090, RTX 5080, RTX 5070 Ti, and RTX 5070. With these new models, Nvidia aims to set new standards for gaming performance and efficiency.
The RTX 5090, as the flagship model, features 21,760 CUDA cores, making it the first GeForce card to surpass the 20,000 core milestone. With 32 GB of GDDR7 memory and a 512-bit memory bus, it boasts a staggering memory bandwidth of 1,792 GB/s. This powerhouse also has a thermal design power (TDP) of 575 watts, which is 125 watts more than its predecessor, the RTX 4090. Such increased power consumption necessitates high-capacity power supplies for optimal performance.
One of the standout features of the RTX 5000 series is the introduction of DLSS 4 with Multi Frame Generation (MFG). This AI-driven technology enables the cards to generate up to three additional frames per rendered image, significantly boosting frame rates. For example, Nvidia demonstrated this with
Cyberpunk 2077, achieving 238 frames per second (FPS) with the RTX 5090 utilizing DLSS 4, compared to just 106 FPS with the RTX 4090 using the previous DLSS 3.5 technology. Such advancements raise the potential for increased performance across various gaming scenarios.
Despite the enhanced performance expectations, the pricing strategy for the RTX 5000 series is eye-catching. The suggested retail prices are as follows: RTX 5090 at 2,329 Euros, RTX 5080 at 1,169 Euros, RTX 5070 Ti at 879 Euros, and the base model RTX 5070 at 649 Euros. The launch dates for the RTX 5090 and RTX 5080 are set for January 30, 2025, with the other two models following later in February.
Notably, the introduction of GDDR7 memory signifies another leap in GPU technology, following the advances seen with previous GDDR generations. The shift from GDDR6X to GDDR7 promises to continue improving bandwidth and efficiency, both of which are necessary to support the increased capabilities of these new cards.
While the RTX 5090 leads the charge with its staggering specs, the RTX 5080 and RTX 5070 models also show promising enhancements. The RTX 5080 offers 10,752 CUDA cores and 16 GB of GDDR7 memory, positioned very competitively against the previous generation. Meanwhile, the RTX 5070 Ti features the same memory configuration but brings down the power consumption to 300 watts.
Cooling solutions have also seen innovation; both the RTX 5090 and RTX 5080 use a new compact cooling design, facilitating two-slot width for the card's physical dimensions. This is significant considering their increased performance output, allowing them to manage heat effectively even under high loads. The RTX 5070, on the other hand, utilizes 6,144 CUDA cores and 12 GB of memory, intended for gamers seeking solid performance at 1080p or even 1440p resolutions.
Alongside its desktop counterparts, Nvidia will be releasing mobile versions of the RTX 5000 series, making these technologies accessible to laptop gamers as well. With 75 games expected to support DLSS 4 by launch, Nvidia is emphasizing its commitment to enhancing gaming experiences through innovative technologies.
Overall, Nvidia’s launch of the RTX 5000 series is poised to redefine gaming graphics, emphasizing not only sheer performance capability but also the seamless integration of AI features aimed at addressing gamer demands for higher frame rates and efficiency. The question remains—will the performance gains justify the rising costs?