New information is starting to circulate about the upcoming NVIDIA GeForce RTX 60 graphics cards and their Rubin architecture. While the performance gains may be relatively modest, the real generational leap could lie elsewhere: in ray tracing and path tracing, where NVIDIA is aiming for a spectacular advancement.
RTX 60: modest gains in rasterization, but massive improvement in ray tracing The latest industry leaks suggest that the RTX 60 could offer an improvement of between 30 and 35% in traditional rasterization compared to the RTX 50 based on the Blackwell architecture. This is a significant increase, but not a groundbreaking one.
Conversely, when it comes to ray tracing, the story changes completely. NVIDIA is reportedly preparing a much more ambitious leap, with performance potentially doubling in this area. This progress is made possible by the introduction of 5th generation RT cores and 6th generation Tensor cores, specifically designed for AI and advanced rendering workloads.
Increased specifications, but no memory revolution In terms of technical specifications, several models are emerging. The GeForce RTX 6090 is said to be based on a 3nm TSMC GR202 GPU, with 24,576 CUDA Cores, roughly 13% more than the RTX 5090. It will retain a similar memory configuration with 32GB of GDDR7 on a 512-bit bus.
Both the RTX 6080 and RTX 6070 are expected to see improvements in bandwidth. The former is rumored to feature 20GB of GDDR7 with a 320-bit bus, while the latter may offer 16GB on a 256-bit interface. Although the core count remains unknown at this point, the increased memory bus and capacities hint at an interesting bandwidth boost.
An architecture designed to support DLSS 5 Beyond the numbers, these advancements are part of a broader strategy: to support the arrival of DLSS 5. This new generation of technology is not just about image reconstruction but aims for AI-assisted rendering.
The initial demonstrations required extreme configurations, with two RTX 5090s working together: one for rendering and the other for neural processing. This approach highlights the high power demands of the technology.
In this context, the RTX 60 series appears to be the first GPUs truly optimized to effectively leverage DLSS 5. With improved IPC, better Tensor Core management, and increased ray tracing power, they are expected to make DLSS 5 viable on a single graphics card.
Uncertain launch timeline The question remains about the launch timeframe. While some sources suggest a release in the second half of 2027, others do not rule out a delay to 2028. This uncertainty is attributed to a challenging industrial environment marked by memory shortages and a focus on AI infrastructure.
Regardless, these initial details suggest a GPU generation less focused on raw power and more on algorithmic efficiency. This transition makes sense as AI gradually becomes the core of NVIDIA’s graphic innovation.
