Nvidia RTX 5060 Ti Specs Leaked in Shipping Manifest
A shipping manifest has inadvertently revealed some details about Nvidia’s upcoming RTX 5060 Ti graphics cards, highlighting that it will have a mere 128-bit memory bus. That’s no change from the last generation, and GDDR7 will help raise memory bandwidth, but it doesn’t suggest that this card will offer anything different from the rest of the RTX 50-series: a modest upgrade.
The RTX 5060 and RTX 5060 Ti have been delayed several times, which has been particularly problematic considering the shortage of other RTX 50 cards and their skyrocketing prices. Fortunately, the shipping manifests suggest that they are almost ready for release, and rumors point to an end-of-April launch for the 5060 Ti and a May launch for the 5060. But the specifications aren’t especially exciting.
The RTX 5060 Ti’s unofficial specifications now says it has 4,608 CUDA cores, a boost clock of 2,572 MHz, 16GB or 8GB of GDDR7 memory running at 28Gbps, and a 128-bit memory bus. As per VideoCardz, that will give it a total of 448GBps of overall bandwidth, with its total board power of 180W. The RTX 5060, by contrast, will allegedly have 3,840 CUDA cores, 8GB of GDDR7 memory running at the same 28Gbps, with a 128-bit memory bus. That will give it the same overall bandwidth as the 5060 Ti, but with an overall board power of just 150W.
That’s more cores than their predecessors but a lower overall percentage of the flagship GPU, suggesting they’re more cut down than the last generation. The new architecture will help raise performance, but we’ve seen many GPUs from Nvidia this generation struggle to offer more than 10% extra raw performance outside of games that support DLSS and frame generation.
Although the RTX 5090 has set a new performance bar with its 21,760 CUDA cores and 32GB of bleeding-edge GDDR7 memory, even it only managed to eclipse its last-generation predecessors by around 30% when DLSS4 and frame generation weren’t in use. Part of this lackluster uplift this generation is down to Nvidia not utilizing a new process node. Shrinking the die size often allows for more transistors in a tighter space, allowing for improvements in efficiency or performance.
Nvidia also refocused a lot of its efforts on AI this generation. While the new Tensor cores are indeed powerful and multi-frame generation in the right setting can be very useful, it’s not been enough to offset the major problems this generation has faced—from shortages to melting cables, missing ROPs, and ongoing driver issues.
READ the latest news shaping the Nvidia market at Newsvidia
Nvidia RTX 5060 Ti Specs Leaked in Shipping Manifest, source
