A Geekbench 5 listing of the GeForce RTX 4060 Ti has surfaced, detailing the specifications of the upcoming Ada Lovelace SKU. Based on the AD106 graphics core, it’ll feature 4,352 FP32 cores spread across 34 SMs with a peak boost clock of 2.54GHz. The GPU will be paired with 8GB of GDDR6 memory clocked at 18Gbps via a 128-bit bus and 32MB of L2 cache.
The GeForce RTX 4060 Ti boasts an external memory bandwidth of 288GB/s and a single-precision compute throughput of 22 TFLOPs. Despite featuring a lower core count than its predecessor, it’s expected to offer RTX 3070 performance levels while drawing 30% less power. The RTX 3060 Ti features 4,864, while the 3070 packs 5888 shaders. Both have an 8GB memory buffer paired with a 256-bit bus.
Regarding the pricing, the GeForce RTX 4060 Ti is expected to cost the same as its predecessor ($399). Tom from MLID certainly thinks so and if you look at this market segment, anything higher won’t make sense. We might see a 16GB variant later on, but for now, the 4060 Ti will be an 8GB card.
AMD plans to launch the Radeon RX 7600 around the same time. Based on the Navi 33 die, it’ll offer affordable 1080p gaming performance for roughly $300. Unlike its siblings, it will be marketed as an RTX 4060 rival and leverage a monolithic design. It’ll also be the only Navi 3x core to leverage the (N6) 6nm process node. Unfortunately, its ray-tracing capabilities will be lower than the rest of the RX 7000 series GPUs and might need a memory upgrade down the road.