NVIDIA takes Turing for a spin again with laptop versions of the GeForce GTX 1650 SUPER and GeForce GTX 1650 Ti
As we exclusively reported in November, NVIDIA is planning to refresh its mobile GPU offering this March. According to our sources, the company is developing mobile versions of the RTX 2060 SUPER, RTX 2070 SUPER and RTX 2080 SUPER, along with two refreshed GTX 16 series GPUs. Last week, we saw the RTX 2080 SUPER Max-Q appear on Geekbench paired with a 10th generation Intel Core H series processor, and it is the turn of its two upcoming GTX 16 series siblings.
Spotted by @_rogame, the two Geekbench listings seemingly confirm that the GTX 1650 SUPER is actually the N18P-G61, rather than the N18P-G62 as we previously asserted. Instead, the latter appears to be the GTX 1650 Ti. Regardless, Geekbench reports both cards as having 4 GB of GDDR6 VRAM, which would be a healthy step-up from the 4 GB of GDDR5 that the existing GTX 1650 offers. Wccftech claims that the GTX 1650 SUPER and GTX 1650 Ti will run on a 128-bit memory bus at 12 Gbps too, yielding a memory bandwidth of 192 GB/s.
Crucially, NVIDIA will distinguish the two cards by GPU. Namely, the GTX 1650 SUPER will remain on the TU117 GPU used in the current GTX 1650, while the GTX 1650 Ti will be based on a TU116 GPU. The GTX 1650 SUPER will also have 14 Compute Units to the 16 of GTX 1650 Ti, with the two having 896 and 1,024 CUDA cores respectively, too.
Expect to see both GPUs launch in March, with NVIDIA making standard and Max-Q configurations available to OEMs.
GeForce GTX 1650 SUPER | GeForce GTX 1650 Ti | |
---|---|---|
Codename | N18P-G61 | N18P-G62 |
GPU | TU117 | TU116 |
Boost clock | 1.56 GHz | 1.49 GHz |
Compute Units (SMs) | 14 | 16 |
CUDA cores | 896 | 1,024 |
Memory bus | 128-bit | 128-bit |
Memory clock | 12 Gbps | 12 Gbps |
Memory bandwidth | 192 GB/s | 192 GB/s |
VRAM | 4 GB GDDR6 | 4 GB GDDR6 |
TGP | 35 W - Max-Q50 W - Standard | 35 W - Max-Q50 W - Standard |