Buying a GeForce GTX 1650 is pretty sketchy already. The older GPU has a wide array of SKUs floating around from many AIBs, and the 1650 line is pretty basic overall. It’s not bad, but it certainly can’t swing above its weight too much and compete with the likes of the SUPER line. But with a new SKU being introduced based on a new silicon run, things are about to get more complicated.
There will now be one more variant of the GeForce GTX 1650 floating around that’s a mild variation of the TU117 silicon from the basic model. The original version sported 896 CUDA cores were attached to 4GB of 8Gbps GDDR5 memory on a 128-bit memory interface bus. There are now newer variants sporting the much fast GDDR6 memory setup. Despite lower clock speed, these updated models had 50% higher memory bandwidth with 12Gbps GDDR6 memory.
And just for the sake of clarity, the TU116 or TU106 denotes the silicon die used. So while it’s a similar chip to other models, new chip revisions will include upgrades like improved memory, better clock speeds, or more power-efficient design. The TU106 die, in particular, is a new die that powers GeForce RTX 20-series cards, such as the GeForce RTX 2060, RTX 2060 Super and RTX 2070. But unlike the TU116 and TU117, the TU106 die comes with Nvidia’s TENSOR and RT cores.
And then things got even stranger. The total number of GPUs in the lineup jumped to four after he GeForce GTX 1650 TU116 and GeForce GTX 1650 TU106 respectively, both got leaked.
And now we get a fifth mainline option based on another chip revision with a GeForce GTX 1650 model sporting a TU 116-150 die. Although with this new die revision, it looks like the new chip will actually be worse for performance compared to more expensive models. It’s a cut-down version, which translates to being a cheaper product thanks to reduced overall core counts on the die.
@momomo_us has posted a picture of this new die, which belongs to a Gigabyte model GPU (GV-N1656OC-4GD R2.0).
The leak suggests that this cheaper model could be supplied to OEMs or be shipped to international markets where there’s less demand for the highest-end cards in some markets. Many countries outside the core Western markets are much poorer, so on average spend less on computer components. So this could mean that Nvidia is releasing some cheaper variants to adjust to that.