Currently the graphics card market is perfectly controlled by NVIDIA. AMD's opposition in this market is limited, as it does not have high-end solutions or support for Ray Tracing. Through the AIDA64 tool we can know that NVIDIA would be preparing two new GTX 1650. One of these graphs would be based on the TU106 silicon used in the RTX 2060 SUPER and RTX 2070 SUPER.
It is at least curious that months after the RTX 3000 in Ampere architecture are launched, NVIDIA makes this move. The GTX 1650 is the simplest graphics for gaming of the brand today. Launching a model with Ray Tracing seems a bit weird, although it would be a solution to clean stock.
- Powered by geforce gtx 1650
- Integrated with 4GB GDDR5 128-bit memory interface
- Windforce 2x cooling system with alternative fans.
- 80mm single blade fan
An NVIDIA GTX 1650 with Ray Tracing?
We have in the market the GTX 1650 GDDR5, the GTX 1650 GDDR6 and the GTX 1650 SUPER, as well as the GTX 1650 Ti for laptops. This indicates four different versions of the same chart, three for desktop, which is rare. One of these new graphics would be based on the TU116 silicon used for the GTX 1650 and the different GTX 1660. But also one based on the TU106 silicon is proposed, so it would be to provide Ray Tracing to the low range.
The new GTX 1650 based on the TU106 silicon would feature 896 CUDA Cores, 56TMU, 32ROP and 4GB GDDR6. Up to this point they are the same specs as the GTX 1650 GDDR6. This new GTX 1650 model with Ray Tracing would add 14 RT Cores and 112 Tensor Cores, being the first GTX to have support for Ray Tracing. All with a TDP of 75W, which is not bad at all.
On the other hand there is the GTX 1650 based on TU116 silicon, which would be traced to the GTX 1650 GDDR6 based on TU117 silicon. It may not implement improvements, just a silicon change to clean up the stock.
Source: DSOG
