The GeForce GTX 1630 is a thoroughly cut GeForce GTX 1650, so it lacks quite a bit of pace.
NVIDIA has officially announced the VGA of the GeForce GTX 1630, the primary purpose of which is to AMD Radeon RX 6400 making your life more uncomfortable. The development uses a chip codenamed TU117 on a 12nm node, along with GDDR6-standard memories.
The specifications for the new VGA are summarized in the table below:
| Type | 1630 |
|---|---|
| GPU code name | TU117 |
| Architecture | Turing |
| Seed clock | 1740 MHz |
| GPU Boost Clock | 1785 MHz |
| Number of CUDA seeds | 512 |
| Number of texturizers | 32 |
| Number of blending units | 16 |
| Number of Integer CUDA cores | 512 |
| Raster hourly performance | 16 pixel |
| Maximum Pixel Fill Speed | 28,6 GPixel/s |
| Maximum Texel fill speed | 57,1 GTexel/s |
| Theoretical computing power (FP16) | 3,66 TFLOPS |
| Theoretical computing power (FP32) | 1,83 TFLOPS |
| Effective memory clock | 12 GHz |
| Memory type | GDDR6 |
| Memory bus | 64 bit |
| Memory bandwidth | 96 GB / s |
| VRAM capacity | 4 GB |
| PCI Express interface | x16-os PCI Express 3.0 |
The system is considered a dull Turing architecture because it does not contain an RT core and also lacks Tensor processors. There is no official reference model in this category, nor is it now, so in terms of power consumption, it is essentially a maximum of 75 watts for the TDP at the default clock, so no six-pin PCI Express power connector is required.
In terms of outputs, DisplayPort 1.4 and HDMI 2.0 are mandatory, and partners can create any configuration from this. NVIDIA does not prohibit DVI either, although its use will be more of a rarity.
The GeForce GTX 1630 is priced at $ 150, but most models on the market include some extras, including better cooling, so there can be a lot of difference in terms of pricing. As mentioned in the first paragraph, the primary competitor in the development is the Radeon RX 6400, but the performance of the new GeForce doesn’t really reach that, as the novelty of the greens is roughly 40% slower than AMD’s entry-level model based on manufacturer tests. This in itself is not tragic if the difference in question is later seen on the price tags.
