Nvidia unveils three GeForce RTX 4060 Family GPUs
Nvidia has released the GeForce RTX 4060 Family, three new mid-range cards in the GeForce RTX 40 Series, the firm’s current generation of consumer GPUs based on its Ada Lovelace GPU architecture.
The $399 8GB GeForce RTX 4060 Ti is due to become available this week, with its $499 16GB counterpart and the $299 8GB GeForce 4060 due in July.
The latest GPUs to use Nvidia’s Ada Lovelace architecture
The GeForce RTX 4060 Family are the latest GPUs based on Nvidia’s new Ada Lovelace architecture.
As well as updates to the cards’ RT ray tracing cores and Tensor AI inferencing cores, it features a new Optical Flow Accelerator, used by DLSS 3.0, Nvidia’s Deep Learning Super Sampling technology.
Originally a render upscaling system, DLSS 3.0 also uses optical flow to generate entire new frames between those rendered traditionally, boosting in-game frame rates by “up to 4x”.
Although DLSS is primarily a games technology, it does have implications for CG artists, since it is also now being adopted in visualisation software like D5 Render.
| Mid-range GeForce RTX 40 Series GPUs | |||||
|---|---|---|---|---|---|
| RTX 4060 | RTX 4060 Ti (8GB) | RTX 4060 Ti (16GB) | RTX 4070 | ||
| Architecture | Ada Lovelace | Ada Lovelace | Ada Lovelace | Ada Lovelace | |
| CUDA cores | 3,072 | 4,352 | 4,352 | 5,888 | |
| Tensor cores* | 96 | 128 | 128 | 184 | |
| RT cores* | 24 | 34 | 34 | 46 | |
| Base clock (GHz) | 1.83 | 2.31 | 2.31 | 1.92 | |
| Boost clock (GHz) | 2.46 | 2.54 | 2.54 | 2.48 | |
| Compute performance FP32 (Tflops)* |
15.1 | 22.1 | 22.1 | 29.2 | |
| GPU memory | 8GB GDDR6 |
8GB GDDR6 |
16GB GDDR6 |
12GB GDDR6X |
|
| TDP | 115W | 160W | 165W | 200W | |
| Release date | 2023 | 2023 | 2023 | 2023 | |
| MSRP at launch | $299 | $399 | $499 | $599 | |
*Data taken from third-party websites.
Key specifications for DCC work
The three new cards flesh out the low- to mid-range of the GeForce RTX 40 Series. Above, you can see their core specifications in comparison to the GeForce RTX 4070, released last month.
Core counts and compute performance are all proportionately lower than the 4070, as are power consumption and price.
The 16GB GeForce RTX 4060 Ti has more GPU memory than either the RTX 4070 or the RTX 4070 Ti, both of which are 12GB cards, offering a trade-off between memory capacity and other specs.
However, all of the GeForce RTX 4060 Family cards use slower GDDR6 memory, with memory bandwidths of 272-288 GB/s, compared to the 504 GB/s for the GDDR6X memory in the 4070 and 4070 Ti.
Nvidia’s performance comparisons of its GeForce 4060 and 3060 Family GPUs and Apple’s M1 Ultra, rendering Nvidia’s Project Sol model in Arnold 7.1.1 and applying a Magic Mask effect in DaVinci Resolve.
Benchmarks and performance in DCC applications
Along with its gaming benchmarks, Nvidia has published performance comparisons between the new cards and its previous-gen GeForce RTX 3060 and 3060 Ti for several DCC applications.
The charts above show rendering speed in Arnold, using Nvidia’s own Project Sol Maya asset, and time taken to apply an AI-based Magic Mask effect to 4K footage in editing and grading app DaVinci Resolve.
Nvidia also includes figures for an Apple Silicon processor: the M1 Ultra available in current Mac Studio workstations, although note that Arnold does not actually support GPU rendering on macOS.
Pricing and release dates
The 8GB GeForce RTX 4060 Ti will be available on 24 May 2023. It has a MSRP of $399. The 16GB GeForce RTX 4060 Ti and GeForce RTX 4060 will be available in July 2023, and have MSRPs of $499 and $299.
Read more about the new GeForce RTX 4060 Family on Nvidia’s product website