The GeForce 40 series is a family of graphics processing units developed by Nvidia, succeeding the GeForce 30 series. The series was announced on September 20, 2022, at the GPU Technology Conference (GTC) 2022 event; the RTX 4090 was released on October 12, 2022, the 16 GB RTX 4080 was released on November 16, 2022. An RTX 4080 12GB was announced in September of 2022, but following some controversy in the media, was "unlaunched" by Nvidia. On January 5, 2023, that model would be released as the RTX 4070 Ti. The RTX 4070 was then released on April 13, 2023. The RTX 4060 Ti was released on May 24th 2023, and the RTX 4060 on June 29th 2023. An RTX 4060 Ti 16GB followed on July 18, 2023. Further models of this generation of desktop and mobile GPUs are to release later in 2023.

The cards are based on Nvidia's Ada Lovelace architecture and feature Nvidia's third-generation RT cores for hardware-accelerated raytracing (RTX), and fourth-generation deep-learning-focussed Tensor Cores.

GeForce 40 series

Details

Architectural highlights of the Ada Lovelace architecture include the following:

GeForce 40 series

CUDA Compute Capability 8.9

TSMC 4Nprocess (custom designed for Nvidia) – not to be confused with N4

Fourth-generation Tensor Cores with FP8, FP16, bfloat16, TensorFloat-32 (TF32) and sparsity acceleration Third-generation Ray Tracing Cores, along with concurrent ray tracing, shading and compute Shader Execution Reordering - needs to be enabled by the developer Dual NVENC with 8K 10-bit 60FPS AV1 fixed function hardware encoding A new generation of Optical Flow Accelerator to aid DLSS 3.0 intermediate AI-based frame generation No NVLink support DisplayPort 1.4a and HDMI 2.1 display connections Double-precision (FP64) performance of the Ada Lovelace chips are 1/64 of single-precision (FP32) performance.

GeForce 40 series

Products

Desktop

RTX 4060 and RTX 4060 Ti use GDDR6 video memory. All other cards use Micron GDDR6X video memory.

Mobile

All models feature GDDR6 memory.

Controversies

RTX 4080 12GB

When the 12GB variant of the RTX 4080 was announced, numerous publications, prominent YouTubers, reviewers, and the community criticized Nvidia for marketing it as an RTX 4080, given the equivalent performance of previous xx80-class cards when compared with other cards of the same series, and the large gap in specifications and performance compared to the 16GB variant. The criticism focussed on Nvidia's naming scheme - two 4080 models were to be offered, an "RTX 4080 12GB" and an "RTX 4080 16GB" - the obvious interpretation was that the only difference between the two products would be a difference in VRAM capacity. However, unlike other examples of Nvidia GPUs with differing memory configurations within the same class which have typically been very close in processing performance, the RTX 4080 12GB uses a completely different processing unit and memory architecture: the 4080 12GB would use the AD104 chip, which features 27% fewer CUDA cores than the 16GB variant's AD103 chip. It would also have a reduced TDP (285W - the 16GB variant was configured to draw up to 320W), and a cut-down 192-bit memory bus. In prior generations (e.g. the GeForce 10, 20, 30 series), a 192-bit bus width had been reserved for xx60-class cards and below. These changes made the card up to 30% slower than the RTX 4080 16GB in memory-agnostic performance, while being priced significantly higher than previous xx70-class cards ($900 vs. $500 for the RTX 3070, approximately 80% more expensive).

On October 14, 2022, Nvidia announced that due to the confusion caused by the naming scheme, it would be "unlaunching"—i.e. postponing the launch of—the RTX 4080 12GB, with the RTX 4080 16GB's launch remaining unaffected.

On January 3, 2023, Nvidia reintroduced the RTX 4080 12GB as the RTX 4070 Ti during CES 2023 and reduced its list price by $100.

12VHPWR connector failures

Some buyers of the RTX 4090 reported that the 12VHPWR—also known as PCIe Gen 5 or 16-pin—connectors of their RTX 4090 were melting, which sparked several theories to explain it. After investigation, several sources reported that the main cause was the 12VHPWR connector not being fully socketed while being put under load that resulted in overheating of the connector's pins, which in turn caused the melting of the plastic housing.

PCI-SIG, the standards organization responsible for the creation of the 12VHPWR connector has decided to make changes to the connector's specifications following the recent failures.