NVIDIA T400E

NVIDIA^® T400E

SKU: VCNT400-LLP

Product Brochure Linecard

The NVIDIA^® T400E packs power and performance in a small form factor so professionals can tackle a range of multi-app workflows with ease. Native support for up to four 5K displays gives you an expansive visual workspace to view your work in stunning detail. The NVIDIA T400E is also part of the PNY Long Life Program*.

* Must be enrolled in the long-life program.

Features
- NVIDIA Turing Architecture
- 4 GB GDDR6
- PCI Express 3.0 x 16
- 4 x Mini Display 1.4a with Latching Mechanism
- 384 CUDA^® Cores
- 30 W Max Power Consumption
Specifications

GPU Memory

4GB GDDR6

Memory Interface

64-bit

Memory Bandwidth

Up to 80GB/s

NVIDIA CUDA^® Cores

384

Single-precision Performance

Up to 1.09 TFLOPS3

System Interface

PCI Express 3.0 x16

View All Specifications

Specifications

GPU Memory

4GB GDDR6

Memory Interface

64-bit

Memory Bandwidth

Up to 80GB/s

NVIDIA CUDA^® Cores

384

Single-precision Performance

Up to 1.09 TFLOPS3

System Interface

PCI Express 3.0 x16

Max Power Consumption

30 W

Thermal Solution

Active

Form Factor

2.713” H x 6.137” L, Single Slot

Display Connectors

4 x mDP 1.4a with Latching Mechanism

Max Simultaneous Displays

4x 3840 x 2160 @ 120Hz | 4x 5120 x 2880 @ 60Hz

Graphics APIs

DirectX 12, Shader Model 6.6, OpenGL 4.64, Vulkan 1.34

Compute APIs

CUDA^® 11.6, OpenCL 3.0, DirectCompute
System Requirements
Hardware Requirements
- Intel Core i5, i7, i9, or Xeon processor or later
- AMD Ryzen or Epyc class processor or later
- PCI Express 4.0 x16 (preferred) expansion slot
- System memory greater than or equal to GPU memory, twice the GPU memory recommended
- Internet connection for driver installation
- DisplayPort 1.4a (preferred), HDMI, or DVI compatible display(s)
Supported Platforms
- Microsoft Windows 11 (64-bit)
- Microsoft Windows 10 (64-bit)
- Red Hat Enterprise Linux 7.x
- SUSE Linux Enterprise Desktop 15.x
- OpenSuse 15
- Fedora 31
- Ubuntu 18.04
- FreeBSD 11.x
- Solaris 11
Package Contains
NVIDIA T400E 4GB Professional Graphics Card

Attached Low-profile (SFF) Bracket

Unattached Full-height (ATX) Bracket

4x mDP to DP NVIDIA-certified Latching Adaptors
Available Accessories

MDP-HDMI-FOUR-PCK

MDP-HDMI-FOUR-PCK connects the NVIDIA T400 4GB (or T400) to HDMI displays at resolutions up to 4K with PNY Part Number DP-HDMI FOUR-PCK. The four included DisplayPort to HDMI adapters are recommended by NVIDIA, provide outstanding image quality, and are built to professional standards.

MDP-HDMI-SINGLE-PCK

MDP-HDMI-SINGLE-PCK connect the NVIDIA T400 4GB to an HDMI display at resolutions up to 4K with PNY Part Number MDP-HDMI-SINGLE-PCK. The included mini-DisplayPort to HDMI adapter is recommended by NVIDIA, provides outstanding image quality, and is built to professional standards with a latching connector.

NVIDIA T400E

Full-size features. Compact design.

The NVIDIA^® T400E, built on the NVIDIA Turing™ GPU architecture, delivers amazing performance and capabilities to power a range of professional workflows. Featuring 384 CUDA^® cores and 4GB of GDDR6 memory, the T400E packs power and performance in a small form factor so professionals can tackle a range of multi-app workflows with ease.

Turing GPU Architecture

The NVIDIA T400E GPU, built on a 12nm FinFET manufacturing process with 384 CUDA cores, offers an efficient, low-profile, single-slot professional solution. Ideal for CAD, DCC, financial services, and visualization, it delivers impressive performance in a compact form factor.

Advanced Shading Technologies

The Turing GPU architecture includes advanced shader technologies: Mesh Shading speeds up geometry processing, offering up to 2x performance. Variable Rate Shading enhances efficiency by adjusting shading rates, cutting shaded pixels by 50%. Texture Space Shading boosts performance in pixel shader-heavy tasks like depth-of-field.

Advanced Streaming Multiprocessor (SM) Architecture

Combined shared memory and L1 cache improve performance significantly while simplifying programming and reducing the tuning required to attain the best application performance. Each SM contains 96 KB of L1 shared memory, which can be configured for various capabilities depending on the compute or graphics workload.

High-Performance GDDR6 Memory

Built with Turing’s vastly optimized GDDR6 memory subsystem for outstanding performance the NVIDIA T400E features 4GB of GPU memory and 80 GB/s of peak bandwidth, more than double the previous generation. NVIDIA T400E is an ideal platform for 3D professionals and multi-display environments, where integrated graphics simply isn’t adequate.

Single Instruction, Multiple Thread (SIMT)

New independent thread scheduling capability enables finer-grain synchronization and cooperation between parallel threads by sharing resources among small jobs.

Mixed-Precision Computing

Double the throughput and reduce storage requirements with 16-bit floating point precision computing to enable the training and deployment of larger neural networks. With independent parallel integer and floating-point data paths, the Turing SM is also much more efficient on workloads with a mix of computation and addressing calculations.

Graphics Preemption

Pixel-level preemption provides more granular control to better support time-sensitive tasks such as VR motion tracking.

Compute Preemption

Preemption at the instruction level provides finer-grain control over compute tasks to prevent long-running applications from either monopolizing system resources or timing out.

H.264 and HEVC Encode/Decode Engines

Deliver faster than real-time performance for transcoding, video editing, and other encoding applications with two dedicated H.264 and HEVC encode engines and a dedicated decode engine that are independent of the 3D/compute pipeline.

NVIDIA GPU BOOST 4.0

Automatically maximize application performance without exceeding the power and thermal envelope of the card. Allows applications to stay within the boost clock state longer under a higher temperature threshold before dropping to a secondary temperature setting base clock.