NVIDIA RTX A40: specs and benchmarks

NVIDIA RTX A40

Summary

NVIDIA started RTX A40 sales 5 October 2020. This is Ampere architecture desktop card based on 8 nm manufacturing process and primarily aimed at gamers. 48 GB of GDDR6 memory clocked at 14.5 GB/s are supplied, and together with 384 Bit memory interface this creates a bandwidth of 695.8 GB/s.

Compatibility-wise, this is dual-slot card attached via PCIe 4.0 x16 interface. Its manufacturer default version has a length of 267 mm. 8-pin EPS power connector is required, and power consumption is at 300 Watt.

We have no data on RTX A40 benchmark results.

General info

Of RTX A40's architecture, market type and release date.
Place in performance rating
not rated
 
Architecture
Ampere
 
Code name
GA102
 
Type
Desktop
 
Release date
5 October 2020
 

Technical specs

RTX A40's general performance parameters such as number of shaders, GPU core clock, manufacturing process, texturing and calculation speed. These parameters indirectly speak of RTX A40's performance, but for precise assessment you have to consider its benchmark and gaming test results.
Pipelines
10752
 
Boost Clock
1755 MHz
 
Transistor count
28,300 million
 
Manufacturing process technology
8 nm
 
Power consumption (TDP)
300 Watt
 
of 900 (Tesla S2050)
Texture fill rate
589.7
 

Compatibility, dimensions and requirements

Information on RTX A40's compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop video cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).
Interface
PCIe 4.0 x16
 
Length
267 mm
 
Supplementary power connectors
8-pin EPS
 

Memory

Parameters of memory installed on RTX A40: its type, size, bus, clock and resulting bandwidth. Note that GPUs integrated into processors don't have dedicated memory and use a shared part of system RAM.
Memory type
GDDR6
 
Maximum RAM amount
48 GB
 
Memory bus width
384 Bit
 
of 6144 (GRID A100A)
Memory clock speed
14.5 GB/s
 
of 19500 (GeForce RTX 3090)
Memory bandwidth
695.8 GB/s
 
of 1024 (Radeon VII)

Video outputs and ports

Types and number of video connectors present on RTX A40. As a rule, this section is relevant only for desktop reference video cards, since for notebook ones the availability of certain video outputs depends on the laptop model.
Display Connectors
3x DisplayPort
 

API support

APIs supported by RTX A40, sometimes including their particular versions.
DirectX
12 Ultimate (12_2)
 
Shader Model
6.5
 
OpenGL
4.6
 
OpenCL
2.0
 
Vulkan
1.2
 
CUDA
8.6
 

AMD equivalent

We believe that the nearest equivalent to RTX A40 from AMD is Radeon RX 6900 XT.

Other GPUs

User rating

Here you can see the user rating of the video card, as well as rate it yourself.

Advices and comments

Here you can ask a question about RTX A40, agree or disagree with our judgements, or report an error or mismatch.