Today Nvidia announced two new desktop graphics cards, the Nvidia RTX A400 and A1000 GPU. Based on Nvidia’s Ampere architecture, one step behind the current Ada Lovelace generation, and designed to consume only 50 watts of power apiece, the new cards seemingly target the very entry level of Nvidia’s workstation graphics portfolio.
And yet, the RTX A400 and A1000 are pitched at professional users. Nvidia’s blog post introducing the new graphics cards says they’re designed to serve architects, engineers, industrial planners, and healthcare and financial professionals, among other users.
“They expand access to AI and ray-tracing technology, equipping professionals with the tools they need to transform their daily workflows,” wrote Nvidia senior product marketing manager Stacy Ozorio.
Specs of the Nvidia RTX A400 and A1000 GPUs
The two new graphics cards share a few features in common. Both the RTX A400 and A1000 are 50-watt, single-slot cards offering four Mini DisplayPort outputs (a new high for the 400-series, which previously had only three display outputs in the Nvidia T400).
The differences between the cards are found it their processing capacity: the number of RT cores for real-time ray tracing, Tensor cores for AI acceleration, CUDA cores for general graphics processing, and the size and bandwidth of their memory.
|
Nvidia RTX A400 |
Nvidia RTX A1000 |
|
|
CUDA cores |
768 |
2304 |
|
RT cores |
6 |
18 |
|
Tensor cores |
24 |
72 |
|
GPU memory |
4 GB GDDR6 |
8 GB GDDR6 |
|
Memory interface |
64-bit |
128-bit |
|
Memory bandwidth |
96 GB/s |
192 GB/s |
These spec differences manifest in a roughly 2.5x performance increase from the A400 to A1000, according to Nvidia’s peak performance data. (You have to give Nvidia credit for the arithmetically accurate product names.)
Are the RTX A400 and A1000 a good fit for engineers?
Nvidia’s new graphics cards clearly improve over their preceding generation, but they themselves are a generation behind Nvidia’s current Ada Generation graphics cards. It was just two months ago that Nvidia announced the RTX 2000 Ada Generation workstation graphics card as the entry point to the Ada generation desktop GPUs, a $625 card that beats the A400 and A1000 on every aspect of performance—most of all in the headline features of real-time ray tracing and AI acceleration.
The new cards seem to be best suited not for engineering workstations but for space- and power-constrained computers that require a bit of extra graphics horsepower—perhaps just the ability to output to four displays at once. “[T]he A400 and A1000 GPUs bring impressive features to compact, energy-efficient workstations,” Ozorio wrote.
The Nvidia RTX A1000 is now available from Nvidia distribution partners including PNY and Ryoyo Electric, and the Nvidia RTX A400 is expected to be available in May. An Nvidia representative told Engineering.com that the estimated street price of the Nvidia RTX A400 is $135 and the RTX A1000 is $365.