NVIDIA has launched two new entries in its Ampere workstation lineup, the RTX A4500 20 GB and the RTX A2000 12 GB graphics cards. Both cards are aimed at a different set of workloads & come equipped with the latest Ampere architecture for faster content creation, AI, & scientific use cases.
NVIDIA's RTX A4500 20 GB & RTX A2000 12 GB Ampere Workstation Graphics Cards Launched
The NVIDIA RTX Ampere Workstation graphics card lineup now consists of six options that users can select from. The two new entrants are positioned in the high-end and entry-level segments.
NVIDIA RTX A4500 20 GB Specifications
Starting with the NVIDIA RTX A4500, the graphics card rocks 7168 CUDA Cores in a total of 56 SMs that are enabled on the GA102 GPU. The core count makes the Ampere GPU featured on the RTX A4500 the most cut-down workstation configuration of the GA102 GPU (with just 66% active cores). Other specifications include 224 Tensor units, 56 RT cores, and a core clock of up to 1.63 GHz. The card features 20 GB of GDDR6 memory that operates across a 320-bit bus interface at pin speeds of up to 16 Gbps, providing the GPU with 640 GB/s of bandwidth. The TDP of the card is rated at 200W.
As for performance, the NVIDIA RTX A4500 offers up to 23.7 TFLOPs FP32, 46.2 TFLOPs RT &182.2 TFLOPs of Tensor performance. The card rocks 4 DisplayPort 1.4 outputs and comes in a dual-slot configuration with an active cooling solution, Power is provided by a single 8-pin connector. NVIDIA also features an NVLINK interconnect on the A4500 which allows for up to two cards to run parallel with a combined memory of 40 GB.
NVIDIA RTX A2000 12 GB Specifications
In addition to the higher-end option, NVIDIA is also offering an upgraded variant of its RTX A2000 graphics card which now rocks twice the memory capacity. The rest of the specs remain the same as the original variant.
The NVIDIA RTX A2000 features the Ampere GPU architecture. In terms of specifications, the RTX A2000 rocks the GA106 GPU which is equipped with 3328 CUDA cores, 104 Tensor cores, and 26 RT cores, all of which offer a nice performance bump over the previous generation offerings. The GPU performance is rated at 8 TFLOPs FP32, 15.6 TFLOPs RT and 63.9 TFLOPs Tensor. The memory is a 192-bit bus design with the chips operating at 12 Gbps pin speeds for an effective 288 GB/s of total bandwidth.
The RTX A2000 itself comes in a small form factor design, featuring a half-height and half-length board. The card even has a small blower type-fan on the shroud. Since this is a 70W TDP card, there are no power connectors. It's a simple plug-and-use card that offers great efficiency in a compact design. There are four Mini DisplayPorts (1.4) near the rear panel IO shroud which also comes with a small vent to exhaust hot air out.
NVIDIA Ampere Workstation Graphics Cards:
|Graphics Card||NVIDIA RTX A2000||NVIDIA RTX A4000||NVIDIA RTX A4500||NVIDIA RTX A5000||NVIDIA RTX A5000||NVIDIA RTX A6000|
|GPU||Ampere GA106 GPU||Ampere GA104 GPU||Ampere GA102 GPU||Ampere GA102 GPU||Ampere GA102 GPU||Ampere GA102 GPU|
|GPU Process||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm|
|Die Size||276 mm²||392.5 mm²||628mm²||628mm²||628mm²||628mm²|
|Boost Clock||1200 MHz||1536 MHz||1635 MHz||1697 MHz||TBD||1455 MHz|
|Single Precision||8.0 TFLOPs||19.2 TFLOPs||23.7 TFLOPs||27.8 TFLOPs||TBD||31.2 TFLOPs|
|VRAM||6/12 GB GDDR6||16 GB GDDR6||20 GB GDDR6||24 GB GDDR6||24 GB GDDR6||48 GB GDDR6|
|NVLINK VRAM||N/A||N/A||40 GB GDDR6||48 GB GDDR6||48 GB GDDR6||96 GB GDDR6|
|Memory Bandwidth||228 GB/s||448 GB/s||640 GB/s||768 GB/s||768 GB/s||768 GB/s|
|Launch Price||TBC||TBC||TBC||TBC||TBD||$4650 US|
|Launch Date||October. 2021||April, 2021||Q4 2021||April, 2021||Q1 2022||December, 2020|