NVIDIA RTX A2000 Laptop GPU
The NVIDIA RTX A2000 Laptop GPU or A2000 Mobile is a professional graphics card for mobile workstations. It is based on the GA107 Ampere chip and offers a similar performance to the consumer GeForce RTX 3050 Ti Laptop GPU. It also offers 2,560 CUDA, 20 Raytracing and 80 Tensor cores paired with a 128 Bit memory bus. It is available in different variants from 35 - 95 Watt (TGP) with different clock speeds (and performance). The GPU supports eDP 1.4b to connect the internal monitor and DisplayPort 1.4 and HDMI 2.1 for external connections.
There is no more Max-Q variant (formerly used for the low power variants) but every OEM can choose to implement Max-Q technologies (Dynamic Boost, WhisperMode).
In 2022 Nvidia released a slightly improved version of the RTX A2000 with more and faster graphics memory. The original 4 GB with 11 Gbits has been doubled to 8 GB GDDR6 clocked at 14 Gbits.
The raw performance should be similar to the GeForce RTX 3050 Ti for laptops at the same TGP level. Both GPUs depend heavily on good cooling and a high TGP for good performance. At a similar power consumption level the RTX A2000 should be clearly faster than the old Quadro T2000 and T2000 Max-Q.
The GA107 chip offers 2.560 FP32 ALUs of which half can also execute INT32 instructions (i.e. 1,280 INT32 ALUs). With Turing all shaders could still execute FP32 or INT32 instructions. The raytracing and tensor cores on the chip were also improved according to Nvidia. The Ampere chips also include an improved 5th generation video encoder (NVENC for H.264 and H.265) and a 7th generation decoder (for various formats now including AV1).
The GA107 chip is manufactured by Samsung in 8nm (8N), which is not quite able to keep up with the 7nm node at TSMC (e.g. used by AMD and also for the professional GA100 Ampere chip). Depending on the TGP, the A2000 can also be used in thin and light laptops (with 35 Watt TGP e.g.).
|RTX A Series|
|Pipelines||2560 - unified|
|Tensor / AI Cores||80|
|Core Speed||893 - 1687 (Boost) MHz|
|Theoretical Performance||9.3 TFLOPS FP32|
|Memory Speed||14000 effective = 1750 MHz|
|Memory Bus Width||128 Bit|
|Max. Amount of Memory||8 GB|
|Memory Bandwidth||192 GB/s|
|API||DirectX 12_2, Shader 7.0, OpenGL 4.6|
|Power Consumption||95 Watt (35 - 95 Watt TGP)|
|Displays||HDMI 2.1, DisplayPort 1.4|
|Features||DisplayPort 1.4, HDMI 2.1, PCIe 4.0 x16, 9.3 SP-FP TFLOPS Peak, 75 Tensor Performance Peak, up to 192 GB/s Memory Bandwidth, Resizable BAR, Support for Modern Standby|
|Date of Announcement||12.04.2021|
|Link to Manufacturer Page||https://www.nvidia.com/content/dam/en-zz...|