Prev
Next
MX3500A-SP
• NVIDIA RTX 3500 Ada Generation Embedded GPU based on NVIDIA Ada Lovelace architecture
• 5120 CUDA cores, 40 RT cores and 160 Tensor cores
• 23.04 TFLOPS peak FP32 performance
• 12GB GDDR6 memory
• PCIe Gen 4 x16 interface
• Support Error Correction Code(ECC)
Overview
The MX3500A-SP MXM 3.1 Type B module features an NVIDIA RTX 3500 Ada Generation Embedded GPU based on the NVIDIA Ada Lovelace architecture. This architecture delivers revolutionary ray tracing and AI performance for graphics and neural networks. With dramatically higher baseline performance tailored for ray tracing and neural graphics workloads, it marks a major advancement in capabilities. The module enables outstanding gaming, content creation, professional graphics, AI acceleration, and compute power.
| ● Advanced GPU engine compared to Ampere Architecture |
| Powered by the NVIDIA RTX 3500 Ada Generation Embedded GPU featuring the NVIDIA Ada Lovelace architecture, the MX3500A-SP module delivers higher performance from its Ray Tracing and Tensor cores compared to GPUs built on the previous architecture. This enables more advanced ray traced graphics and AI capabilities. |
| ● Supports Error Correction Code(ECC) |
| ECC provides higher reliability for compute applications that are sensitive to data corruption, especially important in large-scale cluster computing environments where GPUs process extremely large datasets and/or run applications for extended periods. |
| ● PCIe Gen 4 x16 interface |
| PCIe Gen 4 interface in Ada Lovelace Architecture GPUs delivers faster sequential data transfer than PCIe Gen 3 interface. The increased throughput unlocks higher performance, lower latency and better scalability in Ada Lovelace architecture over previous generations. |
| ● Conformal coating optional service for rugged applications |
| For all series of MXM modules, Aetina offers conformal coating optional service, suitable for harsh environments such as massive-computing embedded applications. |
| ● 5-year long-term supply |
| During long development period, Aetina ensures MXM modules with 5-year availability and exclusive technical support, helping customers to save maintenance cost and time. |
Specifications
Model Number
MX3500A-SP
Bus Type
PCI Express 4.0 x16
Form Factor
MXM graphics module version 3.1, Type B
GPU Engine
GPU Architecture
NVIDIA Ada Lovelace
Graphics Processing Unit
NVIDIA RTX™ 3500 Ada Generation Embedded GPU
NVIDIA CUDA Cores
5120
Graphics Clock (MHz)
1725 / 2250
Floating Point
23.04 TFLOPS
Memory
Memory Size
12GB GDDR6
Memory Clock
18 Gbps
Memory Interface Width
192-bit
Memory Bandwidth
432 GB/sec
Feature Support
DirectX
12 Ultimate
Vulkan
1.2
OpenGL
4.6
Operation System
Windows 10/11 64-bit
Display
Max. Digital Resoultion
7680 x 4320
Outputs
Outputs
DisplayPort 1.4a
Power
Total Graphics Power (TGP)
115 W
Mechanical & Environment
Dimensions (mm)
82.0 x 105.0
Operation Temperature(Standard)
0℃ ~ +55℃ (32°F ~ 131°F)
Storage Temperature
-40°C ~ +85℃ (-40℉ ~ +185℉)
Humidity
10~90%, Non-condensing
Dimensions
Ordering Information
| Model Number | Description |
|---|---|
|
MX3500A-SPA-A1 |
MXM-B, NVIDIA RTX 3500 ADA, Gen4.0 x16, 12GB GDDR6, DP, 115W, 0℃ to +55℃ |
Downloads
| Model Name | File Name | OS | Version | Update Date | Download |
|---|---|---|---|---|---|
|
NVIDIA Ada AI ASIC |
Aetina_AI Accelerator& GPU_brochure | v1.0 | 2024-02-21 | View | |
| MX3500A-SP | Aetina_MX3500A-SP_user manual | v1.0 | 2024-02-05 | View | |
| MX3500A-SP | Aetina_MX3500A-SP_datasheet | v1.1 | 2024-02-01 | View |