13score
#117 of 131
Overall rank
NVIDIAEntry
Arc A310
As an Amazon Associate we may earn from qualifying purchases.
VRAM
4 GB
CUDA
768
FP32
3.072 TF
Bandwidth
124 GB/s
TDP
75W
Boost
2000 MHz
Strengths at a Glance
How it stacks up to the flagship
Each metric is shown as a percentage of the GeForce RTX 5090 D, the strongest card we track.
FP32 compute3.072 TFLOPS3%
vs RTX 5090 D: 104.8 TFLOPS
Memory bandwidth124 GB/s7%
vs RTX 5090 D: 1790 GB/s
VRAM capacity4 GB13%
vs RTX 5090 D: 32 GB
Shading units7684%
vs RTX 5090 D: 21,760
Power efficiency13/10022%
vs RTX 5090 D: 58/100
Synthetic Benchmarks
Estimated benchmark results
Each result is shown as a share of the RTX 5090 D's score in the same test.
3DMark Time Spy4,680pts
vs RTX 5090 D: 36,000 pts
3DMark Port Royal (RT)2,340pts
vs RTX 5090 D: 18,000 pts
Blender (samples/min)676spm
vs RTX 5090 D: 5,200 spm
Geekbench Compute33,600pts
vs RTX 5090 D: 240,000 pts
Performance breakdown
Gaming13
Ray tracing13
AI / Compute14
Creator / 3D13
Power efficiency13
Real-World Gaming
FPS Across Resolutions
1080p · Ultra preset1440p · Ultra preset4K · Ultra preset
Cyberpunk 2077avg 15 fps
1080p
28
1440p
11
4K
6
Counter-Strike 2avg 22 fps
1080p
32
1440p
25
4K
10
Fortniteavg 39 fps
1080p
76
1440p
28
4K
13
Battlefield 5avg 37 fps
1080p
56
1440p
37
4K
18
Far Cry 5avg 29 fps
1080p
43
1440p
28
4K
15
Valorantavg 105 fps
1080p
112
1440p
131
4K
71
Community Feedback
What Owners Say
Owners like it for basic tasks and video playback since it runs cool and quiet. The main complaint is that driver issues make it unreliable for gaming.
Pros
- Sips power, stays whisper quiet
- Fits tiny ITX builds easily
- Handles light gaming at 1080p
- Plays modern codecs for media
Cons
- Only 4GB VRAM limits modern games
- Needs Resizable BAR for full performance
- Driver issues on older games
Supported technologies
Ray TracingDLSSNVENCAV1 Encode
Full specifications
Graphics processor
- Architecture
- Generation 12.7
- Process node
- 6 nm
- Transistors
- 7.2 B
- SM Count
- 6
- Release date
- 2022
Core configuration
- CUDA Cores
- 768
- RT Cores
- 6
- Tensor Cores
- 96
- TMUs
- 32
- ROPs
- 16
- L2 cache
- 4 MB
Memory
- Size
- 4 GB
- Type
- GDDR6
- Bus width
- 64-bit
- Bandwidth
- 124 GB/s
- Memory clock
- 1937 MHz
Clocks & throughput
- Base clock
- 2000 MHz
- Boost clock
- 2000 MHz
- FP32 (float)
- 3.072 TFLOPS
- FP16 (half)
- 3.072 TFLOPS
- Pixel rate
- 32 GPixel/s
- Texture rate
- 64 GTexel/s
Board & power
- TDP
- 75W
- Suggested PSU
- 150W
- Power connectors
- None
- Bus interface
- PCIe 4.0 x8
- Slot width
- 1-slot
Display & outputs
- Max resolution
- 7680×4320
- Outputs
- No outputs
API and SDK support
- DirectX
- 12 Ultimate (12_2)
- Shader Model
- 6.6
- OpenGL
- 4.6
- OpenCL
- 3.0
- Vulkan
- 1.3
Verdict
Our verdict on the Arc A310
An entry-level GPU built for basic display output and light media work, but its 4GB memory limits it severely.
Get it if you need a basic, low-profile GPU for media playback or a simple office PC that sips power and stays dead quiet. Skip it if you plan any gaming or creative work—this card is too weak for even light 3D tasks.
Buy it if…
- You want a dirt-cheap GPU for a basic office PC.
- You need a low-profile card for a small form-factor build.
- You are building a dedicated home media server with Intel Quick Sync.
Leaderboard
Its place in the overall top
3.8
43 votes
Rate this GPU
Add your verdict
Keep exploring