NVIDIA
NVIDIA GeForce RTX 5080
The NVIDIA GeForce RTX 5080 is a high-end desktop graphics card built on the Blackwell architecture. Released in January 2025 at a launch price of $999, it is designed for demanding gamers and content creators seeking top-tier performance.
- Release date
- 2025-01
- Launch price
- 999 USD
- Shader cores
- 10,752 cores
- Video memory
- 16 GB
- Memory type
- Gddr7
- Memory bus width
- 256 bit
Performance and Compute Power
Powered by 10,752 shader cores and a boost clock of 2,620 MHz, the RTX 5080 delivers a heavy-duty FP32 compute of 56.3 TFLOPS. This robust processing muscle easily handles complex visual effects and high frame rates in modern games.
With a base clock of 2,300 MHz, the card maintains steady execution during sustained workloads, ensuring reliable frame delivery without sudden stuttering.
Memory and Bandwidth
Equipped with 16 GB of GDDR7 memory across a 256-bit bus, the card achieves a massive memory bandwidth of 960 GB/s at a speed of 30 Gbps. This rapid data transfer rate keeps up with high-resolution textures and dense asset loading.
The 64 MB last-level cache further cuts down latency by storing frequently accessed data close to the processor, speeding up response times in heavy tasks.
Power and System Requirements
Drawing a board power of 360 W, the RTX 5080 requires a robust power supply unit to operate safely. NVIDIA recommends an 850 W PSU for a PC paired with a Ryzen 9 9950X, alongside a PCIe CEM 5.1 compliant unit using a 12V-2x6 power connector.
While the Founders Edition measures as a 2-slot card with a thickness of 40 mm, NVIDIA's installation guide asks for 3-slot case clearance to accommodate proper airflow and mounting.
Connectivity and Display Setup
The interface relies on PCIe 5.0 with 16 lanes, maximizing data throughput with compatible motherboards. For displays, the card features one HDMI 2.1b port and three DisplayPort 2.1b outputs running at UHBR20 on all three DP connections.
You can drive up to four displays simultaneously, making it simple to run multi-monitor setups for productivity or immersive sim racing rigs.
Full specs
Overview
- Release date
- 2025-01
- Launch price
- 999 USD
Graphics chip
- Architecture
- Blackwell
- Shader cores
- 10,752 cores
- Boost clock
- 2,620 MHz
- Base clock
- 2,300 MHz
- FP32 compute
- 56.3 TFLOPS
Memory
- Video memory
- 16 GB
- Memory type
- Gddr7
- Memory bus width
- 256 bit
- Memory bandwidth
- 960 GB/s
- Memory speed
- 30 Gbps
- Last-level cache
- 64 MB
Power
- Board power
- 360 W
- Recommended PSU
- 850 W
NVIDIA's minimum for a PC with a Ryzen 9 9950X; it recommends a PCIe CEM 5.1 compliant PSU.
- Power connectors
- 12V-2x6
Interface
- PCIe version
- 5.0
- PCIe lanes
- 16 lanes
Display outputs
- HDMI version
- 2.1b
- DisplayPort version
- 2.1b
DisplayPort 2.1b at UHBR20 on all three DP outputs.
- HDMI outputs
- 1 ports
- DisplayPort outputs
- 3 ports
- Max displays
- 4 displays
Video engine
- Hardware encode
- H.264, HEVC, AV1
- Video encoders
- 2 encoders
Size
- Card length
- 304 mm
- Card height
- 137 mm
- Slot width
- 2 slots
The card is 2-slot (40 mm thick), but NVIDIA's install guide asks for 3-slot (61 mm) case clearance.
The Verdict
The RTX 5080 suits enthusiasts building high-performance rigs who want cutting-edge architecture and swift memory speeds for modern displays. Buyers with tight power supplies or restricted PC cases should check clearances and PSU ratings before purchasing. Anyone running older hardware may need a full system upgrade to match its power and slot requirements.
FAQ
What is the launch price of the NVIDIA GeForce RTX 5080?
It launched at $999.
How much video memory does it have?
It includes 16 GB of GDDR7 video memory.
What power supply is recommended?
NVIDIA recommends an 850 W power supply for a PC with a Ryzen 9 9950X using a PCIe CEM 5.1 compliant PSU.
How many displays can be connected at once?
It supports a maximum of 4 displays across its HDMI and DisplayPort outputs.
Spotted something wrong? Tap the next to any section or spec group to let us know. Every report is read.
