Compare · 2 products compared

AMD Radeon RX 9070 GREvsNVIDIA GeForce RTX 4080 Super

These graphics cards offer strong processing power for modern desktop builds while differing significantly in launch cost and underlying memory configurations. Choosing between them comes down to balancing strict power and budget limits against maximum computational throughput.

Report a problem

Intro

What is the problem?

eBay prices refresh hourly — check the listing before you buy. We may earn a commission.

Report a problem

Best for

What is the problem?
Which products? Tick all that apply, or none if you are not sure.

Biggest differences

  • Shader cores

    NVIDIA GeForce RTX 4080 Super leads

    3.33×

    higher is better

    • AMD Radeon RX 9070 GRE3,072 cores
    • NVIDIA GeForce RTX 4080 Super10,240 cores
  • Launch price

    AMD Radeon RX 9070 GRE leads

    −45%

    lower is better

    • AMD Radeon RX 9070 GRE549 USD
    • NVIDIA GeForce RTX 4080 Super999 USD
  • Memory bandwidth

    NVIDIA GeForce RTX 4080 Super leads

    +70%

    higher is better

    • AMD Radeon RX 9070 GRE432 GB/s
    • NVIDIA GeForce RTX 4080 Super736 GB/s
  • Board power

    AMD Radeon RX 9070 GRE leads

    −31%

    lower is better

    • AMD Radeon RX 9070 GRE220 W
    • NVIDIA GeForce RTX 4080 Super320 W

Where each one wins

  • AMD Radeon RX 9070 GRE

    • Launch price
    • Boost clock
    • Board power
    • Recommended PSU
    • PCIe version
    • HDMI version
    • DisplayPort version
  • NVIDIA GeForce RTX 4080 Super

    • Shader cores
    • FP32 compute
    • Video memory
    • Memory type
    • Memory bus width
    • Memory bandwidth
    • Memory speed
    • Last-level cache

What both have

  • HDMI outputs1 ports
  • DisplayPort outputs3 ports

Every spec

Red marks the better value in each row.

SpecAMD Radeon RX 9070 GRENVIDIA GeForce RTX 4080 Super
Overview

Report a problem

Overview specs

What is the problem?
Which specs? Tick all that apply, or none if you are not sure.
Which products? Tick all that apply, or none if you are not sure.
Release date2025-0512024-01
Launch price549 USD2999 USD
Graphics chip

Report a problem

Graphics chip specs

What is the problem?
Which specs? Tick all that apply, or none if you are not sure.
Which products? Tick all that apply, or none if you are not sure.
ArchitectureRdna 4Ada lovelace
Shader cores3,072 cores10,240 cores
Boost clock2,790 MHz2,550 MHz
Base clockNot published2,290 MHz
FP32 compute34.3 TFLOPS52 TFLOPS
Memory

Report a problem

Memory specs

What is the problem?
Which specs? Tick all that apply, or none if you are not sure.
Which products? Tick all that apply, or none if you are not sure.
Video memory12 GB16 GB
Memory typeGddr6Gddr6x
Memory bus width192 bit256 bit
Memory bandwidth432 GB/s736 GB/s
Memory speed18 Gbps23 Gbps
Last-level cache48 MB64 MB
Power

Report a problem

Power specs

What is the problem?
Which specs? Tick all that apply, or none if you are not sure.
Which products? Tick all that apply, or none if you are not sure.
Board power220 W320 W
Recommended PSU650 W750 W3
Power connectors8-pin412V-2x65
Interface

Report a problem

Interface specs

What is the problem?
Which specs? Tick all that apply, or none if you are not sure.
Which products? Tick all that apply, or none if you are not sure.
PCIe version5.04.0
PCIe lanes16 lanesNot published
Display outputs

Report a problem

Display outputs specs

What is the problem?
Which specs? Tick all that apply, or none if you are not sure.
Which products? Tick all that apply, or none if you are not sure.
HDMI version2.1b2.1a
DisplayPort version2.1a1.4a
HDMI outputs1 ports61 ports
DisplayPort outputs3 ports73 ports
Max displaysNot published4 displays
Video engine

Report a problem

Video engine specs

What is the problem?
Which specs? Tick all that apply, or none if you are not sure.
Which products? Tick all that apply, or none if you are not sure.
Hardware encodeH.264, HEVC, AV1AV1
Video encodersNot published2 encoders
Size

Report a problem

Size specs

What is the problem?
Which specs? Tick all that apply, or none if you are not sure.
Which products? Tick all that apply, or none if you are not sure.
Card lengthNot published304 mm
Card heightNot published137 mm
Slot widthNot published3 slots
Notes on 7 values — test conditions and caveats
  1. 1AMD Radeon RX 9070 GRE, Release date: Chinese retail deliveries began May 8, 2025; the global launch followed at Computex 2026.
  2. 2AMD Radeon RX 9070 GRE, Launch price: Global launch MSRP $549; Chinese AIB cards started at ¥4,199 (≈$450 per Notebookcheck).
  3. 3NVIDIA GeForce RTX 4080 Super, Recommended PSU: NVIDIA's 750 W minimum assumes a PC with a Ryzen 9 5900X; needs vary by configuration.
  4. 4AMD Radeon RX 9070 GRE, Power connectors: AIB-only card; most listed partner models use two 8-pin connectors (Yeston Sakura variants use three).
  5. 5NVIDIA GeForce RTX 4080 Super, Power connectors: Card takes a 16-pin 12V-2x6 connector; a 3x PCIe 8-pin adapter ships in the box for PSUs without a PCIe Gen 5 cable.
  6. 6AMD Radeon RX 9070 GRE, HDMI outputs: AMD sells no retail reference card; port counts are for AMD's reference design as listed by VideoCardz.
  7. 7AMD Radeon RX 9070 GRE, DisplayPort outputs: AMD sells no retail reference card; port counts are for AMD's reference design as listed by VideoCardz.

Performance and Compute

The NVIDIA GeForce RTX 4080 Super delivers higher FP32 compute power at 52 TFLOPS compared to the AMD Radeon RX 9070 GRE at 34.3 TFLOPS, helping handle heavy rendering tasks. However, the AMD card achieves a higher boost clock of 2,790 MHz to accelerate demanding instruction loads.

For sheer shader counts, the NVIDIA card includes 10,240 cores against the AMD option's 3,072 cores. This larger core count drives heavier parallel processing workloads efficiently.

Report a problem

Performance and Compute

What is the problem?

Memory Architecture

The GeForce RTX 4080 Super features 16 GB of GDDR6X video memory across a 256-bit bus, generating a bandwidth of 736 GB/s. In contrast, the Radeon RX 9070 GRE offers 12 GB of GDDR6 memory on a 192-bit bus reaching 432 GB/s.

Larger memory capacity and wider buses help maintain high frame rates at elevated resolutions. The NVIDIA card also includes a larger 64 MB last-level cache compared to 48 MB on the AMD card.

Report a problem

Memory Architecture

What is the problem?

Power and Efficiency

The Radeon RX 9070 GRE requires a lower board power of 220 W and a 650 W power supply, easing demands on system components. The GeForce RTX 4080 Super draws 320 W and requires a 750 W minimum power supply.

Lower board power helps keep system temperatures down and reduces thermal management requirements inside the chassis.

Report a problem

Power and Efficiency

What is the problem?

Connectivity and Ports

The Radeon RX 9070 GRE supports PCIe 5.0 for faster interface speeds, whereas the GeForce RTX 4080 Super uses PCIe 4.0. Both designs feature three DisplayPort outputs and one HDMI port.

Display output versions lean in favor of the AMD card with HDMI 2.1b and DisplayPort 2.1a support, compared to HDMI 2.1a and DisplayPort 1.4a on the NVIDIA card.

Report a problem

Connectivity and Ports

What is the problem?

The Verdict

Buy the AMD Radeon RX 9070 GRE if you want lower board power consumption and a more affordable purchase price. Buy the NVIDIA GeForce RTX 4080 Super if you require higher memory bandwidth and maximum compute throughput.

Report a problem

Verdict

What is the problem?

Where to buy

FAQ

What power connectors do these graphics cards require?

The Radeon RX 9070 GRE uses 8-pin connectors depending on the partner model, while the GeForce RTX 4080 Super takes a 16-pin 12V-2x6 connector.

How much video memory does each card have?

The Radeon RX 9070 GRE features 12 GB of GDDR6 memory, whereas the GeForce RTX 4080 Super includes 16 GB of GDDR6X memory.

What are the recommended power supply capacities?

The AMD card recommends a 650 W power supply, while the NVIDIA card requires a 750 W minimum.

Report a problem

FAQ

What is the problem?

Spotted something wrong? Tap the next to any section or spec group to let us know. Every report is read.