Compare · 2 products compared
AMD Radeon RX 9070vsNVIDIA GeForce RTX 4080 Super
The AMD Radeon RX 9070 and the NVIDIA GeForce RTX 4080 Super share 16 GB of video memory and a 256-bit memory bus width. Choosing between them comes down to balancing lower power draw and cost against raw shader power and memory speed.
eBay prices refresh hourly — check the listing before you buy. We may earn a commission.
AMDBest for Budget-conscious builders seeking lower power draw
6of 12 specs better
NVIDIABest for Enthusiasts demanding maximum compute performance
6of 12 specs better
+ Add a product to compare
Biggest differences
Shader cores
NVIDIA GeForce RTX 4080 Super leads
2.86×
higher is better
- AMD Radeon RX 90703,584 cores
- NVIDIA GeForce RTX 4080 Super10,240 cores
Launch price
AMD Radeon RX 9070 leads
−45%
lower is better
- AMD Radeon RX 9070549 USD
- NVIDIA GeForce RTX 4080 Super999 USD
Board power
AMD Radeon RX 9070 leads
−31%
lower is better
- AMD Radeon RX 9070220 W
- NVIDIA GeForce RTX 4080 Super320 W
FP32 compute
NVIDIA GeForce RTX 4080 Super leads
+44%
higher is better
- AMD Radeon RX 907036.1 TFLOPS
- NVIDIA GeForce RTX 4080 Super52 TFLOPS
Where each one wins
AMD Radeon RX 9070
- Launch price
- Board power
- Recommended PSU
- PCIe version
- HDMI version
- DisplayPort version
NVIDIA GeForce RTX 4080 Super
- Shader cores
- Boost clock
- FP32 compute
- Memory type
- Memory bandwidth
- Memory speed
What both have
- Video memory16 GB
- Memory bus width256 bit
- Last-level cache64 MB
- Max displays4 displays
Every spec
Red marks the better value in each row.
| Spec | AMD Radeon RX 9070 | NVIDIA GeForce RTX 4080 Super |
|---|---|---|
| Overview | ||
| Release date | 2025-03 | 2024-01 |
| Launch price | 549 USD1 | 999 USD |
| Graphics chip | ||
| Architecture | Rdna 4 | Ada lovelace |
| Shader cores | 3,584 cores | 10,240 cores |
| Boost clock | 2,520 MHz2 | 2,550 MHz |
| Base clock | Not published | 2,290 MHz |
| FP32 compute | 36.1 TFLOPS | 52 TFLOPS |
| Memory | ||
| Video memory | 16 GB | 16 GB |
| Memory type | Gddr6 | Gddr6x |
| Memory bus width | 256 bit | 256 bit |
| Memory bandwidth | 640 GB/s | 736 GB/s |
| Memory speed | 20 Gbps | 23 Gbps |
| Last-level cache | 64 MB | 64 MB |
| Power | ||
| Board power | 220 W3 | 320 W |
| Recommended PSU | 650 W | 750 W4 |
| Power connectors | 8-pin5 | 12V-2x66 |
| Interface | ||
| PCIe version | 5.0 | 4.0 |
| PCIe lanes | 16 lanes | Not published |
| Display outputs | ||
| HDMI version | 2.1b | 2.1a |
| DisplayPort version | 2.1a7 | 1.4a |
| HDMI outputs | Not published | 1 ports |
| DisplayPort outputs | Not published | 3 ports |
| Max displays | 4 displays | 4 displays |
| Video engine | ||
| Hardware encode | H.264, HEVC, AV1 | AV1 |
| Video encoders | Not published | 2 encoders |
| Size | ||
| Card length | Not published | 304 mm |
| Card height | Not published | 137 mm |
| Slot width | Not published | 3 slots |
Notes on 7 values — test conditions and caveats
- 1AMD Radeon RX 9070, Launch price: AMD's $549 starting MSRP at launch; AMD sells no reference card, so partner models may cost more.
- 2AMD Radeon RX 9070, Boost clock: Peak 'up to' boost; AMD's game clock is 2070 MHz for typical gaming.
- 3AMD Radeon RX 9070, Board power: Partner overclocked cards can exceed 220 W; IGN measured 249 W peaks on a factory-OC card.
- 4NVIDIA GeForce RTX 4080 Super, Recommended PSU: NVIDIA's 750 W minimum assumes a PC with a Ryzen 9 5900X; needs vary by configuration.
- 5AMD Radeon RX 9070, Power connectors: Two 8-pin PCIe power connectors are required (2x 8-pin per AMD and Sapphire).
- 6NVIDIA 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.
- 7AMD Radeon RX 9070, DisplayPort version: Club386 lists the DP 2.1a link rate as UHBR13.5; UHBR20 is not supported.
Performance and Architecture
The NVIDIA GeForce RTX 4080 Super utilizes the Ada Lovelace architecture to deliver significantly more shader cores and a higher FP32 compute capability of 52 TFLOPS compared to the AMD Radeon RX 9070. This raw computing advantage translates into faster processing for heavy workloads and demanding graphics.
In contrast, the AMD Radeon RX 9070 relies on the RDNA 4 architecture, offering 3,584 cores and 36.1 TFLOPS of FP32 compute. While it trails in raw shader count, it delivers a peak boost clock of 2,520 MHz [Peak 'up to' boost; AMD's game clock is 2070 MHz for typical gaming.] that keeps it competitive in everyday gaming scenarios.
Memory and Bandwidth
Both cards feature 16 GB of video memory on a 256-bit bus, making them well-equipped for modern high-resolution textures. They also share a 64 MB last-level cache.
The NVIDIA GeForce RTX 4080 Super pulls ahead by using GDDR6X memory type, achieving a memory speed of 23 Gbps and a bandwidth of 736 GB/s. The AMD Radeon RX 9070 uses standard GDDR6 memory, resulting in a lower memory speed of 20 Gbps and a bandwidth of 640 GB/s.
Power and Efficiency
Power consumption heavily favors the AMD Radeon RX 9070, which has a board power of just 220 W ★ [Partner overclocked cards can exceed 220 W; IGN measured 249 W peaks on a factory-OC card.]. It requires an accessible power setup with an 8-pin [Two 8-pin PCIe power connectors are required (2x 8-pin per AMD and Sapphire).] connector and a 650 W ★ recommended PSU.
Meanwhile, the NVIDIA GeForce RTX 4080 Super demands a much higher board power of 320 W and a 750 W [NVIDIA's 750 W minimum assumes a PC with a Ryzen 9 5900X; needs vary by configuration.] minimum PSU. It also requires a modern 12V-2x6 power connector, though a 3x PCIe 8-pin adapter ships in the box for older power supplies.
Connectivity and Features
The AMD Radeon RX 9070 introduces modern interface standards, supporting PCIe 5.0 ★ and HDMI 2.1b ★ alongside DisplayPort 2.1a ★ [Club386 lists the DP 2.1a link rate as UHBR13.5; UHBR20 is not supported.]. It supports up to 4 displays.
The NVIDIA GeForce RTX 4080 Super sticks to PCIe 4.0 and DisplayPort 1.4a, while offering HDMI 2.1a. Its physical layout includes 1 HDMI port and 3 DisplayPort ports, also driving up to 4 displays.
The Verdict
Buy the AMD Radeon RX 9070 if you want a lower power draw and a starting MSRP of $549 [AMD's $549 starting MSRP at launch; AMD sells no reference card, so partner models may cost more.]. Buy the NVIDIA GeForce RTX 4080 Super if you need maximum compute power and faster GDDR6X memory.
Where to buy
AMD Radeon RX 9070
Best for Budget-conscious builders seeking lower power draw
NVIDIA GeForce RTX 4080 Super
Best for Enthusiasts demanding maximum compute performance
FAQ
What is the launch price difference between these two cards?
The AMD Radeon RX 9070 launched at a starting MSRP of $549 [AMD's $549 starting MSRP at launch; AMD sells no reference card, so partner models may cost more.], while the NVIDIA GeForce RTX 4080 Super launched at $999.
What kind of power supply do I need for the AMD Radeon RX 9070?
A 650 W ★ power supply is recommended, and the card connects using 8-pin [Two 8-pin PCIe power connectors are required (2x 8-pin per AMD and Sapphire).] power connectors.
Do both graphics cards support the same video memory capacity?
Yes, both cards feature 16 GB of video memory on a 256-bit memory bus width.
Which graphics card has better DisplayPort support?
The AMD Radeon RX 9070 features DisplayPort 2.1a ★ [Club386 lists the DP 2.1a link rate as UHBR13.5; UHBR20 is not supported.], whereas the NVIDIA GeForce RTX 4080 Super uses DisplayPort 1.4a.
Spotted something wrong? Tap the next to any section or spec group to let us know. Every report is read.





