GeForce RTX 4070 vs Radeon Pro W6900X
Aggregate performance score
We've compared Radeon Pro W6900X with GeForce RTX 4070, including specs and performance data.
RTX 4070 outperforms Pro W6900X by an impressive 54% based on our aggregate benchmark results.
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 109 | 33 |
| Place by popularity | not in top-100 | 61 |
| Cost-effectiveness evaluation | 3.38 | 61.87 |
| Power efficiency | 10.80 | 24.98 |
| Architecture | RDNA 2.0 (2020−2025) | Ada Lovelace (2022−2024) |
| GPU code name | Navi 21 | AD104 |
| Market segment | Workstation | Desktop |
| Release date | 3 August 2021 (5 years ago) | 12 April 2023 (3 years ago) |
| Launch price (MSRP) | $4,999 | $599 |
Cost-effectiveness evaluation
The higher the ratio, the better. We use the manufacturer's recommended prices.
RTX 4070 has 1730% better value for money than Pro W6900X.
Performance to price scatter graph
Currently popular graphics cards are shown for comparison.
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Real power consumption of some graphics cards can well exceed their nominal TDP, especially if overclocked.
| Pipelines / CUDA cores | 5120 | 5888 |
| Core clock speed | 1825 MHz | 1920 MHz |
| Boost clock speed | 2150 MHz | 2475 MHz |
| Number of transistors | 26,800 million | 35,800 million |
| Manufacturing process technology | 7 nm | 5 nm |
| Power consumption (TDP) | 300 Watt | 200 Watt |
| Texture fill rate | 688.0 | 455.4 |
| Floating-point processing power | 22.02 TFLOPS | 29.15 TFLOPS |
| ROPs | 128 | 64 |
| TMUs | 320 | 184 |
| Tensor Cores | no data | 184 |
| Ray Tracing Cores | 80 | 46 |
| L0 Cache | 1.3 MB | no data |
| L1 Cache | 1 MB | 5.8 MB |
| L2 Cache | 4 MB | 36 MB |
| L3 Cache | 128 MB | no data |
Form factor & compatibility
Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| Length | 267 mm | 240 mm |
| Width | Quad-slot | 2-slot |
| Supplementary power connectors | None | 1x 16-pin |
VRAM capacity and type
Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.
| Memory type | GDDR6 | GDDR6X |
| Maximum RAM amount | 32 GB | 12 GB |
| Memory bus width | 256 Bit | 192 Bit |
| Memory clock speed | 2000 MHz | 1313 MHz |
| Memory bandwidth | 512.0 GB/s | 504.2 GB/s |
| Shared memory | - | - |
| Resizable BAR | + | + |
Connectivity and outputs
This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.
| Display Connectors | 1x HDMI, 4x Thunderbolt | 1x HDMI 2.1, 3x DisplayPort 1.4a |
| HDMI | + | + |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
| Shader Model | 6.5 | 6.7 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.2 | 1.3 |
| CUDA | - | 8.9 |
| DLSS | - | + |
Synthetic benchmarks
Non-gaming benchmark results comparison. The combined score is measured on a 0-100 point scale.
Combined synthetic benchmark score
This is our combined benchmark score.
Passmark
This is the most ubiquitous GPU benchmark. It gives the graphics card a thorough evaluation under various types of load, providing four separate benchmarks for Direct3D versions 9, 10, 11 and 12 (the last being done in 4K resolution if possible), and few more tests engaging DirectCompute capabilities.
Gaming performance
Let's see how good the compared graphics cards are for gaming. Particular gaming benchmark results are measured in FPS.
Average FPS across all PC games
Here are the average frames per second in a large set of popular games across different resolutions:
| Full HD | 120−130 | 198
+65%
|
| 1440p | 65−70
−37.5%
| 104 |
| 4K | 40−45 | 63
+57.5%
|
Cost per frame, $
| 1080p | 41.66 | 3.03
+92.7%
|
| 1440p | 76.91 | 5.76
+92.5%
|
| 4K | 124.98 | 9.51
+92.4%
|
- RTX 4070 has 93% lower cost per frame in 1080p
- RTX 4070 has 93% lower cost per frame in 1440p
- RTX 4070 has 92% lower cost per frame in 4K
FPS performance in popular games
Full HD
Low
| Atomic Heart | 320 | 320 |
| Counter-Strike 2 | 300−350 | 300−350 |
| Cyberpunk 2077 | 216 | 216 |
| Resident Evil 4 Remake | 240 | 240 |
Full HD
Medium
| Atomic Heart | 250 | 250 |
| Battlefield 5 | 170−180 | 170−180 |
| Counter-Strike 2 | 300−350 | 300−350 |
| Cyberpunk 2077 | 174 | 174 |
| Far Cry 5 | 249 | 249 |
| Fortnite | 300−350 | 300−350 |
| Forza Horizon 4 | 250−260 | 250−260 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180 | 170−180 |
| Valorant | 350−400 | 350−400 |
Full HD
High
| Atomic Heart | 148 | 148 |
| Battlefield 5 | 170−180 | 170−180 |
| Counter-Strike 2 | 300−350 | 300−350 |
| Counter-Strike: Global Offensive | 270−280 | 270−280 |
| Cyberpunk 2077 | 143 | 143 |
| Far Cry 5 | 234 | 234 |
| Fortnite | 300−350 | 300−350 |
| Forza Horizon 4 | 250−260 | 250−260 |
| Grand Theft Auto V | 174 | 174 |
| Metro Exodus | 170 | 170 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180 | 170−180 |
| The Witcher 3: Wild Hunt | 351 | 351 |
| Valorant | 350−400 | 350−400 |
Full HD
Ultra
| Battlefield 5 | 170−180 | 170−180 |
| Cyberpunk 2077 | 128 | 128 |
| Far Cry 5 | 218 | 218 |
| Forza Horizon 4 | 250−260 | 250−260 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180 | 170−180 |
| The Witcher 3: Wild Hunt | 171 | 171 |
| Valorant | 350−400 | 350−400 |
Full HD
Epic
| Fortnite | 300−350 | 300−350 |
1440p
High
| Counter-Strike 2 | 190−200 | 190−200 |
| Counter-Strike: Global Offensive | 500−550 | 500−550 |
| Grand Theft Auto V | 137 | 137 |
| Metro Exodus | 104 | 104 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 170−180 | 170−180 |
| Valorant | 450−500 | 450−500 |
1440p
Ultra
| Battlefield 5 | 160−170 | 160−170 |
| Cyberpunk 2077 | 81 | 81 |
| Far Cry 5 | 173 | 173 |
| Forza Horizon 4 | 220−230 | 220−230 |
| The Witcher 3: Wild Hunt | 140 | 140 |
1440p
Epic
| Fortnite | 150−160 | 150−160 |
4K
High
| Atomic Heart | 60−65 | 60−65 |
| Counter-Strike 2 | 85−90 | 85−90 |
| Grand Theft Auto V | 146 | 146 |
| Metro Exodus | 65 | 65 |
| The Witcher 3: Wild Hunt | 117 | 117 |
| Valorant | 300−350 | 300−350 |
4K
Ultra
| Battlefield 5 | 120−130 | 120−130 |
| Counter-Strike 2 | 85−90 | 85−90 |
| Cyberpunk 2077 | 36 | 36 |
| Far Cry 5 | 93 | 93 |
| Forza Horizon 4 | 170−180 | 170−180 |
| PLAYERUNKNOWN'S BATTLEGROUNDS | 95−100 | 95−100 |
4K
Epic
| Fortnite | 75−80 | 75−80 |
This is how Pro W6900X and RTX 4070 compete in popular games:
- RTX 4070 is 65% faster in 1080p
- RTX 4070 is 60% faster in 1440p
- RTX 4070 is 58% faster in 4K
All in all, in popular games:
- there's a draw in 59 tests (100%)
Pros & cons summary
| Performance score | 41.80 | 64.46 |
| Recency | 3 August 2021 | 12 April 2023 |
| Maximum RAM amount | 32 GB | 12 GB |
| Chip lithography | 7 nm | 5 nm |
| Power consumption (TDP) | 300 Watt | 200 Watt |
Pro W6900X has a 167% higher maximum VRAM amount.
RTX 4070, on the other hand, has a 54% higher aggregate performance score, an age advantage of 1 year, a 29% more advanced lithography process, and 33% lower power consumption.
The GeForce RTX 4070 is our recommended choice as it beats the Radeon Pro W6900X in performance tests.
Be aware that Radeon Pro W6900X is a workstation graphics card while GeForce RTX 4070 is a desktop one.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
