GeForce RTX 5070 vs Radeon HD 7990
Aggregate performance score
We've compared Radeon HD 7990 and GeForce RTX 5070, covering specs and all relevant benchmarks.
RTX 5070 outperforms HD 7990 by a whopping 446% based on our aggregate benchmark results.
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 371 | 11 |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | 1.51 | no data |
| Power efficiency | 2.64 | no data |
| Architecture | GCN 1.0 (2011−2020) | Blackwell 2.0 (2025) |
| GPU code name | Malta | GB205 |
| Market segment | Desktop | Desktop |
| Release date | 24 April 2013 (11 years ago) | February 2025 (recently) |
| Launch price (MSRP) | $999 | $549 |
Cost-effectiveness evaluation
The higher the performance-to-price ratio, the better. We use the manufacturer's recommended prices 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. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.
| Pipelines / CUDA cores | 2048 ×2 | 6144 |
| Core clock speed | 950 MHz | 2165 MHz |
| Boost clock speed | 1000 MHz | 2510 MHz |
| Number of transistors | 4,313 million | 31,000 million |
| Manufacturing process technology | 28 nm | 5 nm |
| Power consumption (TDP) | 375 Watt | 250 Watt |
| Texture fill rate | 128.0 ×2 | 481.9 |
| Floating-point processing power | 4.096 TFLOPS ×2 | 30.84 TFLOPS |
| ROPs | 32 ×2 | 64 |
| TMUs | 128 ×2 | 192 |
| Tensor Cores | no data | 192 |
| Ray Tracing Cores | no data | 48 |
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 3.0 x16 | PCIe 5.0 x16 |
| Length | 307 mm | no data |
| Width | 2-slot | 2-slot |
| Supplementary power connectors | 2x 8-pin | 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 | GDDR5 | GDDR7 |
| Maximum RAM amount | 3 GB ×2 | 12 GB |
| Memory bus width | 384 Bit ×2 | 192 Bit |
| Memory clock speed | 1500 MHz | 1750 MHz |
| Memory bandwidth | 288.0 GB/s ×2 | 672.0 GB/s |
Connectivity and outputs
Types and number of video connectors present on the reviewed GPUs. As a rule, data in this section is precise only for desktop reference ones (so-called Founders Edition for NVIDIA chips). OEM manufacturers may change the number and type of output ports, while for notebook cards availability of certain video outputs ports depends on the laptop model rather than on the card itself.
| Display Connectors | 1x DVI, 4x mini-DisplayPort | 1x HDMI 2.1b, 3x DisplayPort 2.1a |
| HDMI | - | + |
API and SDK compatibility
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 (11_1) | 12 Ultimate (12_2) |
| Shader Model | 5.1 | 6.8 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 1.2 | 3.0 |
| Vulkan | 1.2.131 | 1.4 |
| CUDA | - | 10.1 |
| DLSS | - | + |
Synthetic benchmark performance
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.
FPS performance in popular games
Full HD
Low Preset
| Atomic Heart | 1−2
+0%
|
1−2
+0%
|
| Counter-Strike 2 | 7−8
+0%
|
7−8
+0%
|
| Cyberpunk 2077 | 1−2
+0%
|
1−2
+0%
|
Full HD
Medium Preset
| Atomic Heart | 1−2
+0%
|
1−2
+0%
|
| Counter-Strike 2 | 7−8
+0%
|
7−8
+0%
|
| Cyberpunk 2077 | 1−2
+0%
|
1−2
+0%
|
| Forza Horizon 4 | 1−2
+0%
|
1−2
+0%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 6−7
+0%
|
6−7
+0%
|
| Valorant | 24−27
+0%
|
24−27
+0%
|
Full HD
High Preset
| Atomic Heart | 1−2
+0%
|
1−2
+0%
|
| Counter-Strike 2 | 7−8
+0%
|
7−8
+0%
|
| Counter-Strike: Global Offensive | 8−9
+0%
|
8−9
+0%
|
| Cyberpunk 2077 | 1−2
+0%
|
1−2
+0%
|
| Forza Horizon 4 | 1−2
+0%
|
1−2
+0%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 6−7
+0%
|
6−7
+0%
|
| The Witcher 3: Wild Hunt | 3−4
+0%
|
3−4
+0%
|
| Valorant | 24−27
+0%
|
24−27
+0%
|
Full HD
Ultra Preset
| Counter-Strike 2 | 7−8
+0%
|
7−8
+0%
|
| Cyberpunk 2077 | 1−2
+0%
|
1−2
+0%
|
| Forza Horizon 4 | 1−2
+0%
|
1−2
+0%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 6−7
+0%
|
6−7
+0%
|
| The Witcher 3: Wild Hunt | 3−4
+0%
|
3−4
+0%
|
| Valorant | 24−27
+0%
|
24−27
+0%
|
1440p
Ultra Preset
| Forza Horizon 4 | 0−1 | 0−1 |
| The Witcher 3: Wild Hunt | 1−2
+0%
|
1−2
+0%
|
4K
High Preset
| Grand Theft Auto V | 14−16
+0%
|
14−16
+0%
|
| Valorant | 1−2
+0%
|
1−2
+0%
|
4K
Ultra Preset
| Far Cry 5 | 1−2
+0%
|
1−2
+0%
|
| PLAYERUNKNOWN'S BATTLEGROUNDS | 1−2
+0%
|
1−2
+0%
|
4K
Epic Preset
| Fortnite | 2−3
+0%
|
2−3
+0%
|
All in all, in popular games:
- there's a draw in 29 tests (100%)
Pros & cons summary
| Performance score | 14.30 | 78.07 |
| Maximum RAM amount | 3 GB | 12 GB |
| Chip lithography | 28 nm | 5 nm |
| Power consumption (TDP) | 375 Watt | 250 Watt |
RTX 5070 has a 445.9% higher aggregate performance score, a 300% higher maximum VRAM amount, a 460% more advanced lithography process, and 50% lower power consumption.
The GeForce RTX 5070 is our recommended choice as it beats the Radeon HD 7990 in performance tests.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
