NVIDIA GeForce RTX 4070 vs NVIDIA GeForce RTX 5090
The GeForce RTX 5090 has 116% better performance than the GeForce RTX 4070 for the 3DMark 11 Performance GPU benchmark.
The GeForce RTX 4070 has 54% better value for the money than the GeForce RTX 5090 for the 3DMark 11 Performance GPU benchmark.
Summary
About the NVIDIA GeForce RTX 4070 GPU
The NVIDIA GeForce RTX 4070 is a desktop graphics card that launched in Q2 2023 with a MSRP of $599. It is built on the Ada Lovelace GPU microarchitecture (codename AD104) and is manufactured on a 5 nm process.
Memory
The RTX 4070 has 12 GB of GDDR6X memory, with a 1,313 MHz memory clock and a 192 bit interface. This gives it a memory bandwidth of 504.2 Gb/s, which affects how fast it can transfer data to and from memory. GPU memory stores temporary data that helps the GPU with complex math and graphics operations. More memory is generally better, as not having enough can cause performance bottlenecks.
Cores and Clock Speeds
The RTX 4070 includes 5,888 CUDA cores, the processing units for handling parallel computing tasks. The GPU operates at a core clock speed of 1,920 MHz and can dynamically boost its clock speed up to 2,475 MHz. Complementing the processing units are 184 texture mapping units (TMUs) for efficient texture filtering and 64 render output units (ROPs) for pixel processing. Additionally, the GPU features 184 tensor cores optimized for AI-accelerated workloads and 46 RT cores dedicated to real-time ray tracing calculations.
Compatibility & Power Consumption
The RTX 4070 occupies 2 PCIe expansion slots. It supports HDMI 2.1, DisplayPort 1.4a display connections. NVIDIA recommends a power supply of at least 550 W to handle the GPU's thermal design power (TDP) of 200 W.
Benchmark Performance & Value
The RTX 4070 has the 25th best 3DMark 11 Performance GPU score among the 566 benchmarked GPUs in our database. It achieves 46.21% of the performance of the best benchmarked GPU, the NVIDIA GeForce RTX 5090. Its 59,969 score and $599 launch price (MSRP) gives it a performance per dollar of 100.12. This is the 23rd best in value for the 3DMark 11 Performance GPU benchmark.
About the NVIDIA GeForce RTX 5090 GPU
The NVIDIA GeForce RTX 5090 is a desktop graphics card that launched in Q1 2025 with a MSRP of $1,999. It is built on the Blackwell 2.0 GPU microarchitecture (codename GB202) and is manufactured on a 5 nm process.
Memory
The RTX 5090 has 32 GB of GDDR7 memory, with a 1,750 MHz memory clock and a 512 bit interface. This gives it a memory bandwidth of 1.79 Tb/s, which affects how fast it can transfer data to and from memory. GPU memory stores temporary data that helps the GPU with complex math and graphics operations. More memory is generally better, as not having enough can cause performance bottlenecks.
Cores and Clock Speeds
The RTX 5090 includes 21,760 CUDA cores, the processing units for handling parallel computing tasks. The GPU operates at a core clock speed of 2,017 MHz and can dynamically boost its clock speed up to 2,407 MHz. Complementing the processing units are 680 texture mapping units (TMUs) for efficient texture filtering and 176 render output units (ROPs) for pixel processing. Additionally, the GPU features 680 tensor cores optimized for AI-accelerated workloads and 170 RT cores dedicated to real-time ray tracing calculations.
Compatibility & Power Consumption
The RTX 5090 occupies 2 PCIe expansion slots. It supports HDMI 2.1b, DisplayPort 2.1b display connections. NVIDIA recommends a power supply of at least 950 W to handle the GPU's thermal design power (TDP) of 575 W.
Benchmark Performance & Value
The RTX 5090 has the best 3DMark 11 Performance GPU score among the 566 benchmarked GPUs in our database. Its 129,772 score and $1,999 launch price (MSRP) gives it a performance per dollar of 64.92. This is the 60th best in value for the 3DMark 11 Performance GPU benchmark.
General Info
General overview of the GPU, including details like its manufacturer, release date, launch price, and current production status.
| Info | GeForce RTX 4070 | GeForce RTX 5090 |
|---|---|---|
| Manufacturer | NVIDIA | NVIDIA |
| Architecture | Ada Lovelace | Blackwell 2.0 |
| Market Segment | Desktop | Desktop |
| Release Date | Q2 2023 | Q1 2025 |
| Launch Price (MSRP) | $599 | $1,999 |
| Production Status | Active | Active |
| Shop | Check Price | Check Price |
Gaming Performance
Select a game to compare FPS metrics
FPS Benchmarks
This table showcases the average frame rate (FPS) achieved both GPUs in at various resolutions. Frame rate is a crucial indicator of how smoothly the GPU can run the game. A higher FPS generally translates to a smoother gameplay experience.
- Frames Per Second
God of War Ragnarök | GeForce RTX 4070 | GeForce RTX 5090 |
|---|---|---|
| Low - 1080p | -- | -- |
| Medium - 1080p | -- | -- |
| High - 1080p | -- | 387 FPS |
| Ultra - 1080p | -- | 362 FPS |
| QHD - 1440p | -- | 300 FPS |
| 4K UHD - 2160p | -- | 193 FPS |
Compare Frames Per Second (FPS)
The average frame rate (FPS) in can be compared to similar GPUs to assess relative performance. Generally, higher FPS results in a smoother gameplay experience.
| GPU | Frames Per Second | |
|---|---|---|
| GeForce RTX 5090 | 192.7 | |
| GeForce RTX 5080 | 129 | -33% |
| GeForce RTX 5070 Ti | 109 | -43% |
| GeForce RTX 4080 | 100.8 | -48% |
| Radeon RX 9070 | 94 | -51% |
| Radeon RX 7900 XT | 86.1 | -55% |
| GeForce RTX 5070 | 81.5 | -58% |
| GeForce RTX 5060 Ti (16 GB) | 62.7 | -67% |
| GeForce RTX 4060 | 39 | -80% |
Compare Cost Per Frame
The average cost per frame in can be compared to similar GPUs to assess relative value. Generally, a lower cost per frame implies better value for your money.
| GPU | Cost Per Frame | |
|---|---|---|
| GeForce RTX 5070 Ti | 6.87 | -34% |
| Arc B580 | 7.37 | -29% |
| GeForce RTX 4060 | 7.67 | -26% |
| GeForce RTX 5080 | 7.74 | -25% |
| GeForce RTX 5060 | 8.17 | -21% |
| GeForce RTX 5090 | 10.37 | |
| Radeon RX 7900 XT | 10.44 | +1% |
| GeForce RTX 4080 | 11.89 | +15% |
Benchmark Performance
The GeForce RTX 5090 has 116% better performance than the GeForce RTX 4070 for the 3DMark 11 Performance GPU benchmark.
The GeForce RTX 4070 is ranked 25th with a score of 59,969, and the GeForce RTX 5090 is ranked 1st with a score of 129,772.
The GeForce RTX 4070 has 54% better value for the money than the GeForce RTX 5090 for the 3DMark 11 Performance GPU benchmark.
The GeForce RTX 4070 is ranked 23rd with a performance per dollar of 100.12, and the GeForce RTX 5090 is ranked 60th with a performance per dollar of 64.92.
Relative Performance
The average score in the benchmark test can be compared to similar GPUs to assess relative performance. Generally, powerful GPUs tend to have higher scores.
| GPU | Benchmark Performance | |
|---|---|---|
| GeForce RTX 5090 | 129,772 | +116% |
| GeForce RTX 4090 | 104,598 | +74% |
| GeForce RTX 5080 | 100,190 | +67% |
| Radeon RX 9070 XT | 98,912 | +65% |
| Radeon RX 9070 | 88,618 | +48% |
| GeForce RTX 3090 | 61,518 | +3% |
| GeForce RTX 3080 Ti | 60,980 | +2% |
| GeForce RTX 4070 | 59,969 | |
| Radeon RX 6900 XT | 59,119 | -1% |
| Radeon RX 7700 XT | 57,340 | -4% |
Relative Value For Money
The average performance per dollar in the benchmark test can be compared to similar GPUs to assess relative value. A higher score implies a better value for your money.
| GPU | Performance Per Dollar | |
|---|---|---|
| GeForce RTX 4070 SUPER | 114.15 | +14% |
| GeForce RTX 5080 | 100.29 | +0% |
| GeForce RTX 4070 | 100.12 | |
| Radeon RX 6800 | 95.14 | -5% |
| GeForce RTX 4070 Ti | 94.63 | -5% |
| Radeon RX 6950 XT | 65.72 | -34% |
| GeForce RTX 4090 | 65.41 | -35% |
| GeForce RTX 5090 | 64.92 | -35% |
| Radeon RX 6900 XT | 59.18 | -41% |
| GeForce RTX 3080 Ti | 50.86 | -49% |
Benchmark Scores
This table showcases the average performance scores achieved by both GPUs across industry-standard benchmark tests. These scores provide a valuable insight into overall performance. Powerful GPUs tend to have higher scores.
- Popular
| Benchmark | GeForce RTX 4070 | GeForce RTX 5090 |
|---|---|---|
| 3DMark Time Spy Graphics | 17,837 | 47,047.5 (+163.76%) |
| 3DMark Time Spy Score | 17,905 | 37,197 (+107.75%) |
| 3DMark Cloud Gate Graphics | 193,559 | 385,092 (+98.95%) |
| 3DMark Fire Strike Standard Graphics | 43,824 | 89,073 (+103.25%) |
| 3DMark Night Raid Graphics | 187,585 | 228,764 (+21.95%) |
| 3DMark 11 Performance Score | 49,045.5 | 82,518 (+68.25%) |
| PassMark G3D Mark | 26,972 | 39,751 (+47.38%) |
| PassMark G2D Mark | 1,114 | 1,439 (+29.17%) |
Technical Specs
Graphics Processor
General information about the graphics processing unit like their architecture, manufacturing process size, and transistor count. Newer GPU architectures generally bring efficiency improvements and may introduce technologies that enhance graphical capabilities.
| Spec | GeForce RTX 4070 | GeForce RTX 5090 |
|---|---|---|
| Codename | AD104 | GB202 |
| Architecture | Ada Lovelace | Blackwell 2.0 |
| Process Size | 5 nm | 5 nm |
| Transistors | 35,800 million | 92,200 million |
Memory Details
Memory specifications like their capacity, bandwidth, and clock speeds. GPU memory stores graphics data like frames, textures, and shadows which helps display rendered images. These specs are crucial for graphics-intense applications like gaming and 3D modeling.
| Spec | GeForce RTX 4070 | GeForce RTX 5090 |
|---|---|---|
| Memory Size | 12 GB | 32 GB |
| Memory Type | GDDR6X | GDDR7 |
| Memory Bandwidth | 504.2 Gb/s | 1.79 Tb/s |
| Memory Clock | 1,313 MHz | 1,750 MHz |
| Memory Interface | 192 bit | 512 bit |
| L1 Cache | 128 KB | 128 KB |
| L2 Cache | 36 MB | 96 MB |
Board Compatibility
Compatibility information like their slot size, bus interface, power consumption, and display support. These specs are useful for verifying compatibility with your motherboard, power supply, and monitor.
| Spec | GeForce RTX 4070 | GeForce RTX 5090 |
|---|---|---|
| Slots | 2 slots | 2 slots |
| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |
| Thermal Design Power (TDP) | 200 W | 575 W |
| Suggested PSU | 550 W | 950 W |
| Power Connectors | 1x 16-pin | 1x 16-pin |
| Outputs | HDMI 2.1, DisplayPort 1.4a | HDMI 2.1b, DisplayPort 2.1b |
Cores & Clock Speeds
Processing power information like its cores and clock speed. These specs impact how fast they can process graphics. Each type of core or component serves a specific computational purpose.
| Spec | GeForce RTX 4070 | GeForce RTX 5090 |
|---|---|---|
| CUDA Cores | 5,888 | 21,760 |
| Stream Multiprocessors (SM) | 46 | 170 |
| Texture Mapping Units (TMU) | 184 | 680 |
| Render Output Units (ROP) | 64 | 176 |
| Tensor Cores | 184 | 680 |
| Ray Tracing Cores | 46 | 170 |
| Core Clock Speed | 1,920 MHz | 2,017 MHz |
| Core Clock Speed (Boost) | 2,475 MHz | 2,407 MHz |
Theoretical Performance
Theoretical performance numbers derived from the raw specifications of the different components like core count and clock speeds. While these provide a glimpse into peak processing power, they do not represent real-world performance.
| Spec | GeForce RTX 4070 | GeForce RTX 5090 |
|---|---|---|
| Pixel Fill Rate | 158.4 GPixel/s | 423.6 GPixel/s |
| Texture Fill Rate | 455.4 GTexel/s | 1,636.8 GTexel/s |
| FP32 Performance | 29.15 TFLOPS | 104.8 TFLOPS |
| FP64 Performance | 455.4 GFLOPS | 1.64 TFLOPS |
API Support
Graphics API versions supported by these graphics cards. APIs evolve over time, introducing new features and functionalities. Older GPUs may not support recent versions.
| Spec | GeForce RTX 4070 | GeForce RTX 5090 |
|---|---|---|
| DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
| OpenCL | 3.0 | 3.0 |
| OpenGL | 4.6 | 4.6 |
| Shader Model | 6.8 | 6.8 |
Check out these comparisons for similar GPUs:
Looking for alternatives? Check out these similar GPUs:
* Performance rating, performance per dollar, and rankings are based on the 3DMark 11 Performance GPU benchmark and MSRP.