Gpu Compare Benchmark is essential for anyone looking to build or upgrade a gaming PC, workstation, or even a server. The latest GPU comparison benchmarks provide a comprehensive overview of graphics card performance, helping you make an informed decision. At compare.edu.vn, we offer an extensive range of GPU comparisons to suit diverse needs and budgets, from high-end gaming to professional applications.
This article will delve into the latest GPU performance hierarchy, rasterization and ray tracing benchmarks, and key findings for 2025. It also offers a detailed look at our testing methodology, individual game charts, power consumption, clock speeds, and temperatures. This information is designed to help you choose the best graphics card for your specific requirements.
1. Understanding GPU Benchmarks & Performance Hierarchy
GPU benchmarks and the performance hierarchy are essential tools for anyone looking to understand the relative performance of different graphics cards. Tom’s Hardware exhaustively benchmarks current and previous generation GPUs, including all of the best graphics cards. Your graphics card typically plays the biggest role in determining performance, whether it’s playing games, running artificial intelligence workloads like Stable Diffusion, or doing professional video editing — even the best CPUs for Gaming take a secondary role.
Key aspects of GPU benchmarks and performance hierarchy include:
- Ranking GPUs by Performance: The GPU hierarchy ranks graphics cards from the fastest to the slowest based on their performance in various tests and real-world applications.
- Comprehensive Testing: These benchmarks cover a wide range of tests, including gaming, AI workloads, and professional video editing, providing a holistic view of GPU capabilities.
- Performance Factors: The hierarchy doesn’t factor in real-world pricing, graphics card power consumption, overall efficiency, or features into the rankings here.
2. The State of the GPU Market in 2025
The GPU market has seen rapid developments, but supply chain issues and pricing discrepancies continue to be significant concerns. Despite new releases, many GPUs are unavailable at MSRP. According to Nvidia’s record-setting $130 billion in annual revenue, AI is all the rage, commanding billions of dollars in sales, and gaming was just under 9% of the total.
Key observations about the GPU market:
- New GPU Releases: Nvidia announced its Blackwell RTX 50-series architecture at CES 2025, then launched the RTX 5090 and RTX 5080 in late January, the RTX 5070 Ti in February.
- Planning Errors: Nvidia and AMD seemingly made some planning errors with the transitions from the prior generation to the new / upcoming hardware.
- Pricing and Availability: Some of the latest graphics cards are starting to become available at prices closer to their respective MSRPs.
3. April 2025 GPU Update
The April 2025 update brings the latest graphics cards into focus, with older generation parts moved to “legacy” status. Our latest reviews use the new test suite and test PC, and we now have enough of the latest and previous generation GPUs to swap over to the new data for the hierarchy.
Key components of the April 2025 update:
- New Test Suite: Upgraded the test PC to an AMD Ryzen 7 9800X3D and socket AM5 platform, and we’re now in the process of retesting… well, everything, as much as possible.
- Rasterization and Ray Tracing: Our full GPU hierarchy using traditional rendering (aka, rasterization) comes first, and below that we have our ray tracing GPU benchmarks hierarchy.
- Upscaling Technologies: The results are all at native resolution, without enabling DLSS, FSR, or XeSS upscaling or frame generation.
4. GPU Architecture Overview: Nvidia, AMD, and Intel
Nvidia’s Ada Lovelace architecture powers its previous generation RTX 40-series, with new features like DLSS 3 Frame Generation — with DLSS 4 upgrading the quality of both upscaling and ray reconstruction, plus adding MFG (Multi Frame Generation) on the 50-series GPUs. AMD’s RDNA 3 architecture powers the RX 7000-series, with seven desktop cards filling out the product stack. Intel’s Arc Alchemist architecture brings a third player into the dedicated GPU party, even if it’s more of a competitor to the earlier generation midrange offerings (i.e. RTX 3060 and RX 6650 XT).
Architectural highlights:
- Nvidia Ada Lovelace: Powers the RTX 40-series with DLSS 3 Frame Generation. DLSS 4 upgrading the quality of both upscaling and ray reconstruction, plus adding MFG (Multi Frame Generation) on the 50-series GPUs.
- AMD RDNA 3: Powers the RX 7000-series with enhanced performance and efficiency.
- Intel Arc Alchemist: Aims to compete in the midrange market with competitive features.
5. Rasterization GPU Benchmarks Ranking 2025
Our latest GPU benchmarks plan on testing nearly every GPU released in the past seven years, and perhaps we’ll extend that even further. We test every card at 1080p medium and 1080p ultra; where it makes sense, we also test at 1440p ultra and 4K ultra. All the scores are scaled relative to the top-ranking 1080p ultra card, which is the GeForce RTX 5090.
Key games used in the test suite:
- Assassin’s Creed Mirage
- Baldur’s Gate 3
- Black Myth Wukong
- Dragon Age: The Veilguard
- Final Fantasy XVI
- Flight Simulator 2020
- Flight Simulator 2024
- God of War Ragnarök
- Horizon Forbidden West
- The Last of Us Part 1
- A Plague Tale: Requiem
- Spider-Man 2
- Stalker 2
- Starfield
- Warhammer 40,000: Space Marine 2
6. Rasterization GPU Benchmarks: Key Findings
The rasterization benchmarks provide a clear picture of how different GPUs perform in traditional rendering scenarios. This section highlights the key findings from our latest tests.
GPU Benchmarks Hierarchy — Rasterization Performance
| Graphics Card | Lowest Price (MSRP) | 1080p Medium | 1080p Ultra | 1440p Ultra | 4K Ultra | Specifications (Links to Review) |
|---|---|---|---|---|---|---|
| GeForce RTX 5090 | $3,999 ($2,000) | 100.0% (197.5) | 100.0% (157.6) | 100.0% (143.0) | 100.0% | GB202, 21760 shaders, 2407MHz, 32GB GDDR7@28Gbps, 1792GB/s, 575W |
| GeForce RTX 4090 | $2,875 ($1,600) | 98.5% (194.6) | 94.9% (149.6) | 88.2% (126.1) | 80.7% | AD102, 16384 shaders, 2520MHz, 24GB GDDR6X@21Gbps, 1008GB/s, 450W |
| GeForce RTX 5080 | $1,400 ($1,000) | 90.4% (178.5) | 84.9% (133.9) | 78.1% (111.7) | 67.2% | GB203, 10752 shaders, 2617MHz, 16GB GDDR7@30Gbps, 690GB/s, 360W |
| GeForce RTX 4080 Super | $1,600 ($1,000) | 88.2% (174.2) | 82.4% (129.9) | 73.5% (105.1) | 61.5% | AD103, 10240 shaders, 2550MHz, 16GB GDDR6X@23Gbps, 736GB/s, 320W |
| Radeon RX 7900 XTX | $1,270 ($1,000) | 88.1% (174.1) | 79.4% (125.1) | 71.8% (102.6) | 60.8% | Navi 31, 6144 shaders, 2500MHz, 24GB GDDR6@20Gbps, 960GB/s, 355W |
| GeForce RTX 5070 Ti | $960 ($750) | 85.8% (169.4) | 79.1% (124.7) | 71.0% (101.5) | 58.7% | GB203, 8960 shaders, 2452MHz, 16GB GDDR7@28Gbps, 896GB/s, 300W |
| Radeon RX 9070 XT | $850 ($600) | 85.6% (169.0) | 76.1% (119.9) | 68.7% (98.3) | 57.6% | Navi 48, 4096 shaders, 2970MHz, 16GB GDDR6@20Gbps, 640GB/s, 304W |
| GeForce RTX 4070 Ti Super | $1,096 ($800) | 80.8% (159.5) | 73.7% (116.2) | 64.0% (91.5) | 51.4% | AD103, 8448 shaders, 2610MHz, 16GB GDDR6X@21Gbps, 672GB/s, 285W |
| Radeon RX 7900 XT | $894 ($750) | 81.3% (160.5) | 72.3% (114.0) | 63.9% (91.4) | 51.6% | Navi 31, 5376 shaders, 2400MHz, 20GB GDDR6@20Gbps, 800GB/s, 315W |
| Radeon RX 9070 | $670 ($550) | 79.6% (157.1) | 69.6% (109.7) | 61.1% (87.4) | 50.2% | Navi 48, 3584 shaders, 2520MHz, 16GB GDDR6@20Gbps, 640GB/s, 220W |
| GeForce RTX 4070 Ti | $1,096 ($800) | 77.8% (153.5) | 70.3% (110.9) | 59.8% (85.5) | 47.1% | AD104, 7680 shaders, 2610MHz, 12GB GDDR6X@21Gbps, 504GB/s, 285W |
| GeForce RTX 5070 | $600 ($550) | 74.6% (147.4) | 67.5% (106.4) | 56.8% (81.3) | 44.9% | GB205, 6144 shaders, 2512MHz, 12GB GDDR7@28Gbps, 672GB/s, 250W |
| GeForce RTX 4070 Super | $779 ($600) | 73.9% (145.8) | 66.6% (105.0) | 55.8% (79.7) | 43.4% | AD104, 7168 shaders, 2475MHz, 12GB GDDR6X@21Gbps, 504GB/s, 220W |
| Radeon RX 7800 XT | $590 ($500) | 67.3% (132.8) | 57.2% (90.1) | 48.6% (69.5) | 38.5% | Navi 32, 3840 shaders, 2430MHz, 16GB [email protected], 624GB/s, 263W |
| GeForce RTX 4070 | $734 ($550) | 65.6% (129.6) | 58.5% (92.1) | 47.9% (68.5) | 37.2% | AD104, 5888 shaders, 2475MHz, 12GB GDDR6X@21Gbps, 504GB/s, 200W |
| Radeon RX 7600 XT | $420 ($330) | 43.4% (85.6) | 36.7% (57.8) | 29.4% (42.1) | 21.4% | Navi 33, 2048 shaders, 2755MHz, 16GB GDDR6@18Gbps, 288GB/s, 190W |
| Intel Arc B580 | $341 ($250) | 41.4% (81.8) | 35.8% (56.4) | 30.2% (43.2) | 23.7% | BMG-G21, 2560 shaders, 2850MHz, 12GB GDDR6@19Gbps, 456GB/s, 190W |
| GeForce RTX 4060 | $320 ($300) | 42.6% (84.2) | 36.9% (58.1) | 28.4% (40.7) | 16.1% | AD107, 3072 shaders, 2460MHz, 8GB GDDR6@17Gbps, 272GB/s, 115W |
| GeForce RTX 3060 12GB | $294 ($330) | 35.5% (70.1) | 30.7% (48.3) | 24.4% (34.9) | 18.9% | GA106, 3584 shaders, 1777MHz, 12GB GDDR6@15Gbps, 360GB/s, 170W |
| Intel Arc B570 | $296 ($220) | 36.6% (72.3) | 30.3% (47.8) | 24.6% (35.2) | 15.2% | BMG-G21, 2304 shaders, 2750MHz, 10GB GDDR6@19Gbps, 380GB/s, 150W |
| Intel Arc A770 16GB | $439 ($330) | 32.1% (63.4) | 29.8% (46.9) | 25.7% (36.7) | 21.0% | ACM-G10, 4096 shaders, 2400MHz, 16GB [email protected], 560GB/s, 225W |
| Radeon RX 7600 | $260 ($270) | 40.6% (80.2) | 27.1% (42.7) | 19.7% (28.2) | 12.0% | Navi 33, 2048 shaders, 2655MHz, 8GB GDDR6@18Gbps, 288GB/s, 165W |
| Intel Arc A750 | $200 ($200) | 29.0% (57.4) | 25.8% (40.7) | 21.6% (30.9) | 15.9% | ACM-G10, 3584 shaders, 2350MHz, 8GB GDDR6@16Gbps, 512GB/s, 225W |
| Radeon RX 6600 | $210 ($330) | 32.5% (64.1) | 23.6% (37.2) | 16.9% (24.1) | 11.6% | Navi 23, 1792 shaders, 2491MHz, 8GB GDDR6@14Gbps, 224GB/s, 132W |
Key Takeaways:
- Resolution Impact: There is little separation between the fastest GPUs at lower resolutions.
- 5090 vs 4090: While the 5090 is 24% faster than the 4090 at 4K, it’s only 13% faster at 1440p, 5% faster at 1080p ultra, and just 1.5% faster at 1080p medium.
- AMD Performance: The RX 7900 XTX edges out the 9070 XT by 3–6 percent, but it comes in slightly behind Nvidia’s RTX 4080 Super.
- Intel Arc B580: It’s faster than the RTX 4070 and RX 7600 XT, while ostensibly costing less than either of those.
7. Ray Tracing GPU Benchmarks Ranking 2025
Ray tracing can significantly impact framerates, especially in demanding games. Our tests use “medium” and “ultra” ray tracing settings to assess GPU performance under these conditions.
Games used for ray tracing benchmarks:
- Avatar: Frontiers of Pandora (lighter RT title)
- Cyberpunk 2077 (heavy RT)
- F1 24 (lighter RT)
- Spider-Man: Miles Morales (moderate RT)
Cyberpunk 2077 Phantom Liberty screenshots
GPU Benchmarks Hierarchy — Ray Tracing Performance
| Graphics Card | Lowest Price (MSRP) | 1080p Medium | 1080p Ultra | 1440p Ultra | 4K Ultra | Specifications (Links to Review) |
|---|---|---|---|---|---|---|
| GeForce RTX 5090 | $3,999 ($2,000) | 100.0% (200.3) | 100.0% (154.6) | 100.0% (137.7) | 100.0% | GB202, 21760 shaders, 2407MHz, 32GB GDDR7@28Gbps, 1792GB/s, 575W |
| GeForce RTX 4090 | $2,875 ($1,600) | 91.9% (184.1) | 92.3% (142.7) | 85.5% (117.7) | 78.8% | AD102, 16384 shaders, 2520MHz, 24GB GDDR6X@21Gbps, 1008GB/s, 450W |
| GeForce RTX 5080 | $1,400 ($1,000) | 86.2% (172.6) | 84.9% (131.2) | 73.0% (100.5) | 63.7% | GB203, 10752 shaders, 2617MHz, 16GB GDDR7@30Gbps, 690GB/s, 360W |
| GeForce RTX 4080 Super | $1,600 ($1,000) | 83.2% (166.6) | 80.5% (124.4) | 68.5% (94.3) | 60.7% | AD103, 10240 shaders, 2550MHz, 16GB GDDR6X@23Gbps, 736GB/s, 320W |
| **[GeForce RTX 5070 Ti](https://www.amazon.com/s?rh=n%3A17923671011%2Cn |
