SYSTEM ANALYZER

Rate My PC: AMD Ryzen 9 5900XT + NVIDIA GeForce RTX 4070 SUPER

Get a comprehensive performance analysis of your gaming rig with detailed benchmarks, bottleneck detection, and upgrade recommendations

95 / 100
ULTIMATE READY

Apex Performer

Top 5% of systems. Capable of 4K Ultra gaming and advanced rendering.

4K 60+ FPSVR ReadyRay Tracing

System Balance Analysis

CPU vs GPU performance ratio
Well Balanced
CPU
94%
VS
GPU
96%
PROCESSOR

AMD Ryzen 9 5900XT

50,718 Benchmark Score
Top 6% Market Ranking
View Full Specs →
GRAPHICS CARD

NVIDIA GeForce RTX 4070 SUPER

43,223 Benchmark Score
Top 4% Market Ranking
View Full Specs →

Market Position

How your build compares to others
Budget
0-30
Mid-Range
30-60
High-End
60-85
Enthusiast
85-100
Your Build

Game Performance Benchmarks

Real-world 4K FPS in popular titles
View All Games →

Performance Insights

Tips to maximize your system

Optimal Performance

Your system is in the top tier. You can run any modern game at maximum settings.

4K Gaming Ready

Consider a 4K 144Hz monitor to fully utilize your hardware capabilities.

Compatible Games See what you can play Compare CPUs Find upgrades Compare GPUs Find upgrades

Performance Tiers Explained

90-100

Ultimate

4K Ultra gaming, VR ready, ray tracing enabled, professional workloads

4K 60+ FPS VR Ready
70-89

High-End

1440p Ultra or 4K High settings, excellent for modern AAA titles

1440p Ultra 4K High
50-69

Mid-Range

1080p Ultra or 1440p Medium, great value for most gamers

1080p Ultra 1440p Med
30-49

Entry Level

1080p Medium settings, suitable for eSports and older titles

1080p Med eSports
0-29

Legacy

Basic gaming, older titles, consider upgrading for modern games

720p-1080p Low Older Games

The AMD Ryzen 9 5900XT and NVIDIA GeForce RTX 4070 SUPER form a desktop pairing built around a 16-core Zen 3 processor and an Ada Lovelace graphics card. The CPU sits at the 90th percentile among all processors tested, while the GPU lands at the 83rd percentile among all GPUs, placing the combined system at the 87th percentile overall. This is a high-end desktop configuration with no measured FPS data for the exact combination, so all gaming performance figures presented here are estimates derived from the individual component benchmark scores. The processor is a 5000-series part on the AMD Socket AM4 platform, using 16 cores and 32 threads with a base clock of 3.30 GHz and a boost clock of 4.80 GHz, built on TSMC’s 7 nm process with a 105 W TDP. The GPU is a GeForce 40-series part with 12 GB of GDDR6X memory on a 192-bit bus, delivering 504.2 GB/s of bandwidth, and it is marked end-of-life in production status.

CPU Analysis

The Ryzen 9 5900XT is a 16-core, 32-thread processor from AMD’s 5000 series, based on the Zen 3 architecture with the Vermeer codename. It runs on the AMD Socket AM4 platform with dual-channel DDR4 memory support, and it offers 64 MB of L3 cache alongside 512 KB of L2 cache per core and 64 KB of L1 cache per core. The CPU includes 20 PCIe Gen 4 lanes from the processor itself, has ECC memory support, and an unlocked multiplier for overclocking. The transistor count sits at 8,300 million across a dual-chiplet design with two 74 mm² dies, and the process node is 7 nm from TSMC.

Benchmark data shows a clear strength in multi-threaded workloads. The Cinebench R23 multicore score of 37373 is the standout result, reflecting the full 16-core capability. The PassMark multithread score of 43810 and the 3dMark max threads score of 11040 reinforce this pattern. Single-thread performance is comparatively modest but still solid: the Cinebench R23 single-core score is 5276, the 3dMark single-thread score is 942, and the PassMark single-thread score is 3474. The scaling from 2 threads to 16 threads is telling: 3dMark scores go from 1853 at 2 threads to 3589 at 4 threads, 6607 at 8 threads, and 10624 at 16 threads, with max threads reaching 11040. This near-linear scaling indicates the architecture efficiently feeds all cores, making the processor particularly effective for workloads that can use many threads.

The average benchmark score for this CPU is 50718, which places it at the 90th percentile against all CPUs. Compared to its nearest rivals, the margin is razor-thin. The Intel Core i7-13850HX scores 50761, which is 0.1% higher than the 5900XT. The AMD Ryzen AI 9 HX PRO 370 scores 50448, 0.5% lower. The Intel Core i9-14900T scores 51015, 0.6% higher. The Intel Core i9-13980HX scores 50398, 0.6% lower. In practical terms, the 5900XT trades blows with these parts within a single percentage point, meaning the choice between them comes down to platform features rather than raw benchmark dominance.

Real workload interpretation: the Cinebench R15 multicore score of 3767 and the R20 multicore score of 15696 show consistent scaling across renderer versions. PassMark data compression at 597862 and data encryption at 37814 indicate strong throughput for archival and security tasks. Floating point math at 99398 and integer math at 177566 point to heavy compute capability. The extended instructions score of 39141 suggests the AVX and related instruction sets are handling vectorized workloads well. The find prime numbers score of 205 is a lower score, but that test is typically single-thread bound and not representative of the chip’s overall character. The physics score of 1715 in PassMark aligns with the multi-threaded nature of the part.

The memory bandwidth is listed at 51.2 GB/s for dual-channel DDR4, which is not the fastest available but is adequate for the core count. The 64 MB L3 cache is a large pool that helps keep frequently accessed data close to the cores, reducing latency penalties in cache-sensitive workloads. The base clock of 3.30 GHz and boost of 4.80 GHz show a wide dynamic range, allowing the CPU to idle efficiently and ramp up under load. The 105 W TDP is modest for a 16-core part, which speaks to the efficiency of the 7 nm process.

Usage Scenarios

High-refresh gaming: The CPU’s single-thread score of 5276 in Cinebench R23 and the GPU’s PassMark G3D score of 29995 suggest that at lower resolutions, the processor can keep up with the graphics card in most titles. The 16 cores provide headroom for background tasks, but the limiting factor will be the GPU at higher settings. Expect solid frame rates at 1440p, with the CPU rarely being the bottleneck in well-optimized games.

Streaming: With 16 cores and 32 threads, the CPU can handle game encoding and streaming simultaneously without starving the game of processing power. The PassMark multithread score of 43810 indicates that x264 encoding at medium presets is feasible while gaming. The GPU also has NVENC support through its Ada Lovelace architecture, which offloads encoding entirely, making this pairing exceptionally capable for live streaming at high bitrates.

Video editing: The Cinebench R23 multicore score of 37373 is the key metric here. Timeline scrubbing, effect rendering, and export encoding all scale with core count. The 12 GB of VRAM on the GPU helps with GPU-accelerated effects and previews. For 4K video editing, this combination handles multi-layer timelines and long exports without excessive wait times. The PassMark data compression score of 597862 also aids in handling large project files and fast backups.

3D rendering: The CPU’s 16 cores shine in CPU-based renderers, with the Cinebench R20 multicore score of 15696 showing strong performance in Cinema 4D and similar software. The GPU’s FP32 performance of 35.48 TFLOPS and its 56 RT cores make it a competent choice for GPU-accelerated rendering in Blender and Octane. The combination allows for hybrid rendering, where the CPU handles part of the frame and the GPU handles the rest, reducing overall render times.

Software development: The high core count and 64 MB L3 cache are beneficial for compilation workloads. The PassMark integer math score of 177566 indicates strong performance in code compilation and unit testing. The extended instructions score of 39141 supports SIMD-heavy code paths. The 32 threads allow for parallel builds with minimal slowdown. The 105 W TDP makes it a manageable part for long compile sessions without excessive cooling demands.

Student and office work: This is overkill for most student tasks, but the 90th percentile CPU performance ensures that even the most demanding spreadsheet simulations, data analysis in Python, or virtual machine setups run smoothly. The PassMark single-thread score of 3474 handles everyday responsiveness well. The ECC memory support is a bonus for data integrity in research work. The GPU is far more than needed for office applications, but the 12 GB VRAM allows for moderate GPU compute tasks in data science courses.

GPU Analysis

The NVIDIA GeForce RTX 4070 SUPER is based on the AD104 chip with the Ada Lovelace architecture, built on TSMC’s 5 nm process. The die contains 35,800 million transistors across a 294 mm² area, giving a transistor density of 121.8M per mm². The GPU has 7168 shading units, 224 texture mapping units, and 80 raster output units. It includes 56 RT cores for ray tracing and 224 tensor cores for AI acceleration. The base clock is 1980 MHz with a boost clock of 2475 MHz, and the memory runs at 1313 MHz with 21 Gbps effective speed.

Memory configuration: 12 GB of GDDR6X on a 192-bit bus provides 504.2 GB/s of bandwidth. The pixel rate is 198.0 GPixel/s and the texture rate is 554.4 GTexel/s. FP32 performance is rated at 35.48 TFLOPS, with FP16 at the same 35.48 TFLOPS due to a 1:1 ratio. The TDP is 220 W with a suggested PSU of 550 W. The card uses a single 16-pin power connector and is dual-slot in width, measuring 267 mm in length, 112 mm in height, and 42 mm in width. Display outputs include one HDMI 2.1 and three DisplayPort 1.4a connectors, and it interfaces via PCIe 4.0 x16.

Benchmark results show a mixed picture. The 3dMark Steel Nomad DX12 score of 4627 reflects modern DirectX 12 performance. Geekbench OpenCL scores 172795 and Vulkan scores 205624. PassMark G3D scores 29995, which is a strong result. PassMark GPU compute scores 17108. The DirectX 10, 11, 12, and 9 scores are 167, 273, 110, and 344 respectively, which are lower numbers but those tests are older and not directly comparable to modern workloads.

The average benchmark score for the GPU is 43223, placing it at the 83rd percentile among all GPUs. Its nearest rivals are close: the NVIDIA Quadro M6000 24 GB scores 43262 (0.1% higher), the NVIDIA GeForce RTX 5050 Mobile scores 43268 (0.1% higher), the NVIDIA Quadro M6000 scores 43301 (0.2% higher), and the NVIDIA GeForce RTX 4090 Mobile scores 43667 (1% higher). This indicates the 4070 SUPER sits in a performance band where it is effectively matched by high-end mobile parts and older workstation cards, but it holds its own in desktop contexts.

The 12 GB VRAM is a key consideration. At 1440p and 4K, modern games can exceed 12 GB in extreme cases, but for the majority of titles at high settings, this capacity is sufficient. The 504.2 GB/s bandwidth is ample for the 192-bit bus, and the 56 RT cores provide dedicated hardware for ray tracing, which is a step up from the previous generation. The 224 tensor cores enable DLSS and other AI-based features, which can significantly boost frame rates in supported games. The 35.48 TFLOPS FP32 performance indicates strong rasterization capability, and the 1:1 FP16 ratio suggests it can handle some compute workloads efficiently.

Balance and Bottleneck

The combined percentile of 87 places this pairing in the upper tier of systems. The CPU at the 90th percentile is stronger than the GPU at the 83rd percentile, which suggests the processor has more headroom relative to the graphics card. In CPU-bound workloads like compilation, rendering, and data processing, the CPU will be the primary driver, and the GPU will be underutilized. In GPU-bound workloads like gaming at high resolutions, the GPU will be the limiting factor.

The CPU’s single-thread score of 5276 in Cinebench R23 is strong, but not exceptional compared to the latest architectures. In games that are heavily single-thread dependent, the CPU may hold back the GPU at lower resolutions like 1080p. At 1440p and above, the GPU becomes the bottleneck more often, which is the preferred scenario for a balanced system. The 3dMark 16-thread score of 10624 versus the max threads score of 11040 shows that the CPU scales well up to 16 threads, and the 2-thread score of 1853 indicates that even lightly threaded tasks get decent performance.

The GPU’s PassMark G3D score of 29995 is strong, but its nearest rivals include mobile parts like the RTX 4090 Mobile and RTX 5050 Mobile, which suggests that in a desktop context, this GPU is not at the very top of the stack. The 12 GB VRAM can become a limitation at 4K in texture-heavy games, where the CPU’s 64 MB L3 cache cannot compensate for VRAM overflow. In compute workloads, the GPU’s 17108 PassMark compute score is solid, but the CPU’s 43810 multithread score will often be the more relevant number for tasks that are not GPU-accelerated.

The memory bandwidth of the CPU at 51.2 GB/s is a potential bottleneck for certain workloads that require high memory throughput, but for gaming and most productivity tasks, this is not a critical factor. The PCIe Gen 4 x16 interface for the GPU provides sufficient bandwidth for the 12 GB VRAM. The overall balance favors the CPU, meaning that if you upgrade the GPU later, the CPU will still be able to keep up. Conversely, upgrading the CPU would yield diminishing returns in GPU-bound scenarios.

FAQ

Q: What is the CPU’s core and thread count?

A: The AMD Ryzen 9 5900XT has 16 cores and 32 threads, based on the Zen 3 architecture.

Q: How much VRAM does the GPU have and what type?

A: The NVIDIA GeForce RTX 4070 SUPER has 12 GB of GDDR6X memory on a 192-bit bus.

Q: What is the CPU’s benchmark percentile compared to all CPUs?

A: The CPU is at the 90th percentile among all CPUs, with an average benchmark score of 50718.

Q: What is the GPU’s benchmark percentile compared to all GPUs?

A: The GPU is at the 83rd percentile among all GPUs, with an average benchmark score of 43223.

Q: Does this system support ray tracing?

A: Yes, the GPU has 56 RT cores dedicated to ray tracing, and it also has 224 tensor cores for AI acceleration.

Q: What is the CPU’s socket and memory support?

A: The CPU uses the AMD Socket AM4 and supports dual-channel DDR4 memory with ECC support.

Q: Is there measured FPS data for this exact combination?

A: No, the data pack contains no measured FPS rows for this exact CPU+GPU combination, so all FPS figures are estimates.

Q: What is the GPU’s power requirement?

A: The GPU has a TDP of 220 W and a suggested PSU of 550 W.

Q: What is the CPU’s boost clock?

A: The CPU has a boost clock of 4.80 GHz and a base clock of 3.30 GHz.

Gaming Performance

No measured FPS rows exist for the exact combination of the Ryzen 9 5900XT and RTX 4070 SUPER, so all gaming performance figures are estimates derived from the benchmark scores. The CPU’s Cinebench R23 single-core score of 5276 and the GPU’s PassMark G3D score of 29995 provide the basis for these estimates. At 1080p, the CPU’s single-thread performance will be the primary driver, and the 16 cores ensure that background tasks do not interfere with frame pacing. Expect high frame rates in esports titles and competitive shooters, where the CPU’s 4.80 GHz boost clock and 64 MB L3 cache help maintain consistent frame times.

At 1440p, the GPU becomes more relevant. The 12 GB VRAM and 504.2 GB/s bandwidth are sufficient for most games at high settings. The 35.48 TFLOPS FP32 performance suggests strong rasterization throughput, and the 56 RT cores can handle ray tracing at playable frame rates, especially with DLSS enabled via the 224 tensor cores. For AAA titles at 1440p ultra settings, the GPU will likely be the bottleneck, which is a good position to be in because the CPU still has headroom.

At 4K, the 12 GB VRAM may become a limiting factor in the most demanding titles. The 192-bit memory bus and 504.2 GB/s bandwidth are adequate but not exceptional for 4K textures. The GPU’s percentile rank of 83 indicates it is not the top performer, and at 4K it will struggle to maintain 60 FPS in the most demanding games without upscaling. The CPU, at the 90th percentile, will not be the limiting factor at this resolution, but the overall frame rate will be constrained by the GPU.

For VR gaming, the combination of the CPU’s 16 cores and the GPU’s 12 GB VRAM provides a stable base, though the lack of measured data means exact frame times are unknown. The GPU’s 56 RT cores are beneficial for VR titles that use ray-traced reflections or lighting. The CPU’s 32 threads ensure that VR tracking and physics calculations do not compete with the game render thread. Overall, this system is estimated to be well-suited for high-refresh 1440p gaming, with 1080p being CPU-bound in many cases and 4K requiring settings adjustments.

Benchmark Performance

The CPU’s benchmark performance is defined by its 90th percentile rank and an average score of 50718. The Cinebench R23 multicore score of 37373 is the most representative of its multi-threaded capability, and the single-core score of 5276 shows it is not a slouch in lightly threaded tasks. The PassMark multithread score of 43810 and the 3dMark max threads score of 11040 corroborate the multi-core strength. The nearest rival, the Intel Core i7-13850HX, scores 50761, which is only 0.1% higher, indicating that the 5900XT is effectively tied with this mobile processor despite being a desktop part.

The GPU’s benchmark performance is set by its 83rd percentile rank and an average score of 43223. The 3dMark Steel Nomad DX12 score of 4627 is a modern benchmark that shows strong DirectX 12 performance. The Geekbench Vulkan score of 205624 is higher than the OpenCL score of 172795, which is typical for NVIDIA GPUs. The PassMark G3D score of 29995 is a solid result, and the GPU compute score of 17108 indicates decent compute capability. The nearest rival, the NVIDIA Quadro M6000 24 GB, scores 43262, which is 0.1% higher, showing that the 4070 SUPER is competitive with a much older workstation card.

The combined picture shows a system where the CPU is slightly stronger relative to its peers than the GPU. The CPU’s 90th percentile versus the GPU’s 83rd percentile means the processor has more headroom in multi-threaded tasks. The combined percentile of 87 reflects a high-end desktop build. The CPU’s 16 cores and 32 threads are well-suited for productivity, while the GPU’s 12 GB VRAM and Ada Lovelace features (56 RT cores, 224 tensor cores) make it a capable gaming and content creation card. The lack of measured FPS data means that real-world gaming performance must be inferred from these benchmark scores, but the individual scores suggest a balanced system that will not have a significant bottleneck in most workloads.

Build Overview

This is a desktop build class system combining the AMD Ryzen 9 5900XT and the NVIDIA GeForce RTX 4070 SUPER. The CPU is a 16-core, 32-thread processor from the 5000 series, using the Zen 3 architecture on the AM4 socket. The GPU is a GeForce 40-series card with 12 GB of GDDR6X memory and the Ada Lovelace architecture. The combined percentile of 87 places this system in the upper tier of desktop configurations, with the CPU at the 90th percentile and the GPU at the 83rd percentile.

The CPU’s launch date was 2024-07-30 and it is still in active production. The GPU’s launch date was 2024-01-16 and it is marked end-of-life. The CPU has a launch MSRP of $349, and the GPU has a launch MSRP of 599 USD, but this pricing is not a factor in the performance analysis. The CPU’s TDP is 105 W, and the GPU’s TDP is 220 W, which means a standard 550 W PSU is suggested for the GPU alone. The CPU supports DDR4 memory with ECC, and the GPU requires a 16-pin power connector.

This pairing is a strong desktop workstation and gaming combination. The 16 cores of the CPU are well-matched to the GPU’s 12 GB VRAM for content creation tasks that use both CPU and GPU acceleration. The AM4 platform means that this build can use existing DDR4 memory and AM4 motherboards, which is a cost-effective way to get high core counts. The GPU’s end-of-life status means it is no longer in production, but it remains a capable card with 56 RT cores and 224 tensor cores for modern gaming features.

Who Should Build It

Gamers at 1440p will find this system well-suited, as the CPU’s 90th percentile single-thread performance and the GPU’s 83rd percentile G3D score combine to handle high refresh rates at this resolution. For 1080p gaming, the CPU will often be the limiting factor, but the 16 cores provide headroom for streaming or recording. For 4K gaming, the GPU will be the bottleneck, and the 12 GB VRAM may require texture quality reductions in the most demanding titles.

Content creators who work with video editing, 3D rendering, and data processing will benefit from the CPU’s 16 cores and the GPU’s 12 GB VRAM. The Cinebench R23 multicore score of 37373 is strong for CPU-based rendering, and the GPU’s 35.48 TFLOPS FP32 performance accelerates GPU-based renderers. The PassMark data compression score of 597862 helps with large file workflows. Developers will appreciate the 32 threads for parallel compilation tasks, and the ECC memory support provides data integrity for long build processes.

Students in engineering, data science, or computer science fields will find this system more than adequate for their computational needs. The CPU’s 90th percentile rank ensures that even the most demanding simulation software runs smoothly, and the GPU’s 224 tensor cores can accelerate machine learning workloads. Small business workstations that require multi-threaded performance for tasks like financial modeling, database management, or CAD software will benefit from the 16-core processor and the GPU’s compute capability. For users who need a system that can handle both demanding productivity and high-end gaming without compromise, this pairing offers a balanced solution where the CPU has a slight edge over the GPU, leaving room for future GPU upgrades without changing the processor.