SYSTEM ANALYZER

Rate My PC: AMD Ryzen 9 9900X3D + 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
95%
VS
GPU
96%
PROCESSOR

AMD Ryzen 9 9900X3D

54,762 Benchmark Score
Top 5% 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 pairing under review combines the AMD Ryzen 9 9900X3D, a 12-core, 24-thread desktop processor from the 9000 series built on Zen 5 and code-named Granite Ridge, with the NVIDIA GeForce RTX 4070 SUPER, an Ada Lovelace–based GPU with 12 GB of GDDR6X memory. The data set places this build in the 87th combined percentile, reflecting a balanced high-end desktop system. The CPU alone rests in the 91st percentile, the GPU in the 83rd, and the measured ultra settings frame rates across 4K, 1440p, and 1080p games provide a clear picture of what this pair can deliver. No economic or budgetary commentary appears here—only the measured performance, platform properties, and practical system capabilities derived directly from the numbers.

Upgrade Path and Platform

The Ryzen 9 9900X3D sits on AMD Socket AM5, the current desktop socket for Ryzen 9000-series parts. The processor supports DDR5 memory through a dual-channel interface, with a specified memory bandwidth of 89.6 GB/s. The chip also enables ECC memory support, which matters for workstations that require error resilience in long-running threads. The CPU itself provides 24 PCIe Gen 5 lanes, a forward-looking interface that can connect to the fastest current NVMe drives and high-end expansion cards, although the GPU on this build uses its own interface.

The GPU uses a PCIe 4.0 x16 bus connection, which is fully compatible with the CPU’s Gen 5 topology; PCIe 4.0 is still enough to feed a GPU of this class without bottleneck. The suggested power supply for the GPU is 550 W, and the CPU's TDP is 120 W. Combined, these numbers indicate that a typical 850 W PSU (within the 550 W suggested plus CPU power plus other components) will handle the system, though the exact wattage is not part of the data. A sensible next upgrade on this platform would be either to swap in a faster GPU from the successor generation (the data list the GeForce 50 series as the GPU’s successor) or to extend memory capacity, as the DDR5 dual-channel bandwidth is the only memory figure we have. The CPU has an unlocked multiplier, which allows overclocking headroom if the rest of the system provides adequate cooling and power. Because the platform is AM5, the user has a clear upgrade path if they later choose to move to a higher-tier 9000-series processor, though the actual specifications of other processors are not included here.

FAQ

Q: What processor socket does the Ryzen 9 9900X3D use?

A: The AMD Ryzen 9 9900X3D uses the AMD Socket AM5, which is the current desktop socket for the 9000-series Zen 5 architecture.

Q: How much memory and what type does the system support?

A: The CPU supports dual-channel DDR5 memory with a bandwidth of 89.6 GB/s and supports ECC memory. The GPU has its own VRAM pool of 12 GB of GDDR6X with a 192-bit interface and a bandwidth of 504.2 GB/s.

Q: How many cores and threads does the CPU have?

A: The CPU has 12 physical cores and 24 threads, with a base clock of 4.40 GHz and a boost clock of 5.50 GHz.

Q: What is the GPU’s VRAM size and memory type?

A: The RTX 4070 SUPER comes with 12 GB of GDDR6X memory, positioned on a 192-bit bus, delivering 504.2 GB/s of bandwidth.

Q: In which percentile does the CPU rank compared to all classifiers?

A: The CPU ranks in the 91st percentile among all processors, while the GPU ranks in the 83rd percentile, and the full build ranks in the 87th percentile.

Q: Does the system support PCIe Gen 4 or Gen 5?

A: The GPU uses PCIe 4.0 x16, while the CPU offers 24 Gen 5 lanes, meaning the system can connect Gen 5 devices (storage, accelerators) as well as the GPU on Gen 4.

Q: Is the performance data measured or estimated?

A: All frame-rate numbers under “measuredFpsUltraByGame” are designated as measured (true), so they come from actual tests, not simulations.

GPU Analysis

The RTX 4070 SUPER uses the AD104 chip built on a 5 nm process from TSMC, with 35.8 billion transistors and a die size of 294 mm². Its base clock is 1980 MHz, the boost clock is 2475 MHz, and memory runs at 1313 MHz effective 21 Gbps. The GPU has 7168 shading units, 224 texture mapping units, 80 ROPs, and 56 RT (ray tracing) cores, plus 224 tensor cores for the AI acceleration used in DLSS. The pixel fill rate is 198.0 GPixel/s and texture rate is 554.4 GTexel/s. The FP32 performance is 35.48 TFLOPS, with a 1:1 ratio for FP16, meaning the same 35.48 TFLOPS in half-precision.

Benchmark results support these figures. In 3D Steel Nomad (DirectX 12) the GPU scores 4627. GeekBench OpenCL draws a score of 172,795, and Vulkan reaches 205,624. The PassMark G3D score is 29,995, while the compute test returns 17,108. These numbers put the GPU in the 83rd percentile. When compared to its nearest rivals, the RTX 4070 SUPER sits 0.1% below the Quadro M6000 24 GB, 0.1% below the RTX 5050 Mobile, 0.2% below the standard Quadro M6000, and 1.0% below the RTX 4090 Mobile. The tiny deltas among these four demonstrate that the RTX 4070 SUPER is in a tight cluster of similarly performing GPUs, but its 35.48 TFLOPS of FP32 compute and the tensor hardware on Ada Lovelace make it particularly suited for rendering workloads that use ray tracing and DLSS.

Balance and Bottleneck

The CPU outperforms the GPU relative to its peers: the CPU’s 91st percentile versus the GPU’s 83rd percentile suggests the GPU is the weaker link in the pair for most tasks. Frame-rate scaling from measured ultra settings reflects this. In Cyberpunk 2077, the system moves from 83 FPS at 1080p to 62 FPS at 1440p and 41 FPS at 2160p—a clear GPU-limited drop as resolution rises. Similarly, Red Dead Redemption 2 shows 76 FPS at 1080p, 57 at 1440p, and 37 at 4K. At 1080p the CPU’s advantage can push higher in esports titles: Counter-Strike 2 hits 283 FPS at 1080p, 187 at 1440p, and 105 at 2160p. The 4K numbers fall more heavily, which signals GPU bottleneck. At 1080p, the CPU’s single-thread strength is more apparent; however, with a 12-core/24-thread CPU, there is no shortage of compute headroom for most today’s games. Ultimately that for the majority of titles, the GPU becomes the limiting factor at higher resolutions (1440p and 4K), while the CPU can drive machine-fast, high refresh loops at 1080p. Those who prioritize 1080p esports performance might feel the CPU is the primary contributor, but at 4K the GPU defines the ceiling.

Who Should Build It

This is a desktop configuration, and its combined percentile of 87 means top decile performance. The 12 full cores with 24 threads on the CPU, plus the 12 GB GDDR6X GPU with 56 RT cores, target gamers playing at 1440p and 4K with ray tracing enabled, content creators who work with video and 3D rendering, developers who compile large projects, and students or office workers who want a responsive system for multitasking. The GPU’s 56 RT cores and tensor cores make it suitable for real-time ray tracing and AI‑accelerated tasks such as DLSS. The CPU’s 12 cores and 24 threads, with 128 MB of L3 cache, are ideal for parallel workloads like video editing and software compiles. The build also supports ECC memory, which draws in workstation users who need certainty in code correctness. Because the data show measured FPS at 1080p to 4K across all game types, this system can comfortably run games at high settings at 1440p and 144 FPS, or at 4K60 in less demanding titles.

Gaming Performance

The measured figures from the ultra setting are provided across 50+ games at three resolutions. Representative results show that this build is very strong at 1080p and respectable at 1440p, but light at 4K. In esports, Fortnite 142 FPS at 1080p, 86 at 1440p, 46 at 4K; Counter-Strike 2 curves at 267, 172, and 105; Valorant reaches 582 FPS at 1080p, 552 at 1440p, and 442 at 4K. AAA triple titles are both: Cyberpunk 2077 83/62/41 FPS, Red Dead Redemption 2 76/57/37, Call of Duty: Warzone 181/161/136. A few titles are heavier: Ark: Survival Ascended 41 FPS at 1080p, and Microsoft Flight Simulator 102/82/61. The 4K figures generally sit below 60 FPS in such demanding games, but at 1440p maintain near 60, which is the target for high-refresh monitors.

Benchmark Performance

The CPU’s multi-threaded scores indicative of heavy parallel work: Cinebench R23 multi and single core scores are 234,002 and 694 (the provided pack gives multi 47,737 and single 6,739— careful). Actually we must use exactly the pack: Cinebench R23 multi: 47,737, single-core 6,739, Cinebench R20 multi 20,049, single 2,830, Cinebench R15 multi 48,114, single 679. Geekbench multi 22,884, single 3,041. PassMark multithread 56,154, single-thread 4,646. In a three mark comparison the CPU’s score puts it in the 91st percentile. Compared to its closest rivals, it is 0.6% lower than the Intel Core Ultra 5 245KF, 1.3% higher than the Intel Core Ultra 5 245K, 2.0% higher than an Intel Xeon 6505P, and 2.1% higher than an Intel Core i7-14700F. Its average benchmark score is 547,628 (we might want to say the average score), and the percentile gives a “above ” indicator.

The GPU’s performance, at the 83rd percentile, outperformed only slightly by a Quadro M6000 24 GB (0.1% lower). The GPU’s compute power (35.48 TFLOPS) is accompanied by Vulkan score 205,624 and OpenCL 172,795. Combined, the two components in a desktop class give a combined percentile of 87, indicating this pair is among the top 13% of all systems in the dataset.

Build Overview

This is a desktop configuration that pairs the AMD Ryzen 9 9900X3D with the NVIDIA GeForce RTX 4070 SUPER. The processor is a 12-core 24-thread Zen 5 chip from the 9000 series, and the GPU is an Ada Lovelace architecture part with 12 GB of GDDR6X. The build holds the 87th percentile among all systems, with the CPU at the 91st and the GPU at the 83rd percentile. The pair was launched with a launch MSRP of $599 for the CPU and $599 for the GPU, both at release dates in 2025 and 2024, respectively. The system is a desktop built for high-end gaming and content creation, with the CPU’s 12 cores providing strong multi-thread performance and the GPU able to output 4K gaming that hovers above 40 FPS. The launch MSRP is the only financial data point; no other price or cost information is included in the review.

CPU Analysis

The AMD Ryzen 9 9900X3D is built on the Zen 5 architecture’s Granite Ridge code name. It has 12 cores and 24 threads, with a base clock of 4.40 GHz and a boost clock of 5.50 GHz. The CPU uses a 4 nm process from TSMC, counts 16,530 million transistors on a dual-chip die of 70.6 mm². Its cache structure includes 80 KB of L1 per core, 1 MB of L2 per core, and a large 128 MB L3 cache. It supports dual-channel DDR5 memory with 89.6 GB/s bandwidth, ECC on, and has a 120 W TDP. The CPU can be overclocked via an unlocked multiplier.

Benchmark results show strong multi-threaded performance: the CPU scores 47,737 in Cinebench R23 multi-core, 6,739 in single-core, and 22,884 in Geekbench multi-core, 3,041 in single-core. The PassMark multi-thread test yields 56,154, single-thread 4,646 (noting the single-thread duplicate). The data shows the CPU at the 91st percentile among all processors. When compared to close rivals, it is within about 2% above or below its competitor cores, the most notable being a 2.1% lead over the Intel Core i7-14700F. Benchmarks such as PassMark extended instructions (55,426) and floating-point math (119,622) indicate its throughput at practical applications, while the integer math (179,727) and data compression (685 thousand) points to server-like capacity. The CPU is evidently well suited to 3D rendering, compile scale, and scientific simulation, all of which benefit from 12 physical cores, high L3 cache, and the 128 MB capacity.

Usage Scenarios

  • High-refresh gaming: at 1080p this build exceeds 200 FPS in many light games (e.g., 283 in CS:GO, 622 in CS:GO old version, 582 in Valorant) and stays above 120 FPS at 1440p. The CPU and GPU combination is very strong for high-refresh esports and fast-paced shooters.
  • Streaming: with the GPU’s NVENC decoder (implicitly due to Ada Lovelace), the 35.48 TFLOPS of compute can handle video encoding while the CPU’s 12-core 4.40 GHz processing can manage game logic; benchmark scores show enough headroom for multi-tasking.
  • Video editing: The CPU’s Cinebench R23 multi-core score of 47,737 and PassMark multithread 56,154 indicate it can handle 4K timelines, while the GPU’s 12 GB of VRAM and 504.2 GB/s bandwidth assist - decoding and effects. The group’s 4K export tasks are within reach.
  • 3D rendering: The 12 cores accelerate CPU-based Ray tracing and physics in engines, while the 56 RT cores and tensor cores accelerate GPU-based effects. Cinebench results and the GPU’s 4627 Steel Nomad DX12 score show a capable renderer.
  • Software development: The CPU’s 24 threads and 12 cores speed up code compilation (as indicated by high passMark integer math 179,727) and multi-threaded debugging; the large L3 cache maximizes reused data.
  • Students/office: The 12-core CPU, 128 MB L3 cache, and 24 threads make multitasking smooth - running spreadsheets, databases, VMs, and office productivity with headroom. The GPU’s 12 GB ends up capable of running CAD, GIS, and analytics, though no exact benchmark for those exists. As the CPU uses 91st percentile, it makes short work of routine tasks, while the GPU’s 83rd percentile covers the exceptional graphics.