SYSTEM ANALYZER

Rate My PC: Intel Core Ultra 5 225 + NVIDIA GeForce RTX 5090 D

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

94 / 100
ULTIMATE READY

Apex Performer

Top 6% 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
91%
VS
GPU
98%
PROCESSOR

Intel Core Ultra 5 225

36,938 Benchmark Score
Top 9% Market Ranking
View Full Specs →
GRAPHICS CARD

NVIDIA GeForce RTX 5090 D

77,712 Benchmark Score
Top 2% 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

# CPU Analysis

The Intel Core Ultra 5 225 is a 10-core, 10-thread desktop processor built on the Arrow Lake architecture, manufactured on a 3 nm TSMC process node. This is a notably lean configuration for a modern desktop chip — no Hyper-Threading or equivalent, so thread count equals core count. The base clock of 3.30 GHz and boost clock of 4.90 GHz are solid figures for a 65 W TDP part, suggesting efficient power scaling rather than aggressive all-core boosting. The die contains 17,800 million transistors on a 243 mm² package, and the cache hierarchy includes 192 KB of L1 per core, 3 MB of L2 per core, and 20 MB of shared L3.

Benchmark results place this CPU in the 85th percentile among all processors, which is a strong showing for a mainstream 10-core part. The Cinebench R23 multi-core score of 25,891 and single-core score of 3,655 indicate balanced performance across both heavily threaded and lightly threaded workloads. The single-core figure is particularly meaningful: it shows the boost clock is not just a marketing number, and the architecture's IPC is competitive enough to handle gaming and productivity tasks that rely on single-thread responsiveness.

In multi-core workloads, the R23 score of 25,891 suggests this CPU can handle serious content creation tasks, though it will not match high-end 16-core or 24-core parts. The Passmark multi-thread score of 30,459 and integer math score of 65,345 reinforce this — solid throughput for compiling code, rendering, or video encoding, but not class-leading. The floating-point math score of 92,038 is disproportionately high relative to the integer score, indicating the Arrow Lake architecture's FP units are well-optimized for scientific and simulation workloads.

Data compression and encryption scores from Passmark — 302,811 and 22,285 respectively — are interesting. The compression score is exceptionally high, suggesting strong memory bandwidth utilization (the CPU supports dual-channel DDR5 with 102.4 GB/s), while the encryption score is more moderate, which may reflect a lack of dedicated cryptographic acceleration compared to some rivals. The extended instructions score of 27,162 and prime number finding score of 358 round out the picture: the former shows good SIMD throughput, while the latter is relatively low, indicating that the CPU does not excel at highly iterative, branch-heavy integer loops.

The nearest rival comparison shows this CPU sits almost exactly at parity with the AMD Ryzen 5 5600F, with a 0% delta in average benchmark score. It trails the AMD Ryzen 5 5500X3D by 0.2% and the AMD Ryzen AI 7 PRO 450 by 0.4%, while leading the Intel Core i9-12900T by 0.5%. These are all within noise-level margins, meaning the Ultra 5 225 is effectively performance-equivalent to those parts despite their different core counts and architectures. The average benchmark score of 36,938 positions it as a mid-to-upper mainstream CPU, not a flagship but well above budget territory.

The integrated Arc Xe-LPG Graphics with 16 execution units is a capable iGPU for basic display output and light office workloads, but it is clearly not intended for gaming — the discrete GPU in this pairing handles that role entirely. The CPU's 20 PCIe Gen 5 lanes provide ample bandwidth for a high-end GPU and one or two NVMe drives, though the lack of an unlocked multiplier means overclocking is not available for enthusiasts.

# Upgrade Path and Platform

The Intel Core Ultra 5 225 uses the Intel Socket 1851, which is the platform for the Core Ultra Series 2 desktop processors. This socket is relatively new, so users building today have a clear, if limited, upgrade path within the same generation. The CPU supports dual-channel DDR5 memory with a bandwidth of 102.4 GB/s, and does not support ECC memory, which makes it unsuitable for error-critical workstation environments but perfectly fine for consumer and prosumer use.

PCIe Gen 5 support with 20 CPU lanes is a significant feature. This allows a PCIe 5.0 x16 GPU to run at full bandwidth, and the remaining 4 lanes can drive a PCIe 5.0 NVMe SSD. For storage, this means extremely fast boot and load times, particularly for games and applications that can take advantage of Gen 5 throughput. The platform does not support PCIe Gen 4 natively on those CPU lanes, though motherboard chipsets may provide additional Gen 4 lanes for secondary slots.

The 65 W TDP is a relatively low figure for a desktop CPU with 10 cores, which means a modest air cooler is sufficient for stock operation. This also means power supply requirements are primarily driven by the GPU. The suggested PSU for this pairing is 950 W, which provides substantial headroom for the 575 W GPU TDP plus the CPU and other components. A quality 950 W unit will run this system well below its rated capacity, which is beneficial for efficiency and longevity.

For a sensible next upgrade, the data suggests that the CPU is not the primary bottleneck in most scenarios — the GPU is far more powerful relative to its peers. Users on this platform could consider upgrading to a higher-core-count Arrow Lake part if multi-threaded workloads become more demanding, but the 10-core configuration already handles most tasks well. The socket 1851 platform is not forward-compatible with future generations beyond the current Ultra 200 series, so a platform change would be required for a more substantial CPU upgrade. In the meantime, adding more DDR5 memory or a faster PCIe 5.0 SSD would yield more immediate benefits for most users.

# Benchmark Performance

The combined benchmark picture for this Intel Core Ultra 5 225 and NVIDIA GeForce RTX 5090 D pairing is one of strong overall performance, with a combined percentile of 89 across all CPU-GPU pairs. This places the system in the top 11% of all tested builds, which is a high-tier configuration suitable for demanding workloads and high-end gaming.

The CPU's individual percentiles and scores have been covered above, but the interaction with the GPU is what matters for real-world use. The GPU's 3DMark Steel Nomad DX12 score of 14,326 is a modern rasterization benchmark that stresses both the GPU and CPU. The Geekbench OpenCL score of 310,674 and Vulkan score of 376,915 reflect the GPU's raw compute capabilities under different API paradigms — Vulkan is notably higher, suggesting the architecture scales well with driver-level optimizations in that API.

The Passmark GPU tests show an interesting divergence: G3D score of 44,065 is very high, while DirectX 12 score of 219 is relatively low compared to DirectX 11's 371 and DirectX 9's 434. This is unusual and may reflect driver overhead or benchmark methodology quirks rather than actual DX12 performance, since the 3DMark Steel Nomad score (which is DX12) is strong. The GPU compute score of 28,396 is moderate, which may indicate that the Passmark compute test does not fully utilize the RTX 5090 D's tensor cores or specific compute features.

The average FPS across all tested games is 234.6, and the pair ranks 20th out of 3,732 pairs — placing it in the top 0.5% of all CPU-GPU combinations by FPS. This is an exceptional gaming result. The nearest GPU rivals in the benchmark database are the AMD Radeon RX 6650M XT (1.1% ahead), AMD Radeon RX 6850M XT (1.6% behind), and NVIDIA Tesla P100 variants (2.1-2.4% behind), which are all mobile or data-center parts — this underscores that the RTX 5090 D is a desktop flagship with no close desktop competitors in the database.

The CPU's performance relative to its nearest rivals is essentially flat, as noted earlier. This means that in gaming, the GPU will dominate the performance equation, and the CPU will rarely be the limiting factor except in the most CPU-bound titles at lower resolutions. The measured FPS data confirms this: at 1080p, many games show framerates above 300 FPS (CS:GO at 866, Valorant at 553, Roblox at 507), which are classic CPU-limit scenarios, but the GPU still delivers those numbers.

# Usage Scenarios

For high-refresh gaming, this pairing is outstanding at 1080p and 1440p. Counter-Strike 2 achieves 392 FPS at 1080p and 260 FPS at 1440p, while Valorant hits 553 FPS and 524 FPS at those resolutions. Tom Clancy's Rainbow Six Siege reaches 466 FPS at 1080p and 306 FPS at 1440p. These numbers are far beyond any 144 Hz or 240 Hz monitor's refresh rate, meaning the system will never be the bottleneck in competitive esports titles. At 4K, these games still exceed 145 FPS, which is playable on high-refresh 4K displays.

Streaming while gaming is a mixed scenario. The CPU's 10 cores without hyper-threading are sufficient for encoding at 1080p using x264, but the GPU's NVENC (via its tensor and RT cores) would be the preferred choice. The GPU has 680 tensor cores and 170 RT cores, which handle encoding and streaming offloads efficiently. The measured FPS in games like Call of Duty: Warzone (172 FPS at 1080p, 153 FPS at 1440p) shows that there is enough headroom to run a stream encoder without significant frame loss, especially if using GPU-based encoding.

Video editing is well-served by this configuration. The Cinebench R23 multi-core score of 25,891 and Passmark data compression score of 302,811 indicate strong performance for video codecs and timeline rendering. The GPU's 32 GB of VRAM and 1.79 TB/s bandwidth are exceptional for GPU-accelerated effects and 8K video editing. The high FP32 throughput of 104.8 TFLOPS accelerates effects processing in applications that leverage CUDA or DirectCompute.

3D rendering is a strong suit, though with caveats. The CPU's Cinebench R20 multi-core score of 6,317 and R15 multi-core score of 2,609 show it can handle CPU-based rendering in Blender or similar tools, but it will not match high-core-count workstation CPUs. The GPU, however, excels at GPU-based rendering with its massive FP32 and tensor core count. The Geekbench OpenCL score of 310,674 demonstrates the GPU's compute power, which is ideal for renderers like Octane or Redshift that leverage CUDA.

Software development benefits from the CPU's balanced single-thread and multi-thread performance. Compilation is typically multi-threaded, and the R23 multi-core score of 25,891 ensures reasonable build times. The single-core score of 3,655 is critical for IDE responsiveness and quick incremental builds. The Passmark integer math score of 65,345 indicates solid general-purpose integer performance for code execution and testing.

For student and office work, this is massively over-specified. The CPU's integrated Arc Xe-LPG Graphics would handle office tasks without the discrete GPU, though the RTX 5090 D is present in this build. Document editing, web browsing, and spreadsheets are trivial for this hardware, and the system will idle at low power despite the GPU's 575 W TDP when under full load.

# Who Should Build It

The target user for this build is a gamer who demands maximum frame rates at 1440p and 4K. The measured FPS data shows that at 2560x1440, demanding titles like Cyberpunk 2077 run at 86 FPS, Red Dead Redemption 2 at 79 FPS, and Call of Duty: Warzone at 153 FPS. At 3840x2160, these same games run at 57, 52, and 129 FPS respectively. This is a 4K-capable gaming machine that can handle ultra settings in most titles, though the most demanding games may require DLSS or other upscaling to hit 60 FPS at 4K.

Content creators who work with 4K video or large 3D scenes will benefit from the GPU's 32 GB VRAM, which allows entire projects to reside in GPU memory without spilling to system RAM. The CPU's 10 cores are sufficient to keep up with GPU-accelerated workflows, and the 65 W TDP means the system does not require exotic cooling for the CPU, leaving thermal headroom for the GPU.

Software developers who work on large codebases and also game will find this build appealing. The CPU's strong single-thread performance (3,655 in R23 single-core) ensures snappy IDE performance, while the multi-core score of 25,891 handles parallel builds. The GPU is overkill for development but provides a future-proof platform for any GPU-accelerated development tasks like machine learning inference or shader compilation.

Students and small business users are not the target audience for this build — it is far too powerful and expensive for typical office workloads. However, for a student in a demanding field like computer graphics, architecture visualization, or data science, the combination of CPU and GPU provides a workstation-class environment in a desktop form factor. The 950 W suggested PSU indicates that this is a serious power draw, so users should have appropriate electrical infrastructure.

# GPU Analysis

The NVIDIA GeForce RTX 5090 D is a desktop GPU built on the Blackwell 2.0 architecture, manufactured on a 5 nm TSMC process. The GB202 chip contains 92,200 million transistors on a 750 mm² die, making it one of the largest consumer GPUs ever produced. The transistor density of 122.9 million per mm² is impressive but reflects the mature 5 nm node rather than a current-generation process.

The GPU is equipped with 32 GB of GDDR7 memory on a 512-bit bus, yielding a memory bandwidth of 1.79 TB/s. This is a massive amount of VRAM and bandwidth, suitable for 8K texture packs, large AI models, and multi-monitor setups at high resolutions. The memory clock is listed as 1750 MHz with 28 Gbps effective, which is the GDDR7 standard for this generation.

The core configuration includes 21,760 shading units, 680 texture mapping units, and 176 raster operation processors. The GPU also has 170 RT cores for ray tracing and 680 tensor cores for AI acceleration. These numbers are the highest in the GeForce 50-series lineup. The base clock of 2017 MHz and boost clock of 2407 MHz are high for a GPU of this size, contributing to the pixel rate of 423.6 GPixel/s and texture rate of 1,636.8 GTexel/s.

Compute performance is exceptional: FP32 and FP16 throughput are both 104.8 TFLOPS (with FP16 at a 1:1 ratio), making this GPU a compute monster for both single-precision and half-precision workloads. The tensor cores are particularly relevant for DLSS and AI-based rendering techniques, which can significantly boost effective frame rates in supported titles.

Benchmark results confirm the GPU's high-end positioning. The 3DMark Steel Nomad DX12 score of 14,326 indicates strong modern rasterization performance. The Geekbench Vulkan score of 376,915 is higher than the OpenCL score of 310,674, suggesting the driver's Vulkan implementation is more mature or better optimized for this architecture. The Passmark G3D score of 44,065 is very high, though the individual DirectX scores (9, 10, 11, 12) show some variance that may be due to benchmark methodology rather than real-world performance.

The GPU's percentile rank of 92 among all GPUs places it in the top 8% of all graphics cards ever tested. Its average benchmark score of 77,712 is only slightly above the AMD Radeon RX 6650M XT (76,904) and below the AMD Radeon RX 6850M XT (78,940) and NVIDIA Tesla P100 variants (79,396-79,605). These rivals are mobile or data-center parts, so the comparison is not directly relevant to desktop gaming performance. The RTX 5090 D's nearest desktop competitors are not listed in the database, which speaks to its unique positioning.

The GPU requires a 575 W TDP and uses a single 16-pin power connector. The suggested PSU of 950 W provides adequate headroom even with a high-end CPU. The dual-slot design measures 304 mm in length, 137 mm in height, and 48 mm in width, so case compatibility should be checked. Display outputs include 1x HDMI 2.1b and 3x DisplayPort 2.1b, supporting the latest high-refresh and high-resolution monitors.

# FAQ

Q: What is the CPU's core and thread configuration?

A: The Intel Core Ultra 5 225 has 10 cores and 10 threads, with no hyper-threading. It runs at a base clock of 3.30 GHz and a boost clock of 4.90 GHz.

Q: How much VRAM does the RTX 5090 D have?

A: The NVIDIA GeForce RTX 5090 D has 32 GB of GDDR7 memory on a 512-bit bus, providing 1.79 TB/s of memory bandwidth.

Q: What is the combined performance percentile of this CPU-GPU pair?

A: The combined percentile is 89, meaning it outperforms 89% of all tested CPU-GPU pairs in the database. The pair also ranks 20th out of 3,732 pairs by average FPS.

Q: Is this build suitable for 4K gaming?

A: Yes, based on measured data. At 3840x2160, titles like Counter-Strike 2 run at 145 FPS, Call of Duty: Warzone at 129 FPS, and Cyberpunk 2077 at 57 FPS. Many games exceed 60 FPS at 4K ultra settings.

Q: What kind of power supply is recommended?

A: The suggested PSU is 950 W. This provides headroom for the GPU's 575 W TDP and the CPU's 65 W TDP, plus other system components.

Q: Does the CPU support ECC memory?

A: No, the Intel Core Ultra 5 225 does not support ECC memory. It supports dual-channel DDR5 with a bandwidth of 102.4 GB/s.

Q: What is the GPU's ray tracing and tensor core configuration?

A: The RTX 5090 D has 170 RT cores and 680 tensor cores. Its FP32 and FP16 compute performance is 104.8 TFLOPS each.

# Build Overview

This build pairs the Intel Core Ultra 5 225, a 10-core desktop CPU from the Core Ultra Series 2, with the NVIDIA GeForce RTX 5090 D, a desktop GPU from the GeForce 50-series. The build class is desktop, confirming this is a full-tower or similar form factor system, not a laptop or mini-PC.

The CPU is positioned at the 85th percentile among all processors, while the GPU sits at the 92nd percentile among all GPUs. The combined percentile of 89 reflects the overall system performance, which is in the top 11% of all tested configurations. The average benchmark score across all games is 234.6 FPS, and the pair ranks 20th out of 3,732 total pairs — a top 0.5% result.

This is a high-tier desktop build that pairs a capable mid-high-end CPU with an absolute flagship GPU. The CPU is not the fastest available, but it is fast enough to avoid bottlenecking the GPU in most scenarios. The GPU is the clear performance leader in this pairing, and its capabilities far exceed what the CPU could fully saturate in compute-heavy tasks. The system is best described as a GPU-first build, where the CPU provides solid all-around performance for everyday tasks and gaming, while the GPU delivers exceptional frame rates and compute throughput.

# Balance and Bottleneck

The balance between the Intel Core Ultra 5 225 and the NVIDIA GeForce RTX 5090 D is asymmetric, and the data clearly shows where the bottlenecks lie. The CPU's 85th percentile rank is respectable but not elite, while the GPU's 92nd percentile rank is near the top. In gaming workloads, the GPU will be the primary performance driver, and the CPU will only become a limiting factor in extremely CPU-bound scenarios at lower resolutions.

The measured FPS scaling across resolutions provides evidence of this. In Counter-Strike 2, FPS drops from 392 at 1080p to 260 at 1440p to 145 at 4K. If the GPU were the only bottleneck, we would expect more gradual scaling. The steep drop from 1080p to 1440p (33% reduction) suggests the CPU is limiting at 1080p, where the GPU has headroom but the CPU cannot feed frames fast enough. At 4K, the GPU becomes the primary limiter, and the FPS drop is less severe relative to the resolution increase.

Valorant shows an even clearer CPU bottleneck: 553 FPS at 1080p, 524 FPS at 1440p, and 487 FPS at 4K. The minimal FPS scaling across resolutions indicates that the CPU is the limiting factor at all three resolutions, as the game is not GPU-intensive enough to saturate the RTX 5090 D. This is a classic case of a flagship GPU being held back by a mainstream CPU in esports titles.

In contrast, Cyberpunk 2077 shows FPS of 112 at 1080p, 86 at 1440p, and 57 at 4K. The scaling here is more typical of a GPU-bound scenario, where the CPU has enough headroom to feed the GPU at all resolutions. The GPU's performance is the primary constraint, and the CPU is not the bottleneck.

For compute workloads, the balance shifts. The GPU's FP32 throughput of 104.8 TFLOPS dwarfs the CPU's Passmark floating-point score of 92,038. In GPU-accelerated rendering or AI inference, the GPU will be the dominant performer. In CPU-based rendering, the CPU's 10 cores will limit performance, and the GPU will be underutilized. The 32 GB of GPU VRAM is also a significant advantage for large datasets, whereas the CPU's system memory is limited to the DDR5 bandwidth of 102.4 GB/s.

Overall, the CPU is the bottleneck in lightly-threaded, high-FPS gaming scenarios, while the GPU is the bottleneck in graphically demanding games and compute-heavy workloads. This is not an ideal pairing for competitive esports at 1080p, where a faster CPU would yield higher frame rates, but it is excellent for 4K gaming and GPU-accelerated content creation.

# Gaming Performance

The measured FPS data shows that this build delivers exceptional gaming performance across a wide range of titles and resolutions, with dataIsMeasured set to true, meaning these are actual test results rather than estimates.

At 1920x1080, the system achieves extremely high frame rates in competitive titles: Counter-Strike: Global Offensive hits 866 FPS, Valorant reaches 553 FPS, Tom Clancy's Rainbow Six Siege runs at 466 FPS, and Counter-Strike 2 achieves 392 FPS. These are all far beyond the refresh rates of any current monitor, making the system ideal for 360 Hz or even 500 Hz esports displays. Even in more demanding titles, 1080p performance is strong: Cyberpunk 2077 runs at 112 FPS, Red Dead Redemption 2 at 105 FPS, and Microsoft Flight Simulator at 142 FPS.

At 2560x1440, the system remains highly capable. Counter-Strike: Global Offensive still hits 596 FPS, Valorant runs at 524 FPS, and Rainbow Six Siege achieves 306 FPS. AAA titles maintain high frame rates: Call of Duty: Warzone runs at 153 FPS, Cyberpunk 2077 at 86 FPS, and Red Dead Redemption 2 at 79 FPS. Microsoft Flight Simulator runs at 114 FPS, which is excellent for a simulator that is notoriously demanding. The 1440p results are well-suited for 144 Hz and 240 Hz monitors.

At 3840x2160, the system is still capable of high-refresh gaming in many titles, though the most demanding games require compromises. Counter-Strike: Global Offensive runs at 352 FPS, Valorant at 487 FPS, and Rainbow Six Siege at 168 FPS. Call of Duty: Warzone hits 129 FPS, Overpower runs at 119 FPS, and Assetto Corsa achieves 98 FPS. However, Cyberpunk 2077 drops to 57 FPS, Red Dead Redemption 2 to 52 FPS, and Ark: Survival Ascended to 19 FPS. These more demanding titles will benefit from DLSS or other upscaling technologies to reach 60 FPS at 4K.

The average FPS across all tested games is 234.6, with a pair rank of 20th out of 3,732. This indicates that the build is in the top 0.5% of all tested systems by gaming performance. The GPU's 32 GB of VRAM ensures that even games with large texture packs at 4K will not exceed memory capacity. The 1.79 TB/s bandwidth provides ample throughput for high-resolution textures and effects.

Notable outliers include Barony at 331 FPS at 1080p and 109 FPS at 4K, Roblox at 507 FPS at 1080p and 191 FPS at 4K, and Minecraft at 386 FPS at 1080p and 127 FPS at 4K. These lighter titles scale well with the CPU's strong single-core performance. The Medium runs at 82 FPS at 1080p and 54 FPS at 4K, showing that even demanding horror titles are playable at 4K. Ready Or Not achieves 316 FPS at 1080p and 104 FPS at 4K, which is excellent for a tactical shooter. The overall picture is a system that can handle any game on the market at 1080p and 1440p with ease, and at 4K with only the most demanding titles requiring upscaling.