SYSTEM ANALYZER

Rate My PC: Intel Core Ultra 5 245KF + NVIDIA GeForce RTX 5090

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

96 / 100
ULTIMATE READY

Apex Performer

Top 4% 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
98%
PROCESSOR

Intel Core Ultra 5 245KF

55,093 Benchmark Score
Top 5% Market Ranking
View Full Specs →
GRAPHICS CARD

NVIDIA GeForce RTX 5090

79,842 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

The Intel Core Ultra 5 245KF and NVIDIA GeForce RTX 5090 form a desktop build that ranks at the 92nd percentile across all hardware combinations in the database. The processor holds the 91st percentile among all CPUs, while the graphics card occupies the 92nd percentile among all GPUs. This pairing brings together a 14-core Arrow Lake CPU with a 32 GB GDDR7 Blackwell GPU, and the benchmark data indicates a top-tier system for both compute and graphics workloads. No measured FPS rows exist for this exact combination — the FACT PACK contains no measuredFps data — so all frame-rate discussion below is estimated from the benchmark scores.

GPU Analysis

The RTX 5090 is built on the GB202 chip using Blackwell 2.0 architecture, manufactured on a 5 nm process at TSMC with 92,200 million transistors on a 750 mm² die. Memory capacity sits at 32 GB of GDDR7 across a 512-bit bus, delivering 1.79 TB/s of bandwidth. The core clock runs at 2017 MHz base and 2407 MHz boost, with the memory clock at 1750 MHz operating at 28 Gbps effective. The card contains 21,760 shading units, 680 texture mapping units, 176 ROPs, 170 ray-tracing cores, and 680 tensor cores. Pixel throughput is rated at 423.6 GPixel/s and texture rate at 1,636.8 GTexel/s, while FP32 and FP16 both compute 104.8 TFLOPS in a 1:1 ratio. The GPU is a dual-slot card measuring 304 mm in length, 137 mm in height, and 40 mm in width, drawing a TDP of 575 W and requiring a suggested PSU of 950 W with a single 16-pin connector.

Benchmark results show the GPU scoring 18,355 in 3DMark Steel Nomad DX12, 334,370 in Geekbench OpenCL, and 376,728 in Geekbench Vulkan. PassMark G3D returns 39,650, while G2D is 1,413. Legacy DirectX tests yield 226 for DX10, 341 for DX11, 185 for DX12, and 395 for DX9, with GPU compute at 26,756. The GPU sits at the 92nd percentile with an average benchmark score of 79,842. Compared to nearest rivals, the RTX 5090 trails the NVIDIA Tesla P100 PCIe 16 GB (79,605) by 0.3% and the Tesla P100 PCIe 12 GB (79,396) by 0.6%, while leading the AMD Radeon RX 6850M XT (78,940) by 1.1% and the AMD Radeon Pro Vega 64X (80,959) by 1.4% behind. The 32 GB frame buffer and 1.79 TB/s bandwidth support large textures and high-resolution rendering workloads, while the 170 RT cores and 680 tensor cores accelerate ray-traced scenes and AI-accelerated features. The FP32 throughput of 104.8 TFLOPS indicates strong compute capacity for rendering pipelines, and the 92nd percentile ranking confirms the GPU is among the top hardware in the database.

Benchmark Performance

The CPU posts a Cinebench R23 multi-core score of 36,647 and a single-core score of 5,173. In R20, multi-core reaches 15,391 and single-core 2,172, while R15 shows 3,693 multi-core and 521 single-core. PassMark multithread scores 43,110, single-thread 4,715, and the average benchmark score for the CPU is 55,093, placing it at the 91st percentile. The GPU achieves 18,355 in 3DMark Steel Nomad DX12, 334,370 in OpenCL, and 376,728 in Vulkan, with a G3D score of 39,650 and an average of 79,842 at the 92nd percentile. Combined, the pair ranks at the 92nd percentile for all builds.

The CPU's nearest rivals reveal tight competition. The AMD Ryzen 9 9900X3D scores 54,762, sitting 0.6% behind the 245KF, while the Intel Core 7 253PQE reaches 55,919, 1.5% ahead, and the Intel Core i9-14900HX scores 56,004, 1.6% ahead. The Intel Core Ultra 5 245K scores 54,053, 1.9% behind. The GPU's nearest rivals are similarly close, with the Tesla P100 16 GB at 79,605 (0.3% ahead) and the Radeon Pro Vega 64X at 80,959 (1.4% ahead). The combined picture is a system where the GPU leads slightly over the CPU in percentile ranking, yet both sit within a narrow band of top-tier hardware. The delta percentages show that neither component dramatically outpaces its closest alternatives — the CPU sits within 1.9% of its nearest rivals, and the GPU within 1.4% — indicating a balanced pairing of two very strong parts. The CPU's 91st percentile and the GPU's 92nd percentile translate to a system that is positioned just below the absolute peak but well above most desktop configurations.

CPU Analysis

The Intel Core Ultra 5 245KF is a 14-core, 14-thread processor (no hyper-threading) from the Core Ultra Series 2, based on Arrow Lake architecture (Arrow Lake-S). It runs at a base clock of 4.20 GHz and a boost clock of 5.20 GHz, with a TDP of 125 W. Fabricated on a 3 nm process by TSMC, the chip contains 17,800 million transistors on a 243 mm² die. Cache is 192 KB L1 per core, 3 MB L2 per core, and 24 MB shared L3. Memory support covers dual-channel DDR5 with 102.4 GB/s bandwidth, no ECC, and the CPU provides PCIe Gen 5 with 20 lanes (CPU only). There is no integrated graphics, so a discrete GPU is mandatory. The multiplier is unlocked, and the CPU was released on 2024-10-23 with a launch MSRP of $294.

Benchmark data shows the CPU's multi-threaded strength: Cinebench R23 multi-core 36,647 and single-core 5,173; PassMark multithread 43,110 and single-thread 4,715. The PassMark sub-tests reveal the workload profile: integer math at 98,854, floating-point math at 131,546, extended instructions at 37,912, data compression at 460,123, data encryption at 33,381, find prime numbers at 416, random string sorting at 55,206, and physics at 2,998. The CPU's 91st percentile average of 55,093 places it just 0.6% behind the Ryzen 9 9900X3D and 1.9% ahead of the Core Ultra 5 245K. The 14 cores without hyperthreading mean each core handles one thread, which is well-suited for multithreaded workloads that scale linearly, but the single-thread score of 5,173 in R23 indicates strong per-core performance for latency-sensitive tasks. The dual-channel DDR5 bandwidth of 102.4 GB/s supports memory-hungry applications, though the lack of ECC limits server-grade reliability use. The 3 nm process and 17.8 billion transistors position this CPU as a modern, power-efficient part relative to its performance tier.

Who Should Build It

This pairing targets users who need top-tier graphics performance alongside a strong, modern CPU. The 92nd combined percentile and the GPU's 32 GB VRAM make it suitable for 4K gaming and high-resolution content creation. The CPU's 91st percentile and 14 threads handle multithreaded workloads like video rendering and code compilation, while the GPU's 170 RT cores and 680 tensor cores accelerate ray-traced rendering and AI-accelerated effects. Gamers at 4K or 1440p high refresh rates will see the GPU saturate, given the 92nd percentile graphics performance and the CPU's 5.2 GHz boost. Content creators editing 4K video or running 3D rendering software benefit from the 32 GB VRAM and 104.8 TFLOPS FP32 throughput, with the CPU's 36,647 multi-core Cinebench score supporting timeline scrubbing and encoding. Software developers compiling large codebases will use the 14 threads and the 131,546 floating-point math score for heavy computation, while the unlocked multiplier allows tuning for those who want more. Students and small business workstations with heavy simulation or analysis needs can rely on the CPU's 91st percentile and the GPU's compute score of 26,756, though the 575 W TDP requires a robust power supply. The lack of integrated graphics means the RTX 5090 is the only display output path, so this build is not for basic office use without the GPU active.

Usage Scenarios

High-refresh gaming: With the GPU at the 92nd percentile and the CPU at the 91st, the system can drive high frame rates at 1440p and 4K, though the exact FPS is unknown since no measured data exists. The 32 GB VRAM and 1.79 TB/s bandwidth support high-resolution textures without memory pressure.

Streaming: The 14-core CPU handles game encoding and background tasks, while the GPU's tensor cores can accelerate video encoding. The 104.8 TFLOPS FP32 and 680 tensor cores provide headroom for real-time encoding and overlays, though the 575 W GPU TDP requires a capable PSU.

Video editing: The CPU's Cinebench R23 multi-core of 36,647 and the GPU's 32 GB VRAM allow for multi-stream 4K editing and GPU-accelerated effects. The 1.79 TB/s bandwidth moves large frame buffers quickly, and the 680 TMUs handle texture-heavy timelines.

3D rendering: The GPU's 170 RT cores and 104.8 TFLOPS FP32 throughput accelerate ray-traced renders, while the CPU's 14 threads (36,647 multi-core score) handle scene setup and physics. The 32 GB GDDR7 buffer can hold large scenes without spilling to system memory.

Software development: The CPU's single-thread score of 5,173 in R23 handles rapid compilation of single files, while the 14 threads and 43,110 PassMark multithread score speed parallel builds. The GPU is not the primary workhorse here, but its compute cores assist in testing GPU-heavy code.

Student and office work: The system is overkill for typical office tasks, but the CPU's high single-thread performance and 91st percentile ranking make for fast responsiveness. The 125 W CPU TDP and 575 W GPU TDP require a solid power supply, making this more of a professional workstation than a light office machine.

Gaming Performance

No measured FPS rows exist for this exact combination — the FACT PACK contains no measuredFps data — so the following frame-rate expectations are estimated from the benchmark scores. The GPU's 92nd percentile and 3DMark Steel Nomad score of 18,355 indicate very high performance in DX12 titles. The CPU's 91st percentile and 4.2 GHz base / 5.2 GHz boost clock suggest it can feed the GPU without bottlenecking at most resolutions. For 1080p and 1440p, the CPU's single-core performance (R23 single 5,173) will likely be the limiting factor in CPU-bound games, but the GPU should deliver high frame rates. At 4K, the GPU becomes the primary driver, and the 32 GB VRAM and 1.79 TB/s bandwidth prevent memory-related frame drops. Since no measured FPS rows exist, these are qualitative estimates: expect high frame rates in DX12 and Vulkan titles, with the GPU at the 92nd percentile and the CPU at the 91st. The estimated FPS scaling would show a GPU-bound system at 4K, with the CPU's 14 cores providing adequate multi-threaded performance for background tasks.

Balance and Bottleneck

The CPU and GPU are closely matched in percentile rank — 91st for the CPU and 92nd for the GPU — which indicates a well-balanced pairing. In CPU-bound workloads, such as simulation or single-threaded game logic, the CPU's 5.2 GHz boost and R23 single-core of 5,173 provide a strong ceiling, but the lack of hyperthreading means 14 threads may limit heavily threaded tasks. The GPU's 575 W TDP and 32 GB VRAM are the dominant power draw, and in GPU-bound scenarios, the CPU has ample headroom to feed it. The PassMark physics score of 2,998 suggests the CPU handles physics calculations well, but the GPU's compute throughput (26,756 in PassMark GPU compute) is where most rendering load falls. In a game at 4K, the GPU's 92nd percentile will saturate first, making the CPU less of a bottleneck. In a compile or encoding workload, the CPU becomes the limiting factor, with the GPU idle. The 125 W CPU TDP and 575 W GPU TDP together require a 950 W PSU, but the system's balance means neither component is left idle during mixed workloads. The benchmark deltas — CPU within 1.6% of its rivals and GPU within 1.4% — show that neither component is a clear weak point.

Upgrade Path and Platform

The CPU uses Intel Socket 1851, which supports the Core Ultra Series 2 family. Memory support is dual-channel DDR5 with a bandwidth of 102.4 GB/s, and the CPU provides PCIe Gen 5 with 20 lanes (CPU only). The GPU uses PCIe 5.0 x16, so the bus interface is fully compatible. The GPU's suggested PSU is 950 W, with a TDP of 575 W, and the CPU TDP is 125 W, leaving headroom for additional drives or cards. The system's power delivery is sufficient for the current pair, but a future upgrade to a higher-core CPU on the same socket would not require a PSU change. The GPU has 32 GB VRAM, which is high, and the 1.79 TB/s bandwidth is unlikely to bottleneck in the near term. The CPU is unlocked, so overclocking can extend performance without a platform change. A sensible next upgrade would be a CPU with more cores or higher clocks on the same socket, or a future GPU that keeps the same PCIe 5.0 x16 interface. The memory support for dual-channel DDR5 and no ECC means users should stick with standard DDR5 modules. The platform is modern with PCIe Gen 5, so storage and expansion are not limiting.

FAQ

Q: What is the combined percentile of this CPU+GPU build?

A: The build ranks at the 92nd percentile across all hardware.

Q: How much VRAM does the RTX 5090 have?

A: The GPU has 32 GB of GDDR7 memory on a 512-bit bus with 1.79 TB/s bandwidth.

Q: What socket does the CPU use?

A: The Intel Core Ultra 5 245KF uses Intel Socket 1851.

Q: Does the CPU have integrated graphics?

A: No, the CPU has no integrated graphics, so a discrete GPU is required for display output.

Q: What is the GPU's TDP and suggested PSU?

A: The GPU TDP is 575 W, and the suggested PSU is 950 W.

Q: What is the CPU's memory support?

A: The CPU supports dual-channel DDR5 with 102.4 GB/s bandwidth and no ECC.

Q: What PCIe version does the platform support?

A: The CPU provides PCIe Gen 5 with 20 lanes, and the GPU uses PCIe 5.0 x16.

Build Overview

This is a desktop-class build pairing the Intel Core Ultra 5 245KF with the NVIDIA GeForce RTX 5090. The CPU ranks at the 91st percentile with an average benchmark score of 55,093, while the GPU ranks at the 92nd percentile with an average of 79,842. The combined percentile is 92, placing this system in the upper tier of desktop configurations. The CPU was released on 2024-10-23 with a launch MSRP of $294, and the GPU on 2025-01-29 with a launch MSRP of $1,999. The CPU's 14 cores and 4.20 GHz base / 5.20 GHz boost clock, combined with the GPU's 32 GB GDDR7 and 104.8 TFLOPS, create a system that is near the top of the database. The nearest CPU rival is the AMD Ryzen 9 9900X3D, which is 0.6% ahead, and the nearest GPU rival is the Tesla P100, which is 0.3% ahead, indicating that this build is slightly below the absolute peak but still a high-end desktop solution. The lack of measured FPS data means the frame-rate estimates are derived from the benchmark scores, but the overall picture is a powerful, balanced workstation and gaming PC.