SYSTEM ANALYZER

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

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 245K

54,053 Benchmark Score
Top 5% 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

The Intel Core Ultra 5 245K and NVIDIA GeForce RTX 5090 D pairing is a desktop configuration that combines a 91st-percentile CPU with a 92nd-percentile GPU, placing the overall build in the 92nd percentile of all combinations. The data confirms a high-end system, with the CPU's average benchmark score of 54,053 and the GPU's average of 77,712. No measured FPS rows exist for this exact combination in the FACT PACK; all FPS discussion is estimated from the individual benchmark scores.

CPU Analysis

The Intel Core Ultra 5 245K is a 14-core, 14-thread processor from the Core Ultra Series 2, built on the Arrow Lake architecture and manufactured on a 3 nm process by TSMC. It lacks hyper-threading, meaning the 14 cores correspond directly to 14 threads, a design choice that prioritizes efficiency and single-core performance over multi-threaded throughput in heavily threaded workloads. The base clock is 4.20 GHz with a boost clock of 5.20 GHz, and the chip carries a 125 W TDP. The CPU supports DDR5 memory via a dual-channel bus, offering a memory bandwidth of 102.4 GB/s, and includes ECC memory support, which is notable for workstation reliability.

In Cinebench R23, the CPU scores 25,066 in multi-core and 2,132 in single-core. The multi-core score is the headline figure, indicating strong sustained performance for rendering and compilation tasks. The single-core score of 2,132 is competitive, suggesting excellent responsiveness in lightly threaded applications. The CPU's 91st percentile ranking against all CPUs is supported by its nearest rivals: the Intel Xeon 6505P scores 53,701 (0.7% slower), the Intel Core i7-14700F scores 53,620 (0.8% slower), and the AMD Ryzen 9 9900X3D scores 54,762 (1.3% faster). This places the 245K in a tight cluster where performance differences are minimal, but the 245K edges out the Xeon and i7-14700F while trailing the Ryzen 9 9900X3D by a small margin.

PassMark results provide further granularity. The multi-thread score is 43,047, while single-thread is 4,714. The integer math score is 98,524, floating-point math is 130,678, and extended instructions (SIMD) score 37,617. Data compression hits 458,817, while data encryption is 33,286. The physics score is 3,081, and finding prime numbers scores 414. These numbers suggest a CPU that excels in floating-point and integer workloads, with encryption being a relative weak point—though still respectable for a mainstream desktop chip. The L3 cache is 24 MB shared, with L2 at 3 MB per core and L1 at 192 KB per core. The 17,800 million transistors on a 243 mm² die indicate a dense, modern design.

For real-world workloads, the Cinebench R23 multi-core score of 25,066 suggests the CPU can handle video encoding, 3D rendering, and software compilation with ease. The single-core score of 2,132 ensures snappy application launches and responsive gaming. The lack of hyper-threading is a trade-off: power efficiency and thermal headroom improve, but synthetic multi-threaded tests that benefit from additional threads may not reach the levels of competing 14-core/20-thread designs. Still, the 91st percentile ranking confirms it is a top-tier desktop CPU, and its 0.8% advantage over the Intel Core i7-14700F in average score shows it holds its own against a popular alternative.

GPU Analysis

The NVIDIA GeForce RTX 5090 D is a Blackwell 2.0 architecture GPU on a 5 nm TSMC process, featuring the GB202 chip with 92,200 million transistors on a 750 mm² die. It has 32 GB of GDDR7 memory on a 512-bit bus, delivering a bandwidth of 1.79 TB/s. The base clock is 2017 MHz with a boost clock of 2407 MHz, and memory runs at 1750 MHz (28 Gbps effective). The GPU has 21,760 shading units, 680 texture mapping units, 176 ROPs, 170 RT cores, and 680 tensor cores. It achieves a pixel rate of 423.6 GPixel/s and a texture rate of 1,636.8 GTexel/s. FP32 and FP16 performance are both rated at 104.8 TFLOPS, indicating a 1:1 ratio that is unusual and implies strong compute throughput.

The RTX 5090 D sits in the 92nd percentile of all GPUs, with an average benchmark score of 77,712. Its nearest rivals are the AMD Radeon RX 6650M XT (score 76,904, 1.1% slower), the AMD Radeon RX 6850M XT (score 78,940, 1.6% faster), the NVIDIA Tesla P100 PCIe 12 GB (score 79,396, 2.1% faster), and the NVIDIA Tesla P100 PCIe 16 GB (score 79,605, 2.4% faster). The deltaPct values show that the 5090 D is bracketed by mobile AMD GPUs and older Tesla accelerators, but the 5090 D's 32 GB VRAM and 1.79 TB/s bandwidth are far more suited to modern gaming and rendering than those rivals.

In 3DMark Steel Nomad DX12, the GPU scores 14,326, which is a strong result for a modern DirectX 12 title. Geekbench OpenCL scores 310,674, and Vulkan scores 376,915, showing robust compute performance across APIs. PassMark results include a G3D score of 44,065 and a GPU compute score of 28,396. The DirectX 9, 10, 11, and 12 scores are 434, 231, 371, and 219 respectively; these low values are typical for PassMark's legacy DX benchmarks, which do not scale well with modern architectures, but the G3D score of 44,065 is the more representative figure for overall graphics performance. The G2D score of 1,487 indicates solid 2D acceleration.

For rendering workloads, the 104.8 TFLOPS FP32 and FP16 compute, combined with 170 RT cores and 680 tensor cores, make the RTX 5090 D a formidable tool for ray tracing, AI denoising, and neural rendering. The 32 GB VRAM is ample for 8K textures, large scenes in DCC applications, and AI model training that fits on a single GPU. The 1.79 TB/s bandwidth ensures that memory-intensive operations, such as real-time 4K video editing or massive dataset processing, are not bottlenecked by data transfer. The GPU's dual-slot design and 304 mm length require a spacious case, and the 1x 16-pin power connector with a suggested 950 W PSU indicates high power demands, but the performance justifies the hardware requirements.

Benchmark Performance

The CPU's average benchmark score is 54,053, placing it in the 91st percentile of all CPUs. The GPU's average score is 77,712, placing it in the 92nd percentile of all GPUs. The combined system percentile is 92, indicating that this pairing is at the high end of desktop builds. Since there are no measured FPS rows for this exact combination, all gaming performance is estimated from these benchmark scores.

In Cinebench R23, the CPU's multi-core score of 25,066 is 0.8% higher than the Intel Core i7-14700F's average score of 53,620, though the comparison is indirect since the R23 score is a specific test and the rival score is an average across multiple tests. The CPU's single-core score of 2,132 in R23 is strong, suggesting that gaming performance, which often depends on single-thread speed, will be excellent. The GPU's 3DMark Steel Nomad score of 14,326 is a direct measure of DX12 gaming capability, and the Geekbench Vulkan score of 376,915 indicates strong Vulkan performance, which is relevant for modern game engines.

The combined picture shows a system where both components are in the top 10% of their respective categories. The CPU's 91st percentile and GPU's 92nd percentile are nearly matched, suggesting a balanced build where neither component drastically outperforms the other. The PassMark multi-thread score of 43,047 for the CPU and the GPU compute score of 28,396 indicate that the system can handle both CPU-bound and GPU-bound tasks effectively. However, the CPU's 1.3% deficit to the AMD Ryzen 9 9900X3D in average score suggests that in CPU-bound scenarios, there is a slightly faster alternative, but the difference is minor in real-world use.

The data indicates that for 1440p and 4K gaming, the GPU will be the primary driver of FPS, with the CPU providing sufficient single-core performance to avoid bottlenecks. For productivity tasks like video editing, the CPU's multi-core score and the GPU's compute score interact: rendering timelines will benefit from the GPU's 104.8 TFLOPS, while export encoding will leverage the CPU's 25,066 R23 multi-core score. The combined percentile of 92 confirms that this is a high-tier system, but the lack of measured FPS data means that exact frame rates cannot be stated; instead, the scores suggest that the system will perform admirably across demanding workloads.

FAQ

Q: What is the CPU's multi-core performance in Cinebench R23?

A: The Intel Core Ultra 5 245K scores 25,066 in Cinebench R23 multi-core, placing it in the 91st percentile of all CPUs.

Q: How much VRAM does the RTX 5090 D have, and what is its bandwidth?

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

Q: How does the CPU compare to the AMD Ryzen 9 9900X3D?

A: The CPU's average benchmark score is 54,053, which is 1.3% lower than the AMD Ryzen 9 9900X3D's average score of 54,762.

Q: What is the GPU's score in 3DMark Steel Nomad DX12?

A: The RTX 5090 D scores 14,326 in 3DMark Steel Nomad DX12.

Q: Does the CPU support ECC memory?

A: Yes, the Intel Core Ultra 5 245K supports ECC memory, which is beneficial for workstation reliability.

Q: What is the suggested PSU wattage for the RTX 5090 D?

A: The suggested PSU for the RTX 5090 D is 950 W, given its 575 W TDP.

Q: What is the GPU's FP32 compute performance?

A: The RTX 5090 D achieves 104.8 TFLOPS in FP32, with FP16 performance also rated at 104.8 TFLOPS (1:1).

Balance and Bottleneck

The CPU and GPU are closely matched in percentile ranking—91st for the CPU and 92nd for the GPU—which suggests a balanced pairing where neither component is a consistent bottleneck. However, the nature of the workload determines which component limits performance. In CPU-bound tasks, such as physics simulation or data compression, the CPU's PassMark physics score of 3,081 and data compression score of 458,817 are the limiting factors. The CPU's 1.3% deficit to the AMD Ryzen 9 9900X3D in average score means that in heavily threaded CPU workloads, the system may trail by a small margin, but the 0.8% advantage over the Intel Core i7-14700F shows it is competitive.

In GPU-bound tasks, such as high-resolution gaming or 3D rendering with ray tracing, the RTX 5090 D's 92nd percentile ranking and 104.8 TFLOPS FP32 compute dominate. The GPU's 1.1% advantage over the AMD Radeon RX 6650M XT and its 2.4% deficit to the NVIDIA Tesla P100 PCIe 16 GB indicate that its performance is strong but not the absolute highest in every metric. For gaming at 4K, the GPU will be the primary driver, and the CPU's single-core score of 2,132 in Cinebench R23 should be sufficient to feed the GPU without causing a CPU bottleneck, given that the CPU's single-thread PassMark score of 4,714 is robust.

The FPS scaling, estimated from the benchmark scores, suggests that at lower resolutions like 1080p, the CPU's single-core performance becomes more relevant, while at 4K, the GPU's rendering power takes over. The data shows no measured FPS rows for this combination, so the exact scaling cannot be quantified, but the percentile alignment indicates that the system will maintain high frame rates across resolutions. The CPU's 24 MB L3 cache and 102.4 GB/s memory bandwidth are adequate for feeding the GPU's 1.79 TB/s bandwidth, but the dual-channel memory bus may slightly limit CPU-side data throughput in memory-intensive scenarios. Overall, the balance is good, but users should expect the CPU to be the limiting factor in pure physics or compression tasks, and the GPU to be the limiting factor in high-resolution rendering.

Usage Scenarios

High-refresh gaming: The RTX 5090 D's 3DMark Steel Nomad score of 14,326 and 104.8 TFLOPS FP32 compute suggest it can drive high frame rates at 1440p and 4K. The CPU's single-core score of 2,132 in Cinebench R23 and 4,714 in PassMark single-thread ensure that frame pacing is smooth, but the lack of measured FPS data means exact refresh rates cannot be stated. The 32 GB VRAM provides headroom for high-resolution textures.

Streaming: The CPU's 14 cores and 14 threads, with a multi-thread score of 43,047 in PassMark, can handle encoding at moderate presets. The GPU's 680 tensor cores and 1,636.8 GTexel/s texture rate support NVENC-style encoding, though the specific encoder is not detailed. The system's 92nd combined percentile indicates it can handle simultaneous gaming and streaming, but the CPU's lack of hyper-threading may require using the GPU for encoding to maintain FPS.

Video editing: The CPU's Cinebench R23 multi-core score of 25,066 accelerates export rendering, while the GPU's 32 GB VRAM and 1.79 TB/s bandwidth handle 4K and 8K timelines smoothly. The Geekbench OpenCL score of 310,674 indicates strong compute for effects and color grading. The GPU's 104.8 TFLOPS FP16 performance is useful for AI-accelerated editing tools.

3D rendering: The CPU's 25,066 multi-core score in Cinebench R23 is directly relevant to CPU-based rendering, while the GPU's 104.8 TFLOPS FP32 and 170 RT cores accelerate GPU-accelerated renderers like those using ray tracing. The 32 GB VRAM allows large scenes to fit in memory, reducing out-of-core spills. The GPU compute score of 28,396 in PassMark supports this workload.

Software development: The CPU's integer math score of 98,524 and data compression score of 458,817 in PassMark indicate fast compilation and data handling. The 91st CPU percentile ensures quick build times, and the 24 MB L3 cache helps with repeated code access. The system's ECC memory support adds stability for long-running builds.

Student and office work: The CPU's single-thread score of 4,714 in PassMark and 2,132 in Cinebench R23 provide snappy application responsiveness for productivity suites. The integrated Arc Xe-LPG Graphics 64EU can handle basic display output if the GPU is not needed, but the discrete GPU's G2D score of 1,487 ensures smooth 2D acceleration for document work. The 125 W CPU TDP and 575 W GPU TDP mean the system is not power-sipping, but it is appropriate for high-performance tasks.

Upgrade Path and Platform

The Intel Core Ultra 5 245K uses the Intel Socket 1851, which is the platform for the Core Ultra Series 2. The CPU supports DDR5 memory via a dual-channel bus with a bandwidth of 102.4 GB/s, and it has 20 PCIe Gen 5 lanes from the CPU. The GPU uses a PCIe 5.0 x16 interface, which is fully compatible with the CPU's lanes. The motherboard must support Socket 1851 and DDR5 memory; the specific chipset is not listed, but the platform is current as of the CPU's release in late 2024.

The RTX 5090 D has a TDP of 575 W, and the suggested PSU is 950 W, which provides headroom for the CPU's 125 W TDP and other components. The GPU requires a 1x 16-pin power connector, so the PSU must have that connector or an adapter. The dual-slot design and 304 mm length require a case with sufficient clearance. The CPU is multiplier-unlocked, allowing overclocking to boost clocks beyond the 5.20 GHz boost, though the actual headroom is not specified.

For a sensible next upgrade, the CPU can be replaced with a higher-tier Core Ultra Series 2 processor on the same Socket 1851, though no specific models are listed. The GPU can be upgraded to a future GeForce 60-series model, as the successor is listed as "GeForce 60," but that is not yet available. Memory can be expanded from the current dual-channel configuration, and the 24 MB L3 cache is fixed. The platform supports ECC memory, which can be added for workstation use. The 950 W PSU headroom suggests the system can handle a more power-hungry CPU in the future, but the current CPU's 125 W TDP is modest, leaving room for a higher-TDP CPU without a PSU change.

Build Overview

This build pairs the Intel Core Ultra 5 245K, a 14-core desktop CPU with a 91st percentile ranking, with the NVIDIA GeForce RTX 5090 D, a 32 GB VRAM GPU with a 92nd percentile ranking. The combined system percentile is 92, placing it in the top tier of desktop configurations. The CPU's average benchmark score of 54,053 and the GPU's average score of 77,712 confirm that both components are high performers in their respective categories.

The build class is desktop, and the data indicates a balanced, high-end system. The CPU's 91st percentile is nearly matched by the GPU's 92nd percentile, suggesting that this pairing is well-suited for demanding workloads without a significant mismatch. The lack of measured FPS data means that real-world gaming performance must be inferred from the benchmark scores, but the 3DMark Steel Nomad score of 14,326 and the CPU's strong single-core performance suggest a capable gaming machine. The overall tier is high, with the system outperforming the majority of other combinations.

Who Should Build It

This system is targeted at enthusiasts and professionals who require top-tier performance in both CPU and GPU workloads. Gamers at 1440p and 4K resolutions will benefit from the RTX 5090 D's 32 GB VRAM and 1.79 TB/s bandwidth, which can handle high-resolution textures and ray tracing, while the CPU's single-core score of 4,714 in PassMark ensures minimal bottleneck. Content creators, including video editors and 3D renderers, will appreciate the CPU's Cinebench R23 multi-core score of 25,066 and the GPU's 104.8 TFLOPS FP32 compute, which accelerate rendering and export tasks.

Developers compiling large codebases will benefit from the CPU's integer math score of 98,524 and data compression score of 458,817, while the ECC memory support provides stability for long build sessions. Students and office workers in STEM fields may find the system overkill for basic tasks, but it excels in simulation, data analysis, and AI research due to the GPU's 680 tensor cores and the CPU's 91st percentile ranking. Small business workstations handling CAD, video production, or scientific computing will find the combination of 14 CPU cores and 32 GB GPU VRAM sufficient for multi-tasking and large datasets. The system is not for budget-conscious users, but for those who need high performance, the measured scores justify the hardware.