SYSTEM ANALYZER

Rate My PC: Intel Core Ultra 9 386H + NVIDIA GeForce RTX 5090 Mobile

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
93%
VS
GPU
96%
PROCESSOR

Intel Core Ultra 9 386H

43,210 Benchmark Score
Top 7% Market Ranking
View Full Specs →
GRAPHICS CARD

NVIDIA GeForce RTX 5090 Mobile

45,152 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

# CPU Analysis

The Intel Core Ultra 9 386H is a 16-core, 16-thread mobile processor built on Intel's 3 nm Panther Lake architecture, part of the Core Ultra Series 3 generation. It operates with a base clock of 2.10 GHz and a boost clock of 4.90 GHz, drawing a 25 W TDP. This is a notably efficient design for a high-core-count mobile part, and the absence of hyper-threading—16 cores paired with exactly 16 threads—indicates a focus on raw physical core throughput rather than thread duplication. The cache hierarchy is substantial: 192 KB of L1 per core, 2.5 MB of L2 per core, and 18 MB of shared L3 cache. For memory, the chip supports DDR5 and LPDDR5X over a dual-channel bus, delivering 115.2 GB/s of bandwidth. PCIe Gen 5 with 12 CPU lanes provides modern connectivity for storage and accelerators.

Benchmark results place this CPU in the 88th percentile among all processors, a strong showing for a mobile part. The Cinebench R23 multicore score of 20,547 is the headline number, indicating serious multi-threaded capability. The single-core R23 score of 2,071.5 is equally important, showing that the Panther Lake architecture delivers strong per-thread performance despite the power envelope. In Cinebench R20, the multicore score is 12,820 and single-core is 1,809; in the older R15 test, the multicore score is 3,223 and single-core is 303.5. These numbers form a consistent picture: the 386H is a balanced performer that does not sacrifice single-thread responsiveness for multi-core grunt.

The PassMark results add granularity. The multithread score of 35,399 and single-thread score of 4,218 confirm the dual strengths. Data compression at 352,365 and data encryption at 27,150 indicate strong throughput for storage and security workloads. Floating-point math at 108,527 and integer math at 87,284 show a slight bias toward floating-point operations, which is typical for modern x86 cores and beneficial for scientific and rendering tasks. Extended instructions score 29,138, and random string sorting comes in at 42,135, suggesting robust performance in data manipulation and sorting algorithms. The physics score of 3,028 and prime number finding at 341 round out the compute profile.

Comparing to nearest rivals, the 386H sits within a tight cluster. It is 0.3% behind the AMD Ryzen AI Max PRO 385, 0.5% behind the AMD Ryzen AI 9 465, and 0.7% ahead of the Intel Core i9-12900, with a 0.9% lead over the i9-12900KF. These deltas are marginal—less than one percent—which means the 386H is effectively tied with those desktop-class and high-end mobile parts in aggregate benchmark score. For real workloads, this means the 386H will not be the deciding factor in a head-to-head comparison against these rivals; the surrounding platform and GPU will matter more. The average benchmark score of 43,210 reinforces this positioning, placing the chip in the upper tier of mobile processors available today.

# Benchmark Performance

The CPU and GPU in this build each deliver strong individual benchmark results, and the combined picture is one of a high-performance mobile platform. The CPU's average benchmark score is 43,210, placing it in the 88th percentile of all CPUs. The GPU, an NVIDIA GeForce RTX 5090 Mobile, has an average benchmark score of 45,152 and sits in the 84th percentile of all GPUs. The combined percentile for this pairing is 86, indicating that the system as a whole outperforms the vast majority of laptop configurations.

Focusing on the GPU benchmarks, the RTX 5090 Mobile delivers a 3DMark Steel Nomad DX12 score of 5,871, which is a demanding modern rasterization test. In Geekbench compute tests, it scores 201,834 in OpenCL and 198,405 in Vulkan, showing strong cross-API performance. PassMark results include a G3D score of 30,034 and a GPU compute score of 13,401. Legacy DirectX tests show scores of 324 for DX9, 269 for DX11, 183 for DX10, and 138 for DX12; these are less relevant to modern workloads but indicate backward compatibility. The G2D score of 1,057 is modest, reflecting that 2D desktop workloads are not this GPU's primary strength.

The GPU's nearest rivals provide context. The RTX 5090 Mobile is 0.5% behind the AMD Radeon Pro 5500 XT, 0.8% ahead of the NVIDIA GeForce RTX 4070 Ti, 1% behind the Intel Arc A730M, and 1.8% behind the NVIDIA RTX 5880 Ada Generation. These deltas are all within two percent, meaning the RTX 5090 Mobile is performance-equivalent to a range of desktop and mobile GPUs from previous generations. The RTX 4070 Ti comparison is particularly relevant, as that is a well-regarded desktop card; the mobile 5090 essentially matches it in aggregate benchmark terms.

The combined CPU+GPU picture is one of balance. Both components sit in the mid-80s percentile range, and their average scores are within five percent of each other (43,210 vs. 45,152). This suggests that neither component will consistently bottleneck the other in mixed workloads. For gaming, the CPU has enough single-thread performance to feed the GPU, and the GPU has enough raw throughput to handle modern titles at high settings. For productivity, the CPU's multi-thread score and the GPU's compute score complement each other across rendering and simulation tasks.

# Balance and Bottleneck

The data indicates a well-balanced pairing where neither component dominates as a consistent bottleneck. The CPU's 88th percentile and the GPU's 84th percentile are close, and their average scores differ by only about 4.5% (45,152 for GPU vs. 43,210 for CPU). This narrow gap means that in most workloads, the system will scale well with both CPU and GPU utilization rather than hitting a hard wall on one side.

In gaming workloads, the bottleneck typically shifts depending on resolution and settings. The CPU's single-thread score of 4,218 in PassMark and R23 single-core of 2,071.5 are strong enough to drive high frame rates in CPU-bound scenarios, such as esports titles at lower resolutions. The GPU, with its 84th percentile ranking, will become the limiting factor at higher resolutions or with ultra settings, where pixel throughput and memory bandwidth matter more. The RTX 5090 Mobile's 896.0 GB/s memory bandwidth is substantial, which should mitigate bandwidth-related stutters at high resolutions.

For productivity workloads, the balance is more nuanced. The CPU's multithread score of 35,399 in PassMark and Cinebench R23 multicore of 20,547 indicate strong parallel processing capability, but the GPU's compute score of 13,401 in PassMark and 31.80 TFLOPS of FP32 performance suggest that GPU-accelerated tasks like rendering will lean heavily on the graphics processor. In video encoding or 3D rendering with GPU acceleration, the GPU will be the primary engine, while the CPU handles scene management and physics. The CPU's physics score of 3,028 in PassMark is adequate for such tasks, but it is not the strongest component in the system.

The FPS scaling evidence is absent because no measured FPS data exists for this exact combination. However, based on the percentile positions, the system should exhibit balanced scaling: in CPU-light games, the GPU will be the bottleneck; in CPU-heavy simulations or strategy titles, the CPU will take that role. The lack of a significant gap between the two components' percentile rankings (88 vs. 84) suggests that neither will create a severe imbalance in most scenarios. Users should expect a smooth experience across varied workloads, with the understanding that the GPU is marginally the stronger component and will lead in graphics-intensive tasks.

# Who Should Build It

This laptop build targets users who need high-end mobile performance without compromising on either CPU or GPU capability. The 86th combined percentile places it in the upper tier of laptop configurations, making it suitable for professionals and enthusiasts who demand desktop-like performance in a portable form factor.

Gamers at high resolutions and high refresh rates are a primary audience. The RTX 5090 Mobile's 84th GPU percentile and the CPU's strong single-thread performance ensure that modern titles run smoothly at high settings. The 24 GB of GDDR7 memory provides ample headroom for texture-heavy games and future titles that may require more VRAM. Content creators will benefit from the combination of a 16-core CPU for multitasking and a GPU with 31.80 TFLOPS of FP32 performance for accelerated rendering and video encoding. The Cinebench R23 multicore score of 20,547 indicates that video editing timelines and 3D scene assembly will remain responsive.

Software developers will find the CPU's data compression score of 352,365 and integer math score of 87,284 useful for compilation tasks, while the GPU's compute capabilities can accelerate machine learning workloads. Students and small business users in technical fields—engineering, data science, architecture—will appreciate the portability combined with this level of performance. The 25 W CPU TDP and 95 W GPU TDP mean the system can be reasonably powered in a laptop chassis, though the actual battery life will depend on the specific implementation. The lack of a launch MSRP for either component means pricing is not available from the data, but the performance class clearly targets the premium segment.

# Usage Scenarios

High-refresh gaming: With the CPU's single-thread PassMark score of 4,218, the system can drive high frame rates in CPU-bound titles. The GPU's 84th percentile ranking and 3DMark Steel Nomad DX12 score of 5,871 ensure that graphical settings can be pushed high without sacrificing frame times. This combination is suited for 1440p or 4K displays with high refresh rates, assuming the laptop's display supports them.

Streaming: The CPU's 16 cores and 16 threads provide ample headroom for encoding and streaming simultaneously. The PassMark multithread score of 35,399 indicates that background tasks like OBS encoding will not starve the game of CPU resources. The GPU's tensor cores can also assist with AI-based encoding features, though the data does not specify NVENC performance.

Video editing: The Cinebench R23 multicore score of 20,547 accelerates timeline rendering and export. The GPU's 24 GB of VRAM and 896.0 GB/s memory bandwidth handle 4K and 8K footage with multiple effect layers. The Geekbench OpenCL score of 201,834 confirms strong GPU acceleration for effects and color grading.

3D rendering: The RTX 5090 Mobile's FP32 performance of 31.80 TFLOPS and FP16 performance at a 1:1 ratio of 31.80 TFLOPS make it capable for GPU rendering engines. The CPU's extended instructions score of 29,138 supports physics simulations and complex scene calculations. The combined percentile of 86 indicates that this system will handle professional-grade rendering workloads in a mobile form.

Software development: The CPU's integer math score of 87,284 and data encryption score of 27,150 support compilation, code analysis, and secure development workflows. The 18 MB of L3 cache and dual-channel memory bandwidth of 115.2 GB/s reduce latency in large codebases. Developers working on multi-threaded applications will benefit from the 16 physical cores.

Student and office work: The PassMark single-thread score of 4,218 ensures snappy response in office suites and web browsing. The data compression score of 352,365 speeds up file archiving and database tasks. For students in STEM fields, the combination of CPU and GPU performance handles simulation software and data analysis without the need for a separate workstation.

# Upgrade Path and Platform

The Intel Core Ultra 9 386H uses the Intel BGA 2540 socket, which is a soldered mobile platform. This means the CPU is not upgradeable in the traditional sense; users cannot swap the processor for a different model. The same applies to the GPU, which is an integrated graphics processor (IGP) on the board, per the slot width designation. The RTX 5090 Mobile has no power connectors and is a board-integrated component, so GPU upgrades are also not possible. This is typical for laptop builds, where the entire platform is designed around the initial component selection.

Memory support includes DDR5 and LPDDR5X over a dual-channel bus, with a bandwidth of 115.2 GB/s. Depending on the laptop's design, users may be able to upgrade memory if it is socketed rather than soldered. The PCIe Gen 5 interface with 12 CPU lanes provides connectivity for next-generation SSDs, which can be upgraded if the laptop has accessible M.2 slots. The GPU uses a PCIe 5.0 x16 interface, which is separate from the CPU lanes and handled by the chipset.

The CPU's TDP of 25 W and the GPU's TDP of 95 W give a combined thermal envelope of 120 W for the core components. The suggested PSU field is null, so no specific power supply recommendation is available from the data. However, the low TDPs suggest that a laptop power brick in the range of 150-200 W would be adequate, though this is an inference from the TDP values and not a stated fact. A sensible next upgrade for this platform would be increasing memory capacity or adding a high-capacity Gen 5 NVMe SSD, as those are the most likely user-serviceable components. The CPU and GPU are fixed, so the upgrade path is limited to storage and memory, assuming the laptop's design permits access.

# FAQ

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

A: The Intel Core Ultra 9 386H has 16 cores and 16 threads, with no hyper-threading.

Q: How does the CPU compare to the AMD Ryzen AI Max PRO 385?

A: The 386H has an average benchmark score of 43,210, which is 0.3% lower than the AMD Ryzen AI Max PRO 385's score of 43,326.

Q: What is the GPU's memory configuration?

A: The NVIDIA GeForce RTX 5090 Mobile has 24 GB of GDDR7 memory on a 256-bit bus, with a bandwidth of 896.0 GB/s.

Q: Is the RTX 5090 Mobile faster than the RTX 4070 Ti?

A: The RTX 5090 Mobile has an average benchmark score of 45,152, which is 0.8% higher than the RTX 4070 Ti's score of 44,795.

Q: What is the combined performance percentile of this build?

A: The combined percentile for the CPU and GPU pairing is 86, indicating it outperforms 86% of all benchmarked systems.

Q: Does this CPU support ECC memory?

A: No, the Intel Core Ultra 9 386H does not support ECC memory.

Q: What is the GPU's architecture and process node?

A: The RTX 5090 Mobile is based on the Blackwell 2.0 architecture (chip GB203) and is manufactured on a 5 nm process by TSMC.

# Build Overview

This is a laptop build combining the Intel Core Ultra 9 386H CPU with the NVIDIA GeForce RTX 5090 Mobile GPU. The CPU is a 16-core, 16-thread mobile processor from the Panther Lake architecture, and the GPU is a 24 GB GDDR7 mobile graphics processor from the Blackwell 2.0 generation. The system's combined percentile is 86, placing it in the upper tier of laptop configurations. The CPU sits in the 88th percentile of all CPUs, and the GPU sits in the 84th percentile of all GPUs, indicating that both components are high-end relative to the broader market.

The build class is explicitly "laptop," meaning this is a mobile platform with soldered components. The CPU has a 25 W TDP and the GPU has a 95 W TDP, making this a relatively power-efficient high-performance pairing. The average benchmark scores—43,210 for the CPU and 45,152 for the GPU—are closely matched, suggesting a balanced system. This is not a budget configuration; the performance tier, combined with the lack of a launch MSRP for either component, indicates a premium mobile workstation or gaming laptop. The data shows a system that is competitive with desktop parts from the previous generation, such as the Intel Core i9-12900 and NVIDIA GeForce RTX 4070 Ti, but in a portable form factor.

# Gaming Performance

No measured FPS data exists for this exact CPU+GPU combination. The FACT PACK contains no measuredFpsUltraByGame entries, so all frame rate discussions are estimated from the benchmark scores. The data is explicitly marked as not measured, meaning the following analysis is qualitative and based on the components' relative performance.

The RTX 5090 Mobile's 3DMark Steel Nomad DX12 score of 5,871 and PassMark G3D score of 30,034 indicate strong rasterization performance. The 24 GB of GDDR7 memory with 896.0 GB/s bandwidth provides ample headroom for high-resolution textures and modern game assets. The CPU's single-thread PassMark score of 4,218 and Cinebench R23 single-core score of 2,071.5 suggest that frame pacing will be stable in most titles, with the GPU becoming the limiting factor at higher resolutions.

For esports titles at 1080p, the CPU's strong single-thread performance should drive very high frame rates, limited by the GPU's throughput. At 1440p and 4K, the GPU's 84th percentile ranking will be the primary determinant of performance, with the 31.80 TFLOPS FP32 throughput handling modern AAA games at high settings. The GPU's nearest rival, the RTX 4070 Ti, is 0.8% behind in aggregate score, which gives a reference point for expected FPS—roughly equivalent to a desktop RTX 4070 Ti in rasterization. Ray tracing performance is not directly measured in the provided benchmarks, but the 82 RT cores and DirectX 12 Ultimate support indicate capable ray-traced rendering.

# GPU Analysis

The NVIDIA GeForce RTX 5090 Mobile is a high-end mobile GPU based on the Blackwell 2.0 architecture, built on a 5 nm process by TSMC. The chip is designated GB203, with 45,600 million transistors on a 378 mm² die, giving a transistor density of 120.6 million per square millimeter. The GPU has 10,496 shading units, 328 texture mapping units (TMUs), and 112 raster operations pipelines (ROPs). It features 82 RT cores and 328 tensor cores, making it well-equipped for ray tracing and AI-accelerated workloads.

The memory subsystem is a standout feature: 24 GB of GDDR7 on a 256-bit bus, delivering 896.0 GB/s of bandwidth. This is a significant amount of VRAM for a mobile GPU, suitable for large datasets, high-resolution textures, and GPU-compute tasks. The memory clock is 1750 MHz, with an effective speed of 28 Gbps. The base clock is 990 MHz, with a boost clock of 1515 MHz. These clocks are modest compared to desktop parts, reflecting the power constraints of a mobile form factor with a 95 W TDP.

Compute performance is substantial: FP32 throughput is 31.80 TFLOPS, and FP16 is also 31.80 TFLOPS at a 1:1 ratio, indicating no halving for half-precision workloads. The pixel rate is 169.7 GPixel/s and the texture rate is 496.9 GTexel/s. In benchmarks, the GPU scores 5,871 in 3DMark Steel Nomad DX12, 201,834 in Geekbench OpenCL, and 198,405 in Geekbench Vulkan. The PassMark G3D score is 30,034, with a GPU compute score of 13,401. The GPU sits in the 84th percentile of all GPUs, with an average benchmark score of 45,152.

Compared to rivals, the RTX 5090 Mobile is 0.5% behind the AMD Radeon Pro 5500 XT, 0.8% ahead of the NVIDIA GeForce RTX 4070 Ti, 1% behind the Intel Arc A730M, and 1.8% behind the NVIDIA RTX 5880 Ada Generation. These margins are small, meaning the 5090 Mobile is effectively in the same performance class as those GPUs. The RTX 4070 Ti comparison is notable, as that is a desktop card; the fact that a mobile GPU matches it in aggregate benchmarks underscores the efficiency of the Blackwell architecture. The GPU supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, ensuring compatibility with modern APIs. Display outputs are portable-device-dependent, which is standard for a laptop GPU.