SYSTEM ANALYZER

Rate My PC: AMD Ryzen AI Embedded P164 + Intel Arc B370

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

85 / 100
HIGH-END

Power Build

Top 15% of systems. Excellent for 1440p Ultra or 4K High gaming.

1440p Ultra4K High

System Balance Analysis

CPU vs GPU performance ratio
Well Balanced
CPU
95%
VS
GPU
76%
PROCESSOR

AMD Ryzen AI Embedded P164

52,901 Benchmark Score
Top 5% Market Ranking
View Full Specs →
GRAPHICS CARD

Intel Arc B370

1,184 Benchmark Score
Top 24% 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

Strong Performance

Excellent for 1440p gaming. Most games will run at high/ultra settings smoothly.

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

# AMD Ryzen AI Embedded P164 + Intel Arc B370

This pairing combines an 8-core, 16-thread AMD Ryzen AI Embedded P164 processor (codenamed Gorgon Point, built on a 4 nm TSMC process) with Intel’s Arc B370 integrated graphics (Xe3-LPG architecture on a 3 nm Intel process) in a laptop-class build. The CPU sits at the 91st percentile among all CPUs, while the GPU ranks at the 5th percentile among all GPUs, creating a striking performance asymmetry that defines this system’s capabilities. This analysis draws exclusively from the provided benchmark data, with no measured FPS rows available for this exact combination.

CPU Analysis

The AMD Ryzen AI Embedded P164 is a mobile-oriented processor with 8 cores and 16 threads, running at a base clock of 2.00 GHz and boosting up to 5.00 GHz. It belongs to the Ryzen AI Embedded generation featuring Zen 5 / Zen 5c cores, manufactured on TSMC’s 4 nm process with a die size of 233 mm². The cache hierarchy includes 80 KB of L1 per core, 1 MB of L2 per core, and 8 MB of L3 shared across the chip. The processor supports dual-channel DDR5 and LPDDR5X memory, with a memory bandwidth of 89.6 GB/s, and includes ECC memory support—a feature that points toward embedded or workstation reliability use cases.

The CPU’s PassMark multithread score of 25,889 places it at the 91st percentile among all CPUs, an impressive standing for a 28-watt TDP part. This efficiency is noteworthy: the data shows a processor that delivers high multithreaded throughput while maintaining a mobile-class power envelope. The single-thread score of 4,029 further confirms strong per-core performance, which is critical for lightly threaded workloads such as legacy applications, database queries, or interactive development tasks. The floating-point math score of 55,799 and integer math score of 87,940 indicate balanced arithmetic capabilities, while the extended instructions score of 24,193 suggests robust SIMD performance for vectorized code.

The nearest rivals in the benchmark database provide context for this CPU’s positioning. The AMD Ryzen 5 9500F scores 52,873 on average, which is 0.1% below the P164’s average benchmark score of 52,901. The Intel Xeon 634 sits 0.1% higher at 52,974, while the AMD EPYC 7313P and AMD Ryzen 9 7900X score 53,206 and 53,288 respectively, representing 0.6% and 0.7% advantages. These deltas are razor-thin—the P164 trades blows with desktop-class server and enthusiast processors despite its embedded laptop orientation. However, the data also shows that the P164’s L3 cache of 8 MB is modest compared to what those rival parts likely offer, yet the benchmark scores remain competitive, suggesting the Zen 5 architecture’s efficiency compensates for cache capacity.

The data compression score of 327,891 is exceptionally high, indicating strong performance in data-centric workloads like file compression, archiving, or database operations. The random string sorting score of 34,801 reinforces this pattern. In contrast, the find prime numbers score of 71 is low, which is typical of integer-heavy, cache-sensitive workloads that expose memory latency. The physics score of 1,210, while not directly comparable across benchmark suites, suggests moderate performance for physics simulations, which are often memory-bound. The data encryption score of 16,055 indicates capable AES and cryptography throughput, relevant for VPN endpoints, secure file storage, or encrypted communication servers.

GPU Analysis

The Intel Arc B370 is an integrated GPU (IGP) based on the Xe3-LPG architecture, part of the Arc Graphics-M generation for Panther Lake processors. It operates with a base clock of 300 MHz and a boost clock of 2400 MHz, featuring 1,280 shading units, 40 texture mapping units, and 20 raster output units. The GPU includes 10 ray tracing cores, positioning it as hardware-capable for DirectX 12 Ultimate (12_2) features, along with OpenGL 4.6 and Vulkan 1.4 API support. The pixel rate is 48.00 GPixel/s, and the texture rate is 96.00 GTexel/s, while FP32 performance reaches 6.144 TFLOPS and FP16 reaches 12.29 TFLOPS via a 2:1 ratio.

The memory configuration is entirely system-shared, meaning the GPU draws from the host’s DDR5 or LPDDR5X memory rather than dedicated VRAM. The memory bus width and bandwidth are listed as “System Shared” and “System Dependent” respectively, which means the GPU’s memory performance scales directly with the CPU’s 89.6 GB/s dual-channel bandwidth. This shared memory arrangement is typical for integrated graphics and introduces a fundamental constraint: the GPU competes with the CPU for memory bandwidth, which can throttle both under heavy concurrent load.

The GPU’s 3DMark Steel Nomad DX12 score of 1,184 places it at the 5th percentile among all GPUs—a very low standing. The nearest rivals are all legacy or entry-level parts: the ATI Mobility Radeon HD 5570 scores 1,186 (0.2% higher), the ATI Radeon HD 5770 scores 1,190 (0.5% higher), the AMD Radeon HD 7650M scores 1,192 (0.7% higher), and the AMD FirePro M2000 scores 1,168 (1.4% lower). This places the Arc B370’s raw DX12 performance in the range of mid-2010s discrete GPUs, which frames expectations for gaming and rendering workloads. The 10 RT cores provide ray tracing capability, but the low overall score suggests the RT performance will be modest at best—likely sufficient for basic effects at low resolutions and settings, not for demanding RT workloads. The 25 W TDP is remarkably low for the feature set, emphasizing energy efficiency over raw throughput.

Benchmark Performance

The CPU’s average benchmark score is 52,901, with a multithread score of 25,889 and a single-thread score of 4,029. These numbers place the processor at the 91st percentile among all CPUs, indicating that it outperforms roughly nine out of ten CPUs in the database. The nearest rival comparisons show deltas of only ±0.7%, meaning the P164 is statistically tied with the AMD Ryzen 5 9500F, Intel Xeon 634, AMD EPYC 7313P, and AMD Ryzen 9 7900X in overall CPU performance. This is remarkable for a 28 W embedded processor—it achieves desktop-class compute in a power envelope suited for thin-and-light laptops or embedded systems.

The GPU’s single benchmark score of 1,184 in 3DMark Steel Nomad DX12 places it at the 5th percentile, a stark contrast to the CPU’s 91st percentile. The combined percentile for this build is 48, reflecting the severe imbalance between the two components. The data shows a system where the CPU is a top-tier performer but the GPU is an entry-level solution, resulting in a combined profile that lands near the midpoint of all builds.

No measured FPS data exists for this exact CPU+GPU combination—the FACT PACK contains no measuredFpsUltraByGame entries. All gaming performance discussion that follows is therefore estimated from the benchmark scores rather than measured directly. The CPU’s strong single-thread and multithread scores suggest it can feed frames to a more capable GPU without bottlenecking, but the GPU’s low percentile means actual frame rates will be constrained by the integrated graphics. The combined picture is a machine that excels at CPU-bound productivity tasks but will struggle with modern 3D gaming, especially at high settings or resolutions beyond 1080p.

FAQ

Q: Is the AMD Ryzen AI Embedded P164 a desktop or laptop processor?

A: The market segment is listed as “Mobile,” and the build class is “laptop,” so this CPU is designed for portable or embedded systems, not desktop towers.

Q: How does the CPU’s performance compare to the AMD Ryzen 9 7900X?

A: The P164’s average benchmark score is 52,901, while the Ryzen 9 7900X scores 53,288—a delta of only 0.7% in the Ryzen 9’s favor. This means the P164 performs within 1% of that high-end desktop part despite its embedded laptop orientation.

Q: Does the Intel Arc B370 have dedicated VRAM?

A: No. The memory size, type, and bus width are all listed as “System Shared,” meaning it uses the host system’s memory. The bandwidth is “System Dependent,” so it relies on the CPU’s 89.6 GB/s dual-channel DDR5/LPDDR5X bandwidth.

Q: What is the GPU’s ray tracing capability?

A: The Arc B370 includes 10 ray tracing cores and supports DirectX 12 Ultimate (12_2), so it has hardware RT acceleration. However, its 3DMark Steel Nomad score of 1,184 (5th percentile) suggests RT performance will be minimal in practice.

Q: What kind of memory does this build support?

A: The CPU supports DDR5 and LPDDR5X in a dual-channel configuration, with ECC memory support available. The maximum memory bandwidth is 89.6 GB/s.

Q: How does the GPU compare to the AMD FirePro M2000?

A: The Arc B370 scores 1,184, which is 1.4% higher than the AMD FirePro M2000’s 1,168. This places the two GPUs at essentially the same performance level.

Q: What is the CPU’s TDP and what does it imply for cooling?

A: The CPU has a 28 W TDP, which is very low for an 8-core, 16-thread processor. This implies a capable air cooler or even passive cooling might suffice, though the exact cooling solution is not specified in the data.

Who Should Build It

This build targets users who prioritize CPU performance above all else and have modest graphics requirements. The 91st percentile CPU score makes it suitable for developers compiling large codebases, data analysts processing large datasets, or researchers running numerical simulations—all of which benefit from the strong multithreaded score of 25,889 and the high data compression score of 327,891. The 5th percentile GPU, however, limits its appeal for gamers and 3D artists who need discrete-class graphics performance.

Content creators working with video editing or 3D rendering will find the CPU capable, but the GPU will bottleneck any GPU-accelerated rendering pipeline. The 6.144 TFLOPS FP32 performance of the Arc B370 is sufficient for basic image processing or light compositing but will struggle with modern render engines that leverage GPU acceleration heavily. Students and office workers will find this build more than adequate for document processing, spreadsheets, web browsing, and light multitasking, given the CPU’s strong single-thread score of 4,029 and the low 28 W TDP that enables quiet, efficient operation.

Small business workstations that run database servers, ERP systems, or virtualization workloads will benefit from the CPU’s ECC memory support and its competitive standing against server-class parts like the Intel Xeon 634 and AMD EPYC 7313P. The 16 threads provide ample concurrency for multiple VMs or containerized services. However, any workload requiring substantial graphics output—CAD, video editing, gaming—should look elsewhere, as the GPU’s 5th percentile ranking will be a severe limitation.

Gaming Performance

Since no measured FPS data exists for this exact combination, the following figures are estimates derived from the benchmark scores. The Arc B370’s 3DMark Steel Nomad DX12 score of 1,184 places it in the same performance tier as the ATI Mobility Radeon HD 5570 and ATI Radeon HD 5770, which were entry-level GPUs from the early 2010s. This suggests that modern AAA games at 1080p ultra settings will likely run at low, often unplayable frame rates—potentially in the 15-25 FPS range for demanding titles, though the exact numbers cannot be confirmed from the data.

At 720p with low settings, the GPU might manage 30-45 FPS in older or less demanding games, but the system-shared memory architecture will cause performance to vary significantly depending on system memory speed and capacity. The CPU’s strong single-thread score of 4,029 ensures that frame pacing will be consistent, but the GPU is the limiting factor. Esports titles like Counter-Strike 2 or Valorant, which are CPU-bound at competitive settings, may run at playable frame rates due to the CPU’s high performance, but the GPU’s low pixel rate of 48.00 GPixel/s and texture rate of 96.00 GTexel/s will cap resolution and quality settings. Ray tracing is effectively a non-starter for gaming—the 10 RT cores are present, but the 5th percentile score indicates they lack the throughput for real-time RT in any modern title.

Balance and Bottleneck

The data reveals an extreme bottleneck: the CPU operates at the 91st percentile while the GPU sits at the 5th percentile, a difference of 86 percentile points. In CPU-bound workloads like data compression (327,891 score), encryption (16,055), and integer math (87,940), the processor will be the performance driver, and the GPU will remain largely idle. For gaming, however, the GPU is the clear bottleneck—the CPU’s 4,029 single-thread score can easily feed frames to the Arc B370, but the GPU’s 1,184 3DMark score will cap frame rates far below what the CPU could support with a discrete GPU.

The system-shared memory architecture compounds this imbalance. The GPU’s bandwidth is listed as “System Dependent,” meaning it shares the CPU’s 89.6 GB/s dual-channel bandwidth. In memory-intensive workloads where both CPU and GPU are active—such as gaming with background tasks or GPU-accelerated compute—the two components will compete for the same memory bandwidth, likely reducing both CPU and GPU performance. This is evident from the GPU’s low memory bandwidth rating and the CPU’s modest L3 cache of 8 MB, which may increase memory traffic. The combined percentile of 48 reflects this tension: the system’s overall capability is dragged down by the GPU, even though the CPU alone is exceptional.

Build Overview

This is a laptop-class build pairing an AMD Ryzen AI Embedded P164 processor with Intel Arc B370 integrated graphics. The CPU is an 8-core, 16-thread part with a 28 W TDP, socketed in AMD Socket FP8, and built on a 4 nm TSMC process. The GPU is an integrated solution on Intel’s 3 nm process with a 25 W TDP, using system-shared memory. The combined percentile of 48 places this system near the midpoint of all builds in the database, but this masks the extreme variance: the CPU is in the top 9% of all processors, while the GPU is in the bottom 5% of all GPUs.

The CPU’s average benchmark score of 52,901 puts it in the same league as desktop and server parts like the AMD Ryzen 5 9500F (52,873), Intel Xeon 634 (52,974), AMD EPYC 7313P (53,206), and AMD Ryzen 9 7900X (53,288)—all within 0.7% of each other. The GPU’s 1,184 score, however, places it alongside decade-old discrete GPUs like the ATI Radeon HD 5770 (1,190) and AMD Radeon HD 7650M (1,192). This is a compute-first machine with graphics as an afterthought, suited for productivity and embedded applications rather than gaming or graphics-heavy work.

Usage Scenarios

High-refresh gaming: Not viable. The GPU’s 5th percentile score (1,184 in 3DMark Steel Nomad) cannot drive high frame rates at any resolution above minimal settings. The CPU’s 4,029 single-thread score would support high-refresh gaming with a discrete GPU, but the integrated Arc B370 caps performance at entry-level levels.

Streaming: The CPU can handle encoding and streaming workloads thanks to its 16 threads and strong multithread score of 25,889, but the GPU’s low performance means game capture will be limited to 720p or low-quality 1080p. The 12.29 TFLOPS FP16 performance of the GPU may assist with some encode tasks, but the overall graphics capability is insufficient for modern streaming setups.

Video editing: The CPU’s 55,799 floating-point math score and 87,940 integer math score provide solid performance for timeline editing and codec processing, but the GPU’s 6.144 TFLOPS FP32 performance will limit GPU-accelerated effects and rendering. Simple edits and exports will work; complex multi-layer projects will stall on the GPU.

3D rendering: CPU-based rendering will benefit from the 16 threads and competitive standing against the AMD EPYC 7313P (0.6% difference), but GPU-based renderers will hit a hard wall. The Arc B370’s 1,184 score is insufficient for Blender Cycles, V-Ray GPU, or similar GPU-accelerated render engines.

Software development: Excellent. The CPU’s 91st percentile ranking, strong single-thread score of 4,029, and 327,891 data compression score make it ideal for compilation, testing, and data processing. The 16 threads handle parallel builds efficiently, and the 28 W TDP enables quiet, cool operation in developer laptops.

Student and office work: More than sufficient. The CPU’s 4,029 single-thread score handles document processing, spreadsheets, and web applications with ease. The low TDP of 28 W extends battery life, and the integrated GPU is adequate for basic display output and video playback, though the 5th percentile ranking means even light photo editing will feel sluggish.

Upgrade Path and Platform

The CPU uses AMD Socket FP8, a mobile-specific socket that generally does not offer user-upgradeable processors. The platform supports DDR5 and LPDDR5X memory in dual-channel configuration with a maximum bandwidth of 89.6 GB/s, and ECC memory is supported. The CPU provides PCIe Gen 4 with 16 lanes (CPU only), which offers expansion potential for NVMe storage or external GPU enclosures, though the laptop form factor limits physical expansion options.

The GPU has no upgrade path—it is an integrated part of the Panther Lake chip, with no power connectors and a 25 W TDP that is soldered into the system. The suggested PSU is not specified in the data, but the combined CPU TDP of 28 W and GPU TDP of 25 W totals 53 W, implying a modest power supply is sufficient. A sensible next upgrade for this build would be an external GPU (eGPU) via the PCIe Gen 4 lanes, which would allow the CPU’s 91st percentile performance to be fully utilized in graphics workloads. Alternatively, upgrading to a system with a discrete GPU while keeping the same CPU would resolve the bottleneck, as the CPU is clearly capable of feeding a much faster graphics solution. The memory system, with its 89.6 GB/s bandwidth and dual-channel support, should not require immediate upgrading, though adding more LPDDR5X capacity could benefit memory-heavy workloads given the shared memory architecture.