SYSTEM ANALYZER

Rate My PC: AMD Ryzen AI Embedded P185 + Intel Arc B370

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

86 / 100
HIGH-END

Power Build

Top 14% of systems. Excellent for 1440p Ultra or 4K High gaming.

1440p Ultra4K High

System Balance Analysis

CPU vs GPU performance ratio
Well Balanced
CPU
96%
VS
GPU
76%
PROCESSOR

AMD Ryzen AI Embedded P185

62,839 Benchmark Score
Top 4% 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

The AMD Ryzen AI Embedded P185 paired with the Intel Arc B370 is an unusual combination: a near-flagship embedded mobile processor — sitting in the 93rd percentile of all CPUs — matched with an entry-level integrated GPU in the 5th percentile of all GPUs. The CPU benchmark data shows a processor that trades blows with desktop-class Intel parts and AMD's own Ryzen AI 9 PRO 465, while the single available GPU benchmark, a 3DMark Steel Nomad score of 1184, places the graphics side firmly in legacy-entry territory. This is, in effect, a compute-heavy laptop build with a very light graphical load. Note that no measured FPS rows exist for this exact combination in the database — all frame-rate discussion below is estimated from the benchmark scores, not directly measured.

Usage Scenarios

High-refresh gaming: Not a fit. The Arc B370's Steel Nomad result of 1184 sits within a fraction of a percent of GPUs like the ATI Radeon HD 5770 and AMD Radeon HD 7650M, and its 5th-percentile standing among all GPUs rules out modern AAA titles at high frame rates. Estimated from the scores, esports titles at reduced settings are the realistic ceiling, and even then the integrated graphics — with system-shared memory and system-dependent bandwidth — will be the limiter long before the CPU is.

Streaming: The CPU side is more than capable. A passmark multithread score of 31817 on 12 cores and 24 threads comfortably handles encoding alongside gameplay, and the physics score of 1772 indicates strong simulation headroom. The problem is upstream: the GPU cannot render modern games at settings where streaming them would make sense. Streaming lightweight or older titles is feasible; streaming demanding ones is not.

Video editing: A qualified yes. Integer math (117832), floating point math (70587), and data compression (374429) scores all reflect the 93rd-percentile CPU standing, which translates into quick timeline scrubbing, proxy generation, and export throughput for 1080p and moderate 4K workflows. The caveat is the GPU: no dedicated VRAM and system-shared memory bandwidth mean hardware-accelerated effects and timeline playback lean heavily on the CPU's dual-channel DDR5/LPDDR5X subsystem with 89.6 GB/s of bandwidth.

3D rendering: CPU rendering is genuinely strong. The multithread result of 31817 and the extended instructions score of 26544 indicate that CPU-based render engines will exploit all 24 threads effectively, and the 93rd percentile confirms this is upper-tier throughput. GPU-path rendering, however, is off the table for serious work — 6.144 TFLOPS FP32 and a 5th-percentile ranking mean viewport interactivity on complex scenes and GPU render passes will be slow.

Software development: One of the best-fit scenarios in the data. Compilation is a classic multithreaded integer workload, and the integer math score of 117832 combined with the single-thread result of 3977 means both parallel builds and snappy single-threaded toolchain steps — IDE indexing, script execution, test suites — are covered. The find-prime-numbers result of 129 and random string sorting at 40557 reinforce strong general-purpose throughput.

Student and office work: Essentially ideal. Everyday applications are single-thread-bound, and a 3977 single-thread score alongside the 93rd-percentile overall standing guarantees effortless browsing, document work, and multitasking. The 28 W TDP of the CPU and the 25 W TDP of the IGP mean this pairing also suits thin, quiet, long-running portable machines — the build class here is laptop, and the power envelope matches.

FAQ

Q: How strong is the CPU compared to everything else in the database?

A: It sits in the 93rd percentile of all CPUs, with an average benchmark score of 62839 — meaning it outranks roughly the top-tenth boundary of the entire field.

Q: How does it compare to its nearest rivals?

A: Extremely tightly. The Intel Core Ultra 7 255HX is within 0.2 percent (avgScore 62738), the Core i7-13790F is 0.4 percent ahead (63080), the Core Ultra 7 265HX is 0.5 percent ahead (63173), and the AMD Ryzen AI 9 PRO 465 trails by 0.5 percent (62498). Statistically, these five parts are interchangeable in aggregate performance.

Q: Can this build game seriously?

A: No. The Intel Arc B370 ranks in the 5th percentile of all GPUs with a single Steel Nomad score of 1184, placing it alongside legacy parts like the ATI Mobility Radeon HD 5570. No measured FPS data exists for this pairing; estimates from the scores suggest only light or older games are playable.

Q: Are there measured frame rates for this combination?

A: No — the database contains no measuredFps entries for the P185 + B370. Every FPS figure or expectation stated here is an estimate derived from the benchmark scores.

Q: Does the CPU support ECC memory?

A: Yes, ECC memory support is listed as true, which is notable for an embedded/mobile part and relevant for stability-critical deployments.

Q: What is the combined percentile of the build?

A: 49 — almost exactly the median of all pairings, which is the direct arithmetic consequence of a 93rd-percentile CPU dragging a 5th-percentile GPU upward.

Q: Is the multiplier unlocked for overclocking?

A: No. The multiplierUnlocked field is false, so tuning headroom on this embedded part is limited by design.

Benchmark Performance

The CPU's numbers tell a consistent story of high-end mobile and near-desktop throughput. The headline passmark multithread score of 31817, generated across 12 cores and 24 threads, lands the P185 in the 93rd percentile of all CPUs tested. Its average benchmark score of 62839 is the figure used for rival matching, and the closeness of those rivals is striking: the Intel Core Ultra 7 255HX at 62738 (a 0.2 percent gap), the Core i7-13790F at 63080 (0.4 percent), the Core Ultra 7 265HX at 63173 (0.5 percent), and the Ryzen AI 9 PRO 465 at 62498 (0.5 percent). In aggregate terms, the P185 performs like a desktop Intel Core i7-class part.

The sub-scores flesh out the profile. Integer math at 117832 and floating point math at 70587 are both robust, reflecting the Zen 5 / Zen 5c hybrid core layout. Data compression at 374429 is a standout result, directly relevant to archival, packaging, and build-pipeline workloads. Single-thread performance of 3977 — appearing twice in the dataset under duplicate test names — confirms that per-core responsiveness is not sacrificed for density. Physics at 1772 and extended instructions at 26544 round out a balanced set with no weak column.

The GPU side is a single data point: 1184 in 3DMark Steel Nomad (DX12). That score sits in the 5th percentile of all GPUs, and its nearest rivals are instructive — the ATI Mobility Radeon HD 5570 at 1186 (0.2 percent), the ATI Radeon HD 5770 at 1190 (0.5 percent), the AMD Radeon HD 7650M at 1192 (0.7 percent), and the AMD FirePro M2000 at 1168 (1.4 percent). These are legacy-class comparison points, which frames the B370 as an entry-level integrated solution rather than a gaming part. The combined build percentile of 49 is the average of a top-tier CPU and a bottom-tier GPU — a build that is compute-rich and graphics-poor.

Balance and Bottleneck

This is one of the most lopsided pairings in the database. The CPU operates at the 93rd percentile; the GPU at the 5th. In any workload with a graphical output component — games, 3D viewports, GPU-accelerated video effects — the Arc B370 is the bottleneck by an enormous margin. There is no measured FPS data for this combination, but scaling estimated from the scores is unambiguous: the GPU would saturate long before the P185's 24 threads show meaningful utilization in a rendering task. A frame-rate curve for this pairing would be flat regardless of CPU-side settings changes, because the 1184 Steel Nomad score caps the graphical ceiling.

The inverse is true for CPU-bound workloads. Compilation, compression, encryption (19612 in the encryption test), and simulation tasks will fully engage the processor while the IGPU idles. The 28 W CPU TDP and 25 W GPU TDP share a combined envelope suited to portable machines, and under simultaneous CPU + GPU load the system-shared memory design means both components draw from the same 89.6 GB/s dual-channel pool — a further argument that graphics-heavy hybrid workloads are not this build's purpose. The practical takeaway: buy this configuration for the processor, treat the graphics as a display output and fallback.

Upgrade Path and Platform

The platform specifics matter here because this is an embedded mobile part. The AMD Ryzen AI Embedded P185 uses Socket FP8, is built on TSMC's 4 nm process with a 233 mm² die, and carries the Gorgon Point codename. Memory support covers DDR5 and LPDDR5X across a dual-channel bus with 89.6 GB/s of bandwidth, and ECC is supported. PCIe is Gen 4 with 16 CPU lanes — enough for a fast NVMe SSD and a peripheral, but a generational step behind the newest platforms. The CPU also integrates Radeon 890M graphics of its own, which is worth noting: the Arc B370 listed here is itself an IGP (bus interface IGP, slot width IGP, no power connectors), so this pairing exists within a portable-device context rather than a socketed desktop upgrade tree.

Upgrade headroom is therefore constrained by design. No suggestedPsu figure is provided for the GPU, consistent with an integrated solution drawing 25 W with no external power connectors; the combined CPU + GPU power draw of 53 W across both listed TDPs (28 W plus 25 W) means power delivery is a non-issue. Because the multiplier is locked and the socket is FP8, the sensible "upgrade" is not tuning but workload allocation: lean on the P185's 93rd-percentile throughput and, if graphics performance ever becomes a requirement, that requirement points to a different machine class entirely rather than a component swap.

Who Should Build It

Developers and build engineers are the primary audience. The integer math score of 117832, compression result of 374429, and 24-thread multithread throughput of 31817 make this an exceptional compilation and CI-oriented machine in a portable envelope.

Students and general productivity users benefit directly: the 3977 single-thread score guarantees responsive everyday applications, and the low combined power draw suits all-day battery scenarios.

Embedded and industrial system integrators are the nominal target — the "Embedded" designation, ECC support, FP8 socket, and Active production status all point to long-lifecycle deployments where stability and compute density matter more than graphics.

Video editors working at 1080p, or 4K with patience get a capable CPU-driven pipeline, though they should expect GPU-accelerated effects to be limited by the 6.144 TFLOPS FP32 figure and system-shared memory.

Small business workstations handling data processing, encryption (19612), and virtualized or multi-VM workloads will find the 93rd-percentile CPU more than sufficient. Gamers are not the audience — the 5th-percentile GPU makes that clear at any modern resolution, and the estimated FPS picture supports only casual or legacy titles.

CPU Analysis

The Ryzen AI Embedded P185 is a 12-core, 24-thread part combining Zen 5 and Zen 5c cores on TSMC's 4 nm node, with a 2.00 base clock rising to a 5.10 boost. The hybrid layout shows in the benchmark spread: the 3977 single-thread score demonstrates full-height Zen 5 responsiveness, while the 31817 multithread result shows the density-optimized Zen 5c cores contributing meaningfully to aggregate throughput. Cache is generous for the class — 80 KB L1 and 1 MB L2 per core, plus 16 MB of shared L3 — which supports both low-latency single-thread work and sustained all-core loads.

The generation label places this in the Ryzen AI Embedded family, and the performance data validates that positioning: an average score of 62839 within half a percent of the Intel Core Ultra 7 255HX, Core i7-13790F, Core Ultra 7 265HX, and Ryzen AI 9 PRO 465. For real workloads, the numbers translate cleanly. Data compression at 374429 implies fast archival and build-artifact handling; encryption at 19612 suits security-sensitive pipelines; physics at 1772 indicates strong game-engine and simulation logic even though the GPU cannot render the results at high fidelity. The locked multiplier and 28 W TDP confirm this is a fixed-envelope efficiency part, not a tuning target — and at 93rd-percentile performance, it doesn't need to be.

Build Overview

This is a laptop-class pairing (buildClass: laptop) combining a near-flagship embedded mobile CPU with an entry-level integrated GPU. The combined percentile of 49 places the build almost exactly at the median of all tracked combinations — but that median is an artifact of averaging extremes rather than a sign of balance. In CPU terms, this machine competes with desktop Core i7-class hardware; in GPU terms, it competes with decade-old discrete mobile parts. Both components are in Active production status, with the CPU released on 2026-02-28 and the GPU appearing on 2026-01-26, making this a contemporary configuration with no measured FPS entries yet in the database. The honest classification: a mobile compute workstation with display-capable but performance-limited graphics.

GPU Analysis

The Intel Arc B370 is a Panther Lake-based integrated GPU on Intel's 3 nm process, using the Xe3-LPG architecture within the Arc Graphics-M generation. Its specification sheet reads as an efficient entry-level IGP: 1280 shading units, 40 TMUs, 20 ROPs, and 10 RT cores, with a 300 MHz base and 2400 MHz boost clock. Compute output is 6.144 TFLOPS FP32 and 12.29 TFLOPS FP16 at a 2:1 ratio; pixel rate is 48.00 GPixel/s and texture rate 96.00 GTexel/s. Memory is system-shared — size, type, bus width, and bandwidth are all dependent on the host system — meaning real graphical throughput is tied to the 89.6 GB/s dual-channel memory pool of the host platform. API support is modern, spanning DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

The measured reality is the Steel Nomad score of 1184, 5th percentile among all GPUs, bracketed by the ATI Mobility Radeon HD 5570 (1186), ATI Radeon HD 5770 (1190), AMD Radeon HD 7650M (1192), and AMD FirePro M2000 (1168). What this means for rendering: modern game engines at default settings are out of reach, light esports and legacy titles are the playable envelope as estimated from the scores, and the 10 RT cores exist architecturally but cannot deliver meaningful ray-traced performance at this tier. For GPU rendering workloads, the FP16 figure of 12.29 TFLOPS offers some inference and light-compute utility, but the system-shared memory design remains the binding constraint. As a display output and desktop-composition engine paired with a 93rd-percentile CPU, it does its job; as a gaming or rendering engine, the data is unambiguous.