SYSTEM ANALYZER

Rate My PC: AMD Ryzen AI Embedded P164 + AMD Radeon 8065S

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

84 / 100
HIGH-END

Power Build

Top 16% of systems. Excellent for 1440p Ultra or 4K High gaming.

1440p Ultra4K High

System Balance Analysis

CPU vs GPU performance ratio
GPU Bottleneck
CPU
95%
VS
GPU
74%

Your GPU is limiting system performance. Consider upgrading to a more powerful graphics card to better utilize your CPU.

PROCESSOR

AMD Ryzen AI Embedded P164

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

AMD Radeon 8065S

0 Benchmark Score
Top 26% 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.

Bottleneck Detected

GPU Bottleneck - Upgrading the weaker component will improve overall performance.

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 P164 paired with the AMD Radeon 8065S is a laptop-class combination built around a shared silicon philosophy: an embedded, efficiency-focused Zen 5 processor and a memory-sharing RDNA 3.5 integrated GPU, both fabricated on TSMC's 4 nm node. The data places this pairing in the 71st percentile of combined builds — a solid mid-tier result driven almost entirely by the CPU, which sits in the 91st percentile versus all CPUs, while the Radeon 8065S lands at exactly the 50th percentile against all GPUs. Note upfront: the FACT PACK contains no measured FPS rows for this exact combination, so all frame-rate discussion below is estimated from benchmark scores rather than measured in-game results.

Upgrade Path and Platform

The platform story here is defined by the Socket FP8. That is a soldered, embedded-oriented mobile socket — not a desktop socket designed for chip swaps — so the traditional upgrade path of dropping in a faster CPU a few years down the line does not apply in the way it does on an AM-series desktop board. What the platform does offer is modern I/O and memory flexibility within a fixed silicon footprint.

Memory support covers both DDR5 and LPDDR5X on a dual-channel bus, delivering 89.6 GB/s of memory bandwidth. That figure matters more than usual because the Radeon 8065S uses system-shared memory — its VRAM, bus width, and bandwidth are all listed as "System Shared" or "System Dependent." In practice, that means memory choice directly shapes GPU performance: faster shared memory translates into graphics headroom, and the 89.6 GB/s ceiling is the hard limit both the CPU and GPU draw from.

PCIe connectivity is Gen 4 with 16 lanes from the CPU. The iGPU itself connects over a PCIe 5.0 x16 bus interface, though as an IGP with no power connectors and a slot width of "IGP," it draws everything from the host package. There is no discrete card to add and no suggested PSU figure to plan around — total thermal design sits at a 28 W CPU TDP and a 55 W GPU TDP, figures consistent with a mobile platform engineered for efficiency rather than socketed expandability. ECC memory is supported, which is a meaningful differentiator for embedded and workstation-adjacent deployments where data integrity matters.

A sensible "upgrade" for this platform, then, is not a CPU swap. It is memory and storage: populating the fastest supported LPDDR5X configuration and using the Gen 4 lanes for high-throughput NVMe storage. Anyone needing more graphics horsepower will be looking at a different machine, not an upgrade to this one.

Balance and Bottleneck

This is an unbalanced pairing by design, and the percentile numbers make that explicit. The CPU operates in the 91st percentile of all processors — upper-tier territory — while the GPU sits at the exact midpoint of the GPU distribution at the 50th percentile. The combined build percentile of 71 reflects that average: strong compute, middling graphics.

The bottleneck is clearly the graphics side, and it is structural. With a shared-memory design, the Radeon 8065S competes with the CPU for the same 89.6 GB/s of bandwidth. Workloads that hammer both subsystems simultaneously — gaming while background tasks run, or GPU compute feeding off CPU-prepared data — will feel that contention. The CPU's multithreaded score of 25889 against a single-thread score of 4029 shows a healthy roughly 6.4x scaling ratio across its 16 threads, indicating the Zen 5 cores are rarely the limiting factor in productivity or creation workloads.

Conversely, in rendering-heavy tasks, the GPU's 15.36 TFLOPS of FP32 throughput and 2560 shading units define the ceiling. The CPU cannot hand off work fast enough to change that; the 8065S will finish when it finishes. The practical takeaway from the percentile spread: this build is compute-rich and graphics-mid, and every workload choice should account for that asymmetry.

Benchmark Performance

The CPU benchmark picture is remarkably strong for a 28 W embedded part. The passmark multithread result of 25889 and average benchmark score of 52901 place it in rare company. The nearestRivals data shows just how tightly clustered the competition is: the AMD Ryzen 5 9500F averages 52873, a delta of just 0.1 percent; the Intel Xeon 634 averages 52974, a delta of -0.1 percent; the AMD EPYC 7313P averages 53206 at -0.6 percent; and the AMD Ryzen 9 7900X averages 53288 at -0.7 percent. In other words, the P164 trades blows within a fraction of a percent of desktop and server silicon — including a 12-thread-class workstation Xeon and a high-core-count desktop Ryzen 9 — while operating at a 28 W TDP.

Individual test results reinforce the all-around strength. Integer math scores 87940, floating point math 55799, data compression 327891, random string sorting 34801, and data encryption 16055. The extended instructions result of 24193 indicates solid throughput in AVX-type workloads. The single-thread score of 4029 is particularly telling: it confirms the 5.00 GHz boost clock delivers genuine per-core snap, not just aggregate throughput from threading. The physics score of 1210 and find-prime-numbers result of 71 round out a profile with no glaring weak spot.

The GPU side has no benchmark scores recorded in the FACT PACK — avgBenchmarkScore is 0 and the benchmarks array is empty — so its 50th percentile placement is the only positioning datum available. The combined picture: a 91st-percentile CPU dragging a 50th-percentile GPU to an overall 71st-percentile build.

Who Should Build It

Given the measured profile, this combination targets several distinct user groups.

Content creators working with CPU-bound pipelines are the clearest fit. The encryption, compression, and integer scores rival desktop and server parts, so code compilation, asset compression, and batch processing all run at near-workstation pace. Developers fall into the same bucket: the multithreaded and single-thread results support both fast parallel builds and snappy IDE responsiveness, and ECC support adds a reliability margin for long-running jobs.

Gamers at lower resolutions are served acceptably but not impressively. A 50th-percentile GPU with system-shared memory is a 1080p-class proposition, with settings likely below "Ultra" in demanding titles — an inference from the benchmark positioning, since no measured FPS data exists for this pairing.

Students and general-purpose users get the best of the deal: 91st-percentile CPU performance means everything from browser-heavy multitasking to local data analysis feels fast, and the 28 W CPU TDP points toward battery-friendly operation in a laptop chassis.

Small business workstations are arguably the intended audience — the "Embedded" branding, ECC support, and FP8 socket all point there. The Xeon 634 comparison at -0.1 percent delta is directly relevant: this platform delivers Xeon-adjacent compute in a laptop power envelope, suitable for edge deployments, industrial controllers, and compact integrity-sensitive systems.

CPU Analysis

The Ryzen AI Embedded P164 is an 8-core, 16-thread processor from the Gorgon Point family, mixing Zen 5 and Zen 5c cores on TSMC's 4 nm process, with a 233 mm² die. Base clock is 2.00 GHz with boosts to 5.00 GHz, and the multiplier is locked — consistent with its embedded positioning and a 28 W TDP that leaves little thermal room for manual tuning anyway.

Cache configuration is generous per core: 80 KB of L1 and 1 MB of L2 per core, with 8 MB of L3 shared across the chip. That per-core L2 sizing favors latency-sensitive work — exactly what the 4029 single-thread score demonstrates. The clock range is wide, and the benchmark data suggests the boost algorithm spends meaningful time near the top of it, because the per-core results compete with unlocked desktop silicon in the rivals list.

What the scores mean for real workloads: the 87940 integer math result maps to everyday application performance and general computing; the 55799 floating point score covers scientific and simulation workloads; the 327891 compression score indicates fast file archiving and data pipeline throughput; and the 24193 extended instructions result suggests strong vectorized performance for media encoding and cryptographic operations. The integrated Radeon 880M handles display duties when the 8065S is not the active graphics device, though in this pairing the 8065S is the graphics solution of record.

The memory subsystem deserves emphasis: dual-channel DDR5/LPDDR5X at 89.6 GB/s, shared with the GPU. For a CPU at the 91st percentile, that bandwidth is adequate but not extravagant — a deliberate trade-off of the shared-memory design.

GPU Analysis

The Radeon 8065S is a Gorgon Halo chip on the RDNA 3.5 architecture, part of the Navi Mobile (RX 8000M) generation, built on TSMC 4 nm with a 308 mm² die. Its predecessor platform designation is Polaris Mobile. The configuration: 2560 shading units, 160 texture mapping units, 64 render output units, and 40 ray tracing cores. Tensor cores are not listed — RDNA 3.5 handles AI workloads through its shader architecture rather than dedicated tensor hardware, at least per the data available here.

Clocks run from a 1295 MHz base to a 3000 MHz boost, with memory clocked at "System Shared" — there is no dedicated VRAM. That is the defining characteristic. Memory size, type, bus width, and bandwidth are all system-dependent, which means GPU performance scales with host memory configuration and is bounded by that 89.6 GB/s figure.

Raw throughput numbers: 15.36 TFLOPS FP32, FP16 at 1:1 for the same 15.36 TFLOPS, a pixel fill rate of 192.0 GPixel/s, and a texture rate of 480.0 GTexel/s. The 40 RT cores enable hardware ray tracing, though at this shader count ray-traced workloads will demand significant upscaling or setting reductions. API support is current — DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6 — and the bus interface is PCIe 5.0 x16. Display outputs are portable-device dependent, as expected for an integrated mobile part, and there are no power connectors; the 55 W TDP is package-delivered.

For rendering work, the picture is a mid-pack GPU: capable of standard rasterization pipelines at moderate resolutions, with acceleration hardware present but modest. The 50th percentile placement says median GPU performance across the entire distribution — which includes everything from entry laptop graphics to flagship discrete cards.

Gaming Performance

To restate clearly: the FACT PACK contains no measured FPS data for this CPU and GPU combination. Every frame-rate statement in this section is an estimate derived from the benchmark scores and percentile placements, not from in-game measurement.

Working from the data available: a GPU at the 50th percentile with 2560 shading units, 15.36 TFLOPS of FP32 throughput, and system-shared memory bounded by 89.6 GB/s of bandwidth describes a 1080p-class gaming experience. At 1080p with Ultra presets, expect playable but not high-refresh performance in modern AAA titles; more demanding engines will require settings reductions or upscaling to hold smooth frame rates. At 1440p, the shared-memory bandwidth becomes the likely limiting factor first, and Ultra settings should be considered out of reach for graphically heavy games. At 4K, the estimate is straightforward: not the intended use case for this silicon.

The CPU will not be the constraint. A 4029 single-thread score in the 91st percentile is more than sufficient to feed a 50th-percentile GPU in virtually any engine, and the physics score of 1210 supports simulation-heavy titles. Esports and competitive titles — lighter rendering loads — are where this pairing should feel most comfortable, with the GPU's boost clock reaching 3000 MHz providing headroom in less demanding scenes. The 40 RT cores technically enable ray tracing via DirectX 12 Ultimate, but the estimate here is cautious: treat RT as a toggle for lighter titles rather than a general-purpose capability.

Build Overview

This is a laptop-class build — the buildClass field confirms it — pairing an embedded Zen 5/Zen 5c processor with an RDNA 3.5 integrated GPU, both on TSMC 4 nm silicon. The combined percentile of 71 positions it in the upper-middle tier of all tracked combinations: meaningfully above the median build, but held back from premium territory by the GPU's median placement.

The CPU is the star. At the 91st percentile, within a fraction of a percent of the Ryzen 5 9500F, Xeon 634, EPYC 7313P, and Ryzen 9 7900X on aggregate benchmark scores, it delivers performance that competes with desktop and server parts at a 28 W TDP. That efficiency-to-performance ratio is the defining virtue of the platform. The Radeon 8065S, at the 50th percentile with 55 W of package power and shared memory, is a competent mid-range graphics solution that neither embarrasses nor boosts the pairing.

The overall tier: a compute-forward mobile platform for users whose workloads prioritize processing over pixel-pushing.

FAQ

Q: Can this CPU compete with desktop processors?

A: The data says yes, closely. Its average benchmark score of 52901 sits within 0.1 percent of the AMD Ryzen 5 9500F (52873), 0.1 percent of the Intel Xeon 634 (52974), 0.6 percent of the AMD EPYC 7313P (53206), and 0.7 percent of the AMD Ryzen 9 7900X (53288) — all while operating at a 28 W TDP.

Q: Does the GPU have its own video memory?

A: No. Memory size, type, bus width, and bandwidth are all listed as "System Shared" or "System Dependent." The GPU draws from the same dual-channel DDR5/LPDDR5X pool as the CPU, capped at 89.6 GB/s of total bandwidth.

Q: Is ray tracing supported?

A: Yes, technically. The Radeon 8065S includes 40 ray tracing cores and supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6. With 2560 shading units, however, heavy ray-traced workloads should be approached with modest expectations.

Q: Are there measured game FPS figures for this combination?

A: No. The FACT PACK contains no measured FPS data for this exact pairing, and dataIsMeasured is false. All frame-rate expectations discussed are estimates derived from benchmark scores and percentile placements.

Q: Can the CPU be overclocked?

A: No. The multiplier is locked, which is consistent with the part's embedded positioning and its 28 W thermal envelope on Socket FP8.

Q: Does the platform support ECC memory?

A: Yes, ECC memory support is listed — a notable feature for workstation, edge, and integrity-sensitive embedded deployments.

Q: What PCIe connectivity is available?

A: The CPU provides PCIe Gen 4 with 16 lanes, while the GPU itself uses a PCIe 5.0 x16 bus interface. There are no external power connectors; the graphics solution is fully integrated.