SYSTEM ANALYZER

Rate My PC: Intel Core i9-13900H + Intel Arc A370M

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

91 / 100
ULTIMATE READY

Apex Performer

Top 9% 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
88%
VS
GPU
93%
PROCESSOR

Intel Core i9-13900H

28,886 Benchmark Score
Top 12% Market Ranking
View Full Specs →
GRAPHICS CARD

Intel Arc A370M

29,175 Benchmark Score
Top 7% 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 i9-13900H paired with the Intel Arc A370M is an all-Intel mobile platform: a 14-core Raptor Lake-H processor sitting in the 81st percentile of all CPUs, matched with a compact Alchemist discrete GPU in the 74th percentile of all GPUs. The combined percentile of 78 places this laptop-class pairing firmly in upper-mid territory of the database — a CPU-forward configuration where the processor clearly outranks the graphics solution. Notably, the FACT PACK contains no measured FPS data for this exact combination, so all frame-rate discussion below is estimated from the benchmark scores rather than drawn from direct game testing. The i9-13900H carries a launch MSRP of $617.

CPU Analysis

The Core i9-13900H is a 14-core, 20-thread mobile processor built on Intel's 10 nm process, with a 257 mm² die fabricated in-house by Intel. It is a Raptor Lake-H design, which means a hybrid arrangement of performance and efficiency cores feeding twenty threads, with a base clock of 2.60 GHz and a boost clock reaching 5.40 GHz. That 5.40 GHz figure is the headline spec, and the single-thread benchmarks confirm it translates into real responsiveness: a Cinebench R23 single-core score of 1964.5 and a PassMark single-thread score of 3754 put the chip's per-core throughput comfortably among strong performers, reflected in its 81st percentile standing against all CPUs in the database.

The cache hierarchy tells a similar story of a premium mobile part. Each core carries 80 KB of L1 and 2 MB of L2, while a shared 24 MB block of L3 serves the whole die. For real workloads, that shared L3 matters most in content creation and compilation tasks where working sets repeatedly cross cores.

The multi-threaded scaling is where the data gets interesting. In Cinebench R23 the chip scores 17471 multi-core against 1964.5 single-core — roughly nine times the single-core figure from fourteen cores and twenty threads. The 3DMark CPU profile makes the scaling curve explicit: 1078 single-thread, 2002 at two threads, 3591 at four, 5395 at eight, 7063 at sixteen, and 7670 at maximum threads. Two things stand out. First, scaling from one to four threads is nearly linear, which is excellent for games and lightly threaded applications. Second, the curve flattens appreciably beyond eight threads — the jump from sixteen to max threads adds barely more than 600 points. This is typical behavior for a 45 W-class mobile part where power and thermal budgets cap all-core sustained clocks, and it implies that heavily threaded render jobs will complete quickly but not at the rate the raw thread count suggests.

Against its nearest rivals, the picture is remarkably tight. The i9-13900H's average benchmark score of 28886 sits within a fraction of a percent of the Intel Core i5-12600H (28882, deltaPct 0), the AMD Ryzen 5 5600XT (28940, deltaPct -0.2), the Intel Core i7-12650H (28815, deltaPct 0.2), and the Intel Core Ultra 5 135H (29093, deltaPct -0.7). That cluster is worth dwelling on. The i9 branding suggests a clear lead, yet the aggregate data shows this chip performing essentially dead-even with a previous-generation Core i5 mobile part and a mid-range desktop Ryzen 5. What the average obscures is where the i9-13900H earns its keep: the 5.40 GHz boost and the 1964.5 R23 single-core score likely represent the strongest single-thread result in that rival set, while the aggregate is dragged toward parity by the mobile power envelope. For buyers, the implication is that this chip is best understood as an excellent single-thread performer with good — not exceptional — multi-core throughput.

The PassMark subtests flesh out the workload profile. Integer math scores 97159 and floating point 69611; data compression reaches 326425 while encryption manages 18974; extended instructions score 19330, random string sorting 35637, physics 1745, and finding prime numbers a modest 104. The compression and integer figures indicate snappy file archiving and general system responsiveness, while the physics score is the relevant one for game simulation load — at 1745 it sits consistently with the chip's overall tier.

The platform details round out a modern mobile foundation: support for both DDR4 and DDR5 on a dual-channel bus, PCIe Gen 5 with 8 CPU-attached lanes, Iris Xe Graphics 96EU integrated graphics, and a BGA 1744 socket — meaning the chip is soldered and non-replaceable. The multiplier is locked, so no overclocking; the 45 W TDP defines the thermal class of the laptop chassis it lives in.

FAQ

Q: How many cores and threads does the Core i9-13900H have?

A: 14 cores and 20 threads, arranged in Intel's hybrid Raptor Lake-H architecture with a shared 24 MB L3 cache.

Q: How does this CPU compare to its rivals?

A: Its average benchmark score of 28886 is statistically tied with the Intel Core i5-12600H (28882), AMD Ryzen 5 5600XT (28940), Intel Core i7-12650H (28815), and Intel Core Ultra 5 135H (29093) — all within a percent of each other.

Q: Is the Arc A370M a strong gaming GPU?

A: It sits in the 74th percentile of all GPUs with an average score of 29175, essentially level with the AMD Radeon RX Vega M GH and Radeon RX 470 in the database. It is a capable entry-level discrete mobile GPU, not a high-refresh solution at high settings.

Q: How much VRAM does the Arc A370M have, and why does it matter?

A: 4 GB of GDDR6 on a 64-bit bus, delivering 112.0 GB/s of bandwidth. That capacity is the practical ceiling for texture quality in modern games and the hard limit for large GPU compute workloads.

Q: Are there measured game frame rates for this exact pairing?

A: No. The database contains no measured FPS rows for the i9-13900H combined with the Arc A370M; all frame-rate discussion here is estimated from the benchmark scores.

Q: Can this CPU be overclocked or upgraded later?

A: No on both counts. The multiplier is locked and the chip uses the Intel BGA 1744 socket, meaning it is soldered to the motherboard.

Q: What memory does the platform support?

A: Both DDR4 and DDR5, dual-channel, per the CPU's memory support specification.

Benchmark Performance

The combined picture starts with the CPU numbers. Cinebench across three generations shows consistent placement: R15 multi 2666.5 / single 279; R20 multi 9676 / single 1365; R23 multi 17471 / single 1964.5. The single-to-multi ratio holds near nine to one in R23, confirming sustained all-core output rather than burst-only performance. PassMark multithread lands at 27673 with the single-thread figure at 3754, and the chip's average benchmark score of 28886 across the suite places it in the 81st percentile of all CPUs — a high placement that reflects a strong upper-mid mobile performer.

The GPU side is thinner on test coverage but coherent. Geekbench OpenCL returns 29676 and Geekbench Vulkan 28673, for an average of 29175 and the 74th percentile against all GPUs. The rival set frames that number precisely: the AMD Radeon RX Vega M GH scores 29197 (deltaPct -0.1), the AMD FirePro W8000 29211 (deltaPct -0.1), the AMD Radeon RX 470 28996 (deltaPct 0.6), and — most tellingly — the AMD Radeon RX 6800M 28874 (deltaPct 1). That last comparison deserves scrutiny. The RX 6800M is a much larger discrete mobile GPU on paper, yet in these compute-oriented Geekbench tests the Arc A370M edges it by a percent. The lesson is that Geekbench OpenCL and Vulkan reward compute throughput and API efficiency, not gaming rasterization; the A370M's 4.198 TFLOPS FP32 figure and modern Xe-HPG architecture punch above their gaming weight in compute contexts.

The combined percentile of 78 for the pairing sits between the CPU's 81 and the GPU's 74, which is the arithmetic one would expect from a balanced-but-CPU-leaning combination. Neither component is remotely a bottleneck to the other in desktop productivity; the question is only whether the GPU can keep up in 3D workloads — and there the answer is that it defines the ceiling.

Gaming Performance

To restate clearly: no measured FPS data exists in the database for this exact CPU+GPU combination, so every frame-rate statement in this section is an estimate derived from the benchmark scores, not a recorded result.

What do the scores permit us to infer? The Arc A370M's average of 29175 and 74th percentile, benchmarked essentially even with the Radeon RX 470, place it in entry-level discrete territory. With 1024 shading units, 64 TMUs, 32 ROPs, a 1550 MHz base and 2050 MHz boost clock, and 4 GB of GDDR6 on a 64-bit bus, the realistic expectation is playable frame rates at 1080p in esports and older or lighter titles, with settings reductions necessary in modern AAA games — and the 4 GB VRAM buffer becoming the binding constraint at high texture quality before raw shader throughput does. Ray tracing, supported by the 8 RT cores and DirectX 12 Ultimate (12_2) compatibility, is architecturally present but impractical to enable at this performance tier; the estimated cost in frames would be severe.

The CPU, by contrast, is overqualified for the GPU in gaming terms. The 3DMark single-thread score of 1078 and two-thread score of 2002 indicate the processor can feed game logic comfortably; any frame-rate ceiling in games will come from the A370M. At 1440p and 4K, where the GPU load scales up, the estimated frame rates would drop toward unplayable at ultra settings, and resolution scaling or upscaling would be advisable. In short: estimated as a 1080p, medium-to-high settings machine for most titles, with high-refresh esports plausible at reduced settings.

Balance and Bottleneck

The bottleneck analysis is unambiguous. With the CPU at the 81st percentile and the GPU at the 74th, the pairing is imbalanced in favor of the processor — and in 3D workloads, that imbalance is decisive. The 3DMark CPU scaling data shows the i9-13900H scoring 5395 at eight threads, more than enough headroom for any game engine's main and worker threads, while the A370M's 112.0 GB/s memory bandwidth and 4 GB VRAM cap texture and asset throughput well before the CPU breaks a sweat.

The evidence cuts the other way in compute and creation workloads. Cinebench R23 multi at 17471 and PassMark multithread at 27673 show the CPU is the workhorse for rendering, encoding, and compilation, with the GPU contributing little beyond what its OpenCL score of 29676 permits in accelerable tasks. The A370M's FP16 throughput of 8.397 TFLOPS (2:1) gives it some utility in inference-adjacent and media workloads, but the 4 GB buffer limits model and asset sizes.

One curious nuance: the GPU's compute scores being tied with the RX 6800M in Geekbench while its gaming-class hardware is far smaller suggests driver and API efficiency strengths in the Xe-HPG architecture. In Vulkan 1.4-supporting titles, estimated performance may exceed what raw specs imply; in older DirectX paths, possibly the reverse. Without measured FPS rows, these remain hypotheses the data invites but cannot confirm.

Upgrade Path and Platform

On a laptop, the traditional upgrade path largely does not exist, and the specs confirm it. The BGA 1744 socket means the CPU is soldered; the Arc A370M is an integrated mobile GPU with IGP slot classification and PCIe 4.0 x8 as its bus interface — also not user-replaceable. The locked multiplier forecloses overclocking.

What remains upgradeable is memory and storage. Dual-channel DDR4 or DDR5 support means populating both channels is the single most impactful configuration decision, particularly for the Iris Xe Graphics 96EU iGPU, which shares system memory bandwidth. PCIe Gen 5 with 8 CPU lanes supports a fast primary SSD, with additional lanes presumably chipset-provided in a given laptop implementation.

Power is defined by the component TDPs: 45 W for the CPU and 35 W for the GPU, a combined 80 W before platform overhead — well within any chassis designed for this pairing, with no suggested PSU figure required for a laptop. A sensible "next upgrade" is therefore a whole-platform move rather than a part swap: the rival data already names the Intel Core Ultra 5 135H (deltaPct -0.7, i.e., marginally faster on average) as the successor-era equivalent, suggesting that generational jumps within this class are small and a meaningful step up requires moving to a higher GPU tier.

Usage Scenarios

High-refresh gaming: Estimated, not measured — but the A370M's 74th percentile and 4 GB VRAM argue against high-refresh AAA gaming. Esports titles at 1080p with reduced settings are the plausible fit, and even then the GPU is the ceiling.

Streaming: The CPU's 20 threads and 17471 R23 multi-core score can handle encoding alongside gameplay, but since the GPU is already the gaming bottleneck, dedicating encode load to the CPU is the only viable arrangement — and the combined load may exceed what a 45 W part sustains. Casual streaming is realistic; simultaneous high-quality streaming and gaming is not.

Video editing: Well matched for CPU-bound editing. The 17471 R23 multi score, 326425 compression result, and DDR5-capable dual-channel memory support smooth timeline scrubbing and proxy workflows. GPU-accelerated effects are constrained by the 4 GB buffer and 112.0 GB/s bandwidth.

3D rendering: Viable as a CPU render node in miniature — Cinebench R23 multi of 17471 places it in solid mobile territory. GPU rendering is limited by 4 GB VRAM despite OpenCL competence (29676).

Software development: A standout scenario for this build. The 3754 PassMark single-thread score makes IDEs and compilers feel fast, while 27673 multithreaded handles parallel builds, container workloads, and test suites with room to spare.

Student and office work: Overkill in the best sense. The single-core figures deliver instant application launches and responsiveness, the 24 MB L3 keeps multi-app multitasking fluid, and the Iris Xe 96EU iGPU handles displays without engaging the discrete GPU.

Build Overview

This is a laptop-class build (buildClass: laptop) pairing a mobile flagship-branded CPU with an entry-tier discrete mobile GPU. The i9-13900H contributes an 81st-percentile CPU profile — strong single-thread, good multi-thread, modern platform features including PCIe Gen 5 and DDR5 support. The Arc A370M contributes a 74th-percentile GPU profile with modern APIs (DirectX 12 Ultimate, Vulkan 1.4, OpenGL 4.6) but modest rasterization resources. The combined 78th percentile positions the machine as an upper-mid laptop overall: a productivity and creation powerhouse relative to its class, with gaming ability that trails its processing muscle. The CPU remains in active production; the GPU is flagged end-of-life, which colors long-term driver expectations even though Intel lists Vulkan 1.4 support.

Who Should Build It

Since this is a BGA laptop platform, "building it" means choosing a notebook that implements this pairing. The measured data suits it to specific users. Developers and computer science students are the clearest fit: the 3754 single-thread and 27673 multi-thread PassMark results, plus 24 MB of shared L3, cover everything from light scripting to parallel compilation. Content creators doing 1080p video work benefit from the 17471 Cinebench R23 multi score and the 326425 compression figure, provided their projects fit within the GPU's 4 GB when acceleration is needed. Gamers should calibrate expectations: this is a 1080p, settings-adjusted machine by estimate, not a 1440p ultra rig — the 74th-percentile GPU and its RX 470-level benchmark company say as much. Small business workstations doing office productivity, light data work (the 18974 encryption and 97159 integer scores are relevant here), and presentations will find the CPU dramatically more than sufficient. Users in 3D-heavy professions — architecture visualization, GPU rendering farms, high-refresh AAA gaming — should look elsewhere; the data does not support those workloads at this tier.

GPU Analysis

The Arc A370M is Intel's Alchemist-generation (Arc 3 Mobile) discrete GPU, built on TSMC's 6 nm process as the DG2-128 chip. The physical stats are compact: 157 mm² of silicon carrying 7,200 million transistors at a density of 45.9M per mm². The render backend comprises 1024 shading units, 64 texture mapping units, and 32 render output units, with 8 RT cores for hardware ray tracing and no dedicated tensor cores listed. Clocks run from a 1550 MHz base to a 2050 MHz boost, with memory at 1750 MHz (14 Gbps effective).

The throughput figures define the chip's character: 4.198 TFLOPS FP32, 8.397 TFLOPS FP16 at a 2:1 ratio, a pixel rate of 65.60 GPixel/s, and a texture rate of 131.2 GTexel/s. The memory subsystem is the tightest constraint — 4 GB of GDDR6 across a 64-bit bus for 112.0 GB/s of bandwidth. For rendering workloads, that bandwidth figure means texture-heavy scenes saturate quickly, and the 4 GB capacity sets a hard wall on asset size in both games and GPU compute.

Yet the benchmark scores complicate a purely spec-based dismissal. Geekbench OpenCL at 29676 and Vulkan at 28673 tie the chip with the Radeon RX Vega M GH (29197, deltaPct -0.1), the FirePro W8000 (29211, deltaPct -0.1), and — remarkably — the Radeon RX 6800M (28874, deltaPct 1). For compute kernels and modern-API workloads, the A370M delivers far beyond its rasterization class, suggesting Xe-HPG's architectural efficiency is real even in its smallest mobile form. In this pairing, that makes the GPU a modest gaming component but a surprisingly credible light-compute accelerator — one whose 35 W TDP and end-of-life status are the final practical considerations. As a partner to the 81st-percentile i9-13900H, it is the defining limit of the build's 3D performance and a quiet overachiever in everything else.