SYSTEM ANALYZER

Rate My PC: Intel Core i9-14900HX + 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

Intel Core i9-14900HX

56,004 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

The Intel Core i9-14900HX paired with Intel Arc B370 is an unusual entry in the benchmark database: a laptop-class flagship CPU bolted to one of the weakest GPUs currently indexed, an integrated Xe3-LPG graphics solution sharing system memory. The data shows a processor sitting in the 91st percentile against all CPUs — 24 cores, 32 threads, a 5.80 GHz boost clock — while the Arc B370 languishes in the 5th percentile of all GPUs. That gap defines this build. It is a compute-heavy machine with a graphics pipeline that cannot keep pace, and every score in the fact pack reinforces that reading. Note that no measured FPS rows exist for this exact combination, so all frame-rate discussion below is estimated from the benchmark scores rather than drawn from measured game data.

CPU Analysis

The i9-14900HX is a Raptor Lake-HX Refresh part, fabricated on Intel's 10 nm node, with a die size of 257 mm² and a part number of SRMXF. It is a BGA 1964 socket mobile chip, which means it is soldered to its motherboard — this is not a desktop CPU wearing a laptop badge, but a genuine mobile flagship. The core topology is 24 cores and 32 threads, a hybrid arrangement in the Raptor Lake family, with 2 MB of L2 cache per core and a shared 36 MB L3 pool. The base clock is 2.20 GHz and the boost clock reaches 5.80 GHz, with an unlocked multiplier — an unusually enthusiast-friendly feature for a mobile part, even if thermal and power constraints in a laptop chassis will govern what that headroom is worth in practice.

The benchmark scores tell a clear story about workload behavior. In Cinebench R23, the chip posts 30055 in multi-core and 2181.5 in single-core — a ratio of roughly thirteen to one, which indicates that heavily threaded software scales well across the full core count. The R15 and R20 results follow the same pattern: 4574 multi / 310.5 single and 15616 multi / 2204 single respectively. Geekbench lands at 17511 multi-core and 2330 single-core, and PassMark reports 43778 on its multithread test against 4175 single-threaded. In every suite, the multi-core figure towers over the single-core one, meaning this CPU delivers its strongest results in rendering, compilation, encoding, and other parallel workloads rather than in lightly threaded tasks.

Single-thread performance is still strong in absolute terms. A Geekbench single-core score of 2330 and PassMark single-thread of 4175 place the chip comfortably in the 91st percentile of all CPUs, which means everyday application responsiveness — browser tabs, office suites, IDEs — is excellent even before parallelism comes into play. The PassMark subtests add texture: 157375 in integer math, 112273 in floating point math, 539965 in data compression, 32619 in encryption, 31247 in extended instructions, 60861 in string sorting, and 193 in prime number search. The compression and encryption numbers in particular point to strong file-archiving and disk-encryption throughput. A physics score of 2638 suggests solid simulation performance for game physics and engine workloads on the CPU side.

Benchmark Performance

The combined picture is stark. The CPU's average benchmark score is 56004, placing it in the 91st percentile versus all CPUs; the Arc B370's average score is 1184 from its single indexed result — 1184 in 3DMark Steel Nomad DX12 — placing it in the 5th percentile of all GPUs. The combined build percentile is 48, which is the mathematical average of two very different realities: near-flagship compute and near-bottom graphics.

Against its nearest rivals, the i9-14900HX is in a dead heat. The Intel Core 7 253PQE averages 55919, just 0.2 percent behind. The AMD Ryzen AI Max 390 averages 56273, 0.5 percent ahead; the AMD Ryzen AI 9 HX PRO 470 averages 56306, also 0.5 percent ahead; and the AMD Ryzen Threadripper PRO 3955WX averages 56555, 1 percent ahead. These are razor-thin margins. Notably, matching a Threadripper PRO workstation chip within 1 percent on average score — in a mobile package — underlines how much raw throughput this CPU carries for its class.

The GPU's rival set tells the opposite story. The Arc B370 sits against the ATI Mobility Radeon HD 5570 (1186, 0.2 percent behind the Arc), ATI Radeon HD 5770 (1190, 0.5 percent ahead), AMD Radeon HD 7650M (1192, 0.7 percent ahead), and AMD FirePro M2000 (1168, 1.4 percent behind). When the nearest comparable parts are mobility GPUs from over a decade ago, the 5th percentile placement is not an anomaly — it is the honest position of the hardware. The Arc B370 is a 3 nm Panther Lake integrated design with 1280 shading units, 40 TMUs, 20 ROPs, and 10 RT cores, producing 6.144 TFLOPS of FP32 compute, a 48.00 GPixel/s pixel rate, and a 96.00 GTexel/s texture rate. Those are entry-level figures by any modern standard.

Balance and Bottleneck

This is one of the most lopsided CPU-GPU pairings in the database. The CPU occupies the 91st percentile; the GPU occupies the 5th. In any workload that renders frames on the GPU, the Arc B370 is the bottleneck by an enormous margin — there is no resolution, no setting, and no scenario in which the i9-14900HX's compute capacity is the limiting factor for gaming. The combined percentile of 48 hides the imbalance because it averages the two components; the real-world experience will be almost entirely dictated by the weaker half.

The evidence for this is structural as well as numerical. The Arc B370 has "System Shared" memory — its bandwidth and capacity are dependent on system RAM via a dual-channel memory bus. That means GPU performance is hostage to system memory configuration, which is a second soft bottleneck: the same shared-bandwidth pool that feeds the CPU's DDR4/DDR5 support also feeds the graphics pipeline. Meanwhile, the CPU's 24-core throughput goes almost entirely unused in games, where frame generation is GPU-bound. In CPU-bound tasks — and only in those — this build performs like a near-top-tier machine. The 25 W TDP of the integrated GPU versus the 55 W TDP of the CPU further sketches the power budget split: this chassis was designed around compute, not graphics.

Gaming Performance

No measured FPS data exists for this exact combination — the fact pack contains no measuredFps rows, and dataIsMeasured is false. All frame-rate figures in this section are therefore estimates derived from the benchmark scores, not measurements.

Working from a Steel Nomad DX12 score of 1184 and the 5th percentile GPU placement, the expectations are modest. At 1080p with low-to-medium settings, lightweight and older titles should be playable; modern AAA games at ultra settings are not a realistic target for this graphics solution. The presence of 10 RT cores and DirectX 12 Ultimate support means ray-tracing features are technically available, but with 1280 shading units and system-shared memory, enabling them would push frame rates well below playable estimates in demanding engines.

The CPU side contributes nothing to the ceiling here. With a 2638 PassMark physics score and top-decile multi-core throughput, the processor could feed far stronger GPUs hundreds of frames per second; the Arc B370 simply cannot consume them. Scaling across resolutions will also be unusually flat in relative terms — because the GPU is the constraint at every resolution, dropping from 1080p to 720p yields less benefit than it would on a balanced system, as shared-memory bandwidth saturates. Esports-class and retro titles are the realistic gaming diet; anything chasing visual fidelity should be considered out of scope.

Usage Scenarios

High-refresh gaming: Not viable. The GPU's 5th percentile placement and 1184 Steel Nomad score put high-refresh, high-fidelity gaming out of reach. Lightweight esports titles at reduced settings are the only path to smooth frame rates, and even then expectations should stay conservative.

Streaming: Partially viable. CPU encoding via the 24-core processor is comfortably supported — a 30055 Cinebench R23 multi-core score means x264-style encoding loads leave headroom. But if the stream includes game rendering, the GPU bottleneck dominates. Streaming desktop content, video playback, or creative work is fine; streaming demanding games is not.

Video editing: Strong on the CPU side. Rendering timelines, scrubbing proxy footage, and exporting benefit from 24 cores and 32 threads, and the 539965 data-compression score indicates fast archive and export throughput. Final-quality playback and GPU-accelerated effects are limited by the Arc B370's shared-memory design and modest compute.

3D rendering: Excellent for CPU renderers. Cinebench R23's 30055 and R20's 15616 are workstation-adjacent figures — within 1 percent of a Ryzen Threadripper PRO 3955WX on average score. GPU-accelerated viewport work and GPU rendering are the weak links, at 6.144 TFLOPS FP32 with system-shared bandwidth.

Software development: A top-tier scenario. Compilation is inherently parallel, and the multi-to-single-core scaling shown across every benchmark suite means large builds complete quickly. A 2330 Geekbench single-core score keeps IDEs, emulators, and containers responsive. The DDR5 support and ECC capability add stability and bandwidth advantages for serious development machines.

Student and office work: Overkill in a good way. Documents, browsers, video calls, and research multitasking barely register on a 91st-percentile CPU. The integrated UHD Graphics 770 also exists as a secondary display path, and the machine will remain responsive under any realistic productivity load.

FAQ

Q: Is the Intel Arc B370 a dedicated graphics card? A: No. It is an integrated GPU on the 3 nm Panther Lake platform with an IGP bus interface, no power connectors, system-shared memory, and display outputs that are portable-device dependent. Its 25 W TDP is part of the platform power budget, not a discrete card's draw.

Q: How does the i9-14900HX compare to its closest rivals? A: Extremely closely. It averages 56004, versus 55919 for the Intel Core 7 253PQE (0.2 percent gap), 56273 for the AMD Ryzen AI Max 390 and 56306 for the AMD Ryzen AI 9 HX PRO 470 (both 0.5 percent ahead), and 56555 for the AMD Ryzen Threadripper PRO 3955WX (1 percent ahead).

Q: Can this build game at 1080p ultra settings? A: Not realistically. No measured FPS data exists for this pairing, and estimates based on a 1184 Steel Nomad DX12 score and 5th percentile GPU ranking point to settings reductions being necessary in modern titles even at 1080p.

Q: Does the CPU support overclocking? A: Yes. The multiplier is unlocked, which is notable for a mobile BGA 1964 part — though actual headroom depends on the laptop's cooling solution.

Q: What memory does the platform support? A: DDR4 and DDR5 on a dual-channel bus, with ECC memory supported — a rare combination that suits workstation-oriented use.

Q: Is this a desktop build? A: No. The build class is laptop. The CPU is a mobile market-segment chip on the BGA 1964 socket, and the GPU is integrated.

Q: What is the combined performance percentile of this pairing? A: 48 — but that figure averages a 91st-percentile CPU with a 5th-percentile GPU, so the combined number understates the CPU and overstates the gaming experience.

Upgrade Path and Platform

There is no conventional upgrade path here, and the database data makes that explicit. The BGA 1964 socket means the CPU is soldered in place — component swaps are not possible, so the platform is fixed at purchase. Memory is the one lever with real effect: because the Arc B370 draws its bandwidth and capacity from system RAM over a dual-channel bus, populating the fastest supported DDR5 configuration directly improves GPU performance as well as CPU throughput. This is the single most impactful "upgrade" available for a laptop of this design.

The platform also carries 16 lanes of PCIe Gen 5 from the CPU, which is relevant for external enclosures or high-speed storage in machines that expose those lanes. On power, the CPU's TDP is 55 W and the GPU's is 25 W; the GPU requires no power connectors and has no suggested PSU value, consistent with an integrated design where the chassis power delivery is engineered as a whole. A sensible next upgrade, in practical terms, is a different machine entirely: this CPU deserves a platform with a dedicated GPU many percentiles higher, while this GPU suits a far lighter compute load.

Build Overview

This is a laptop-class build: Intel's Core i9-14900HX, a Raptor Lake-HX Refresh mobile flagship released in January 2024, paired with the Intel Arc B370, a Panther Lake Xe3-LPG integrated GPU on a 3 nm process released in January 2026. The tier split is the defining characteristic. On compute, this is an upper-echelon machine — 91st percentile among all CPUs, trading blows with Ryzen AI Max and Threadripper PRO parts within a 1 percent band. On graphics, it is 5th percentile, benchmarked against mobility GPUs of a much earlier era.

The pairing is best understood as a mobile workstation without workstation graphics: 24 cores, 36 MB of shared L3, DDR5 and ECC support, and DirectX 12 Ultimate / Vulkan 1.4 / OpenGL 4.6 API coverage on the display side. The combined percentile of 48 is technically mid-tier; functionally, this build is high-end for computation and entry-level for rendering.

Who Should Build It

Buyers of this configuration — it cannot truly be built, only chosen — should be compute-first users. Software developers will get the most from it: the multi-core scaling across every indexed suite makes it an excellent compilation and virtualization machine, and ECC support suits long-running stable workloads. Video editors working with CPU-side rendering and proxy workflows will find the 30055 Cinebench R23 multi-core score productive, provided they do not depend on GPU-accelerated effects. 3D artists using CPU render engines get Threadripper-adjacent throughput in a portable chassis; those using GPU renderers should look elsewhere.

Small business workstations benefit similarly: encryption-heavy operations (32619 PassMark), compression-heavy backups (539965), and general multitasking all land on this CPU's strengths, with ECC memory adding data-integrity reassurance. Students in technical fields get a machine that will outlast a degree cycle on compute. What no buyer should expect is gaming performance: gamers targeting 1080p high settings or beyond are not served by a 5th-percentile GPU, no matter how strong the 24-core processor beside it is. This configuration is for people whose workload ends at the CPU — and who accept that the graphics pipeline exists merely to put pixels on a screen.