Intel Arc G3 vs NVIDIA GeForce RTX 5090 D Comparison

Where Each One Wins

The Intel Arc G3 and NVIDIA GeForce RTX 5090 D occupy entirely different performance strata, as the recorded data confirms. The Arc G3 is an integrated graphics processor (IGP) built for the Panther Lake mobile platform, while the RTX 5090 D is a discrete dual-slot desktop card with a 575 W TDP. The head-to-head benchmark results are not close; the RTX 5090 D holds every measurable advantage across all recorded tests, and the Arc G3 has zero wins in the database.

The RTX 5090 D delivers 104.8 TFLOPS of FP32 compute versus 6.144 TFLOPS for the Arc G3, a 17.1x gap in raw shader throughput. Its pixel rate of 423.6 GPixel/s compares to 48.00 GPixel/s for the Intel part, and the texture rate of 1,636.8 GTexel/s dwarfs the Arc G3's 96.00 GTexel/s. These are not marginal differences; they represent entirely different product categories.

For rasterization workloads, the RTX 5090 D's 21760 shading units, 680 texture mapping units, and 176 ROPs provide the foundation for its benchmark dominance. The Arc G3's 1280 shading units, 40 TMUs, and 20 ROPs are appropriate for an integrated solution targeting low-power portable devices, but they cannot compete with a dedicated card that draws 575 W and requires a 950 W suggested power supply.

In ray tracing, the RTX 5090 D includes 170 RT cores versus 10 for the Arc G3. The NVIDIA card also carries 680 tensor cores, while the Intel part lists none in the database. The RTX 5090 D supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, matching the Arc G3's API support, but the hardware resources behind those APIs are vastly larger.

Memory capacity and bandwidth further separate the two. The RTX 5090 D uses 32 GB of GDDR7 on a 512-bit bus with 1.79 TB/s bandwidth. The Arc G3 relies on system shared memory with bandwidth described as system dependent, meaning its memory performance scales with the host platform rather than dedicated VRAM.

The RTX 5090 D's benchmark scores reflect this hardware advantage. In PassMark G3D, it scores 44065; in Geekbench Vulkan, 376915; in Geekbench OpenCL, 310674; and in 3DMark Steel Nomad DX12, 14326. The Arc G3 has no recorded benchmark scores in the database, making direct numerical comparison impossible, but the architecture and specification data leave no ambiguity about which product is faster.

The Verdict

Based strictly on the recorded data, the NVIDIA GeForce RTX 5090 D is the performance leader by every available metric. Its 104.8 TFLOPS FP32 throughput, 1.79 TB/s memory bandwidth, and 423.6 GPixel/s pixel rate place it in the top tier of the database's GPU rankings, with a 92nd percentile position among all GPUs. Its average benchmark score of 77712 confirms this standing.

The Arc G3 sits at the 50th percentile among all GPUs, with an average benchmark score of 0 in the database. Its 25 W TDP and integrated design target a completely different use case: portable devices where power consumption and physical footprint matter more than raw performance. The RTX 5090 D's 575 W TDP and dual-slot form factor make it unsuitable for such applications, while the Arc G3's IGP nature makes it incapable of matching desktop discrete performance.

Users requiring maximum frame rates, high-resolution rendering, or compute-intensive workloads should select the RTX 5090 D. The data shows it leads in every benchmark category recorded, including PassMark DirectX 10 (231), DirectX 11 (371), DirectX 12 (219), DirectX 9 (434), G2D (1487), G3D (44065), and GPU compute (28396). The 3DMark Steel Nomad DX12 score of 14326 further validates its DirectX 12 performance.

The Arc G3 suits systems where integrated graphics are the only option, such as thin-and-light laptops or compact portable devices. Its 300 MHz base clock and 2400 MHz boost clock, combined with 6.144 TFLOPS FP32, provide adequate graphics capability for basic tasks, but the database contains no benchmark scores to suggest it can handle demanding workloads.

The RTX 5090 D's nearest rivals in the database are all AMD and NVIDIA professional or mobile parts: the AMD Radeon RX 6650M XT scores 76904 (1.1% ahead), the AMD Radeon RX 6850M XT scores 78940 (1.6% behind), and the NVIDIA Tesla P100 PCIe variants score 79396 and 79605 (2.1% and 2.4% behind respectively). This clustering indicates the RTX 5090 D performs in the same range as high-end mobile and professional workstation GPUs, rather than at the absolute top of the desktop consumer segment.

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark entries between the Intel Arc G3 and the NVIDIA GeForce RTX 5090 D. However, the RTX 5090 D's individual benchmark results provide a clear picture of its capabilities, while the Arc G3's absence of recorded scores means any comparison must rely on architectural specifications.

The RTX 5090 D's PassMark G3D score of 44065 represents its strongest single benchmark result, indicating substantial rasterization throughput. Its PassMark GPU compute score of 28396 shows strong general-purpose compute performance, while the Geekbench OpenCL score of 310674 and Vulkan score of 376915 demonstrate exceptional cross-API compute and graphics capability. The 3DMark Steel Nomad DX12 score of 14326 provides a modern DirectX 12 workload result.

The PassMark DirectX 9 score of 434 versus DirectX 11 score of 371 versus DirectX 12 score of 219 shows an interesting pattern: the RTX 5090 D performs relatively better on older DirectX versions in this specific benchmark suite, though all scores remain far above any integrated solution's expected range. The DirectX 10 score of 231 and G2D score of 1487 round out the benchmark profile.

For the Arc G3, the 6.144 TFLOPS FP32 rate and 48.00 GPixel/s pixel rate define its theoretical maximum throughput. The 96.00 GTexel/s texture rate and 12.29 TFLOPS FP16 (2:1 ratio) indicate balanced compute resources for its power envelope. The 10 RT cores provide basic ray tracing support, a feature that was once exclusive to discrete GPUs but has become available in integrated parts.

The RTX 5090 D's FP16 performance of 104.8 TFLOPS at a 1:1 ratio versus the Arc G3's 12.29 TFLOPS at 2:1 shows another significant gap. The NVIDIA part maintains full FP16 throughput without the 2:1 penalty the Intel part incurs, making it substantially more capable for half-precision workloads.

Memory bandwidth is perhaps the most decisive differentiator. The RTX 5090 D's 1.79 TB/s GDDR7 bandwidth versus the Arc G3's system-dependent shared memory means the NVIDIA card can feed its 21760 shading units without starvation, while the Intel part must compete with CPU memory traffic for bandwidth.

FAQ

Q: What is the performance difference between the Intel Arc G3 and the NVIDIA GeForce RTX 5090 D?

A: The RTX 5090 D produces 104.8 TFLOPS FP32 versus 6.144 TFLOPS for the Arc G3, a 17.1x gap. Its pixel rate is 423.6 GPixel/s versus 48.00 GPixel/s, and its texture rate is 1,636.8 GTexel/s versus 96.00 GTexel/s. The RTX 5090 D has no recorded benchmark losses to the Arc G3 in the database.

Q: Which GPU has more memory and bandwidth?

A: The RTX 5090 D has 32 GB of GDDR7 on a 512-bit bus with 1.79 TB/s bandwidth. The Arc G3 uses system shared memory with system dependent bandwidth, meaning its memory performance depends on the host platform's RAM configuration.

Q: What are the power requirements for each GPU?

A: The Arc G3 has a 25 W TDP and uses no power connectors, functioning as an integrated graphics processor. The RTX 5090 D has a 575 W TDP, requires a 1x 16-pin power connector, and has a 950 W suggested power supply.

Q: How do their ray tracing capabilities compare?

A: The RTX 5090 D includes 170 RT cores, while the Arc G3 has 10 RT cores. The NVIDIA card also includes 680 tensor cores, while the Intel part lists no tensor cores in the database.

Q: What is the release timeline for these products?

A: The RTX 5090 D was released on 2025-01-29, while the Arc G3 was released on 2026-05-31. The RTX 5090 D belongs to the GeForce 50 series with a predecessor in the GeForce 40 series, and the Arc G3 belongs to the Arc Graphics-M (Panther Lake) generation.

Q: How does the RTX 5090 D rank among all GPUs in the database?

A: The RTX 5090 D sits at the 92nd percentile among all GPUs with an average benchmark score of 77712. Its nearest rivals are the AMD Radeon RX 6650M XT (76904, 1.1% ahead), AMD Radeon RX 6850M XT (78940, 1.6% behind), and two NVIDIA Tesla P100 PCIe variants (79396 and 79605, 2.1% and 2.4% behind respectively).

Architecture Differences

The Intel Arc G3 uses the Panther Lake chip with Xe3-LPG architecture built on Intel's 3 nm process. It is an integrated graphics processor with no separable slot width, using an IGP bus interface. The RTX 5090 D uses the GB202 chip with Blackwell 2.0 architecture fabricated by TSMC on a 5 nm process, containing 92,200 million transistors on a 750 mm² die with a density of 122.9M transistors per mm².

The Arc G3's clock behavior differs fundamentally from the discrete NVIDIA part. Its base clock is 300 MHz with a boost of 2400 MHz, while the RTX 5090 D runs at 2017 MHz base and 2407 MHz boost. The NVIDIA memory clock is 1750 MHz with 28 Gbps effective data rate, whereas the Arc G3's memory clock is listed as system shared.

Shader resources show the scale difference: the Arc G3 has 1280 shading units, 40 TMUs, and 20 ROPs, while the RTX 5090 D has 21760 shading units, 680 TMUs, and 176 ROPs. The RT core count of 170 versus 10 and the tensor core count of 680 versus none further distinguish their compute capabilities. The FP32 throughput of 104.8 TFLOPS for NVIDIA versus 6.144 TFLOPS for Intel, and FP16 of 104.8 TFLOPS at 1:1 versus 12.29 TFLOPS at 2:1, reflect both the hardware count and the clock differences.

Physical specifications diverge completely. The RTX 5090 D measures 304 mm in length, 137 mm in height, and 48 mm in width, occupying a dual-slot form factor. The Arc G3 has no listed dimensions, consistent with its IGP integration into a host processor. The RTX 5090 D's display outputs include 1x HDMI 2.1b and 3x DisplayPort 2.1b, while the Arc G3's outputs are described as portable device dependent.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, meaning software compatibility is identical even though hardware performance differs by orders of magnitude. The production status for both is listed as active, with the RTX 5090 D's launch MSRP recorded as 2,299 USD. The Arc G3 has no launch MSRP in the database.

The RTX 5090 D's power delivery infrastructure reflects its performance class: a 575 W TDP, 1x 16-pin connector, and 950 W suggested power supply. The Arc G3's 25 W TDP and no power connectors underscore its integration into a mobile platform where thermal and power budgets are constrained. These architectural differences explain why the benchmark results, where recorded, show the RTX 5090 D in a completely different performance tier.