Intel Arc A770 vs NVIDIA GeForce RTX 5090 D Comparison

NVIDIA GeForce RTX 5090 D and Intel Arc A770 occupy completely different corners of the GPU market, and the benchmark data reflects that split. The RTX 5090 D is a 575 W flagship with 92,200 million transistors on TSMC 5 nm, while the Arc A770 is an end-of-life 225 W part built on TSMC 6 nm with 21,700 million transistors. Across the three shared benchmarks, the RTX 5090 D wins all three, with deltas ranging from 184.6% to 382.5%. Its average benchmark score of 77,712 places it in the 92nd percentile of all GPUs, while the A770's 68,809 average sits in the 90th percentile. The data does not paint a close contest; it paints a generational and market-tier gap. However, the A770 remains a distinct product with its own strengths in specific workloads, and the choice between them depends entirely on what the user prioritizes: raw compute and modern feature support versus a lower-power, legacy-friendly option.

The Verdict

From the data, the RTX 5090 D is the clear choice for anyone needing maximum performance in DirectX 12 workloads or compute-heavy tasks. Its 3DMark Steel Nomad score of 14,326 is 382.5% higher than the A770's 2,969, and its Geekbench OpenCL score of 310,674 is 184.6% ahead. The RTX 5090 D also leads in Geekbench Vulkan with 376,915 versus 94,284, a 299.8% advantage. If the priority is frame rates in modern games or accelerating professional compute, the RTX 5090 D wins outright.

The Arc A770 is the pick only if the user's priorities are entirely different: lower power draw (225 W vs 575 W), a smaller PSU requirement (550 W vs 950 W), and a much lower launch MSRP of 329 USD. The A770's 16 GB GDDR6 memory is half the 32 GB GDDR7 found on the RTX 5090 D, and its 512.0 GB/s bandwidth is a fraction of the 1.79 TB/s. For a user who does not need the RTX 5090 D's extreme compute and is constrained by power or budget, the A770 is the only logical option from this comparison. The data shows no scenario where the A770 wins a benchmark, so its appeal is purely practical rather than performance-based.

Architecture Differences

The two GPUs are built on fundamentally different architectures. The RTX 5090 D uses Blackwell 2.0 on a 5 nm TSMC process, with a die size of 750 mm² and a transistor density of 122.9M per mm². The A770 uses Xe-HPG on a 6 nm TSMC process, with a 406 mm² die and a transistor density of 53.4M per mm². This means the RTX 5090 D packs more than four times the transistors into less than twice the die area, explaining its massive compute lead.

The RTX 5090 D features 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. The A770 has 4,096 shading units, 256 TMUs, 128 ROPs, and 32 RT cores, with no tensor cores listed. The RTX 5090 D's FP32 throughput is 104.8 TFLOPS, while the A770 manages 19.66 TFLOPS. In FP16, the RTX 5090 D delivers 104.8 TFLOPS (1:1), while the A770 offers 39.32 TFLOPS (2:1). The RTX 5090 D also has a higher pixel rate (423.6 GPixel/s vs 307.2 GPixel/s) and texture rate (1,636.8 GTexel/s vs 614.4 GTexel/s). Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is identical.

Memory architecture differs sharply. The RTX 5090 D uses 32 GB of GDDR7 on a 512-bit bus, while the A770 uses 16 GB of GDDR6 on a 256-bit bus. The RTX 5090 D's memory clock is listed as 1750 MHz (28 Gbps effective), while the A770 runs at 2000 MHz (16 Gbps effective). The resulting bandwidth gap is enormous: 1.79 TB/s versus 512.0 GB/s. The RTX 5090 D also uses PCIe 5.0 x16, whereas the A770 uses PCIe 4.0 x16.

Where Each One Wins

The RTX 5090 D wins in every benchmark where both cards have data. Its strongest relative showing is in 3DMark Steel Nomad DX12, where it scores 14,326 versus 2,969, a 382.5% lead. This suggests the RTX 5090 D is exceptionally well-suited to modern DirectX 12 gaming and ray-traced workloads. In Geekbench Vulkan, the RTX 5090 D scores 376,915 versus 94,284, a 299.8% lead, indicating strong performance in Vulkan-based games and compute. In Geekbench OpenCL, the lead is 184.6% (310,674 vs 109,175), showing a clear advantage in general-purpose compute tasks.

The A770 does not win any of the three head-to-head benchmarks. Its wins are limited to its physical attributes: lower TDP (225 W vs 575 W), lower suggested PSU (550 W vs 950 W), and a more modest power connector setup (1x 6-pin + 1x 8-pin vs 1x 16-pin). The A770 also has a higher base clock (2100 MHz vs 2017 MHz) and a slightly higher boost clock (2400 MHz vs 2407 MHz, a 7 MHz difference). However, these clock advantages do not translate into any benchmark victory. The A770's nearest rival is the NVIDIA CMP 90HX, with a delta of -0.3%, meaning its average score is essentially tied with that card. The RTX 5090 D's nearest rival is the AMD Radeon RX 6650M XT, with a delta of 1.1%, indicating it is only slightly ahead of that mobile part in average score.

FAQ

Q: Which card is faster in DirectX 12 workloads?

A: The RTX 5090 D is dramatically faster. In 3DMark Steel Nomad DX12, it scores 14,326 versus 2,969 for the A770, a 382.5% lead.

Q: Does the Intel Arc A770 win any benchmark against the RTX 5090 D?

A: No. Across the three shared benchmarks (3DMark Steel Nomad DX12, Geekbench OpenCL, Geekbench Vulkan), the RTX 5090 D wins all three. The A770 has zero wins in the head-to-head data.

Q: What is the memory capacity difference?

A: The RTX 5090 D has 32 GB of GDDR7, while the A770 has 16 GB of GDDR6. The RTX 5090 D also has a 512-bit bus versus a 256-bit bus, giving it 1.79 TB/s bandwidth compared to the A770's 512.0 GB/s.

Q: How do their power requirements compare?

A: The RTX 5090 D has a TDP of 575 W and requires a 950 W PSU. The A770 has a TDP of 225 W and requires a 550 W PSU. The RTX 5090 D uses a single 16-pin connector, while the A770 uses one 6-pin and one 8-pin.

Q: Which GPU has a higher launch MSRP?

A: The RTX 5090 D has a launch MSRP of 2,299 USD. The A770 has a launch MSRP of 329 USD.

Q: Are both cards still in production?

A: No. The RTX 5090 D has an active production status, while the A770 is end-of-life. The A770's successor is Battlemage, and the RTX 5090 D's successor is GeForce 60.

Head-to-Head Benchmarks

The biggest single win for the RTX 5090 D comes in 3DMark Steel Nomad DX12. The RTX 5090 D scores 14,326, while the A770 scores 2,969, yielding a delta of 382.5%. This is not a marginal improvement; it is a 3.8x performance gap in a modern DirectX 12 test. For any user running current games or DX12-based applications, this result alone separates the two cards decisively.

In Geekbench Vulkan, the RTX 5090 D scores 376,915 versus 94,284 for the A770, a delta of 299.8%. This result is slightly closer in relative terms than the DX12 test, but still represents a 4x lead in raw score. Vulkan is used in many cross-platform games and compute APIs, so this advantage extends beyond a single test. The RTX 5090 D's 680 tensor cores and 170 RT cores likely contribute to this lead, though the benchmark data does not isolate those factors.

The smallest relative win for the RTX 5090 D is in Geekbench OpenCL, where it scores 310,674 versus 109,175, a delta of 184.6%. OpenCL is a general-purpose compute API, and this result shows the RTX 5090 D is still nearly 3x faster than the A770 in this workload. The A770's FP32 throughput of 19.66 TFLOPS is only about 19% of the RTX 5090 D's 104.8 TFLOPS, which aligns with this benchmark outcome. Across all three tests, the RTX 5090 D's wins are consistent and overwhelming, with no single test where the A770 comes close.

Specification Differences

The specification sheets differ in nearly every measurable field. The RTX 5090 D uses a GB202 chip on Blackwell 2.0 architecture, while the A770 uses a DG2-512 chip on Xe-HPG. The process node is 5 nm versus 6 nm, and the RTX 5090 D has 92,200 million transistors on a 750 mm² die, while the A770 has 21,700 million on a 406 mm² die. Transistor density is 122.9M per mm² for the RTX 5090 D compared to 53.4M per mm² for the A770.

Clock speeds are similar on paper: the RTX 5090 D has a base clock of 2017 MHz and a boost of 2407 MHz, while the A770 has a base of 2100 MHz and a boost of 2400 MHz. Memory clocks differ: 1750 MHz (28 Gbps effective) for the RTX 5090 D versus 2000 MHz (16 Gbps effective) for the A770. The RTX 5090 D has 32 GB of GDDR7 on a 512-bit bus, while the A770 has 16 GB of GDDR6 on a 256-bit bus. Bandwidth is 1.79 TB/s versus 512.0 GB/s.

Compute resources are heavily skewed: 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores for the RTX 5090 D, versus 4,096 shading units, 256 TMUs, 128 ROPs, and 32 RT cores for the A770. The RTX 5090 D's FP32 is 104.8 TFLOPS, and its FP16 is 104.8 TFLOPS (1:1); the A770's FP32 is 19.66 TFLOPS, and its FP16 is 39.32 TFLOPS (2:1). Pixel rate is 423.6 GPixel/s versus 307.2 GPixel/s, and texture rate is 1,636.8 GTexel/s versus 614.4 GTexel/s.

Power and physical specs differ: TDP is 575 W versus 225 W, suggested PSU is 950 W versus 550 W, and power connectors are 1x 16-pin versus 1x 6-pin + 1x 8-pin. Both are dual-slot, but the RTX 5090 D measures 304 mm in length, 137 mm in height, and 48 mm in width, while the A770 has no listed dimensions. The RTX 5090 D uses PCIe 5.0 x16, the A770 uses PCIe 4.0 x16. Display outputs are 1x HDMI 2.1b and 3x DisplayPort 2.1b for the RTX 5090 D, versus 1x HDMI 2.1 and 3x DisplayPort 2.0 for the A770. Release dates are 2025-01-29 for the RTX 5090 D and 2022-10-11 for the A770. Production status is Active versus End-of-life.