Intel Arc A380E vs NVIDIA GeForce RTX 5090 D Comparison

Head-to-Head Benchmarks

The recorded data shows a one-sided comparison. The Intel Arc A380E has no benchmark entries in the database, while the NVIDIA GeForce RTX 5090 D has ten recorded scores. Consequently, the RTX 5090 D wins every comparison where data exists. The Arc A380E’s average benchmark score is listed as 0, and it holds a 50th percentile position among all GPUs. The RTX 5090 D, by contrast, posts an average benchmark score of 77,712 and sits at the 92nd percentile, a gap of 42 percentile points that places the two cards in entirely different performance strata.

The RTX 5090 D’s most decisive results come from compute-heavy workloads. In Geekbench Vulkan, it scores 376,915, and in Geekbench OpenCL, it scores 310,674. These figures dwarf the Arc A380E’s complete absence of recorded scores. The PassMark G3D test returns 44,065 for the RTX 5090 D, while PassMark GPU Compute returns 28,396. The 3DMark Steel Nomad DX12 test yields 14,326. Even the lowest recorded RTX 5090 D result, PassMark DirectX 12 at 219, has no counterpart from the Intel part. The data confirms that the Arc A380E cannot be positioned against the RTX 5090 D on any measured workload; the NVIDIA card leads by definition across all ten benchmark categories.

Relative to its own nearest rivals, the RTX 5090 D is closely matched in aggregate scoring. It trails the AMD Radeon RX 6850M XT by 1.6%, sits 2.1% behind the NVIDIA Tesla P100 PCIe 12 GB, and is 2.4% behind the Tesla P100 PCIe 16 GB. It leads the AMD Radeon RX 6650M XT by 1.1%. These deltas, all within a few percentage points, indicate that the RTX 5090 D’s average score of 77,712 places it in a tight competitive cluster. The Arc A380E has no nearest rival entries at all, reinforcing its status as a product without comparable measurements in this database.

Architecture Differences

The two GPUs come from different architectural generations and process nodes. The Intel Arc A380E uses the DG2-128 chip built on the Xe-HPG architecture, part of the Alchemist (Arc 3) generation, fabricated on a 6 nm process at TSMC. The NVIDIA GeForce RTX 5090 D uses the GB202 chip with Blackwell 2.0 architecture, belonging to the GeForce 50 series, also fabricated at TSMC but on a 5 nm node. The process difference is small in absolute terms, yet the transistor counts are not. The Arc A380E packs 7,200 million transistors on a 157 mm² die, giving a transistor density of 45.9 million per square millimeter. The RTX 5090 D integrates 92,200 million transistors on a 750 mm² die, yielding 122.9 million per square millimeter. That is roughly 2.7 times the density and over 12 times the raw transistor count.

Compute resources differ by an order of magnitude. The Arc A380E has 1,024 shading units, 64 texture mapping units, 32 ROPs, and 8 ray tracing cores. The RTX 5090 D has 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. The Intel part lists no tensor core count in the data. Floating-point throughput tells the same story: the Arc A380E delivers 4.096 TFLOPS FP32 and 8.192 TFLOPS FP16 at a 2:1 ratio, while the RTX 5090 D delivers 104.8 TFLOPS for both FP32 and FP16 at a 1:1 ratio. The NVIDIA card’s FP32 figure is roughly 25.6 times higher. Pixel and texture rates follow suit: 64.00 GPixel/s and 128.0 GTexel/s for Intel versus 423.6 GPixel/s and 1,636.8 GTexel/s for NVIDIA.

Memory architecture is equally divergent. The Arc A380E uses 6 GB of GDDR6 on a 96-bit bus, delivering 186.0 GB/s of bandwidth. The RTX 5090 D uses 32 GB of GDDR7 on a 512-bit bus, delivering 1.79 TB/s. That is over five times the capacity and roughly 9.6 times the bandwidth. Clock behavior also differs. The Intel GPU runs at a fixed 2000 MHz base and boost, with memory at 1937 MHz (15.5 Gbps effective). The NVIDIA GPU has a 2017 MHz base and 2407 MHz boost, with memory at 1750 MHz (28 Gbps effective). The boost clock advantage is modest, but the memory speed difference is substantial.

Power and physical design diverge sharply. The Arc A380E carries a 75 W TDP, requires no power connectors, and fits a single slot. The RTX 5090 D has a 575 W TDP, uses a single 16-pin connector, and occupies a dual-slot design. The suggested power supply ratings are 250 W for Intel and 950 W for NVIDIA. Dimensions reflect the power gap: the Arc A380E measures 254 mm by 127 mm by 20 mm, while the RTX 5090 D measures 304 mm by 137 mm by 48 mm. The NVIDIA card is longer, taller, and more than twice as thick. Both support PCIe, but the Intel card uses PCIe 4.0 x8 while the NVIDIA card uses PCIe 5.0 x16. Display outputs also differ: the Arc A380E has four DisplayPort 2.0 connections, while the RTX 5090 D has one HDMI 2.1b and three DisplayPort 2.1b outputs. API support is identical for both: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Production status separates the two further. The Arc A380E is end-of-life, released on 2024-03-31, with its predecessor listed as Xe Graphics and its successor as Battlemage. The RTX 5090 D is active, released on 2025-01-29, with its predecessor listed as GeForce 40 and its successor as GeForce 60. The Intel card has no launch MSRP in the database, while the NVIDIA card carries a launch MSRP of 2,299 USD.

The Verdict

The data supports only one conclusion for performance-centric workloads: the RTX 5090 D dominates every measurable category. With an average benchmark score of 77,712 versus 0 for the Arc A380E, and a 92nd versus 50th percentile ranking, the NVIDIA card is the clear choice for anyone prioritizing raw compute, rendering, or gaming performance. The RTX 5090 D’s 32 GB GDDR7 memory, 1.79 TB/s bandwidth, and 104.8 TFLOPS FP32 throughput provide capabilities the Intel part cannot approach. The Arc A380E, with 6 GB GDDR6, 186.0 GB/s bandwidth, and 4.096 TFLOPS FP32, targets a fundamentally different class of use.

The Arc A380E does hold advantages in efficiency and physical footprint. Its 75 W TDP requires no auxiliary power connectors and a 250 W suggested PSU, compared to 575 W, a 16-pin connector, and a 950 W suggested PSU for the RTX 5090 D. It is also single-slot, shorter, and lighter in dimensions. These traits make it suitable for constrained systems, but the benchmark data offers no evidence that it can compete on performance. The RTX 5090 D’s nearest rivals, all within roughly 2.5% of its average score, show that it competes in a crowded high-end field, but the Arc A380E is not part of that field.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce RTX 5090 D has an average benchmark score of 77,712, while the Intel Arc A380E has an average benchmark score of 0.

Q: How much memory does each card have?

A: The Intel Arc A380E has 6 GB of GDDR6 on a 96-bit bus, while the NVIDIA GeForce RTX 5090 D has 32 GB of GDDR7 on a 512-bit bus.

Q: What is the transistor count difference?

A: The Arc A380E has 7,200 million transistors on a 157 mm² die, while the RTX 5090 D has 92,200 million transistors on a 750 mm² die.

Q: Which card has a higher boost clock?

A: The RTX 5090 D has a boost clock of 2407 MHz, compared to the Arc A380E’s 2000 MHz boost clock.

Q: What are the TDP requirements?

A: The Arc A380E has a 75 W TDP with no power connectors and a 250 W suggested PSU, while the RTX 5090 D has a 575 W TDP with a 16-pin connector and a 950 W suggested PSU.

Q: Are the APIs supported the same?

A: Yes, both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Where Each One Wins

The RTX 5090 D wins on every benchmark with recorded data. Its ten scores span 3DMark Steel Nomad DX12 at 14,326, Geekbench OpenCL at 310,674, Geekbench Vulkan at 376,915, and multiple PassMark tests including G3D at 44,065 and GPU Compute at 28,396. It also wins on memory bandwidth, compute throughput, texture and pixel rates, and ray tracing core count. For any workload that depends on these resources, the RTX 5090 D is the only option with evidence of capability.

The Arc A380E wins on power efficiency and system integration. It draws 75 W versus 575 W, needs no power connector, and fits a single slot. Its 250 W suggested PSU is 700 W lower than the RTX 5090 D’s 950 W requirement. The Intel card is also shorter (254 mm versus 304 mm), less tall (127 mm versus 137 mm), and much thinner (20 mm versus 48 mm). It offers four DisplayPort 2.0 outputs, one more physical display connection than the RTX 5090 D’s combination of one HDMI 2.1b and three DisplayPort 2.1b. For systems where space, power delivery, or display count matter more than raw performance, the Arc A380E has structural advantages. The RTX 5090 D also trails its nearest rivals slightly, sitting 1.6% behind the AMD Radeon RX 6850M XT, 2.1% behind the Tesla P100 12 GB, and 2.4% behind the Tesla P100 16 GB, which contextualizes its performance within the high-end tier.

Specification Differences

The two cards differ in nearly every hardware specification. The Arc A380E uses a 6 nm process and the Xe-HPG architecture, while the RTX 5090 D uses a 5 nm process and Blackwell 2.0. Transistor counts are 7,200 million versus 92,200 million, and die sizes are 157 mm² versus 750 mm². Shading units number 1,024 versus 21,760, TMUs 64 versus 680, ROPs 32 versus 176, and RT cores 8 versus 170. The RTX 5090 D adds 680 tensor cores; the Arc A380E has none listed. FP32 performance is 4.096 TFLOPS versus 104.8 TFLOPS, and FP16 performance is 8.192 TFLOPS versus 104.8 TFLOPS. Memory capacity is 6 GB versus 32 GB, type is GDDR6 versus GDDR7, bus width is 96-bit versus 512-bit, and bandwidth is 186.0 GB/s versus 1.79 TB/s. Base clocks are 2000 MHz versus 2017 MHz, boost clocks 2000 MHz versus 2407 MHz, and memory clocks 1937 MHz versus 1750 MHz. TDP is 75 W versus 575 W, slot width is single-slot versus dual-slot, power connectors are none versus one 16-pin, and suggested PSU is 250 W versus 950 W. Bus interface is PCIe 4.0 x8 versus PCIe 5.0 x16. Dimensions are 254 mm by 127 mm by 20 mm versus 304 mm by 137 mm by 48 mm. Production status is end-of-life versus active, release dates are 2024-03-31 versus 2025-01-29, and the RTX 5090 D carries a launch MSRP of 2,299 USD while the Arc A380E has none recorded.