Intel Arc A380 vs NVIDIA GeForce GTX 750 Comparison

The Intel Arc A380 and NVIDIA GeForce GTX 750 represent two distinct eras of graphics hardware. The Arc A380 is a modern entry-level discrete GPU built on a 6 nm process with ray tracing support, while the GTX 750 is a legacy Maxwell-based card from 2014. The recorded data shows a decisive performance gap, but the comparison still yields interesting insights into architectural evolution, feature sets, and use-case suitability.

Head-to-Head Benchmarks

The database contains two direct head-to-head benchmark results between these cards. Both are compute and API-level tests, and the Intel Arc A380 wins both by a substantial margin.

In the Geekbench OpenCL test, the Arc A380 scores 38,224 points against the GTX 750's 9,315 points. That is a delta of 310.3% in favor of Intel. OpenCL is a compute workload that stresses raw shading unit throughput and memory bandwidth. The Arc A380 has 1,024 shading units, 64 texture mapping units, and 32 ROPs, compared to the GTX 750's 512 shading units, 32 TMUs, and 16 ROPs. The doubling of every major execution resource explains the scale of the gap. The A380 also has a memory bandwidth of 186.0 GB/s on a 96-bit GDDR6 interface, while the GTX 750 offers 80.19 GB/s on a 128-bit GDDR5 bus. Although the GTX 750 has a wider bus, the Arc A380's much higher effective memory clock (15.5 Gbps effective versus 5 Gbps effective) gives it more than double the bandwidth.

The Vulkan head-to-head result is even more lopsided. The Arc A380 scores 36,736 points, while the GTX 750 scores 8,078 points. The delta is 354.8% in favor of Intel. Vulkan is a low-level graphics API that benefits from modern driver overhead management and hardware features. The Arc A380 supports Vulkan 1.4 and DirectX 12 Ultimate (12_2), which includes features like mesh shaders and variable rate shading. The GTX 750 also lists Vulkan 1.4 support in the database, but its DirectX 12 support is limited to 12 (11_0), meaning it cannot access the full DirectX 12 feature set. The architecture differences matter here: the GTX 750's Maxwell design predates hardware-level features that current APIs can exploit.

The average benchmark score in the database reinforces this trend. The Arc A380 has an average benchmark score of 8,558 across all its recorded tests, while the GTX 750 averages 7,222. The Arc A380 sits at the 44th percentile of all GPUs in the database, and the GTX 750 sits at the 40th percentile. That is a narrow overall percentile gap, but the head-to-head tests show the Arc A380 pulling far ahead in compute and modern API workloads. The delta between their average scores is about 18.5%, a much smaller margin than the individual head-to-head tests suggest, because the GTX 750's recorded benchmark suite is limited to three Geekbench tests while the Arc A380 has ten tests across different workloads.

The Arc A380's nearest rivals in the database are the AMD FirePro W5170M (average score 8,595, delta -0.4%), AMD Radeon HD 8870M (8,462, delta 1.1%), NVIDIA GeForce MX330 (8,458, delta 1.2%), and AMD Radeon 880M (8,436, delta 1.4%). The GTX 750's nearest rivals are the AMD Radeon Vega 8 Mobile (7,203, delta 0.3%), NVIDIA GeForce GTX 560 SE (7,171, delta 0.7%), Intel Iris Pro Graphics P580 (7,170, delta 0.7%), and NVIDIA GeForce GTX 970 (7,157, delta 0.9%). These clusters show that the Arc A380 competes with a higher tier of mobile and integrated GPUs, while the GTX 750 sits among older discrete and integrated parts.

FAQ

Q: How much faster is the Intel Arc A380 than the GTX 750 in OpenCL compute?

A: The Arc A380 scores 38,224 in Geekbench OpenCL versus 9,315 for the GTX 750. That is a 310.3% performance lead for the Intel card, driven by double the shading units, TMUs and ROPs, plus 186.0 GB/s of memory bandwidth against 80.19 GB/s.

Q: Does the GTX 750 support ray tracing hardware?

A: No. The database lists 8 ray tracing cores for the Arc A380 and no ray tracing cores for the GTX 750. The GTX 750's architecture predates hardware ray tracing entirely.

Q: Which card supports newer DirectX features?

A: The Arc A380 supports DirectX 12 Ultimate (12_2), while the GTX 750 supports DirectX 12 (11_0). The "12_2" feature level is the higher tier, enabling advanced rendering techniques that the older Maxwell GPU cannot handle.

Q: What is the memory capacity difference?

A: The Arc A380 has 6 GB of GDDR6 memory on a 96-bit bus. The GTX 750 has 1024 MB (1 GB) of GDDR5 memory on a 128-bit bus. Even with a narrower bus, the Arc A380's bandwidth is 186.0 GB/s versus 80.19 GB/s for the GTX 750.

Q: Which card has a higher transistor density?

A: The Arc A380 is built on a 6 nm TSMC process with 7,200 million transistors on a 157 mm² die, yielding a density of 45.9M transistors per mm². The GTX 750 uses a 28 nm TSMC process with 1,870 million transistors on a 148 mm² die, yielding 12.6M per mm².

Q: Are both cards still in production?

A: No. Both are listed as end-of-life in the database. The Arc A380 was released on 2022-06-13 and the GTX 750 on 2014-02-17.

Architecture Differences

The Arc A380 uses the DG2-128 chip with Intel's Xe-HPG architecture, part of the Alchemist generation (Arc 3). It is fabricated by TSMC on a 6 nm process. The die contains 7,200 million transistors across 157 mm², which works out to a transistor density of 45.9 million per square millimeter. The GTX 750 uses the GM107 chip with NVIDIA's Maxwell architecture, part of the GeForce 700 generation. It is also fabricated by TSMC, but on a 28 nm process. The die contains 1,870 million transistors across 148 mm², for a density of 12.6 million per square millimeter. The process node difference alone, 6 nm versus 28 nm, is massive and explains the density gap.

The execution resources differ sharply. The Arc A380 has 1,024 shading units, 64 TMUs, 32 ROPs, and 8 ray tracing cores. The GTX 750 has 512 shading units, 32 TMUs, and 16 ROPs, with no ray tracing cores. Every execution unit count is exactly doubled in the Arc A380. The pixel rate for the Arc A380 is 65.60 GPixel/s and the texture rate is 131.2 GTexel/s. The GTX 750's pixel rate is 17.36 GPixel/s and texture rate is 34.72 GTexel/s. The FP32 compute throughput is 4.198 TFLOPS for the Arc A380, while the GTX 750 manages 1,111.0 GFLOPS. The Arc A380 also lists FP16 throughput of 8.397 TFLOPS with a 2:1 ratio, a feature absent from the GTX 750's spec sheet.

The memory subsystems are also fundamentally different. The Arc A380 uses 6 GB of GDDR6 memory clocked at 1937 MHz with 15.5 Gbps effective speed on a 96-bit bus. The GTX 750 uses 1024 MB of GDDR5 memory clocked at 1253 MHz with 5 Gbps effective speed on a 128-bit bus. The bandwidth figures are 186.0 GB/s for Intel and 80.19 GB/s for NVIDIA. The Arc A380 draws a 75 W TDP and requires a single 8-pin power connector, while the GTX 750 has a 55 W TDP and requires no power connectors at all. Both cards list a 250 W suggested PSU. The physical design differs too: the Arc A380 is dual-slot with dimensions of 222 mm (8.7 inches) by 114 mm (4.5 inches) by 42 mm (1.7 inches), while the GTX 750 is single-slot and 145 mm (5.7 inches) long with no recorded height or width.

The API support shows the generational gap. The Arc A380 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The GTX 750 also supports OpenGL 4.6 and Vulkan 1.4, but its DirectX support is capped at 12 (11_0). The Arc A380's display outputs are 1x HDMI 2.1 and 3x DisplayPort 2.0, while the GTX 750 offers 2x DVI and 1x mini-HDMI 1.4a. The bus interface also differs: the Arc A380 uses PCIe 4.0 x8, and the GTX 750 uses PCIe 3.0 x16.

The Verdict

The data points to clear conclusions. For anyone looking at compute performance in OpenCL or Vulkan, the Arc A380 is the superior card by a wide margin: 310.3% in OpenCL and 354.8% in Vulkan. The Arc A380 also has 6 GB of memory versus 1 GB on the GTX 750, ray tracing cores where the GTX 750 has none, DirectX 12 Ultimate support where the GTX 750 is limited to 12 (11_0), and roughly four times the FP32 throughput.

The GTX 750 does retain a few advantages worth noting from the specification data. It has a lower TDP at 55 W versus 75 W, requires no power connectors, and needs no power connectors of any kind on the board. That makes it a viable option for legacy systems without auxiliary power capability. It is also a single-slot card with a shorter length of 145 mm, compared to the Arc A380's dual-slot, 222 mm footprint. For someone constrained by a compact case or an older motherboard with a PCIe 3.0 x16 slot, the GTX 750 is the more flexible physical fit.

However, the performance gap is too large to ignore. The Arc A380's nearest rival cluster sits around 8,400 to 8,600 average benchmark score, and the GTX 750 cluster sits around 7,150 to 7,200. The Arc A380 is in a different performance tier entirely. The GTX 750's nearest rival list even includes the NVIDIA GeForce GTX 970 with a delta of 0.9%, a card that was itself a high-end part in its generation, which shows how compressed the older GPU performance range is in the database. For modern workloads that leverage Vulkan features, ray tracing, large textures, or high-resolution assets, the GTX 750 will struggle while the Arc A380 has headroom.

Specification Differences

The two cards differ in every major specification category except foundry, OpenGL version, Vulkan version, suggested PSU wattage, and production status. Both are made by TSMC, both support OpenGL 4.6, both list Vulkan 1.4, both suggest a 250 W PSU, and both are end-of-life products.

The process node changes from 6 nm on the Arc A380 to 28 nm on the transistor count goes from 7,200 million to 1,870 million. The die size is 157 mm² for the Arc A380 and 148 mm² for the GTX 750 million transistors per mm². The base clock speeds are 2000 MHz base and 2050 MHz boost for the Arc A380, versus 1020 MHz base and 1080 MHz boost for the GTX 750. Memory speed is 1937 MHz with 15.5 Gbps effective for the Arc A380 and 1253 MHz with 5 Gbps effective for GTX 750 GB of GDDR6 on a 96-bit bus versus 1024 MB of GDDR5 on 128-bit bus. Bandwidth is 186.0 GB/s for the Arc A380 and 80.0 GB/s for GTX 750 shading units count is 1,024 versus 512 TMUs 64 versus 32 32 ROPs 8 ray tracing cores for the Arc A380 with nulls for GTX 750. Pixel rates are 65.60 GPixel/s for Arc A380 and 17.36 GPixel/s GTX 750 texture rate 131.2 GTexel/s for the Arc A380 34.72 GTexel/s GTX 750 FP32 is 4.198 TFLOPS Arc A380 and 1,110.0 GFLOPS for GTX 750. FP16 is 8.397 TFLOPS with 2:1 ratio for the Arc A380 and null for GTX 750 watts TDP 75 W Arc A380 versus GTX 750 slot width dual-slot Arc A380 and single-slot GTX 750 power connectors and 1x 8-pin Arc A380, none for GTX 750 bus interface PCIe 4.0 x8 for Arc A380 and a PCIe 3.0 x16 for GTX 750 display outputs 1x HDMI 2.1, 3x DisplayPort 2.0 for the Arc A380, and 2x DVI plus 1x mini-HDMI 1.4a for the GTX 750. DirectX support is 12 Ultimate (12_2) for the Arc A380 and 12 (11_0) for GTX 750 dimensions length 222 mm (8.7 inches) for the Arc A380, 145 mm (5.7 inches) GTX 750, height 114 mm (4.5 inches) for the Arc A380 with null for GTX 750, width 42 mm (1.7 inches) for the Arc A380 with null for GTX 750 launch MSRP is 149 USD for Arc A380 and 119 USD for GTX 750 release date 2022-06-13 for Arc A380 and 2014-02-17 for GTX 750 predecessor Xe Graphics for Arc A380 and GeForce 600 for GTX 750 successor Battlemage for Arc A380 and GeForce 900 for GTX 750.

Where Each One Wins

The Arc A380 wins in compute-bound workloads. The Geekbench OpenCL result of 38,224 against 9,315 shows a 310.3% advantage. Applications using OpenCL for rendering tasks, physics simulations, image processing, or data-heavy workloads will see massive gains on the Arc A380. The Vulkan test result of 36,736 against 8,078 shows a 354.8% advantage. Games and engines that use Vulkan for low-level GPU control will favor the Arc A380 heavily. The Arc A380's 6 GB of GDDR6 memory and 186.0 GB/s bandwidth make it suitable for texture-heavy scenarios, high-resolution rendering, or workloads that need to keep large datasets in GPU-visible memory. Ray tracing workloads are exclusively the Arc A380's domain since the GTX 750 has no ray tracing cores. The Arc A380's DirectX 12 Ultimate (12_2) feature level also gives access to modern rendering pipeline features unavailable on the GTX 750's 12 (11_0) support.

The GTX 750 wins in power efficiency at the board level. Its 55 W TDP is lower than the Arc A380's 75 W TDP. It requires no power connectors, which is a significant advantage in systems with no PCIe auxiliary power cables. Its single-slot design and 145 mm length make it easier to fit in small form factor cases or legacy chassis. The GTX 750's PCIe 3.0 x16 interface is compatible with older motherboards, though the Arc A380's PCIe 4.0 x8 is backward compatible in practice. For users with a pre-2014 system that has an available slot and no power connector, the GTX 750 is the only functional choice. The GTX 750 also has a lower launch MSRP of 119 USD versus the Arc A380's 149 USD launch MSRP, though both are end-of-life products now.

The GTX 750's recorded benchmark suite is limited to Geekbench Metal, OpenCL, and Vulkan, where it scores 4,274 in Metal, 9,315 in OpenCL, and 8,078 in Vulkan. The Arc A380 has no Metal score in the database, so Apple-centric workloads cannot be directly compared. That is one narrow area where the GTX 750 has recorded data and the Arc A380 does not, but it is not a performance advantage, merely a data gap. For every overlapping benchmark in the database, the Arc A380 wins decisively. The conclusion from the recorded measurements is that the Arc A380 is the stronger card for essentially all modern workloads, while the GTX 750 retains relevance only in legacy system integration and ultra-low-power scenarios.