Intel Arc A750 vs Intel Arc A770M Comparison
Intel Arc A750
Arc A770M
PERFORMANCE BENCHMARKS
Analysis: Intel Arc A750 vs Intel Arc A770M
FAQ
Q: Which GPU has the higher average benchmark score?
A: The Intel Arc A750 scores 20,582, while the Intel Arc A770M scores 18,383. That places the A750 in the 66th percentile of all GPUs, while the A770M sits in the 62nd percentile.
Q: How far apart are the two in raw compute performance?
A: The Arc A750 reaches 17.20 TFLOPS FP32, slightly above the A770M's 16.79 TFLOPS FP32. The A750 also holds a 10.1% lead in Geekbench OpenCL (98,554 versus 89,494) and a 15.1% lead in Geekbench Vulkan (85,631 versus 74,422).
Q: Do the two cards have the same memory bandwidth?
A: Yes, both use GDDR6 memory with a 256-bit bus and 512.0 GB/s bandwidth. The difference is capacity: the A750 has 8 GB, while the A770M has 16 GB.
Q: Which GPU is more power-efficient?
A: The Intel Arc A770M has a 120 W TDP, compared to the A750's 225 W TDP. The A770M achieves nearly the same FP32 throughput at a much lower power envelope.
Q: Are they based on the same architecture?
A: Yes, both use the Xe-HPG architecture, the DG2-512 chip, a 6 nm TSMC process, and 21,700 million transistors on a 406 mm² die.
Q: Which GPU wins more benchmark comparisons?
A: The Intel Arc A750 wins all 10 head-to-head tests. Its biggest margins are 16.1% in Passmark DirectX 10 and 15.1% in Geekbench Vulkan, with only a 0% delta in Passmark DirectX 12 (a tie at 70).
Architecture Differences
Both GPUs share the DG2-512 silicon, Xe-HPG architecture, 6 nm TSMC process, 21,700 million transistors, and a 406 mm² die. The transistor density is identical at 53.4M per mm². From there, the two diverge in configuration and implementation.
The Arc A750 is the desktop variant, built as a dual-slot card with a PCIe 4.0 x16 interface. It draws 225 W and requires both a 6-pin and an 8-pin power connector, with a suggested PSU of 550 W. Its display outputs include 1x HDMI 2.1 and 3x DisplayPort 2.0.
The Arc A770M is the mobile part, packaged as an IGP (integrated graphics processor) with no slot width, no power connectors, and no suggested PSU. Its display outputs are listed as portable device dependent. The A770M runs at a 120 W TDP, which is 105 W lower than the A750.
Clock speeds differ notably. The A750 has a base clock of 2050 MHz and a boost of 2400 MHz. The A770M drops to 1650 MHz base and 2050 MHz boost. Despite the lower clocks, the A770M has a wider execution configuration: 4096 shading units, 256 TMUs, 128 ROPs, and 32 ray tracing cores. The A750 has 3584 shading units, 224 TMUs, 112 ROPs, and 28 ray tracing cores.
The extra hardware on the A770M does not translate to higher throughput. The A750 records 268.8 GPixel/s pixel rate and 537.6 GTexel/s texture rate, versus 262.4 GPixel/s and 524.8 GTexel/s for the A770M. The same story holds for FP32 and FP16: the A750 posts 17.20 TFLOPS and 34.41 TFLOPS (2:1), while the A770M posts 16.79 TFLOPS and 33.59 TFLOPS (2:1). The A770M's higher core count is offset by lower clocks, leaving the A750 with the edge in every measured throughput category.
Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Memory is GDDR6 on a 256-bit bus with 512.0 GB/s bandwidth for both. The A750 ships with 8 GB; the A770M doubles that to 16 GB.
The A750 is marked as end-of-life with a release date of October 11, 2022, a predecessor of Xe Graphics, and a successor of Battlemage. The A770M has no recorded release date, predecessor, or successor in the database, but it is also end-of-life.
The Verdict
The data points to a clear winner for raw performance: the Intel Arc A750. It leads in average benchmark score (20,582 versus 18,383), percentile rank (66th versus 62nd), and wins all 10 head-to-head comparisons. The largest margins are 16.1% in Passmark DirectX 10 and 15.1% in Geekbench Vulkan, with a 14.7% lead in 3DMark Steel Nomad DX12. For any workload that stresses shader throughput, DirectX 10/11/9, or Vulkan, the A750 is the stronger pick.
The Intel Arc A770M is not without reason to exist. It delivers 16 GB of memory versus the A750's 8 GB, which matters for large datasets or texture-heavy workloads. It also draws 120 W versus 225 W, making it the more power-conscious choice in a portable context. The A770M's nearest rivals include the AMD Radeon RX 460 (0.1% behind) and the NVIDIA GeForce RTX 3060 Mobile (1.2% behind), so it sits in the midrange mobile segment.
The choice comes down to context. If the priority is peak performance in a desktop system, the A750 is the answer. If the priority is memory capacity and power efficiency in a mobile platform, the A770M is the only option with 16 GB and a 120 W envelope. The benchmark data does not show a single scenario where the A770M outperforms the A750, but its lower TDP and double memory capacity are structural advantages that no benchmark delta can erase.
Specification Differences
| Specification | Intel Arc A750 | Intel Arc A770M |
| --- | --- | --- |
| Base Clock | 2050 MHz | 1650 MHz |
| Boost Clock | 2400 MHz | 2050 MHz |
| Memory Size | 8 GB | 16 GB |
| Shading Units | 3584 | 4096 |
| TMUs | 224 | 256 |
| ROPs | 112 | 128 |
| Ray Tracing Cores | 28 | 32 |
| Pixel Rate | 268.8 GPixel/s | 262.4 GPixel/s |
| Texture Rate | 537.6 GTexel/s | 524.8 GTexel/s |
| FP32 | 17.20 TFLOPS | 16.79 TFLOPS |
| FP16 | 34.41 TFLOPS (2:1) | 33.59 TFLOPS (2:1) |
| TDP | 225 W | 120 W |
| Slot Width | Dual-slot | IGP |
| Power Connectors | 1x 6-pin + 1x 8-pin | None |
| Suggested PSU | 550 W | None |
| Display Outputs | 1x HDMI 2.1, 3x DisplayPort 2.0 | Portable Device Dependent |
Identical between the two: DG2-512 chip, Xe-HPG architecture, 6 nm process, TSMC foundry, 21,700 million transistors, 406 mm² die, 53.4M/mm² transistor density, 2000 MHz memory clock with 16 Gbps effective, GDDR6 type, 256-bit bus, 512.0 GB/s bandwidth, PCIe 4.0 x16 interface, DirectX 12 Ultimate (12_2), OpenGL 4.6, Vulkan 1.4, and end-of-life production status.
Head-to-Head Benchmarks
The Arc A750 wins every recorded comparison. The strongest margin is in Passmark DirectX 10, where the A750 scores 65 against the A770M's 56, a 16.1% gap. That result signals a substantial lead in legacy DirectX 10 workloads, not just modern APIs.
Geekbench Vulkan shows a 15.1% advantage: 85,631 versus 74,422. The A750's Vulkan result is its second-largest win and demonstrates strong driver performance in cross-platform compute and graphics. 3DMark Steel Nomad DX12 follows at 14.7%, with scores of 2612 and 2278. This is a modern DirectX 12 test, and the A750's lead here reinforces that the desktop card is not merely faster in older APIs.
Passmark GPU Compute gives the A750 a 12.3% win (5368 versus 4778). Geekbench OpenCL shows a 10.1% margin (98,554 versus 89,494). Both tests point to the A750's higher FP32 throughput and clock advantage being the deciding factor.
Smaller wins include Passmark G3D at 6.5% (12,534 versus 11,774) and Passmark DirectX 11 at 4.3% (72 versus 69). Passmark G2D is 3% higher (732 versus 711), and Passmark DirectX 9 is 1.7% higher (181 versus 178). The only tie is Passmark DirectX 12 at 70 for both, a 0% delta.
The A750's average benchmark score of 20,582 places it 0.1% above the Intel Arc B570 (20,556) and 0.2% above the NVIDIA GeForce RTX 3070 Mobile (20,534). The A770M's average of 18,383 sits 0.1% above the AMD Radeon RX 460 (18,373) and 1.2% above the NVIDIA GeForce RTX 3060 Mobile (18,159). These nearest-rival comparisons show the A750 competing with higher-tier mobile GPUs, while the A770M aligns with midrange parts.
Where Each One Wins
The Intel Arc A750 wins in every performance metric recorded. It is the choice for anyone prioritizing raw speed in DirectX 10, Vulkan, OpenCL, or 3DMark Steel Nomad. Its 10.1% OpenCL lead and 15.1% Vulkan lead make it suited for compute-heavy applications that use those APIs. The 14.7% margin in 3DMark Steel Nomad DX12 indicates strong modern gaming performance relative to the A770M. The 16.1% DirectX 10 win is the most decisive split, so legacy DirectX 10 titles or workloads will show the largest gap.
The Intel Arc A770M wins in two structural categories that benchmarks do not capture: memory capacity and power draw. It offers 16 GB of GDDR6 versus the A750's 8 GB, with identical 512.0 GB/s bandwidth. That makes it the better fit for tasks that exceed 8 GB of VRAM, such as large texture sets or compute buffers. Its 120 W TDP is 105 W lower than the A750's 225 W, which makes it the only viable option for portable or power-limited systems. The A770M also has more shading units (4096 versus 3584), TMUs (256 versus 224), ROPs (128 versus 112), and ray tracing cores (32 versus 28), but the A750's higher clocks overcome that hardware advantage in every recorded test.
For mobile users, the A770M is the sole choice, as the A750 requires dual-slot cooling, external power connectors, and a 550 W PSU. For desktop users who can supply 225 W and have no memory capacity concerns, the A750 is the stronger performer across all benchmark categories. The A770M's 16 GB frame buffer is its one clear advantage, and the A750's 17.20 TFLOPS FP32 versus 16.79 TFLOPS FP32 is a modest but consistent lead. The data does not show a scenario where the A770M beats the A750 in speed, so the decision hinges on form factor, power, and memory requirements.