Intel Iris Xe MAX Graphics vs NVIDIA T400 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA T400 has an average benchmark score of 16508, while the Intel Iris Xe MAX Graphics scores 14315. The T400 sits in the 60th percentile of all GPUs, compared to the 56th percentile for the Intel part.

Q: How large is the performance gap in the shared benchmark?

A: In the Geekbench OpenCL test, the NVIDIA T400 scores 17039 against 14315 for the Intel Iris Xe MAX Graphics, a 19% difference in favor of the T400.

Q: What are the memory configurations of these two cards?

A: The NVIDIA T400 has 2 GB of GDDR6 memory on a 64-bit bus with 80.00 GB/s bandwidth. The Intel Iris Xe MAX Graphics has 4 GB of LPDDR4X memory on a 128-bit bus with 68.26 GB/s bandwidth.

Q: Which GPU offers higher raw compute throughput?

A: The Intel Iris Xe MAX Graphics has higher FP32 performance at 2.534 TFLOPS, compared to 1,094.4 GFLOPS for the NVIDIA T400. The Intel part also leads in texture rate at 79.20 GTexel/s versus 34.20 GTexel/s.

Q: Which GPU has a higher boost clock?

A: The Intel Iris Xe MAX Graphics boosts to 1650 MHz, while the NVIDIA T400 boosts to 1425 MHz. The base clocks are 300 MHz for Intel and 420 MHz for NVIDIA.

Q: What are the closest rivals to each GPU according to the database?

A: The NVIDIA T400's nearest rivals include the NVIDIA GeForce RTX 5090 D V2 (delta 0%), AMD Radeon PRO W7500 (delta 0.6%), and NVIDIA RTX PRO 6000 Blackwell (delta 0.6%). The Intel Iris Xe MAX Graphics is closest to the AMD Radeon Vega 11 (delta -0.3%) and NVIDIA GeForce GTX 1070 Ti (delta 0.3%).

Architecture Differences

The NVIDIA T400 is built on the Turing architecture using the TU117 chip, manufactured by TSMC on a 12 nm process. The die measures 200 mm² and contains 4,700 million transistors, yielding a transistor density of 23.5M per mm². It belongs to the Quadro Turing generation and is the successor to Quadro Volta, with Workstation Ampere as its successor. The card is described as end-of-life, released on 2021-05-05.

The Intel Iris Xe MAX Graphics uses the DG1 chip based on Generation 12.1 architecture, part of the Xe Graphics generation. It is fabricated by Intel on a 10 nm process, with a die size of 95 mm². The database lists no transistor count or density for this part. It was released on 2020-10-30, is end-of-life, and its successor is Alchemist.

The process node difference is significant: Intel's 10 nm versus TSMC's 12 nm. The Intel die is substantially smaller at 95 mm² versus 200 mm², which reflects the simpler integration as an IGP (integrated graphics processor). The T400 is a discrete single-slot card with three mini-DisplayPort 1.4a outputs, while the Iris Xe MAX has no display outputs, indicating it is designed to work alongside another GPU.

Both GPUs support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. Neither has ray tracing cores or tensor cores listed. The shading unit counts differ sharply: the Intel part has 768 shading units, 48 TMUs, and 24 ROPs, while the T400 has 384 shading units, 24 TMUs, and 16 ROPs. The Intel architecture doubles the shading hardware, yet the benchmark results show it does not translate into a win in the measured OpenCL test.

Where Each One Wins

The NVIDIA T400 wins the only head-to-head benchmark in the database, the Geekbench OpenCL test, with a 19% margin. This gives it 1 win versus 0 for the Intel part. The T400 also holds a higher average benchmark score (16508 versus 14315) and a better percentile rank (60th versus 56th).

The Intel Iris Xe MAX Graphics wins in several architectural categories despite losing the benchmark. It has higher FP32 throughput at 2.534 TFLOPS, more than double the T400's 1,094.4 GFLOPS. It also has higher pixel rate (39.60 GPixel/s versus 22.80 GPixel/s) and texture rate (79.20 GTexel/s versus 34.20 GTexel/s). The Intel part doubles the memory capacity (4 GB versus 2 GB) and has a wider 128-bit bus compared to 64-bit, though its bandwidth is lower at 68.26 GB/s versus 80.00 GB/s.

The T400 wins on memory bandwidth, offering 80.00 GB/s from GDDR6 memory running at 1250 MHz (10 Gbps effective), versus 68.26 GB/s from LPDDR4X at 2133 MHz (4.3 Gbps effective). The T400 also has a lower TDP at 30 W versus 25 W, though both have the same suggested PSU of 200 W.

In practical terms, the T400 is the stronger performer in compute workloads as measured by Geekbench OpenCL. The Iris Xe MAX offers higher theoretical compute but the database shows it trails in the actual test. The T400's discrete nature with dedicated display outputs makes it suitable for workstation tasks, while the Intel part's no-output design points to a secondary or embedded role.

Specification Differences

| Specification | NVIDIA T400 | Intel Iris Xe MAX Graphics |

|---|---|---|

| Architecture | Turing | Generation 12.1 |

| Chip | TU117 | DG1 |

| Process Node | 12 nm | 10 nm |

| Die Size | 200 mm² | 95 mm² |

| Transistors | 4,700 million | Not listed |

| Base Clock | 420 MHz | 300 MHz |

| Boost Clock | 1425 MHz | 1650 MHz |

| Memory Size | 2 GB | 4 GB |

| Memory Type | GDDR6 | LPDDR4X |

| Memory Bus Width | 64 bit | 128 bit |

| Memory Bandwidth | 80.00 GB/s | 68.26 GB/s |

| Shading Units | 384 | 768 |

| TMUs | 24 | 48 |

| ROPs | 16 | 24 |

| Pixel Rate | 22.80 GPixel/s | 39.60 GPixel/s |

| Texture Rate | 34.20 GTexel/s | 79.20 GTexel/s |

| FP32 | 1,094.4 GFLOPS | 2.534 TFLOPS |

| FP16 | 2.189 TFLOPS (2:1) | 5.069 TFLOPS (2:1) |

| TDP | 30 W | 25 W |

| Slot Width | Single-slot | IGP |

| Power Connectors | None | Not listed |

| Bus Interface | PCIe 3.0 x16 | PCIe 4.0 x8 |

| Display Outputs | 3x mini-DisplayPort 1.4a | No outputs |

| Release Date | 2021-05-05 | 2020-10-30 |

The Intel part has double the shading units, TMUs, and ROPs. It also has a 128-bit memory bus versus 64-bit. However, the T400 compensates with faster GDDR6 memory and higher bandwidth. The T400 uses PCIe 3.0 x16 while the Intel part uses PCIe 4.0 x8. The T400 is a single-slot card with power connectors listed as none, while the Intel part is an IGP with no outputs.

Head-to-Head Benchmarks

The only head-to-head result in the database is the Geekbench OpenCL test. The NVIDIA T400 scores 17039, and the Intel Iris Xe MAX Graphics scores 14315. The delta is 19% in favor of the T400. This is a decisive win for NVIDIA, giving it all 1 of the 1 recorded head-to-head wins.

The T400's average benchmark score of 16508 is higher than its OpenCL score, suggesting consistency across tests. The Intel part's average score equals its only benchmark result at 14315, meaning no other tests are recorded for it.

The nearest rival data places the T400 in a competitive cluster. Its avg score of 16508 is within 0.9% of the AMD Radeon RX 5700 XT (16361) and within 0.6% of both the AMD Radeon PRO W7500 (16415) and NVIDIA RTX PRO 6000 Blackwell (16408). The Intel Iris Xe MAX Graphics sits close to the AMD Radeon Vega 11 (14352, delta -0.3%) and the NVIDIA GeForce GTX TITAN (14373, delta -0.4%), with a slight edge over the NVIDIA GeForce GTX 1070 Ti (14277, delta 0.3%).

The 19% delta between the two cards is substantial. The T400 is closer to the RTX 5090 D V2 (16504, delta 0%) than it is to the Intel part. The Intel Iris Xe MAX is nearly 2000 points behind the T400's OpenCL score, which places it in a lower performance tier alongside older Vega and GTX 10-series parts.

The Verdict

The data clearly favors the NVIDIA T400 in actual benchmark performance. It wins the only direct comparison by 19%, holds a higher average score (16508 versus 14315), and ranks in a better percentile (60th versus 56th). For any workload measured by Geekbench OpenCL, the T400 is the stronger choice.

The Intel Iris Xe MAX Graphics does offer advantages on paper: higher FP32 compute (2.534 TFLOPS), more memory (4 GB), and a wider memory bus (128-bit). However, these specifications do not translate into a benchmark win. The T400's GDDR6 memory at 80.00 GB/s bandwidth appears to provide a more effective memory subsystem despite the smaller 2 GB capacity and 64-bit bus.

For users needing a discrete workstation GPU with display outputs (three mini-DisplayPort 1.4a), the NVIDIA T400 is the appropriate pick. It is a single-slot card with no power connectors and a 30 W TDP, making it easy to integrate. The Intel part, with no outputs and an IGP form factor, is not a direct replacement for a display adapter.

The choice depends on the use case. If the requirement is a functioning GPU with OpenCL performance, the T400 leads decisively. If the requirement is raw compute density with double the FP32 and FP16 throughput, the Intel part has the architectural edge, but the database shows no benchmark to confirm that advantage in practice. The recorded data indicates the T400 is the better performer, and the verdict follows the measurements: choose the NVIDIA T400 for benchmark-verified performance, choose the Intel Iris Xe MAX only if its 4 GB memory and higher theoretical compute are more important than the recorded 19% deficit in OpenCL.