NVIDIA GeForce RTX 3090 vs NVIDIA T600 Mobile Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA T600 Mobile has an average benchmark score of 32,849, while the NVIDIA GeForce RTX 3090 has an average score of 27,565. The T600 Mobile sits at the 77th percentile among all GPUs, while the RTX 3090 is at the 73rd percentile.

Q: How far apart are the two in the Geekbench OpenCL test?

A: The RTX 3090 scores 172,758 points versus 35,486 for the T600 Mobile, a difference of 79.5% in favor of the RTX 3090. This is the largest margin in any recorded head-to-head benchmark.

Q: In which test does the T600 Mobile come closer to the RTX 3090?

A: The Vulkan test shows a narrower gap. The RTX 3090 scores 53,927, while the T600 Mobile scores 30,211, a delta of 44%. This still favors the RTX 3090, but the relative difference is much smaller than in OpenCL.

Q: What are the nearest rivals to the T600 Mobile in the database?

A: The closest competitors include the NVIDIA T550 Mobile, with an average score of 33,161 and a delta of -0.9%, the NVIDIA P104-100 at 32,982 (-0.4%), the AMD Radeon RX 590 GME at 32,601 (0.8%), and the AMD FirePro S9300 X2 at 32,540 (0.9%).

Q: Which GPUs sit closest to the RTX 3090 in average score?

A: The AMD Radeon RX 7800M is 1.1% ahead, the AMD Radeon Pro Vega 20 is 1% ahead, the NVIDIA GeForce RTX 4070 Mobile is 0.5% ahead, and the AMD Radeon RX 6700 XT is 0.5% ahead.

Q: What is the RTX 3090's launch MSRP, and does the T600 Mobile have one?

A: The RTX 3090 has a launch MSRP of 1,499 USD. The T600 Mobile has no launch MSRP recorded in the database.

Architecture Differences

The two GPUs belong to different NVIDIA generations and target entirely different product segments. The T600 Mobile uses the TU117 chip, built on the Turing architecture, and is part of the Quadro Turing-M generation. The RTX 3090 uses the GA102 chip, built on the Ampere architecture, and belongs to the GeForce 30 series.

The manufacturing process differs substantially. The T600 Mobile is fabricated on a 12 nm process at TSMC, with 4,700 million transistors on a 200 mm² die. The RTX 3090 is built on an 8 nm process at Samsung, packing 28,300 million transistors into a 628 mm² die. This results in a transistor density of 23.5 million per square millimeter for the T600 Mobile versus 45.1 million per square millimeter for the RTX 3090.

The memory subsystems are completely different. The T600 Mobile has 4 GB of GDDR6 memory on a 128-bit bus, delivering 192.0 GB/s of bandwidth. The RTX 3090 has 24 GB of GDDR6X memory on a 384-bit bus, providing 936.2 GB/s of bandwidth. The memory clock rates also differ: the T600 Mobile runs at 1500 MHz (12 Gbps effective), while the RTX 3090 runs at 1219 MHz (19.5 Gbps effective).

The compute resources are dramatically different. The T600 Mobile has 896 shading units, 56 texture mapping units, and 32 ROPs. It has no dedicated RT cores or tensor cores. The RTX 3090 has 10,496 shading units, 328 TMUs, 112 ROPs, 82 RT cores, and 328 tensor cores. The pixel rate is 45.12 GPixel/s for the T600 Mobile versus 189.8 GPixel/s for the RTX 3090. The texture rate is 78.96 GTexel/s versus 556.0 GTexel/s.

The API support also differs. The T600 Mobile supports DirectX 12 (12_1), while the RTX 3090 supports DirectX 12 Ultimate (12_2). Both support OpenGL 4.6 and Vulkan 1.4.

The physical and power characteristics are polar opposites. The T600 Mobile has a TDP of 40 W, is an IGP slot width, and requires no power connectors. The RTX 3090 has a TDP of 350 W, is a triple-slot card, uses a 1x 12-pin power connector, and requires a 750 W suggested PSU. The RTX 3090 is also physically large at 336 mm in length, 140 mm in height, and 61 mm in width. The T600 Mobile has no recorded dimensions, as it is portable device dependent.

The bus interfaces differ as well. The T600 Mobile uses PCIe 3.0 x16, while the RTX 3090 uses PCIe 4.0 x16. The display outputs are portable-device dependent for the T600 Mobile, while the RTX 3090 offers 1x HDMI 2.1 and 3x DisplayPort 1.4a.

The Verdict

The data shows a clear split between these two products. The T600 Mobile is a low-power, mobile-focused GPU that appears in the database with a higher average benchmark score than the RTX 3090. That result is driven by the specific benchmark suites used for each card. The RTX 3090, however, wins both recorded head-to-head comparisons by wide margins.

For users who need raw compute throughput and memory bandwidth, the RTX 3090 is the only rational choice. It has 6 times the memory capacity, 4.9 times the bandwidth, and 14 times the FP32 throughput of the T600 Mobile. The RTX 3090 also has 82 RT cores and 328 tensor cores, which the T600 Mobile lacks entirely. Any workload involving ray tracing or AI acceleration will go to the RTX 3090.

For users who require a low-power, compact solution that can be integrated into a portable device, the T600 Mobile is the relevant product. Its 40 W TDP, IGP slot width, and lack of external power connectors make it suitable for systems where the RTX 3090's 350 W TDP and triple-slot footprint are impossible.

The benchmark data itself suggests that the T600 Mobile performs admirably for its class. Its average score of 32,849 places it above the RTX 3090's average of 27,565. However, this is misleading for direct comparison because the two GPUs are not tested in the same workloads. The only common tests are in head-to-head benchmarks, and the RTX 3090 wins both. The T600 Mobile's advantage in average score is likely due to the specific mix of tests in the database, not to an actual performance advantage.

The verdict is straightforward. The RTX 3090 is for high-performance desktop compute and gaming. The T600 Mobile is for low-power mobile systems where the RTX 3090 cannot physically be installed. The data does not support choosing the T600 Mobile over the RTX 3090 for any performance reason.

Specification Differences

The two GPUs differ in nearly every specification category.

| Specification | T600 Mobile | RTX 3090 |

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

| Chip | TU117 | GA102 |

| Architecture | Turing | Ampere |

| Process node | 12 nm | 8 nm |

| Foundry | TSMC | Samsung |

| Transistors | 4,700 million | 28,300 million |

| Die size | 200 mm² | 628 mm² |

| Transistor density | 23.5M / mm² | 45.1M / mm² |

| Base clock | 780 MHz | 1395 MHz |

| Boost clock | 1410 MHz | 1695 MHz |

| Memory clock | 1500 MHz (12 Gbps effective) | 1219 MHz (19.5 Gbps effective) |

| Memory size | 4 GB | 24 GB |

| Memory type | GDDR6 | GDDR6X |

| Memory bus | 128 bit | 384 bit |

| Memory bandwidth | 192.0 GB/s | 936.2 GB/s |

| Shading units | 896 | 10,496 |

| TMUs | 56 | 328 |

| ROPs | 32 | 112 |

| RT cores | None | 82 |

| Tensor cores | None | 328 |

| Pixel rate | 45.12 GPixel/s | 189.8 GPixel/s |

| Texture rate | 78.96 GTexel/s | 556.0 GTexel/s |

| FP32 | 2.527 TFLOPS | 35.58 TFLOPS |

| FP16 | 5.053 TFLOPS (2:1) | 35.58 TFLOPS (1:1) |

| TDP | 40 W | 350 W |

| Slot width | IGP | Triple-slot |

| Power connectors | None | 1x 12-pin |

| Suggested PSU | None | 750 W |

| Bus interface | PCIe 3.0 x16 | PCIe 4.0 x16 |

| DirectX | 12 (12_1) | 12 Ultimate (12_2) |

| Release date | 2021-04-11 | 2020-08-31 |

Head-to-Head Benchmarks

The head-to-head data contains only two common tests: Geekbench OpenCL and Geekbench Vulkan. The RTX 3090 wins both.

In the Geekbench OpenCL test, the RTX 3090 scores 172,758, while the T600 Mobile scores 35,486. The delta is 79.5% in favor of the RTX 3090. This is the largest performance gap in the recorded data. The T600 Mobile performs at only about 20% of the RTX 3090's level in this test.

In the Geekbench Vulkan test, the RTX 3090 scores 53,927, while the T600 Mobile scores 30,211. The delta is 44% in favor of the RTX 3090. This is a smaller gap than OpenCL, but still a decisive victory for the RTX 3090. The T600 Mobile performs at about 56% of the RTX 3090's level in Vulkan.

The Win count is 2 for the RTX 3090 and 0 for the T600 Mobile. The data is one-sided. No benchmark in the head-to-head set shows the T600 Mobile ahead.

The RTX 3090's advantage in OpenCL is far larger than its advantage in Vulkan. This suggests the RTX 3090's compute resources are particularly impactful in OpenCL workloads, while the Vulkan test may be more balanced by other factors.

Where Each One Wins

The RTX 3090 wins in every recorded comparison. It wins the OpenCL test by 79.5%, which is the largest margin. It wins the Vulkan test by 44%. Its FP32 throughput of 35.58 TFLOPS is 14 times that of the T600 Mobile's 2.527 TFLOPS. Its memory bandwidth is 4.9 times higher. Its texture rate is 7 times higher. Its pixel rate is 4.2 times higher.

The RTX 3090 is the only option for workloads that use RT cores or tensor cores. It has 82 RT cores and 328 tensor cores. The T600 Mobile has none. Any ray tracing, AI inference, or deep learning workload will go to the RTX 3090. The RTX 3090 also has 24 GB of memory versus 4 GB, which is essential for large datasets.

The T600 Mobile has no head-to-head wins. However, its average benchmark score of 32,849 is higher than the RTX 3090's 27,565. This is not a head-to-head result, but it does suggest that the T600 Mobile performs well in the specific tests that contribute to its average. The T600 Mobile also has a 77th percentile ranking versus the RTX 3090's 73rd percentile.

The T600 Mobile wins in practical deployment scenarios. It has a 40 W TDP, no power connectors, and an IGP slot width. It can be installed in portable devices. The RTX 3090 has a 350 W TDP, requires a 750 W PSU, and takes up a triple-slot footprint. The T600 Mobile is the only one of the two that can be used in a laptop or other compact system.

For pure compute and graphics performance, the RTX 3090 is the clear winner. For mobile integration and power efficiency, the T600 Mobile is the only choice. The data shows no scenario where the T600 Mobile outperforms the RTX 3090 in raw performance, but the T600 Mobile's existence is justified by its form factor and power constraints.