NVIDIA GeForce RTX 4070 AD103 vs Lisuan Tech LX ULTRA Comparison

Where Each One Wins

The recorded data shows a split personality between these two cards. The NVIDIA GeForce RTX 4070 AD103 is built around a 192-bit memory interface with 12 GB of GDDR6X memory, delivering 504.2 GB/s of bandwidth. The Lisuan Tech LX ULTRA uses the same 192-bit bus width but carries 24 GB of GDDR6 memory at 432.0 GB/s. That bandwidth deficit is the LX ULTRA's primary structural weakness, while its 24 GB capacity is its clearest advantage.

For rasterization throughput, the LX ULTRA posts a higher pixel rate of 192.0 GPixel/s against the RTX 4070 AD103's 158.4 GPixel/s. The LX ULTRA also fields 96 ROPs versus 64 on the NVIDIA card, a 50% advantage in pixel processing hardware. Texture rate tells the opposite story: the RTX 4070 AD103 reaches 455.4 GTexel/s while the LX ULTRA sits at 384.0 GTexel/s, a 18.6% gap in favor of NVIDIA. The FP32 compute race goes to NVIDIA as well, 29.15 TFLOPS versus 24.58 TFLOPS, a 18.6% lead.

The LX ULTRA wins in FP16 throughput outright. Its 49.15 TFLOPS at 2:1 ratio doubles its FP32 output, whereas the RTX 4070 AD103 delivers 29.15 TFLOPS at a 1:1 ratio. That means the LX ULTRA holds a 68.6% advantage in half-precision compute, a decisive margin for workloads that use FP16 accumulation. The RTX 4070 AD103 counters with 184 tensor cores and 46 RT cores, hardware the LX ULTRA does not list at all. Ray tracing and AI-accelerated workloads therefore fall entirely to NVIDIA's card by default.

Memory capacity is the LX ULTRA's other clear win. Its 24 GB doubles the RTX 4070 AD103's 12 GB, which matters for large datasets, high-resolution textures, or multi-application workflows. Bandwidth still favors NVIDIA at 504.2 GB/s versus 432.0 GB/s, a 16.7% advantage. The LX ULTRA's memory clock runs at 2250 MHz (18 Gbps effective) against the RTX 4070 AD103's 1313 MHz (21 Gbps effective), so the NVIDIA card uses faster GDDR6X to overcome the capacity deficit.

Power draw differs modestly. The RTX 4070 AD103 is rated at 200 W TDP, the LX ULTRA at 225 W. Both use a single 16-pin connector and both recommend a 550 W PSU. The LX ULTRA runs hotter in terms of rated power but delivers more pixel throughput per watt? No, that is not derivable from the data. What the data shows is a 25 W TDP gap in favor of NVIDIA.

Architecture Differences

The RTX 4070 AD103 uses the AD103 chip built on TSMC's 5 nm process, part of the Ada Lovelace architecture from the GeForce 40 generation. It packs 45,900 million transistors into a 379 mm² die, giving a transistor density of 121.1M per mm². The LX ULTRA uses the 7G105 chip on TSMC's 6 nm node, part of the TrueGPU architecture in the 7G100 generation. Its transistor count and die size are not recorded in the database.

Shading unit counts are close: 5888 on the RTX 4070 AD103 versus 6144 on the LX ULTRA, a 4.3% advantage for the latter. TMU counts follow the same pattern, 184 versus 192, again favoring the LX ULTRA by 4.3%. ROPs diverge further, 64 on NVIDIA versus 96 on Lisuan, a 50% gap. The LX ULTRA's higher ROP count directly explains its higher pixel rate.

The RTX 4070 AD103 includes dedicated ray tracing cores (46 of them) and tensor cores (184), while the LX ULTRA lists neither. That is a structural difference, not a minor spec gap. NVIDIA's card is built for RT and DLSS-style acceleration; the LX ULTRA has no such hardware in the recorded data. API support also differs: the RTX 4070 AD103 supports Vulkan 1.4, the LX ULTRA supports Vulkan 1.3. Both support DirectX 12 Ultimate (12_2) and OpenGL 4.6.

Memory technology separates the two clearly. The RTX 4070 AD103 uses GDDR6X, the LX ULTRA uses GDDR6. Both run on a 192-bit bus, but the GDDR6X modules on the NVIDIA card reach 21 Gbps effective versus 18 Gbps on the LX ULTRA. That explains the bandwidth gap. The LX ULTRA compensates with twice the capacity.

Physical dimensions differ. The RTX 4070 AD103 measures 240 mm in length, 110 mm in height, 40 mm in width. The LX ULTRA is longer at 268 mm, slightly taller at 112 mm, same width at 40 mm. Both are dual-slot cards. The LX ULTRA offers four DisplayPort 1.4a outputs; the RTX 4070 AD103 offers one HDMI 2.1 and three DisplayPort 1.4a outputs.

Production status separates them further. The RTX 4070 AD103 is end-of-life, released on 2024-02-29, with the GeForce 30 as predecessor and GeForce 50 as successor. The LX ULTRA is active, released on 2026-03-16, with no predecessor or successor recorded. The NVIDIA card has a launch MSRP of 599 USD; the LX ULTRA has no recorded launch MSRP.

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark runs between these two cards. The wins and benchmark arrays are empty, and the average benchmark score for both is zero. That means the comparison must rest on the recorded specification data and the derived performance metrics.

The largest numerical advantage belongs to the LX ULTRA in FP16 compute. Its 49.15 TFLOPS at 2:1 ratio exceeds the RTX 4070 AD103's 29.15 TFLOPS at 1:1 by 68.6%. That is the single biggest performance gap in either direction. For any workload that uses FP16 arithmetic, the LX ULTRA is substantially ahead.

The RTX 4070 AD103 takes the FP32 crown at 29.15 TFLOPS versus 24.58 TFLOPS, a 18.6% lead. Texture rate also favors NVIDIA: 455.4 GTexel/s versus 384.0 GTexel/s, again 18.6%. Bandwidth favors NVIDIA at 504.2 GB/s versus 432.0 GB/s, a 16.7% margin. The LX ULTRA leads in pixel rate at 192.0 GPixel/s versus 158.4 GPixel/s, a 21.2% advantage.

The ROP difference explains the pixel rate gap. With 96 ROPs against 64, the LX ULTRA has 50% more pixel-processing units. The shading unit and TMU counts are nearly even, so the FP32 and texture rate gaps come from clock speeds. The RTX 4070 AD103 boosts to 2475 MHz with a 1920 MHz base; the LX ULTRA has no base or boost clocks recorded, so its 24.58 FP32 TFLOPS with 6144 shading units implies a lower operating frequency than the NVIDIA card.

Memory latency and capacity trade-offs are visible in the specs. The LX ULTRA's 24 GB capacity is double the RTX 4070 AD103's 12 GB, but its bandwidth is 16.7% lower. For workloads that fit within 12 GB, the NVIDIA card moves data faster. For workloads that exceed 12 GB, the LX ULTRA avoids paging or spills that would cripple the smaller card.

Ray tracing hardware exists only on the RTX 4070 AD103. The 46 RT cores and 184 tensor cores have no counterpart on the LX ULTRA. Any ray-traced scene, any AI upscaling, any tensor-core workload runs only on the NVIDIA card. The LX ULTRA's FP16 throughput might assist some AI tasks, but the dedicated tensor cores on the RTX 4070 AD103 are specialized silicon for that purpose.

Power efficiency is not directly recorded, but TDP figures are. The RTX 4070 AD103 draws 200 W, the LX ULTRA draws 225 W. The NVIDIA card delivers higher FP32 and texture throughput at 25 W lower power, indicating better efficiency per watt in those domains. The LX ULTRA delivers higher pixel and FP16 throughput at the higher power envelope.

FAQ

Q: Which card has more memory?

A: The Lisuan Tech LX ULTRA has 24 GB of GDDR6 memory, exactly double the 12 GB of GDDR6X on the NVIDIA GeForce RTX 4070 AD103.

Q: Which card has higher memory bandwidth?

A: The NVIDIA GeForce RTX 4070 AD103 reaches 504.2 GB/s with its GDDR6X memory at 21 Gbps effective, while the Lisuan Tech LX ULTRA delivers 432.0 GB/s from GDDR6 at 18 Gbps effective, a 16.7% gap.

Q: Does the Lisuan Tech LX ULTRA support ray tracing?

A: The database records no ray tracing cores for the LX ULTRA. The NVIDIA GeForce RTX 4070 AD103 lists 46 RT cores and 184 tensor cores, so ray tracing and tensor workloads are only supported by the NVIDIA card.

Q: Which card has higher FP32 compute?

A: The NVIDIA GeForce RTX 4070 AD103 delivers 29.15 TFLOPS of FP32, while the Lisuan Tech LX ULTRA delivers 24.58 TFLOPS, a 18.6% advantage for NVIDIA.

Q: Which card has higher FP16 compute?

A: The Lisuan Tech LX ULTRA reaches 49.15 TFLOPS at a 2:1 ratio, versus 29.15 TFLOPS at 1:1 on the NVIDIA GeForce RTX 4070 AD103, a 68.6% lead for the LX ULTRA.

Q: What is the power draw difference?

A: The NVIDIA GeForce RTX 4070 AD103 is rated at 200 W TDP, the Lisuan Tech LX ULTRA at 225 W TDP. Both use a single 16-pin connector and recommend a 550 W PSU.

Specification Differences

| Specification | NVIDIA GeForce RTX 4070 AD103 | Lisuan Tech LX ULTRA |

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

| Chip | AD103 | 7G105 |

| Architecture | Ada Lovelace | TrueGPU |

| Generation | GeForce 40 | 7G100 |

| Process node | 5 nm | 6 nm |

| Transistors | 45,900 million | Unknown |

| Die size | 379 mm² | Unknown |

| Transistor density | 121.1M / mm² | Not recorded |

| Base clock | 1920 MHz | Not recorded |

| Boost clock | 2475 MHz | Not recorded |

| Memory clock | 1313 MHz, 21 Gbps effective | 2250 MHz, 18 Gbps effective |

| Memory size | 12 GB | 24 GB |

| Memory type | GDDR6X | GDDR6 |

| Memory bandwidth | 504.2 GB/s | 432.0 GB/s |

| Shading units | 5888 | 6144 |

| TMUs | 184 | 192 |

| ROPs | 64 | 96 |

| RT cores | 46 | Not recorded |

| Tensor cores | 184 | Not recorded |

| Pixel rate | 158.4 GPixel/s | 192.0 GPixel/s |

| Texture rate | 455.4 GTexel/s | 384.0 GTexel/s |

| FP32 | 29.15 TFLOPS | 24.58 TFLOPS |

| FP16 | 29.15 TFLOPS (1:1) | 49.15 TFLOPS (2:1) |

| TDP | 200 W | 225 W |

| Display outputs | 1x HDMI 2.1, 3x DisplayPort 1.4a | 4x DisplayPort 1.4a |

| Vulkan | 1.4 | 1.3 |

| Length | 240 mm | 268 mm |

| Height | 110 mm | 112 mm |

| Width | 40 mm | 40 mm |

| Production status | End-of-life | Active |

| Release date | 2024-02-29 | 2026-03-16 |

| Predecessor | GeForce 30 | Not recorded |

| Successor | GeForce 50 | Not recorded |

| Launch MSRP | 599 USD | Not recorded |

The two cards share the same 192-bit memory bus, dual-slot width, single 16-pin power connector, 550 W suggested PSU, PCIe 4.0 x16 interface, DirectX 12 Ultimate (12_2), and OpenGL 4.6 support. Everything else in the specification sheets diverges, with NVIDIA leading in bandwidth, texture rate, FP32, and API version, while Lisuan leads in memory capacity, ROPs, pixel rate, and FP16.