NVIDIA GeForce RTX 4070 AD103 vs Lisuan Tech LX PRO Comparison

Head-to-Head Benchmarks

The recorded database contains no benchmark scores for either the NVIDIA GeForce RTX 4070 AD103 or the Lisuan Tech LX PRO. Both cards register a percentile rank of 50 against all GPUs, with an average benchmark score of zero. This means no direct performance comparison can be drawn from measured results. The absence of data does not indicate equivalence; it simply reflects that no workloads have been logged for either product.

What can be compared is the theoretical throughput derived from each card's specified hardware. The NVIDIA GeForce RTX 4070 AD103 delivers a peak FP32 rate of 29.15 TFLOPS, while the Lisuan Tech LX PRO reaches 24.58 TFLOPS. That places the NVIDIA card roughly 18.6% higher in single-precision compute. In FP16 workloads, the relationship flips. The LX PRO outputs 49.15 TFLOPS with a 2:1 ratio, while the RTX 4070 AD103 manages 29.15 TFLOPS at a 1:1 ratio. The Lisuan card holds a 68.6% advantage in half-precision throughput. The texture rate favors NVIDIA at 455.4 GTexel/s versus 384.0 GTexel/s, a margin of 18.6%. The pixel rate, however, belongs to Lisuan at 192.0 GPixel/s against 158.4 GPixel/s, a 21.2% lead.

These figures indicate that the RTX 4070 AD103 is stronger in traditional rasterization workload metrics like texture fill and FP32 shader work. The LX PRO counters with higher pixel throughput and double-rate FP16, which points toward compute-heavy or AI-oriented tasks where half-precision matters. Memory bandwidth also differs: NVIDIA lists 504.2 GB/s from 12 GB of GDDR6X on a 192-bit bus, while Lisuan lists 432.0 GB/s from 24 GB of GDDR6 on the same 192-bit bus. The NVIDIA card has 16.7% more bandwidth, but the Lisuan card has twice the capacity.

FAQ

Q: Which card has higher FP32 compute performance?

A: The NVIDIA GeForce RTX 4070 AD103 reaches 29.15 TFLOPS, which is 18.6% higher than the Lisuan Tech LX PRO's 24.58 TFLOPS.

Q: Does the Lisuan Tech LX PRO support half-precision compute?

A: Yes, it achieves 49.15 TFLOPS FP16 using a 2:1 ratio, which is 68.6% higher than the NVIDIA card's 29.15 TFLOPS FP16 at a 1:1 ratio.

Q: How much memory does each card have?

A: The RTX 4070 AD103 has 12 GB of GDDR6X, while the LX PRO has 24 GB of GDDR6. Both use a 192-bit memory bus.

Q: Which card has higher memory bandwidth?

A: The NVIDIA card delivers 504.2 GB/s, compared to 432.0 GB/s for the Lisuan card, a 16.7% advantage.

Q: What is the power draw difference?

A: The RTX 4070 AD103 has a TDP of 200 W, while the LX PRO has a TDP of 225 W. Both recommend a 550 W PSU and use a single 16-pin connector.

Q: Are both cards compatible with DirectX 12 Ultimate?

A: Yes, both support DirectX 12 Ultimate (12_2) and OpenGL 4.6. The NVIDIA card supports Vulkan 1.4, while the Lisuan card supports Vulkan 1.3.

Architecture Differences

The NVIDIA GeForce RTX 4070 AD103 is built on the Ada Lovelace architecture using a 5 nm process at TSMC. The die measures 379 mm² and houses 45,900 million transistors, yielding a density of 121.1 million transistors per square millimeter. The chip is designated AD103 and belongs to the GeForce 40 generation. Its predecessor is the GeForce 30 series, and its successor is the GeForce 50 series. The architecture includes 46 ray tracing cores and 184 tensor cores. These dedicated units support hardware-accelerated ray tracing and tensor-based operations, which are standard for the GeForce 40 lineup.

The Lisuan Tech LX PRO uses the TrueGPU architecture with a chip labeled 7G105, fabricated on a 6 nm process at TSMC. The transistor count and die size are not recorded in the database. The generation is listed as 7G100. No ray tracing cores or tensor cores are specified for this card. This suggests the LX PRO does not include dedicated hardware for either ray tracing or tensor acceleration, at least according to the recorded specifications. The architecture appears focused on general-purpose shader throughput, which aligns with its higher FP16 capability and larger memory pool.

Both cards share the same foundry, TSMC, but differ in process node: 5 nm versus 6 nm. The NVIDIA card uses a smaller node, which typically allows higher transistor density and lower power per transistor. The LX PRO uses a slightly larger node but compensates with a simpler architecture lacking specialized cores. The RTX 4070 AD103 includes 46 ray tracing cores and 184 tensor cores, while the LX PRO has none recorded. This is a fundamental architectural divergence: one card is designed for hybrid rendering with ray tracing and AI acceleration, the other is a pure rasterization and compute device.

The API support also differs. Both cards support DirectX 12 Ultimate and OpenGL 4.6, but the NVIDIA card supports Vulkan 1.4, while the Lisuan card supports Vulkan 1.3. This is a minor but measurable difference in graphics API compatibility.

Specification Differences

The two cards diverge across nearly every major specification category. The NVIDIA GeForce RTX 4070 AD103 uses a 5 nm process, while the Lisuan Tech LX PRO uses 6 nm. The NVIDIA card has 5,888 shading units, 184 texture mapping units, and 64 raster output units. The Lisuan card has 6,144 shading units, 192 TMUs, and 96 ROPs. This gives the Lisuan card 4.4% more shaders, 4.4% more TMUs, and 50% more ROPs. The NVIDIA card counters with higher clocks: 1920 MHz base and 2475 MHz boost, while the Lisuan card has no base or boost clock recorded. Memory clocks also differ: NVIDIA lists 1313 MHz with 21 Gbps effective, while Lisuan lists 2250 MHz with 18 Gbps effective.

Memory configuration shows a clear split. The RTX 4070 AD103 has 12 GB of GDDR6X with a 192-bit bus and 504.2 GB/s bandwidth. The LX PRO has 24 GB of GDDR6 with a 192-bit bus and 432.0 GB/s bandwidth. The NVIDIA card uses faster memory with higher bandwidth but half the capacity. Power consumption differs: the NVIDIA card is rated at 200 W, the Lisuan card at 225 W. Both are dual-slot designs with a single 16-pin power connector and a 550 W recommended PSU.

Physical dimensions vary slightly. The NVIDIA card measures 240 mm in length, 110 mm in height, and 40 mm in width. The Lisuan card measures 248 mm in length, 118 mm in height, and 48 mm in width. The Lisuan card is longer, taller, and thicker. Display outputs also differ: NVIDIA provides 1x HDMI 2.1 and 3x DisplayPort 1.4a, while Lisuan provides 4x DisplayPort 1.4a. The NVIDIA card has no HDMI output on the Lisuan card, and the Lisuan card has no HDMI output at all.

Production status differs as well. The RTX 4070 AD103 is marked as end-of-life, released on 2024-02-29. The LX PRO is active, with a release date of 2026-03-16. The NVIDIA card has a recorded launch MSRP of 599 USD. The Lisuan card has no launch MSRP recorded. The NVIDIA card lists a predecessor and successor, while the Lisuan card lists none. The PCIe interface is identical: PCIe 4.0 x16 for both.

The Verdict

The data reveals two cards with opposing strengths. The NVIDIA GeForce RTX 4070 AD103 delivers higher FP32 compute, higher texture rate, higher memory bandwidth, and more advanced API support with Vulkan 1.4. It also includes dedicated ray tracing and tensor cores, which the Lisuan card lacks entirely. The NVIDIA card is smaller, consumes less power, and comes from a well-documented architecture with a clear lineage from GeForce 30 to GeForce 50.

The Lisuan Tech LX PRO offers double the memory capacity at 24 GB, higher pixel rate, more shading units, more TMUs, more ROPs, and significantly higher FP16 throughput. It also has a larger physical footprint and higher power draw. The lack of ray tracing and tensor cores means it cannot accelerate those specific workloads. Its Vulkan support is one version behind.

For users prioritizing raw single-precision rendering performance, texture-heavy workloads, memory bandwidth, and ray tracing capability, the NVIDIA card is the stronger choice based on specifications. For users who need large memory capacity for datasets that exceed 12 GB, or who rely on half-precision compute for machine learning or scientific workloads, the Lisuan card offers clear advantages. The absence of benchmark scores means these conclusions rest entirely on theoretical peak rates and hardware configurations, not measured results.

The production status matters for availability. The NVIDIA card is end-of-life, while the Lisuan card is active. This suggests the Lisuan card is currently available, while the NVIDIA card may be harder to source. The release dates place the NVIDIA card in early 2024 and the Lisuan card in early 2026, indicating a generational gap. The NVIDIA card has a known launch MSRP of 599 USD, but no pricing information exists for the Lisuan card.

Where Each One Wins

The NVIDIA GeForce RTX 4070 AD103 wins in FP32 compute with 29.15 TFLOPS versus 24.58 TFLOPS. It wins in texture rate with 455.4 GTexel/s versus 384.0 GTexel/s. It wins in memory bandwidth with 504.2 GB/s versus 432.0 GB/s. It wins in power efficiency, drawing 200 W versus 225 W while delivering higher FP32 and texture performance. It wins in API support with Vulkan 1.4 versus 1.3. It wins in ray tracing and tensor acceleration, thanks to its 46 ray tracing cores and 184 tensor cores, features absent from the Lisuan card. It wins in physical compactness, measuring 240 mm by 110 mm by 40 mm versus 248 mm by 118 mm by 48 mm. It wins in production maturity, with a defined predecessor and successor in the GeForce 30 and 50 series.

The Lisuan Tech LX PRO wins in memory capacity with 24 GB versus 12 GB. It wins in pixel rate with 192.0 GPixel/s versus 158.4 GPixel/s. It wins in shading units with 6,144 versus 5,888. It wins in TMUs with 192 versus 184. It wins in ROPs with 96 versus 64. It wins in FP16 compute with 49.15 TFLOPS versus 29.15 TFLOPS. It wins in display outputs by offering 4x DisplayPort 1.4a, while the NVIDIA card offers 3x DisplayPort and 1x HDMI. It wins in production status as an active product versus end-of-life. It wins in release recency, launching in 2026 versus 2024.

The use-case split follows these strengths. For gaming and real-time rendering where FP32 and texture throughput dominate, the NVIDIA card is superior. For compute workloads that require FP16 precision, such as certain AI inference or training tasks, the Lisuan card offers more than double the throughput. For rendering to high-resolution displays with many outputs, the Lisuan card's four DisplayPort connections provide more flexibility. For applications that fit within 12 GB of memory, the NVIDIA card's higher bandwidth is beneficial. For datasets or models that exceed 12 GB, the Lisuan card's 24 GB capacity is the only option between these two. For ray-traced scenes, the NVIDIA card has dedicated hardware; the Lisuan card has none. For users seeking a current production card, the Lisuan card is active, while the NVIDIA card is no longer in production.