Intel UHD Graphics 770 Mobile vs Lisuan Tech LX ULTRA Comparison

Intel UHD Graphics 770 Mobile is an integrated graphics processor built on Intel’s Raptor Lake architecture, using a 10 nm process and the Generation 12.2 design. Lisuan Tech LX ULTRA is a discrete dual-slot graphics card based on the 7G105 chip, fabricated by TSMC on a 6 nm node, using the TrueGPU architecture from the 7G100 generation. The database records both parts as active production status, with the Intel part released on 2023-01-03 and the Lisuan part on 2026-03-16. No benchmark scores are recorded for either product, and the head-to-head benchmark array is empty, so the analysis below relies on the recorded hardware specifications and the performance metrics derived from them.

Head-to-Head Benchmarks

The database contains no direct benchmark results for either the Intel UHD Graphics 770 Mobile or the Lisuan Tech LX ULTRA. The head-to-head benchmark array is empty, and both products show an average benchmark score of zero. The percentile versus all GPUs is recorded at 50 for both parts, but this value is not accompanied by any measured performance data. Without recorded frame rates, compute scores, or thermal results, the comparison must proceed through the theoretical peak rates and hardware resources that are present in the specification records.

The FP32 throughput is the clearest differentiator. Intel UHD Graphics 770 Mobile delivers 819.2 GFLOPS of single-precision compute. Lisuan Tech LX ULTRA delivers 24.58 TFLOPS, which is 30 times higher on paper. In FP16 compute, the Intel part reaches 1.638 TFLOPS using a 2:1 ratio, while the Lisuan part reaches 49.15 TFLOPS using the same 2:1 ratio, a ratio of exactly 30 again. The texture rate follows the same pattern: Intel records 25.60 GTexel/s, while Lisuan records 384.0 GTexel/s, a 15-fold difference. Pixel throughput shows Intel at 12.80 GPixel/s versus Lisuan at 192.0 GPixel/s, a 15-fold difference as well.

The shading unit count reinforces the gap. Intel has 256 shading units, 16 texture mapping units, and 8 raster operation units. Lisuan has 6144 shading units, 192 TMUs, and 96 ROPs. That is 24 times more shading units, 12 times more TMUs, and 12 times more ROPs. The memory subsystem is even more lopsided. Intel uses system shared memory with a system dependent bandwidth figure, while Lisuan has 24 GB of dedicated GDDR6 memory on a 192-bit bus, with a recorded bandwidth of 432.0 GB/s. The memory clock for Lisuan is listed as 2250 MHz with 18 Gbps effective data rate.

Clock speeds do not appear for Lisuan, as both base and boost clocks are null in the record. Intel has a base clock of 300 MHz and a boost clock of 1600 MHz. The lack of clock data for Lisuan means the performance ratios above are based on the recorded throughput rates, not on clock comparisons. The Intel part’s boost clock is modest, and its 15 W TDP reflects its integrated nature. Lisuan carries a 225 W TDP, a 15-fold difference in power draw that matches the pixel rate and texture rate ratios.

The API support shows a generation gap. Intel supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. Lisuan supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.3. The DirectX feature level difference, 12_1 versus 12_2, indicates the Lisuan part supports the full DirectX 12 Ultimate feature set, including mesh shaders and variable rate shading, while the Intel part stops at the earlier feature level. Vulkan support is higher on Intel, 1.4 versus 1.3, but the DirectX difference is more significant for modern game workloads.

The Verdict

The recorded data indicates that Lisuan Tech LX ULTRA is the dominant part in every measurable compute and memory metric. FP32 performance is 30 times higher, FP16 performance is 30 times higher, texture rate is 15 times higher, pixel rate is 15 times higher, and memory bandwidth is 432.0 GB/s versus system dependent. The shading unit count is 24 times higher, the TMU count is 12 times higher, and the ROP count is 12 times higher. The Lisuan part also has a 192-bit memory bus and 24 GB of GDDR6, whereas Intel uses system shared memory with an unspecified bus width.

Intel UHD Graphics 770 Mobile holds advantages in two areas: power consumption and API version for Vulkan. The Intel part draws 15 W, while Lisuan draws 225 W, a 15-fold difference. The Intel part supports Vulkan 1.4, while Lisuan supports Vulkan 1.3. These are the only recorded metrics where Intel leads. For any workload that uses the GPU’s compute units, texture units, or rasterizers, the Lisuan part is ahead by an order of magnitude.

The database shows no benchmark scores to confirm real-world behavior, so the verdict is based on the recorded peak rates and hardware specifications. The Lisuan part’s 24.58 TFLOPS FP32 and 432.0 GB/s bandwidth place it in a completely different performance class. The Intel part’s 819.2 GFLOPS and system shared memory place it in the entry-level integrated segment. Users who need raw GPU throughput should select Lisuan Tech LX ULTRA. Users who need an integrated solution with low power draw should select Intel, provided the workload fits within its 819.2 GFLOPS limit.

Where Each One Wins

The Intel UHD Graphics 770 Mobile wins in scenarios where power consumption is the primary constraint. Its 15 W TDP is 15 times lower than the Lisuan part’s 225 W TDP, making it suitable for portable devices where the display outputs are listed as portable device dependent. The Intel part is an IGP with a ring bus interface, meaning it does not require a separate power connector and fits into a system with shared memory. Its Vulkan 1.4 support is newer than the Lisuan part’s Vulkan 1.3, which could matter for applications that use Vulkan 1.4-specific extensions.

The Lisuan Tech LX ULTRA wins in every performance-oriented scenario. Its FP32 throughput of 24.58 TFLOPS is suited for compute-heavy rendering, simulation, or machine learning inference. Its FP16 throughput of 49.15 TFLOPS doubles the FP32 rate, which benefits mixed-precision workloads. The 432.0 GB/s memory bandwidth, 24 GB capacity, and 192-bit bus allow large datasets to stay on the GPU. The 4x DisplayPort 1.4a outputs support multi-monitor setups, and the PCIe 4.0 x16 interface provides a full-bandwidth connection to the host. The 1x 16-pin power connector and 550 W suggested PSU indicate a high-power discrete card for desktop systems.

The DirectX 12 Ultimate (12_2) support on Lisuan enables hardware features that the Intel part’s DirectX 12 (12_1) cannot use. Games or applications that require DirectX 12 Ultimate features will run only on the Lisuan part. The Lisuan part’s dual-slot form factor, with dimensions of 268 mm length, 112 mm height, and 40 mm width, requires a desktop chassis with adequate clearance, while the Intel IGP has no physical dimensions recorded because it is integrated into the processor package.

The memory type also decides use cases. Intel uses system shared memory, which means the GPU borrows from the main system RAM, and bandwidth is system dependent. Lisuan has dedicated GDDR6 memory with a fixed bandwidth of 432.0 GB/s, so performance is not affected by other system memory traffic. For workloads that stream large textures or data sets, the dedicated memory on Lisuan provides predictable performance, while the Intel part’s system dependent bandwidth could become a bottleneck.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: Lisuan Tech LX ULTRA records 24.58 TFLOPS FP32, while Intel UHD Graphics 770 Mobile records 819.2 GFLOPS. The Lisuan part is 30 times higher.

Q: What is the memory configuration of each GPU?

A: Intel UHD Graphics 770 Mobile uses system shared memory with a system dependent bandwidth and no dedicated size, type, or bus width. Lisuan Tech LX ULTRA has 24 GB of GDDR6 memory on a 192-bit bus with 432.0 GB/s bandwidth.

Q: Do the two GPUs support the same DirectX version?

A: No. Intel supports DirectX 12 (12_1), while Lisuan supports DirectX 12 Ultimate (12_2). The Lisuan part has a higher feature level.

Q: Which GPU has a higher texture fill rate?

A: Lisuan Tech LX ULTRA records 384.0 GTexel/s, while Intel UHD Graphics 770 Mobile records 25.60 GTexel/s. The Lisuan part is 15 times higher.

Q: What are the power consumption figures?

A: Intel UHD Graphics 770 Mobile has a TDP of 15 W. Lisuan Tech LX ULTRA has a TDP of 225 W and a suggested PSU of 550 W.

Q: Which GPU has more shading units?

A: Lisuan Tech LX ULTRA has 6144 shading units. Intel UHD Graphics 770 Mobile has 256 shading units. The Lisuan part has 24 times more.

Architecture Differences

The two GPUs come from fundamentally different design philosophies. Intel UHD Graphics 770 Mobile uses the Raptor Lake chip and Generation 12.2 architecture, fabricated on Intel’s 10 nm process. Lisuan Tech LX ULTRA uses the 7G105 chip and TrueGPU architecture from the 7G100 generation, fabricated by TSMC on a 6 nm process. The process node difference, 10 nm versus 6 nm, gives the Lisuan part a density advantage, although the transistor count and die size are listed as unknown for the Lisuan part and are not recorded for the Intel part.

The compute resources differ by large factors. Intel has 256 shading units, 16 TMUs, and 8 ROPs. Lisuan has 6144 shading units, 192 TMUs, and 96 ROPs. The Lisuan part therefore has 24 times the shading units, 12 times the TMUs, and 12 times the ROPs. Neither part records any RT cores or tensor cores, so ray tracing and tensor acceleration are not present in the specification records for either product.

The memory architecture is entirely different. Intel uses system shared memory, with the type, bus width, and size all listed as system shared, and bandwidth listed as system dependent. The memory clock is also listed as system shared. Lisuan has dedicated 24 GB GDDR6 memory with a 192-bit bus, a memory clock of 2250 MHz with 18 Gbps effective data rate, and a fixed bandwidth of 432.0 GB/s. This dedicated memory arrangement removes the dependency on system RAM performance.

The clock behavior differs as well. Intel records a base clock of 300 MHz and a boost clock of 1600 MHz. Lisuan records no base or boost clock, leaving the clock speeds unlisted, but the memory clock is specified at 2250 MHz. The lack of GPU clock data for Lisuan means the throughput numbers come from the recorded pixel rate, texture rate, and FP32 values, not from clock calculations.

The interface and form factor separate the two parts. Intel uses a ring bus interface and is classified as an IGP, meaning it is integrated into the processor and has no slot width, power connectors, or dimensions recorded. Lisuan uses a PCIe 4.0 x16 interface, is a dual-slot card, requires a 1x 16-pin power connector, and has a suggested PSU of 550 W. The physical dimensions are recorded for Lisuan: 268 mm length, 112 mm height, and 40 mm width. Display outputs differ too: Intel lists portable device dependent, while Lisuan lists 4x DisplayPort 1.4a.

The API support shows the architectural generation gap. Intel supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. Lisuan supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.3. The DirectX 12_2 feature level on Lisuan indicates support for the full DirectX 12 Ultimate feature set, which includes hardware-accelerated ray tracing, mesh shaders, and variable rate shading, even though no dedicated RT cores are listed. Intel’s DirectX 12_1 support is one feature level lower, excluding those Ultimate features. Vulkan support is the reverse: Intel has 1.4, Lisuan has 1.3.

The production status and release dates place the parts in different eras. Intel was released on 2023-01-03 and is listed as active, with a successor named Arc Graphics-M. Lisuan was released on 2026-03-16 and is also active, with no successor recorded. The Intel part has a predecessor of null, while Lisuan also has null. The Lisuan part is newer by over three years, which aligns with its larger resource counts and higher feature level. The Intel part’s successor, Arc Graphics-M, indicates a planned or actual replacement within the Intel lineup, but no performance data for that successor is present in the record.

The power envelope explains the performance gap. Intel draws 15 W, which is suitable for integrated graphics in a mobile or low-power desktop context. Lisuan draws 225 W, which is 15 times higher, and requires a 550 W power supply. The 15-fold difference in TDP matches the 15-fold difference in pixel rate and texture rate, although the FP32 difference is 30-fold. This suggests the Lisuan part achieves higher compute efficiency per watt in FP32, but the data does not include measured efficiency figures. The power connectors and slot width confirm the discrete nature of Lisuan, while Intel’s IGP classification means no separate power delivery is needed.