Intel Arc 130V Mobile vs NVIDIA B200 SXM6 Comparison

Where Each One Wins

The Intel Arc 130V Mobile and NVIDIA B200 SXM6 occupy entirely different segments of the GPU market, and the recorded data reflects this split clearly. Intel's part is an integrated graphics processor (IGP) built for portable devices, while NVIDIA's B200 SXM6 is a server module designed for compute-heavy workloads. The benchmark wins are not distributed evenly; rather, each product dominates in the categories that match its design intent.

The Intel Arc 130V Mobile uses the Xe2-LPG architecture on the Lunar Lake chip, fabricated on a 3 nm process at TSMC. It carries 896 shading units, 56 texture mapping units, and 28 raster output pipelines. Its boost clock reaches 1850 MHz from a 300 MHz base. The pixel rate is 51.80 GPixel/s, and the texture rate is 103.6 GTexel/s. Floating-point performance is rated at 3.315 TFLOPS for FP32 and 6.630 TFLOPS for FP16 with a 2:1 ratio. The power envelope is 37 W, which makes it suitable for thin-and-light laptops. The device supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, and it has 7 ray tracing cores. Its percentile ranking against all GPUs is 50, placing it at the median of the database.

The NVIDIA B200 SXM6, by contrast, is a massive server accelerator. It uses the GB100 chip with the Blackwell architecture, built on a 5 nm process at TSMC. The die size is 1628 mm², and the transistor count is 208,000 million, yielding a density of 127.8M transistors per mm². The B200 has 18,944 shading units, 592 texture mapping units, and 24 raster output pipelines. Its boost clock reaches 1830 MHz from a 120 MHz base. Pixel rate is 43.92 GPixel/s, and texture rate is 1,083.4 GTexel/s. FP32 throughput is 69.34 TFLOPS, and FP16 throughput is also 69.34 TFLOPS, but at a 1:1 ratio, meaning no rate reduction for half precision. The memory subsystem is the defining feature: 180 GB of HBM3e on an 8192-bit bus, delivering 8.19 TB/s of bandwidth. The power draw is 1000 W, with a suggested power supply of 1400 W. The B200 has 592 tensor cores. Its percentile is also 50, and the launch MSRP is 34,999 USD.

The wins are categorical. The Intel part wins on portability, power efficiency, and graphics API support for consumer applications. The NVIDIA part wins on raw compute throughput, memory capacity, memory bandwidth, and tensor core availability. Neither product encroaches on the other's territory. The Intel Arc 130V Mobile is a graphics solution for everyday rendering and light workloads. The B200 SXM6 is a data center accelerator for AI training, scientific simulation, and high-performance computing. The data shows that the Intel part has a higher pixel rate (51.80 GPixel/s versus 43.92 GPixel/s), but the B200 has a texture rate over ten times higher (1,083.4 GTexel/s versus 103.6 GTexel/s). The FP32 gap is roughly 21 times in favor of the B200 (69.34 TFLOPS versus 3.315 TFLOPS). The memory bandwidth difference is even starker: 8.19 TB/s versus a system-dependent figure for the Intel IGP.

Architecture Differences

The two GPUs diverge at every architectural level, from process node to memory interface to feature set. The Intel Arc 130V Mobile is built on a 3 nm TSMC process, while the NVIDIA B200 SXM6 uses a 5 nm TSMC process. The Intel die is 172 mm², whereas the NVIDIA die spans 1628 mm², nearly ten times larger. Transistor counts reflect this scale: the B200 packs 208,000 million transistors, while the Intel part's transistor count is listed as unknown. The transistor density for the B200 is 127.8M per mm²; no density figure is recorded for the Intel part.

The memory architecture is fundamentally different. The Intel Arc 130V Mobile uses system shared memory, with the type, bus width, and bandwidth all marked as system dependent. This means the GPU borrows from the host system's RAM, and performance scales with the platform's memory configuration. The NVIDIA B200 SXM6, in contrast, has dedicated HBM3e memory totaling 180 GB, wired to an 8192-bit bus. The bandwidth is fixed at 8.19 TB/s, a figure that does not depend on any external factor. This dedicated, high-bandwidth memory is essential for the B200's intended workloads, which involve moving large datasets between compute units and memory.

The compute units differ in both count and capability. The Intel part has 896 shading units, 56 TMUs, and 28 ROPs. The B200 has 18,944 shading units, 592 TMUs, and 24 ROPs. The B200 has far more shading units and TMUs, but fewer ROPs. The Intel part includes 7 ray tracing cores, while the B200's ray tracing core count is not recorded. The B200 includes 592 tensor cores, which are absent from the Intel part's specification sheet. These tensor cores are critical for matrix operations used in AI and deep learning, and their presence distinguishes the B200 as a compute accelerator rather than a graphics-oriented GPU.

Clock behavior also differs. The Intel Arc 130V Mobile has a base clock of 300 MHz and a boost clock of 1850 MHz. The B200 SXM6 has a base clock of 120 MHz and a boost clock of 1830 MHz. The Intel part runs at a higher base clock, but the boost clocks are nearly identical. The B200 compensates for its lower base clock with a massive number of compute units and a 1000 W power budget. The Intel part operates within a 37 W envelope, which constrains its sustained performance but allows integration into portable devices.

The API support is another clear division. The Intel part supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, making it suitable for consumer gaming and graphics applications. The B200 lists DirectX, OpenGL, and Vulkan as N/A, confirming that it is not designed for conventional graphics rendering. The B200 has no display outputs, while the Intel part's display outputs are portable device dependent. The bus interfaces reflect the form factors: the Intel part uses an IGP interface, while the B200 uses PCIe 6.0 x16 and comes as an SXM module.

The release dates are close: the Intel part launched on 2024-09-23, and the B200 launched on 2024-10-31. The Intel predecessor is HD Graphics-M, and the B200's predecessor is Server Hopper. The B200's successor is Server Rubin, while the Intel part has no recorded successor. The B200's launch MSRP is 34,999 USD, while the Intel part has no listed MSRP, consistent with its role as an integrated component rather than a standalone purchase.

The Verdict

The data supports a clear verdict: each product is the correct choice for its respective use case, and neither can substitute for the other. The Intel Arc 130V Mobile is the appropriate part for a portable device requiring graphics output, DirectX 12 Ultimate support, and modest power consumption. Its 37 W power envelope and IGP form factor mean it can be embedded in a laptop without a separate power connector or additional cooling beyond what the chassis provides. Its FP32 performance of 3.315 TFLOPS and texture rate of 103.6 GTexel/s are sufficient for standard graphics workloads, and its 7 ray tracing cores enable hardware-accelerated ray tracing in supported applications.

The NVIDIA B200 SXM6 is the appropriate choice for a server environment where compute density and memory bandwidth are paramount. Its 69.34 TFLOPS FP32 throughput, 69.34 TFLOPS FP16 throughput at 1:1 ratio, and 592 tensor cores make it a formidable accelerator for AI training and inference. The 180 GB HBM3e memory with 8.19 TB/s bandwidth allows it to hold large models and datasets in memory, reducing the need for frequent host-device transfers. The 1000 W power draw and 1400 W suggested PSU are acceptable in a data center context, where power and cooling infrastructure are designed for such loads.

The benchmark data, while sparse in head-to-head comparisons, reinforces this separation. The Intel part wins on pixel rate (51.80 GPixel/s versus 43.92 GPixel/s), which reflects its ROP configuration and clock speed. The B200 wins decisively on texture rate, FP32, FP16, and memory bandwidth. The B200 also has a much larger die and transistor count, indicating its role as a high-end compute part. The Intel part's percentile of 50 and the B200's percentile of 50 both indicate median standing in the full GPU database, but this parity is misleading: the database includes both consumer and server parts, so a median ranking for a 1000 W server accelerator and a 37 W integrated GPU does not imply comparable performance in any specific workload.

Users should select the Intel Arc 130V Mobile for portable, graphics-centric tasks where power efficiency and API compatibility matter. Users should select the NVIDIA B200 SXM6 for server-side compute workloads where raw throughput and memory capacity are the limiting factors. The data does not support a recommendation for one over the other because they are not competitors; they serve different markets with different requirements.

FAQ

Q: Which GPU has higher FP32 performance?

A: The NVIDIA B200 SXM6 has 69.34 TFLOPS FP32, while the Intel Arc 130V Mobile has 3.315 TFLOPS FP32. The B200 is roughly 21 times faster.

Q: Does the Intel Arc 130V Mobile support DirectX 12 Ultimate?

A: Yes, the Intel part supports DirectX 12 Ultimate (12_2), along with OpenGL 4.6 and Vulkan 1.4. The NVIDIA B200 SXM6 lists all three APIs as N/A.

Q: What is the memory configuration of the NVIDIA B200 SXM6?

A: The B200 has 180 GB of HBM3e memory on an 8192-bit bus, with a bandwidth of 8.19 TB/s. The Intel Arc 130V Mobile uses system shared memory, with bandwidth listed as system dependent.

Q: How many tensor cores does each GPU have?

A: The NVIDIA B200 SXM6 has 592 tensor cores. The Intel Arc 130V Mobile does not list a tensor core count in the database.

Q: What are the power requirements for each GPU?

A: The Intel Arc 130V Mobile has a TDP of 37 W. The NVIDIA B200 SXM6 has a TDP of 1000 W and a suggested PSU of 1400 W.

Q: Which GPU has more texture mapping units?

A: The NVIDIA B200 SXM6 has 592 TMUs, compared to 56 TMUs on the Intel Arc 130V Mobile. The B200 also has a higher texture rate: 1,083.4 GTexel/s versus 103.6 GTexel/s.

Head-to-Head Benchmarks

The head-to-head benchmark data for these two GPUs is empty, and the wins count is zero for both sides. The nearest rivals lists are also empty for both parts. The database records some numerical specifications that can serve as a proxy for comparative analysis, but there are no direct benchmark scores to walk through. The following comparison uses the recorded specifications to illustrate the magnitude of the differences.

The largest win for the NVIDIA B200 SXM6 is in FP32 throughput. The B200 delivers 69.34 TFLOPS, while the Intel Arc 130V Mobile delivers 3.315 TFLOPS. This is a 66.025 TFLOPS difference in favor of the B200. In FP16, the B200 again delivers 69.34 TFLOPS at a 1:1 ratio, while the Intel part delivers 6.630 TFLOPS at a 2:1 ratio. The B200's FP16 figure is over ten times higher, and the 1:1 ratio means no performance penalty for half-precision work.

The texture rate is another decisive B200 win. The B200 processes 1,083.4 GTexel/s, compared to 103.6 GTexel/s for the Intel part. This is a 979.8 GTexel/s gap. The B200's 592 TMUs drive this advantage, though the Intel part has a higher pixel rate: 51.80 GPixel/s versus 43.92 GPixel/s, a 7.88 GPixel/s difference in favor of the Intel Arc 130V Mobile.

Memory bandwidth is the most lopsided specification. The B200's 8.19 TB/s is a fixed, dedicated figure, while the Intel part's bandwidth is system dependent, meaning it cannot be compared as a static number. The B200's 180 GB memory capacity and 8192-bit bus are far beyond anything an integrated GPU can access. The B200 also has 592 tensor cores, a feature entirely absent from the Intel part's recorded specifications.

The die size difference is substantial. The B200 measures 1628 mm², while the Intel part measures 172 mm². The B200 has 208,000 million transistors, a figure the Intel part does not record. The B200's transistor density is 127.8M per mm².

Clock speeds are similar at boost: 1850 MHz for the Intel part and 1830 MHz for the B200. The base clocks differ, with the Intel part at 300 MHz and the B200 at 120 MHz. The Intel part has more ROPs (28 versus 24), which contributes to its higher pixel rate. The B200 has far more shading units (18,944 versus 896) and TMUs (592 versus 56).

The power envelope is a win for the Intel part. Its 37 W TDP is a fraction of the B200's 1000 W. The B200 requires a 1400 W suggested PSU, while the Intel part uses an IGP interface with no power connectors listed. The B200 uses PCIe 6.0 x16 and comes as an SXM module, while the Intel part is an IGP with no slot width beyond that designation.

The API support is a win for the Intel part. DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 are all listed as supported. The B200 lists all three as N/A and has no display outputs. The Intel part's display outputs are portable device dependent, meaning it can drive displays in a laptop context.

The release timing is close, with the Intel part launching on 2024-09-23 and the B200 on 2024-10-31. Both parts are marked as active in production. The B200 has a recorded launch MSRP of 34,999 USD, while the Intel part has no MSRP, consistent with its integrated nature. The B200's predecessor is Server Hopper, and its successor is Server Rubin. The Intel part's predecessor is HD Graphics-M, with no successor recorded.

The data confirms that these are not comparable products in any meaningful benchmark sense. The B200 SXM6 is a data center compute accelerator with massive throughput and memory resources. The Intel Arc 130V Mobile is a low-power integrated GPU for portable systems. The specifications align with these roles, and the empty head-to-head benchmark table reflects the lack of overlap in their target workloads.