Intel Arc Pro B370 vs NVIDIA GeForce RTX 4060 Ti AD104 Comparison

Where Each One Wins

The Intel Arc Pro B370 and the NVIDIA GeForce RTX 4060 Ti AD104 occupy distinctly different positions in the performance spectrum, and the recorded data confirms that each holds advantages in separate use cases. The Intel part is an integrated graphics processor (IGP) built for the Panther Lake mobile platform, while the NVIDIA part is a discrete, dual-slot add-in board. The data shows that the RTX 4060 Ti AD104 is the dominant compute and rendering solution across nearly every measurable metric, while the Arc Pro B370 wins exclusively in power efficiency and integration flexibility.

The Arc Pro B370 delivers 6.144 TFLOPS of FP32 compute and 12.29 TFLOPS of FP16 compute, figures that place it at the 50th percentile among all GPUs in the database. Its 25 W TDP and IGP form factor make it the clear choice for systems where physical space, cooling capacity, and power delivery are constrained. The NVIDIA part, by contrast, delivers 22.06 TFLOPS of FP32 and 22.06 TFLOPS of FP16, which is roughly 3.6 times the FP32 throughput of the Intel part. The RTX 4060 Ti AD104 also reaches the 50th percentile in the database, but that percentile is achieved with far higher absolute performance because the distribution of GPUs includes many weaker integrated parts.

For rasterization throughput, the RTX 4060 Ti AD104 posts a pixel rate of 121.7 GPixel/s and a texture rate of 344.8 GTexel/s. The Arc Pro B370 posts 48.00 GPixel/s and 96.00 GTexel/s. These are not close figures. The NVIDIA card is approximately 2.5 times faster in pixel fill and 3.6 times faster in texture fill. For any workload that depends on fill-rate limited rendering, the discrete NVIDIA solution wins outright.

The Intel part wins in the power envelope. At 25 W versus 160 W, the Arc Pro B370 consumes 135 W less under load. It requires no power connectors and no suggested PSU rating, whereas the RTX 4060 Ti AD104 needs a single 16-pin connector and a 450 W suggested PSU. The Intel part is also smaller by nature: it is an IGP with no dimensions listed, versus the NVIDIA card at 240 mm length, 111 mm height, and 40 mm width. For compact portable devices, the Intel part is the only viable option among these two.

The NVIDIA part wins in memory bandwidth decisively. The RTX 4060 Ti AD104 has 8 GB of GDDR6 on a 128-bit bus, delivering 288.0 GB/s. The Arc Pro B370 uses system shared memory, with bandwidth described as system dependent. The Intel part cannot match the dedicated memory bandwidth of the discrete card, and the gap in memory subsystem capability is a fundamental architectural difference rather than a clock speed difference.

Architecture Differences

The two GPUs come from different manufacturers, different foundries, and different design philosophies. The Intel Arc Pro B370 uses the Panther Lake chip with Xe3-LPG architecture, built on Intel's 3 nm process. The NVIDIA GeForce RTX 4060 Ti AD104 uses the AD104 chip with Ada Lovelace architecture, built on TSMC's 5 nm process. The process node difference gives Intel a manufacturing advantage in transistor density potential, though specific transistor counts for the Intel part are listed as unknown. The NVIDIA chip contains 35,800 million transistors on a 294 mm² die, with a transistor density of 121.8M per mm².

The shader configuration differs substantially. The Intel part has 1,280 shading units, 40 texture mapping units, and 20 raster operation units. The NVIDIA part has 4,352 shading units, 136 TMUs, and 48 ROPs. That means the NVIDIA card has 3.4 times the shading units, 3.4 times the TMUs, and 2.4 times the ROPs. These ratios align closely with the FP32 and texture rate differences observed in the benchmark data.

Ray tracing hardware also differs. The Arc Pro B370 includes 10 RT cores, while the RTX 4060 Ti AD104 includes 34 RT cores. The NVIDIA part has more than three times the ray tracing hardware. Additionally, the NVIDIA part includes 136 tensor cores, which accelerate AI and deep learning workloads. The Intel part lists no tensor core count in the database, indicating that AI acceleration hardware is either absent or not specified.

Clock behavior differs significantly. The Intel part has a base clock of 300 MHz and a boost clock of 2400 MHz, a wide dynamic range that suits power-constrained mobile operation. The NVIDIA part has a base clock of 2310 MHz and a boost clock of 2535 MHz, a much narrower range because it is designed to run near its maximum frequency continuously. Memory clocks follow a similar pattern: the Intel part uses system shared memory with no fixed clock, while the NVIDIA part runs its GDDR6 at 2250 MHz, 18 Gbps effective.

The bus interface and form factor reveal the intended deployment. The Arc Pro B370 connects via IGP, meaning it is integrated into the processor package and shares system memory. The RTX 4060 Ti AD104 uses PCIe 4.0 x8, a dedicated slot interface. Display outputs also differ: the Intel part outputs are portable device dependent, while the NVIDIA card provides 1x HDMI 2.1 and 3x DisplayPort 1.4a. API support is identical for both, with DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Production status differs as well. The Arc Pro B370 is listed as active production, released on 2026-01-26, with its predecessor being HD Graphics-WM. The RTX 4060 Ti AD104 is end-of-life, released on 2024-03-31, with its predecessor being GeForce 30 and successor being GeForce 50. The Intel part represents a current generation integrated solution, while the NVIDIA part is a previous-generation discrete card that has already been succeeded.

Head-to-Head Benchmarks

The recorded data does not include direct head-to-head benchmark runs between these two parts, but the specification-derived performance metrics provide a clear comparative picture. The FP32 compute difference is the largest single gap. The RTX 4060 Ti AD104 delivers 22.06 TFLOPS versus 6.144 TFLOPS for the Arc Pro B370. This is a 3.59 times advantage for the NVIDIA card. In FP16 compute, the NVIDIA part again delivers 22.06 TFLOPS, while the Intel part reaches 12.29 TFLOPS through a 2:1 ratio. The NVIDIA advantage in FP16 is 1.79 times.

Texture fill rate shows the NVIDIA card at 344.8 GTexel/s versus 96.00 GTexel/s for the Intel part. That is a 3.59 times advantage, matching the FP32 ratio because texture rate scales directly with shading unit count and clock. Pixel fill rate shows 121.7 GPixel/s versus 48.00 GPixel/s, a 2.54 times advantage for the NVIDIA card. The smaller pixel rate advantage relative to texture rate reflects the ROP count difference: 48 ROPs versus 20 ROPs, which is a 2.4 times ratio.

Memory bandwidth is the most lopsided category in absolute terms. The RTX 4060 Ti AD104 has 288.0 GB/s of dedicated bandwidth. The Arc Pro B370 has no dedicated memory bandwidth figure because it relies on system shared memory, with bandwidth described as system dependent. In any configuration, the Intel part cannot approach the dedicated GDDR6 bandwidth of the NVIDIA card. For memory-intensive workloads such as high-resolution textures or large data sets, the NVIDIA card has a structural advantage that no clock or efficiency improvement could close.

Clock speeds tell a different story. The Intel part boosts to 2400 MHz, which is close to the NVIDIA boost of 2535 MHz. The base clocks are far apart: 300 MHz for Intel versus 2310 MHz for NVIDIA. The Intel part's low base clock reflects its power management strategy as an integrated component, while the NVIDIA card operates at a consistently high frequency. The boost clock gap is only 5.6 percent, but the shading unit count gap is 3.4 times, so the NVIDIA card's performance advantage comes from hardware resources, not clock speed.

The TDP difference is stark. The Arc Pro B370 consumes 25 W, while the RTX 4060 Ti AD104 consumes 160 W. The NVIDIA card delivers 3.59 times the FP32 throughput while consuming 6.4 times the power. In performance per watt, the Intel part is superior: 0.246 TFLOPS per watt versus 0.138 TFLOPS per watt for the NVIDIA card. However, in absolute performance, the NVIDIA card is the clear winner. The data indicates that the Intel part trades raw performance for efficiency, while the NVIDIA part trades efficiency for maximum throughput.

The Verdict

The data supports a clear split between these two products. The NVIDIA GeForce RTX 4060 Ti AD104 is the superior choice for any workload that demands maximum compute throughput, high fill rates, dedicated memory bandwidth, or ray tracing performance. It offers 22.06 TFLOPS FP32, 344.8 GTexel/s texture rate, 121.7 GPixel/s pixel rate, 288.0 GB/s memory bandwidth, and 34 RT cores. It is a dual-slot discrete card with a 160 W TDP, requiring a 16-pin connector and a 450 W suggested PSU. It is end-of-life, having been succeeded by the GeForce 50 series, and its launch MSRP was 399 USD.

The Intel Arc Pro B370 is the appropriate choice for systems that cannot accommodate a discrete GPU. It is an IGP with a 25 W TDP, no power connectors, and no suggested PSU requirement. It delivers 6.144 TFLOPS FP32 and 12.29 TFLOPS FP16, which is sufficient for lightweight graphics workloads but far below the NVIDIA card. Its performance is competitive in efficiency: it delivers more FP32 throughput per watt than the NVIDIA card. It is in active production and was released more recently.

The percentile data places both parts at the 50th percentile of all GPUs, but this should not be read as equivalent performance. The database's percentile ranking reflects the distribution of all GPUs, and the Intel part reaches the median with far lower absolute scores because it is an integrated solution among many other integrated and low-power parts. The NVIDIA card reaches the same percentile with much higher absolute performance because the distribution includes many weaker integrated parts that pull down the median.

For a user with a fixed desktop platform that has a PCIe 4.0 x8 slot and a 450 W PSU, the RTX 4060 Ti AD104 is the only sensible choice from these two. For a user on a Panther Lake mobile platform with no room for a discrete card, the Arc Pro B370 is the only option. These products do not compete in the same market segment, and the data confirms that the NVIDIA card is the higher-performance part in every absolute metric except power consumption and physical footprint.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The NVIDIA GeForce RTX 4060 Ti AD104 delivers 22.06 TFLOPS FP32, while the Intel Arc Pro B370 delivers 6.144 TFLOPS FP32. The NVIDIA card is approximately 3.59 times faster.

Q: What is the power consumption difference between the two?

A: The Intel Arc Pro B370 has a 25 W TDP and requires no power connectors. The NVIDIA GeForce RTX 4060 Ti AD104 has a 160 W TDP, requires a single 16-pin connector, and has a suggested PSU of 450 W.

Q: Do both GPUs support the same graphics APIs?

A: Yes. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: How much memory bandwidth does each GPU have?

A: The NVIDIA GeForce RTX 4060 Ti AD104 has 8 GB of GDDR6 memory on a 128-bit bus with 288.0 GB/s bandwidth. The Intel Arc Pro B370 uses system shared memory, and its bandwidth is system dependent.

Q: Which GPU has more ray tracing cores?

A: The NVIDIA GeForce RTX 4060 Ti AD104 has 34 RT cores. The Intel Arc Pro B370 has 10 RT cores.

Q: What is the production status of each GPU?

A: The Intel Arc Pro B370 is in active production and was released on 2026-01-26. The NVIDIA GeForce RTX 4060 Ti AD104 is end-of-life and was released on 2024-03-31.