Intel Arc G3 vs Intel Arc Pro B60 Dual Comparison

Head-to-Head Benchmarks

The recorded data for the Intel Arc G3 and the Intel Arc Pro B60 Dual contains no direct head-to-head benchmark results. The database shows zero wins for either part, and the average benchmark score for both is zero. Because no measured performance deltas exist, the comparison must rest entirely on the architectural specifications and derived throughput figures recorded for each product.

The Arc Pro B60 Dual delivers a raw compute advantage that is strictly two times the Arc G3 in several key metrics. The Pro B60 has 2560 shading units versus 1280 on the G3, exactly doubling the shader count. The texture rate follows the same pattern: 384.0 GTexel/s versus 96.00 GTexel/s, a 4x difference. The pixel rate shows the largest gap, with the Pro B60 reaching 192.0 GPixel/s against 48.00 GPixel/s for the G3, a 4x advantage. FP32 throughput is 12.29 TFLOPS on the Pro B60 versus 6.144 TFLOPS on the G3, exactly double. FP16 performance is 24.58 TFLOPS versus 12.29 TFLOPS, again double.

Clock behavior complicates the raw comparison. The G3 has a base clock of 300 MHz and a boost clock of 2400 MHz. The Pro B60 has a base clock of 2000 MHz and the same 2400 MHz boost. At boost, both parts run at the same frequency, so the G3's lower base clock does not limit its peak performance in the recorded data. The G3's 25 W TDP versus the Pro B60's 400 W TDP indicates the G3 achieves its throughput at a fraction of the power envelope, but the database records no efficiency score, so this remains a specification observation rather than a measured result.

Memory is a decisive separation point. The G3 uses System Shared memory, with bandwidth listed as System Dependent and no dedicated VRAM capacity. The Pro B60 uses 24 GB of GDDR6 on a 192-bit bus, delivering 456.0 GB/s of bandwidth. The G3's memory bandwidth cannot be quantified from the record, making the Pro B60's 456.0 GB/s the only concrete bandwidth figure available. For workloads that scale with memory capacity or sustained bandwidth, the Pro B60's dedicated 24 GB pool is categorically different from the G3's shared-memory design.

The G3's ray tracing hardware is also half the Pro B60's: 10 RT cores versus 20. The G3 has 40 TMUs and 20 ROPs, while the Pro B60 has 160 TMUs and 80 ROPs, each a 4x difference. These ratios hold consistently across the texture and pixel pipelines, suggesting the Pro B60 is architecturally scaled to a higher throughput class rather than simply clocked higher.

Both cards support identical API feature sets: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither part records tensor core counts, so AI acceleration comparisons cannot be quantified from this data.

The Verdict

The database records no benchmark scores, no percentile ranks above the neutral 50th percentile for both parts, and no nearest rivals. Both products sit at the 50th percentile versus all GPUs, which reflects the absence of measured data rather than a performance equivalence. Without measured results, the verdict must be drawn from the specification sheet alone.

The Intel Arc Pro B60 Dual is the clear choice for any workload that demands dedicated memory, high bandwidth, or high fill rates. It doubles the G3's shader count, doubles its FP32 throughput, quadruples its pixel and texture rates, and provides 24 GB of dedicated GDDR6 memory at 456.0 GB/s. The 400 W TDP, 800 W suggested PSU, dual-slot cooler, and 300 mm length indicate a card designed for a desktop workstation with substantial power delivery. Its launch MSRP is 1,199 USD.

The Intel Arc G3 is an integrated graphics processor with a 25 W TDP, no power connectors, and an IGP bus interface. It uses system shared memory, so its bandwidth and capacity are System Dependent and cannot be compared numerically to the Pro B60. For a portable device or a low-power platform, the G3 offers the same API support as the Pro B60 with a fraction of the power draw, but the database cannot quantify its real-world performance because no benchmarks are recorded.

The data indicates these are not competing products. The G3 is an integrated solution for low-power platforms, while the Pro B60 is a dual-slot, 400 W discrete workstation card. Users requiring the Pro B60's compute and memory characteristics should select it based on the specification advantages. Users constrained to an integrated part with no power connectors have only the G3 as an option in this comparison.

Architecture Differences

The two GPUs come from different generations and foundries. The Intel Arc G3 uses the Panther Lake chip built on Xe3-LPG architecture, belongs to the Arc Graphics-M (Panther Lake) generation, and is fabricated on a 3 nm process at Intel. The Intel Arc Pro B60 Dual uses the BMG-G21 chip built on Xe2-HPG architecture, belongs to the Battlemage (Pro Series) generation, and is fabricated on a 5 nm process at TSMC. The G3 is the newer process node at 3 nm versus 5 nm, though the G3's transistor count and die size are recorded as unknown. The Pro B60 has a recorded transistor count of 19,600 million on a 272 mm² die, yielding a transistor density of 72.1M / mm².

The memory architecture differs fundamentally. The G3 uses System Shared memory with a System Shared bus width and System Dependent bandwidth. The Pro B60 uses 24 GB of GDDR6 on a 192-bit bus with 2375 MHz memory clock and 19 Gbps effective speed, delivering 456.0 GB/s. The G3's memory clock is listed as System Shared, meaning no fixed memory clock is recorded.

The compute configuration scales by roughly a factor of two to four across units. The G3 has 1280 shading units, 40 TMUs, 20 ROPs, and 10 RT cores. The Pro B60 has 2560 shading units, 160 TMUs, 80 ROPs, and 20 RT cores. Both lack recorded tensor core counts. FP32 and FP16 rates on the Pro B60 are exactly double the G3's, while pixel and texture rates are exactly four times higher.

Power and physical design are strongly divergent. The G3 is an IGP with a 25 W TDP, no power connectors, and no slot width beyond IGP. The Pro B60 is a dual-slot card with a 400 W TDP, one 16-pin power connector, an 800 W suggested PSU, and a PCIe 5.0 x8 bus interface. Physical dimensions for the Pro B60 are 300 mm length, 110 mm height, and 40 mm width. The G3 has no recorded dimensions. Display outputs also differ: the G3 is Portable Device Dependent, while the Pro B60 provides 4x mini-DisplayPort 2.1.

Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The G3 was released on 2026-05-31, while the Pro B60 was released on 2025-09-04, making the Pro B60 the earlier launch by roughly nine months. Both are recorded as Active production status, and neither has a predecessor or successor in the database.

FAQ

Q: Which GPU has higher FP32 throughput?

A: The Intel Arc Pro B60 Dual records 12.29 TFLOPS FP32, exactly double the Intel Arc G3's 6.144 TFLOPS.

Q: How much memory bandwidth does the Arc G3 have?

A: The Arc G3 uses System Shared memory with bandwidth listed as System Dependent. No fixed bandwidth figure is recorded for it. The Arc Pro B60 Dual, by contrast, records 456.0 GB/s from 24 GB of GDDR6 on a 192-bit bus.

Q: Do both GPUs support the same graphics APIs?

A: Yes. Both the Intel Arc G3 and Intel Arc Pro B60 Dual support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What is the power draw for each card?

A: The Intel Arc G3 has a 25 W TDP and requires no power connectors. The Intel Arc Pro B60 Dual has a 400 W TDP, uses one 16-pin power connector, and has an 800 W suggested PSU.

Q: How many ray tracing cores does each GPU have?

A: The Intel Arc G3 has 10 RT cores. The Intel Arc Pro B60 Dual has 20 RT cores, exactly double.

Q: Which GPU was released earlier?

A: The Intel Arc Pro B60 Dual was released on 2025-09-04. The Intel Arc G3 was released on 2026-05-31, roughly nine months later.

Where Each One Wins

The Intel Arc Pro B60 Dual wins on every quantified compute metric in the database. It doubles the G3's shading units (2560 versus 1280), doubles FP32 throughput (12.29 TFLOPS versus 6.144 TFLOPS), doubles FP16 throughput (24.58 TFLOPS versus 12.29 TFLOPS), doubles RT core count (20 versus 10), and quadruples pixel rate (192.0 GPixel/s versus 48.00 GPixel/s) and texture rate (384.0 GTexel/s versus 96.00 GTexel/s). It is the only part in this comparison with a dedicated memory subsystem: 24 GB GDDR6, 192-bit bus, 456.0 GB/s bandwidth. Its 2000 MHz base clock is far above the G3's 300 MHz base, and both boost to 2400 MHz. The Pro B60 also provides four mini-DisplayPort 2.1 outputs, while the G3's display outputs are Portable Device Dependent. The Pro B60's 400 W TDP, 800 W suggested PSU, and dual-slot design indicate a workstation card intended for sustained high-throughput operation.

The Intel Arc G3 wins on power efficiency and integration. Its 25 W TDP is 16 times lower than the Pro B60's 400 W TDP. It requires no power connectors, uses an IGP bus interface, and has no physical dimensions recorded, consistent with an integrated part. It uses System Shared memory, which removes the need for a dedicated VRAM pool and allows the platform to allocate memory dynamically. The G3 is built on a 3 nm Intel process versus the Pro B60's 5 nm TSMC process, which may contribute to its lower power draw, though no efficiency metric is recorded. It also has a later release date of 2026-05-31 versus the Pro B60's 2025-09-04. For a portable device or an integrated platform where add-in card space and power delivery are unavailable, the G3 provides the same DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 API support as the Pro B60, but at a 25 W envelope.

The recorded data does not include any benchmark scores, so the G3's real-world performance in its intended integrated role cannot be quantified. Its 50th percentile rank versus all GPUs is identical to the Pro B60's, and both have an average benchmark score of zero. The G3's advantage is therefore limited to the specification-level traits of power, integration, and process node. The Pro B60's advantage covers every measurable performance metric in the database, from shading and ray tracing to memory bandwidth and fill rate.