GPU Comparison

NVIDIA GeForce GT 545 and Intel Arc Pro B60 occupy vastly different points in the GPU landscape, separated by over a decade of architectural evolution. The GT 545 is an end-of-life Fermi-era part from 2011, while the Arc Pro B60 is an active Battlemage-generation workstation card from 2025. Benchmark data shows the Arc Pro B60 holds a commanding lead in raw compute, but the GT 545’s single OpenCL score places it surprisingly close in that specific metric, creating a nuanced comparison that hinges on workload and API support.

Head-to-Head Benchmarks

The most direct numerical comparison comes from the Geekbench OpenCL test, where the GT 545 scores 3594. The Arc Pro B60 does not have a Geekbench OpenCL result in the data, so direct apples-to-apples comparison on that specific test is impossible. However, the GT 545’s score places it within 0.1% of the NVIDIA RTX 5000 Mobile Ada Generation (3596) and 0.6% behind the GeForce GT 735M (3616). This suggests the GT 545’s OpenCL performance is competitive with much newer entry-level parts, though its overall percentile rank of 21 indicates it sits below the majority of GPUs.

The Arc Pro B60’s benchmark suite is far more extensive, covering DirectX and compute workloads. Its Passmark G3D score of 14580 is its strongest result, dwarfing its Passmark DirectX scores: 179 in DirectX 9, 61 in DirectX 10, 122 in DirectX 11, and 76 in DirectX 12. The GPU compute score of 7029 is also substantial, more than double the DirectX 11 score and nearly half the G3D score. In 3DMark Steel Nomad DX12, it posts 2646. These results paint a picture of a card that excels in modern compute and general 3D rendering but shows weaker legacy DirectX performance relative to its own capabilities.

Comparing the two directly requires using the Arc Pro B60’s average benchmark score of 3182 against the GT 545’s average of 3594. This shows the GT 545 is 12.9% higher on average, a surprising result given the architectural gap. However, this average is heavily influenced by the GT 545 having only one benchmark while the Arc Pro B60 has eight, many of which are legacy DirectX tests where modern architectures often underperform relative to their compute potential. The Arc Pro B60’s nearest rivals include the NVIDIA Quadro P1000 (3163, deltaPct 0.6) and the GeForce GT 640 (3210, deltaPct -0.9), indicating its average score sits in a similar range to older mid-range parts despite its 24 GB GDDR6 memory and 12.29 TFLOPS FP32 throughput.

The GT 545’s nearest rivals tell a different story: it trails the GTX 1050 by 1% (3629) and the AMD Radeon HD 6770 by 1.5% (3649), but leads the RTX 5000 Mobile Ada by 0.1%. This clustering suggests the GT 545’s OpenCL score is competitive with a wide range of GPUs from different eras, though its overall percentile rank of 21 versus the Arc Pro B60’s 20 shows both cards sit near the bottom quartile of all GPUs.

The Verdict

The data supports a clear split recommendation. For users running OpenCL-based workloads that rely on the GT 545’s specific driver optimizations, its 3594 score is respectable and falls within 1.5% of several modern entry-level cards. However, the GT 545 has no DirectX 12 Ultimate support, no Vulkan support, and only 1536 MB of DDR3 memory, making it unsuitable for any modern gaming or professional 3D workload beyond the most basic compute tasks.

The Arc Pro B60 is the only viable option for modern workloads. Its 24 GB GDDR6 memory with 456.0 GB/s bandwidth, 2560 shading units, and 20 ray tracing cores provide the hardware foundation for contemporary applications. Its Passmark G3D score of 14580 is its standout result, suggesting strong performance in general 3D rendering. The DirectX 12 score of 76 is low in absolute terms, but the card supports DirectX 12 Ultimate (12_2), meaning it can run the latest APIs even if its legacy performance lags. The 3DMark Steel Nomad DX12 score of 2646 further confirms its modern rendering capability.

Anyone considering the GT 545 should recognize its 21st percentile ranking and end-of-life status. The Arc Pro B60, despite its 20th percentile ranking, offers a modern feature set that the GT 545 cannot match: PCIe 5.0 x8 interface, four mini-DisplayPort 2.1 outputs, and Vulkan 1.4 support. The GT 545’s PCIe 2.0 x16 and single DVI/HDMI 1.3a/VGA outputs limit it to legacy displays and slow data transfer. For any new purchase, the Arc Pro B60 is the data-supported choice; for preservation or legacy compute, the GT 545 holds a narrow edge in its sole benchmark.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GT 545 has an average benchmark score of 3594, which is 12.9% higher than the Intel Arc Pro B60’s average of 3182. However, this comparison is limited because the GT 545 has only one benchmark result (Geekbench OpenCL), while the Arc Pro B60 has eight results across DirectX and compute tests.

Q: How does the GT 545 compare to the RTX 5000 Mobile Ada Generation?

A: The GT 545 scores 3594 in Geekbench OpenCL, which is 0.1% lower than the RTX 5000 Mobile Ada Generation’s score of 3596. This places the two cards nearly identical in this specific test, despite the RTX 5000 being a much newer mobile workstation part.

Q: What is the Arc Pro B60’s best benchmark result?

A: The Arc Pro B60’s highest score is 14580 in the Passmark G3D test, which significantly exceeds its other results. Its next best is Passmark GPU Compute at 7029, followed by 3DMark Steel Nomad DX12 at 2646 and Passmark G2D at 763.

Q: Does the GT 545 support modern APIs?

A: The GT 545 supports DirectX 12 (11_0) and OpenGL 4.6, but has no Vulkan support listed. The Arc Pro B60 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, making it far more future-proof for API compatibility.

Q: What memory configurations do these cards use?

A: The GT 545 has 1536 MB of DDR3 memory on a 192-bit bus with 38.40 GB/s bandwidth. The Arc Pro B60 has 24 GB of GDDR6 memory on a 192-bit bus with 456.0 GB/s bandwidth, representing a 11.9x increase in bandwidth.

Q: Which card has a higher pixel fill rate?

A: The Arc Pro B60 has a pixel rate of 192.0 GPixel/s, which is 44.4 times higher than the GT 545’s 4.320 GPixel/s. This massive difference reflects the Arc Pro B60’s 80 ROPs versus the GT 545’s 16 ROPs.

Specification Differences

The two cards differ across nearly every specification. The GT 545 uses 1536 MB of DDR3 memory, while the Arc Pro B60 uses 24 GB of GDDR6. Both have a 192-bit memory bus, but bandwidth differs drastically: 38.40 GB/s versus 456.0 GB/s. The GT 545 has 144 shading units, 24 TMUs, and 16 ROPs, while the Arc Pro B60 has 2560 shading units, 160 TMUs, and 80 ROPs. The Arc Pro B60 adds 20 ray tracing cores; the GT 545 has none.

Clock speeds also diverge, with the GT 545 listing only a memory clock of 800 MHz (1600 Mbps effective), while the Arc Pro B60 has a base clock of 2000 MHz and boost clock of 2400 MHz, with memory at 2375 MHz (19 Gbps effective). Power requirements favor the GT 545 at 70 W TDP with no power connectors and a 250 W suggested PSU, versus the Arc Pro B60’s 200 W TDP, 1x 8-pin connector, and 550 W suggested PSU. The GT 545 is single-slot and 145 mm long, while the Arc Pro B60 is dual-slot, measuring 167 mm x 69 mm x 40 mm. Bus interfaces differ: PCIe 2.0 x16 for the GT 545 versus PCIe 5.0 x8 for the Arc Pro B60. Display outputs are 1x DVI, 1x HDMI 1.3a, and 1x VGA for the GT 545, versus 4x mini-DisplayPort 2.1 for the Arc Pro B60.

Architecture Differences

The GT 545 is built on NVIDIA’s Fermi 2.0 architecture using the GF116 chip, manufactured on a 40 nm process at TSMC with 1,170 million transistors on a 238 mm² die (transistor density of 4.9M / mm²). The Arc Pro B60 uses Intel’s Xe2-HPG architecture with the BMG-G21 chip, built on a 5 nm process at TSMC with 19,600 million transistors on a 272 mm² die (transistor density of 72.1M / mm²). This represents a 16.7x increase in transistor count and a 14.7x increase in density.

The GT 545’s FP32 performance is 414.7 GFLOPS, while the Arc Pro B60 delivers 12.29 TFLOPS, a 29.6x difference. The Arc Pro B60 also lists FP16 performance of 24.58 TFLOPS (2:1), which the GT 545 does not provide. Texture rates differ by 22.2x (17.28 GTexel/s versus 384.0 GTexel/s), and pixel rates differ by 44.4x (4.320 GPixel/s versus 192.0 GPixel/s). The Arc Pro B60 supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the GT 545 only reaches DirectX 12 (11_0) with no Vulkan. The GT 545 belongs to the GeForce 500 generation with the GeForce 400 as predecessor and GeForce 600 as successor; the Arc Pro B60 is part of the Battlemage (Pro Series) generation with no listed predecessor or successor.

Where Each One Wins

The GT 545 wins in the single benchmark where both have comparable data: its Geekbench OpenCL score of 3594 exceeds the Arc Pro B60’s average of 3182. It also wins on power efficiency, with a 70 W TDP versus 200 W, and requires no power connectors or specialized PSU. For legacy systems with PCIe 2.0 slots, the GT 545’s x16 interface may be more compatible than the Arc Pro B60’s x8 PCIe 5.0 slot. Its single-slot design and shorter length (145 mm versus 167 mm) make it easier to fit in compact cases.

The Arc Pro B60 wins in every other measurable category. Its 24 GB GDDR6 memory with 456.0 GB/s bandwidth is essential for large datasets, and its 12.29 TFLOPS FP32 performance enables compute workloads the GT 545 cannot handle. The 20 ray tracing cores provide hardware acceleration for ray-traced rendering, a feature entirely absent from the GT 545. Its Passmark G3D score of 14580 indicates strong general 3D performance, and the DirectX 12 Ultimate support ensures compatibility with the latest games and professional applications. The four mini-DisplayPort 2.1 outputs support modern high-resolution displays, while the GT 545’s HDMI 1.3a and VGA are obsolete. For any workload involving modern APIs, large textures, or ray tracing, the Arc Pro B60 is the only choice. The GT 545 remains relevant only for basic OpenCL compute tasks where its single high score can be leveraged.