Intel Arc B570 vs NVIDIA P106-100 Comparison

Head-to-Head Benchmarks

The benchmark data across all three recorded tests shows a decisive sweep for the Intel Arc B570. In the 3DMark Steel Nomad DX12 test, the Arc B570 scores 2649 against the NVIDIA P106-100's 899, a delta of -66.1% from the winner's perspective. This indicates the Intel part delivers roughly three times the raw graphics throughput in this modern DirectX 12 workload, a gap that reflects not just generational distance but also architectural intent.

The compute-oriented results reinforce the same pattern. In Geekbench OpenCL, the Arc B570 posts 83514 while the P106-100 manages 35951, a -57% delta. The Vulkan benchmark widens further: 96844 for Intel versus 32897 for NVIDIA, a -66% delta. Across all three tests, the P106-100 fails to secure a single win. The head-to-head record stands at 0 wins for NVIDIA and 3 wins for Intel, with every margin comfortably exceeding 50 percentage points.

It is worth noting what these deltas mean in practical terms. A 66% deficit in Steel Nomad suggests the P106-100 cannot sustain comparable frame rates in geometry-heavy or shading-heavy scenes typical of recent DirectX 12 titles. The Vulkan gap of 66% similarly points to fundamental throughput differences in draw call processing and async compute pathways. The OpenCL result, while slightly smaller at 57%, still places the P106-100 at roughly 43% of the Arc B570's compute output. These are not marginal differences; they are category-level separations.

The database also places each card within a broader context. The P106-100 holds a 68th percentile ranking among all GPUs, with an average benchmark score of 23249. Its nearest rivals include the AMD Radeon Pro Vega 16 at 23250 (0% delta), the AMD Radeon RX 6600M at 23273 (-0.1%), and the AMD Radeon R9 M290X at 23276 (-0.1%). The Arc B570, despite winning every head-to-head test, sits at the 65th percentile with an average score of 20556. Its nearest rivals include the NVIDIA GeForce RTX 3070 Mobile at 20534 (0.1% delta), the Intel Arc A750 at 20582 (-0.1%), and the AMD Radeon R9 M390X at 20662 (-0.5%). This apparent contradiction, a lower average score yet a sweep in head-to-head tests, stems from the different benchmark mixes. The Arc B570's PassMark scores, which include low DirectX 9 and DirectX 10 figures (164 and 65 respectively), drag down its average despite strong modern API results.

The Verdict

The recorded data leaves little room for ambiguity. The Intel Arc B570 is the superior product in every benchmark that the database contains. The P106-100, an end-of-life mining GPU with no display outputs, cannot compete on any measurable axis. Its best relative showing is in OpenCL where it reaches 43% of the Arc B570's score, but even that outcome is a decisive loss.

Buyers should choose the Arc B570 if they need a functional graphics card with modern API support, display connectivity, and competitive performance in current workloads. The card ships with 10 GB of GDDR6 memory, a 160-bit bus, and 380.0 GB/s of bandwidth, figures that dwarf the P106-100's 6 GB GDDR5, 192-bit bus, and 192.2 GB/s. The Arc B570 also carries 18 ray tracing cores, a feature the Pascal-based P106-100 lacks entirely. Its DirectX 12 Ultimate support (12_2) versus the older 12_1 on the P106-100 further separates the two for contemporary gaming and compute applications.

The P106-100 should only be considered by someone who has no alternative and needs raw compute in a very specific legacy context. Its 4.375 TFLOPS of FP32 performance, 136.7 GTexel/s texture rate, and 82.03 GPixel/s pixel rate are all far below the Arc B570's 11.52 TFLOPS, 360.0 GTexel/s, and 200.0 GPixel/s. The P106-100 also requires a PCIe 1.0 x16 interface, a relic that severely limits bandwidth in modern systems, and has no display outputs, rendering it useless for standard desktop use. The Arc B570, by contrast, uses PCIe 4.0 x8 and provides 1x HDMI 2.1a plus 3x DisplayPort 2.1 outputs.

Where Each One Wins

From the benchmark data, the Intel Arc B570 wins in every tested category: 3DMark Steel Nomad DX12, Geekbench OpenCL, and Geekbench Vulkan. There is no test in the database where the P106-100 comes out ahead. This makes the use-case split straightforward.

The Arc B570 is the choice for anyone running modern DirectX 12 or Vulkan applications. Its Steel Nomad score of 2649 suggests it can handle recent game engines with tessellation, variable rate shading, and ray tracing effects at playable settings, particularly given its 18 RT cores. Its Vulkan score of 96844 indicates strong driver efficiency in low-level APIs, which benefits both gaming and compute workloads. The OpenCL score of 83514 also supports general-purpose GPU computing tasks such as image processing, physics simulations, or machine learning inference at a modest scale.

The P106-100 has no winning use cases in the recorded data. Its 3DMark Steel Nomad score of 899 places it far below even entry-level modern GPUs. Its Geekbench OpenCL score of 35951 and Vulkan score of 32897 are similarly uncompetitive. The card's 16 nm process node, 4,400 million transistors, and 200 mm² die size all point to a 2017-era design that has been superseded by multiple generations. Its production status is end-of-life, and its release date of 2017-06-18 places it over seven years behind the Arc B570's 2025-01-15 debut.

For legacy compute tasks that rely on CUDA or specific Pascal-era optimizations, the P106-100 might theoretically offer compatibility, but the database provides no benchmark evidence of such an advantage. The average benchmark score of 23249 for the P106-100 versus 20556 for the Arc B570 reflects the PassMark test suite's weighting, but the head-to-head tests are unambiguous.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA P106-100 has an average benchmark score of 23249, which is higher than the Intel Arc B570's 20556. However, this average includes different test suites, and the head-to-head benchmarks show the Arc B570 winning all three tests.

Q: What is the biggest performance gap between the two cards?

A: The largest delta appears in both the 3DMark Steel Nomad DX12 test and the Geekbench Vulkan test, where the Intel Arc B570 leads by 66.1% and 66% respectively. The OpenCL gap is slightly smaller at 57%.

Q: Does the P106-100 support ray tracing?

A: No. The P106-100 has no ray tracing cores listed in the database. The Intel Arc B570 includes 18 ray tracing cores.

Q: Can the P106-100 be used as a display adapter?

A: No. The database lists the P106-100 as having no display outputs. The Intel Arc B570 provides 1x HDMI 2.1a and 3x DisplayPort 2.1 outputs.

Q: How do their memory configurations differ?

A: The P106-100 has 6 GB of GDDR5 memory on a 192-bit bus with 192.2 GB/s bandwidth. The Arc B570 has 10 GB of GDDR6 memory on a 160-bit bus with 380.0 GB/s bandwidth.

Q: What are the power requirements for each card?

A: The P106-100 has a TDP of 120 W with a 1x 6-pin power connector and a suggested PSU of 300 W. The Arc B570 has a TDP of 150 W with a 1x 8-pin connector and a suggested PSU of 450 W.

Architecture Differences

The two GPUs come from vastly different eras and design philosophies. The NVIDIA P106-100 uses the GP106 chip built on the Pascal architecture, fabricated by TSMC on a 16 nm process. It contains 4,400 million transistors on a 200 mm² die, giving a transistor density of 22.0 million per square millimeter. The Intel Arc B570 uses the BMG-G21 chip based on the Xe2-HPG architecture, also from TSMC but on a 5 nm process. It packs 19,600 million transistors into a 272 mm² die, achieving a density of 72.1 million per square millimeter. That density difference, over three times higher, reflects the architectural and process advancements between 2017 and 2025.

The compute resources differ substantially. The P106-100 has 1280 shading units, 80 texture mapping units, and 48 render output units. The Arc B570 more than doubles the shading units to 2304, increases TMUs to 144, and raises ROPs to 80. The Arc B570 also adds 18 RT cores, which the P106-100 lacks entirely. Neither card lists tensor cores in the database, but the Arc B570's FP16 throughput of 23.04 TFLOPS (2:1 ratio) vastly exceeds the P106-100's 68.36 GFLOPS (1:64 ratio), indicating the Intel part can handle mixed-precision workloads far more efficiently.

Memory architecture also separates the two. The P106-100 uses GDDR5 at 2002 MHz (8 Gbps effective) across a 192-bit bus, yielding 192.2 GB/s. The Arc B570 uses GDDR6 at 2375 MHz (19 Gbps effective) across a 160-bit bus, achieving 380.0 GB/s despite the narrower interface. The larger 10 GB capacity on the Arc B570 versus 6 GB on the P106-100 further benefits modern games and compute tasks that require larger working sets.

Clock behavior differs as well. The P106-100 has a 1506 MHz base and 1709 MHz boost. The Arc B570 runs a flat 2500 MHz for both base and boost, a higher sustained frequency that contributes to its throughput advantage. The bus interface also reflects generational progress: the P106-100 uses PCIe 1.0 x16, while the Arc B570 uses PCIe 4.0 x8, offering far greater bandwidth for data transfers to and from the host system.

API support is another clear differentiator. The P106-100 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The Arc B570 supports DirectX 12 Ultimate (12_2), which includes features like mesh shaders and enhanced ray tracing, alongside the same OpenGL 4.6 and Vulkan 1.4. The Arc B570's 5 nm process, larger transistor count, higher density, and modern Xe2-HPG architecture all contribute to its benchmark dominance. The P106-100 was designed for mining with no display outputs, while the Arc B570 is a fully featured consumer graphics card. This fundamental purpose difference, combined with the seven-year gap in release dates, explains the complete sweep in head-to-head results.