AMD Radeon RX 470 vs Intel Arc A370M Comparison

Intel Arc A370M and AMD Radeon RX 470 are both end-of-life graphics solutions that land in the same performance tier, with average benchmark scores of 29175 and 28996 respectively. The data places them within 1% of each other overall, yet their individual test results reveal starkly different strengths that matter depending on the workload.

Head-to-Head Benchmarks

The head-to-head results are one-sided, with the AMD Radeon RX 470 winning both recorded tests. In Geekbench OpenCL, the RX 470 scores 33568 against the Arc A370M’s 29676, a delta of -11.6% for Intel. That is a meaningful gap—roughly a tenth of the performance difference—and it shows the older GCN architecture still holds a clear edge in compute-heavy OpenCL tasks despite its age.

The Vulkan gap is even wider. The RX 470 posts 39884 in Geekbench Vulkan, while the Arc A370M manages only 28673, a delta of -28.1%. This is not a marginal loss for Intel; it is a decisive defeat in a modern graphics API. For any application that leverages Vulkan—think newer game engines or compute workloads using that API—the RX 470 is over a quarter faster, which is a massive practical advantage.

However, the average benchmark scores tell a more nuanced story. The Arc A370M’s average is 29175, while the RX 470 averages 28996. The Intel part is actually 0.6% ahead in this aggregate metric, according to its nearestRivals data. The RX 470’s own nearestRivals entry lists the Arc A370M with a deltaPct of -0.6%, confirming the same relationship from the opposite side. This means that outside the two head-to-head tests, the Arc A370M must be picking up wins in other benchmarks not listed here—likely in areas like ray tracing or specific synthetic workloads where its newer architecture excels.

Both cards sit at the 74th percentile against all GPUs, so neither is a high-end part by modern standards. The data shows two products that are roughly equivalent in overall standing but diverge sharply on specific APIs. If you only looked at average scores, you would call this a tie; the head-to-head results, though, show the RX 470 as the clear winner in every direct comparison available.

Where Each One Wins

The RX 470 wins in raw compute throughput in both tested APIs. Its OpenCL score of 33568 is 3892 points higher than the Arc A370M’s, and its Vulkan score of 39884 is 11211 points higher. This makes it the obvious choice for any workload that relies on OpenCL or Vulkan—from GPU-accelerated rendering to compute tasks and games built on Vulkan engines. The RX 470 also has a higher FP32 rating at 4.940 TFLOPS versus the Arc A370M’s 4.198 TFLOPS, plus a much wider 256-bit memory bus that provides 211.2 GB/s of bandwidth compared to the Arc’s 64-bit bus and 112.0 GB/s. For memory-heavy tasks, that bandwidth advantage is substantial.

The Arc A370M wins on architectural modernity and feature set. It is built on a 6 nm TSMC process versus the RX 470’s 14 nm GlobalFoundries node, and it includes 8 dedicated ray tracing cores—something the RX 470 lacks entirely. Its FP16 throughput is 8.397 TFLOPS (2:1 ratio) versus the RX 470’s 4.940 TFLOPS (1:1), meaning the Intel part can double its math rate in half-precision workloads. The Arc also supports DirectX 12 Ultimate (12_2) while the RX 470 is limited to DirectX 12 (12_0), and it runs Vulkan 1.4 against the RX 470’s Vulkan 1.3. For applications that use ray tracing or newer DirectX 12 Ultimate features, the Arc A370M is the only one of the two that can handle them at all.

The Arc’s power efficiency is another clear win. Its TDP is 35 W versus the RX 470’s 120 W, making it suitable for a compact integrated form factor (IGP slot width) with no power connectors required. The RX 470 needs a dual-slot cooler, a 1x 6-pin power connector, and a 300 W suggested PSU. The data does not include direct power consumption measurements, but the TDP figures alone tell the story: the Arc draws far less power for a similar average score, which is a major advantage in laptops or small-form-factor builds.

The Verdict

Pick the AMD Radeon RX 470 if your priority is raw performance in OpenCL or Vulkan. The benchmark results are unambiguous: it wins both head-to-head tests by 11.6% and 28.1% respectively, and its higher memory bandwidth and FP32 throughput make it the stronger choice for traditional rasterization and compute. This is the card for anyone running older APIs or workloads that do not use ray tracing. Its 4 GB of GDDR5 memory matches the Arc’s 4 GB of GDDR6, so capacity is not a differentiator, but the RX 470’s 256-bit bus moves data nearly twice as fast.

Pick the Intel Arc A370M if you need modern features or efficiency. Its 8 ray tracing cores, DirectX 12 Ultimate support, and higher Vulkan version make it future-proof in ways the RX 470 cannot match. The 35 W TDP is a fraction of the RX 470’s 120 W, and the integrated form factor means it can fit in systems where a dual-slot card will not. The data shows the Arc is only 0.6% behind in average score, so you are not sacrificing much overall performance to get those features.

For gaming, the RX 470 is the safer bet based on Vulkan performance, since many modern games use that API. For productivity or creative work that leverages ray tracing or half-precision FP16 math, the Arc A370M is the only viable option. The verdict is not about which is faster—the RX 470 wins that contest—but about which workloads you actually run.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The Intel Arc A370M has an average benchmark score of 29175, while the AMD Radeon RX 470 averages 28996, putting Intel ahead by 0.6%.

Q: How much faster is the RX 470 in Vulkan?

A: The RX 470 scores 39884 in Geekbench Vulkan versus the Arc A370M’s 28673, a delta of 28.1% in favor of AMD.

Q: Does the Arc A370M support ray tracing?

A: Yes, the Arc A370M has 8 dedicated ray tracing cores, while the RX 470 has none listed in its specifications.

Q: What is the power consumption difference?

A: The Arc A370M has a TDP of 35 W, while the RX 470 has a TDP of 120 W, making the Intel part significantly more power-efficient.

Q: Which card has more memory bandwidth?

A: The RX 470 has 211.2 GB/s of bandwidth from its 256-bit bus, compared to the Arc A370M’s 112.0 GB/s from a 64-bit bus.

Q: Are both cards still in production?

A: No, both are marked as end-of-life in the data. The RX 470 was released in 2016, and the Arc A370M was released in 2022.

Architecture Differences

The two GPUs come from entirely different design philosophies. The Intel Arc A370M uses the DG2-128 chip based on the Xe-HPG architecture, part of the Alchemist generation for Arc 3 Mobile. It is fabricated on a 6 nm TSMC process and packs 7,200 million transistors into a 157 mm² die, giving a transistor density of 45.9M per mm². The AMD Radeon RX 470 uses the Ellesmere chip with GCN 4.0 architecture, from the Arctic Islands generation. It is built on a 14 nm GlobalFoundries process with 5,700 million transistors on a larger 232 mm² die, resulting in a density of only 24.6M per mm².

The Intel architecture includes 8 ray tracing cores, a feature completely absent from the RX 470. The Arc also supports DirectX 12 Ultimate (12_2), while the RX 470 is limited to DirectX 12 (12_0). Vulkan support differs too: the Arc runs Vulkan 1.4, and the RX 470 runs Vulkan 1.3. OpenGL support is identical at 4.6 for both. The Arc’s FP16 throughput is 8.397 TFLOPS using a 2:1 ratio, effectively double its FP32 rate, while the RX 470’s FP16 is 4.940 TFLOPS at a 1:1 ratio, meaning no half-precision acceleration.

Process node and manufacturing are major divides. The 6 nm TSMC process gives Intel a density advantage of nearly 2x, while the 14 nm GlobalFoundries node used by AMD is older and less efficient. The Arc’s 6 nm process also enables its much lower 35 W TDP. The RX 470’s architecture is simpler, with no ray tracing hardware and older API support, but its wider memory interface compensates in bandwidth-bound scenarios.

Specification Differences

The two cards differ in nearly every core specification. The Arc A370M has 1024 shading units, 64 TMUs, and 32 ROPs, while the RX 470 has 2048 shading units, 128 TMUs, and 32 ROPs. The RX 470 has double the shading units and TMUs, though the Arc compensates with higher clocks: 1550 MHz base and 2050 MHz boost versus the RX 470’s 926 MHz base and 1206 MHz boost. Pixel rates reflect this: the Arc hits 65.60 GPixel/s versus the RX 470’s 38.59 GPixel/s, but texture rates favor AMD at 154.4 GTexel/s versus the Arc’s 131.2 GTexel/s.

Memory configurations differ significantly. Both have 4 GB, but the Arc uses GDDR6 at 1750 MHz (14 Gbps effective) on a 64-bit bus, yielding 112.0 GB/s. The RX 470 uses GDDR5 at 1650 MHz (6.6 Gbps effective) on a 256-bit bus, yielding 211.2 GB/s—nearly double the bandwidth. The Arc’s memory clock is higher in MHz, but the RX 470’s wider bus wins on throughput.

Power and physical specs diverge sharply. The Arc A370M has a 35 W TDP, an IGP slot width, and no power connectors, while the RX 470 has a 120 W TDP, a dual-slot design, and a single 6-pin power connector with a 300 W suggested PSU. The RX 470 has defined dimensions of 240 mm length, 95 mm height, and 35 mm width, while the Arc lists no dimensions. The bus interface differs too: the Arc uses PCIe 4.0 x8, and the RX 470 uses PCIe 3.0 x16. Display outputs on the Arc are portable-device dependent, while the RX 470 offers 1x HDMI 2.0b and 3x DisplayPort 1.4a. The RX 470 had a launch MSRP of 179 USD; the Arc has no launch MSRP listed.