AMD Radeon 820M vs Intel Arc B580 Comparison

The Verdict

The AMD Radeon 820M is an integrated graphics processor designed for thin-and-light portable devices, drawing just 15 W and sharing system memory. The Intel Arc B580 is a discrete dual-slot add-in card with a 190 W TDP, 12 GB of dedicated GDDR6 memory, and a 450 W suggested power supply. The recorded data shows no head-to-head benchmark wins for the Radeon 820M, while the Arc B580 holds every recorded advantage. The Arc B580 sits at the 68th percentile of all GPUs in the database, with an average benchmark score of 23021. The Radeon 820M has no recorded benchmark scores and sits at the 50th percentile, which reflects its role as an IGP rather than a performance part.

The Arc B580 is the only choice for anyone assembling a desktop with a dedicated graphics card, as it delivers a full suite of measured results across DirectX 9 through 12, OpenCL, Vulkan, and compute workloads. The Radeon 820M is for portable systems where the GPU is fused into the processor, uses no power connector, and relies entirely on system memory. The data does not support using the 820M for anything beyond what an integrated solution can do in a 15 W envelope. The Arc B580, by contrast, is a standalone card that outperforms the Radeon RX 580 2048SP by 0.2% in average score, trails the GeForce RTX 2080 by 0.6%, and sits 0.7% behind the RTX 3080. Those deltas place it in the same performance class as high-end cards from the previous several generations, while the 820M has no comparable measured rivals.

Architecture Differences

The AMD Radeon 820M uses the RDNA 3.5 architecture, built on a 4 nm TSMC process, and belongs to the Navi III IGP generation for Strix Point Mobile. It is built on the Krackan Point 2 chip. The Intel Arc B580 uses the Xe2-HPG architecture, built on a 5 nm TSMC process, and belongs to the Battlemage generation under the Arc 5 family, with the BMG-G21 chip. The process node difference is small, 4 nm versus 5 nm, but the two parts are fundamentally different classes of hardware.

The Radeon 820M has 128 shading units, 8 texture mapping units, 4 render output units, and 2 ray tracing cores. The Arc B580 has 2560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores. That is a 20x difference in shading units and a 10x difference in RT cores. The Radeon 820M carries no transistor count or die size in the database, while the Arc B580 uses 19,600 million transistors on a 272 mm² die with a transistor density of 72.1 million per square millimeter.

Memory architecture separates these two completely. The Radeon 820M uses system shared memory for capacity, type, and bus width, with bandwidth described as system dependent. The Arc B580 has 12 GB of GDDR6 on a 192-bit bus with 456.0 GB/s of bandwidth. The Radeon 820M has no tensor cores listed, and the Arc B580 also has no tensor cores listed, but the Arc B580 has dedicated RT cores in far greater quantity.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Radeon 820M uses a PCIe 4.0 x8 interface, and the Arc B580 uses the same PCIe 4.0 x8 interface. The Radeon 820M has no power connectors because it is an IGP, while the Arc B580 requires one 8-pin connector and a 450 W suggested power supply.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark entries between these two GPUs. The Radeon 820M has zero recorded benchmark scores in its profile, while the Arc B580 has ten recorded results. The wins tally shows 0 wins for the AMD part and 0 wins for the Intel part in direct head-to-head tests, but the absence of shared tests means the comparison must rely on the Arc B580's absolute scores and the 820M's lack of any measured output.

The Arc B580 delivers a 3DMark Steel Nomad DX12 score of 3068. In Geekbench, it scores 92821 in OpenCL and 109672 in Vulkan. Passmark results show a G3D score of 15748, a GPU compute score of 7729, and legacy API scores of 183 in DirectX 9, 128 in DirectX 11, 76 in DirectX 10, and 76 in DirectX 12. The Passmark G2D score is 709. The average benchmark score across all recorded tests is 23021.

The nearest rivals in the database for the Arc B580 put its performance in context. The AMD Radeon RX 580 2048SP averages 23061, which is 0.2% higher than the B580. The NVIDIA GeForce RTX 2080 averages 22895, which is 0.6% lower. The NVIDIA GeForce RTX 3080 averages 23172, which is 0.7% higher. The NVIDIA P106-100 averages 23249, which is 1% higher. The Arc B580 therefore lands within a 1% band of four established discrete GPUs, including an RTX 3080. The Radeon 820M has no such comparison points, with no scores and no nearest rivals recorded.

The pixel rate and texture rate figures reinforce the class gap. The Radeon 820M outputs 11.20 GPixel/s and 22.40 GTexel/s. The Arc B580 outputs 213.6 GPixel/s and 427.2 GTexel/s. Floating point performance shows a similar divide: the Radeon 820M reaches 716.8 GFLOPS in FP32 with FP16 at a 1:1 ratio, while the Arc B580 reaches 13.67 TFLOPS in FP32 and 27.34 TFLOPS in FP16 at a 2:1 ratio.

FAQ

Q: Which GPU has more shading units?

A: The Intel Arc B580 has 2560 shading units. The AMD Radeon 820M has 128 shading units.

Q: How much memory does each GPU have?

A: The Intel Arc B580 has 12 GB of GDDR6 on a 192-bit bus with 456.0 GB/s bandwidth. The AMD Radeon 820M uses system shared memory for capacity, type, and bus width, with bandwidth listed as system dependent.

Q: What is the power draw of each part?

A: The AMD Radeon 820M has a 15 W TDP and uses no power connectors. The Intel Arc B580 has a 190 W TDP, requires one 8-pin power connector, and has a 450 W suggested power supply.

Q: Does the Arc B580 support the same APIs as the Radeon 820M?

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

Q: What does the Arc B580's average benchmark score compare to?

A: The Arc B580 averages 23021. It sits 0.2% behind the AMD Radeon RX 580 2048SP, 0.6% ahead of the NVIDIA GeForce RTX 2080, 0.7% behind the NVIDIA GeForce RTX 3080, and 1% behind the NVIDIA P106-100.

Q: Are there any recorded benchmark scores for the Radeon 820M?

A: No. The Radeon 820M has zero recorded benchmark scores in the database. The Arc B580 has scores for 3DMark Steel Nomad DX12, Geekbench OpenCL, Geekbench Vulkan, and multiple Passmark tests.

Where Each One Wins

The Intel Arc B580 wins in every measured category where data exists. It delivers 3068 in 3DMark Steel Nomad DX12, 92821 in Geekbench OpenCL, and 109672 in Geekbench Vulkan. Its Passmark G3D score of 15748 and GPU compute score of 7729 show strong rasterization and compute capability. The 12 GB GDDR6 frame buffer with 456.0 GB/s bandwidth supports modern game loads and high-resolution textures, while 20 RT cores provide hardware ray tracing support. The 13.67 TFLOPS FP32 throughput and 27.34 TFLOPS FP16 throughput position it as a capable discrete card for gaming and general GPU compute.

The Radeon 820M wins in power efficiency and physical integration. Its 15 W TDP is dramatically lower than the Arc B580's 190 W TDP. It requires no power connector, no separate slot, and no suggested PSU, because it is an IGP inside a mobile processor. The 4 nm process node is smaller than the Arc B580's 5 nm node. Its display outputs are described as portable device dependent, meaning the laptop or handheld manufacturer controls connectivity. The 820M is the appropriate part for a system where the GPU must be built into the processor, run off shared memory, and stay within a 15 W thermal budget.

The Arc B580 also has the advantage in memory bandwidth by a wide margin. The B580's 456.0 GB/s dedicated bandwidth versus the 820M's system dependent shared memory means the discrete card will not compete with the CPU for memory access. The B580's 213.6 GPixel/s pixel rate is roughly 19 times the 820M's 11.20 GPixel/s. The B580's 427.2 GTexel/s texture rate is roughly 19 times the 820M's 22.40 GTexel/s. The B580's 20 RT cores versus the 820M's 2 RT cores gives it a 10x advantage in hardware ray tracing resources.

Specification Differences

The two GPUs differ on nearly every specification field in the database. The AMD Radeon 820M uses the Krackan Point 2 chip with RDNA 3.5 architecture, while the Intel Arc B580 uses the BMG-G21 chip with Xe2-HPG architecture. The 820M is built on a 4 nm TSMC process; the B580 is built on a 5 nm TSMC process.

Transistor count and die size are unknown for the 820M, while the B580 lists 19,600 million transistors on a 272 mm² die with a density of 72.1 million per square millimeter. Clock behavior differs: the 820M has a 400 MHz base and 2800 MHz boost, while the B580 has a flat 2670 MHz base and boost, with memory clocked at 2375 MHz or 19 Gbps effective.

Memory specifications are entirely different. The 820M uses system shared memory for size, type, and bus width, with system dependent bandwidth. The B580 has 12 GB GDDR6, a 192-bit bus, and 456.0 GB/s bandwidth. The 820M has 128 shading units, 8 TMUs, 4 ROPs, and 2 RT cores. The B580 has 2560 shading units, 160 TMUs, 80 ROPs, and 20 RT cores.

Pixel rate is 11.20 GPixel/s on the 820M versus 213.6 GPixel/s on the B580. Texture rate is 22.40 GTexel/s versus 427.2 GTexel/s. FP32 performance is 716.8 GFLOPS versus 13.67 TFLOPS. FP16 performance is 716.8 GFLOPS at 1:1 ratio versus 27.34 TFLOPS at 2:1 ratio.

The 820M has a 15 W TDP, IGP slot width, no power connectors, and no suggested PSU. The B580 has a 190 W TDP, dual-slot width, one 8-pin power connector, and a 450 W suggested PSU. The 820M uses a PCIe 4.0 x8 interface, and so does the B580. Display outputs are portable device dependent on the 820M, while the B580 has one HDMI 2.1a and three DisplayPort 2.1 outputs.

Physical dimensions are absent for the 820M, while the B580 measures 272 mm in length, 115 mm in height, and 45 mm in width. The B580 also has a launch MSRP of 249 USD. The 820M has no launch MSRP recorded. Release dates differ: the 820M was released on 2025-02-28, while the B580 was released on 2024-12-12. The 820M's predecessor is Navi II IGP, and the B580's predecessor is Alchemist. Both are listed as active in production. The B580 has a recorded average benchmark score of 23021 and sits at the 68th percentile, while the 820M has no score and sits at the 50th percentile.