AMD Steam Machine GPU vs Intel Arc Pro B370 Comparison

FAQ

Q: What are the core differences in GPU architecture between the AMD Steam Machine GPU and the Intel Arc Pro B370?

A: The AMD Steam Machine GPU uses the Navi 33 chip with RDNA 3.0 architecture and a 6 nm TSMC process, while the Intel Arc Pro B370 uses the Panther Lake chip with Xe3-LPG architecture and a 3 nm Intel process. The AMD part is a discrete console GPU designed for Valve, whereas the Intel part is an integrated graphics processor (IGP) for mobile or portable systems.

Q: How do the two GPUs compare in raw compute performance?

A: The AMD Steam Machine GPU delivers 17.56 TFLOPS of FP32 compute, while the Intel Arc Pro B370 delivers 6.144 TFLOPS. This means the AMD GPU has roughly 2.86 times the single-precision floating-point throughput of the Intel part. In FP16, AMD maintains 17.56 TFLOPS with a 1:1 ratio, while Intel reaches 12.29 TFLOPS with a 2:1 ratio.

Q: What are the memory configurations of each GPU?

A: The AMD Steam Machine GPU has 8 GB of GDDR6 memory on a 128-bit bus with 288.0 GB/s bandwidth. The Intel Arc Pro B370 uses system shared memory, with type, bus width, and bandwidth all listed as system dependent. The AMD GPU offers dedicated memory bandwidth, while the Intel IGP relies on the host system's memory.

Q: How do the power requirements differ?

A: The AMD Steam Machine GPU has a TDP of 110 W, while the Intel Arc Pro B370 has a TDP of 25 W. Neither GPU requires power connectors, but the AMD part is a discrete card with dimensions of 156 mm by 152 mm by 162 mm, while the Intel part is an IGP with no separate dimensions listed.

Q: What API support do both GPUs provide?

A: Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. They are identical in their API feature sets, meaning software compatibility at the API level should be similar for both parts.

Q: What are the release dates and production statuses?

A: The AMD Steam Machine GPU was released on June 28, 2026, and the Intel Arc Pro B370 was released on January 26, 2026. Both are listed as active in production. The Intel part has a predecessor named HD Graphics-WM, while the AMD part has no predecessor listed.

Architecture Differences

The AMD Steam Machine GPU and the Intel Arc Pro B370 represent fundamentally different design philosophies. The AMD part is a discrete console GPU based on the Navi 33 chip, built on TSMC's 6 nm process. It integrates 13,300 million transistors on a 204 mm² die, yielding a transistor density of 65.2 million per square millimeter. The Intel Arc Pro B370 is an integrated graphics processor built on the Panther Lake chip, manufactured on Intel's 3 nm process. Its transistor count and die size are listed as unknown, which limits direct comparison of manufacturing scale.

The compute architectures diverge significantly. AMD's RDNA 3.0 design uses 1792 shading units, 112 texture mapping units, and 64 render output units. Intel's Xe3-LPG design uses 1280 shading units, 40 TMUs, and 20 ROPs. The AMD GPU has 28 ray tracing cores, while the Intel part has 10. Neither GPU has dedicated tensor cores listed. The AMD GPU operates at a base clock of 1720 MHz, a game clock of 2250 MHz, and a boost clock of 2450 MHz. The Intel part has a much lower base clock of 300 MHz but a boost clock of 2400 MHz, suggesting the Intel design relies more heavily on boost behavior.

Memory architecture presents the largest structural difference. The AMD Steam Machine GPU uses dedicated 8 GB GDDR6 memory with a 128-bit interface and 288.0 GB/s bandwidth. The Intel Arc Pro B370 uses system shared memory, meaning its performance depends entirely on the host platform's memory subsystem. This makes the AMD GPU self-contained for graphics workloads, while the Intel IGP competes with the CPU for memory bandwidth.

The process technology gap is notable: 6 nm versus 3 nm. The AMD GPU's larger process node requires 110 W of power, while the Intel IGP operates at 25 W. The Intel part uses the Xe3-LPG architecture, which is designed for low-power integrated scenarios, while AMD's RDNA 3.0 targets a more traditional discrete GPU role. The AMD GPU has a generation designation of Console GPU (Valve), indicating its purpose as a dedicated console component, while the Intel part is categorized as Arc Graphics-WM (Panther Lake), an integrated solution for portable devices.

Head-to-Head Benchmarks

The database contains no recorded head-to-head benchmark scores for these two GPUs. The wins count for each part is zero, and the average benchmark score for both is zero. The percentile versus all GPUs is 50 for each, placing both at the median of the database's tracked GPU population, though this percentile is calculated without specific benchmark data behind it.

Without direct measurements, the recorded specifications provide the only basis for comparison. The AMD Steam Machine GPU delivers 17.56 TFLOPS of FP32 compute, which is 2.86 times the 6.144 TFLOPS of the Intel Arc Pro B370. This is the largest single-specification gap between the two parts. The pixel rate tells a similar story: the AMD GPU processes 156.8 GPixel/s versus 48.00 GPixel/s for Intel, a 3.27 times advantage. Texture rate shows the AMD GPU at 274.4 GTexel/s versus 96.00 GTexel/s for Intel, a 2.86 times difference.

The memory bandwidth gap is even more pronounced in structural terms. The AMD GPU has a fixed 288.0 GB/s of dedicated bandwidth, while the Intel part's bandwidth is system dependent and cannot be quantified from the database. The AMD GPU's 8 GB of GDDR6 memory provides predictable capacity, while the Intel IGP's memory size is likewise system dependent.

Ray tracing hardware favors the AMD part with 28 RT cores versus 10 for Intel. The shading unit count favors AMD at 1792 versus 1280, a 1.4 times difference. The AMD GPU also has more TMUs (112 versus 40) and ROPs (64 versus 20), which contributes to its higher pixel and texture throughput.

The Intel part does hold advantages in certain areas. Its FP16 throughput of 12.29 TFLOPS exceeds its FP32 figure due to the 2:1 ratio, while the AMD GPU maintains a 1:1 ratio at 17.56 TFLOPS. The Intel part's 3 nm process node is smaller than AMD's 6 nm node, and its 25 W TDP is dramatically lower than the AMD GPU's 110 W. The Intel part also has a higher boost clock relative to its base clock, suggesting different power management characteristics.

The Verdict

The recorded data indicates that the AMD Steam Machine GPU is the stronger performer across nearly every measurable specification. Its FP32 compute is 2.86 times higher, its pixel rate is 3.27 times higher, and its texture rate is 2.86 times higher than the Intel Arc Pro B370. The AMD part has more shading units, more TMUs, more ROPs, and more ray tracing cores. It has dedicated memory with fixed bandwidth, while the Intel part depends on system memory.

The Intel Arc Pro B370 is the more power-efficient option. Its 25 W TDP is less than one quarter of the AMD GPU's 110 W. It uses a smaller 3 nm process node and is designed as an integrated processor, which eliminates the need for a separate card or power connectors. For systems where power consumption is the primary constraint, the Intel part is the clear choice.

The AMD GPU includes a game clock of 2250 MHz, a specification the Intel part does not list. The AMD part also has defined dimensions of 156 mm by 152 mm by 162 mm, while the Intel IGP has no listed dimensions. The production status for both parts is active, and both share identical API support.

The data suggests that the AMD Steam Machine GPU targets scenarios requiring maximum graphics throughput in a dedicated console form factor. The Intel Arc Pro B370 targets scenarios where integration and low power draw are paramount. Neither part has benchmark scores in the database, so the verdict rests entirely on specification analysis.

Specification Differences

Process Node: AMD uses 6 nm (TSMC), Intel uses 3 nm (Intel).

Transistors: AMD has 13,300 million, Intel is unknown.

Die Size: AMD is 204 mm², Intel is unknown.

Transistor Density: AMD is 65.2M per mm², Intel is not listed.

Base Clock: AMD is 1720 MHz, Intel is 300 MHz.

Boost Clock: AMD is 2450 MHz, Intel is 2400 MHz.

Game Clock: AMD is 2250 MHz, Intel does not list one.

Memory Size: AMD has 8 GB GDDR6, Intel uses system shared.

Memory Type: AMD is GDDR6, Intel is system shared.

Memory Bus Width: AMD is 128 bit, Intel is system shared.

Memory Bandwidth: AMD is 288.0 GB/s, Intel is system dependent.

Shading Units: AMD has 1792, Intel has 1280.

TMUs: AMD has 112, Intel has 40.

ROPs: AMD has 64, Intel has 20.

RT Cores: AMD has 28, Intel has 10.

Pixel Rate: AMD is 156.8 GPixel/s, Intel is 48.00 GPixel/s.

Texture Rate: AMD is 274.4 GTexel/s, Intel is 96.00 GTexel/s.

FP32: AMD is 17.56 TFLOPS, Intel is 6.144 TFLOPS.

FP16: AMD is 17.56 TFLOPS (1:1), Intel is 12.29 TFLOPS (2:1).

TDP: AMD is 110 W, Intel is 25 W.

Slot Width: AMD is not listed, Intel is IGP.

Bus Interface: AMD is not listed, Intel is IGP.

Display Outputs: AMD has 1x HDMI 2.1a and 1x DisplayPort 2.1, Intel is portable device dependent.

Dimensions: AMD is 156 mm by 152 mm by 162 mm, Intel has none listed.

Release Date: AMD is June 28, 2026, Intel is January 26, 2026.

Predecessor: AMD has none, Intel has HD Graphics-WM.

Foundry: AMD uses TSMC, Intel uses Intel.

Architecture: AMD uses RDNA 3.0, Intel uses Xe3-LPG.

Chip: AMD uses Navi 33, Intel uses Panther Lake.

Generation: AMD is Console GPU (Valve), Intel is Arc Graphics-WM (Panther Lake).

Memory Clock: AMD is 2250 MHz (18 Gbps effective), Intel is system shared.

Where Each One Wins

The AMD Steam Machine GPU wins in raw compute throughput. Its FP32 figure of 17.56 TFLOPS is nearly three times the Intel part's 6.144 TFLOPS. This translates to advantages in pixel rate, texture rate, and shading unit count. Applications that stress floating-point math, texture filtering, or rasterization will favor the AMD part. The dedicated 8 GB GDDR6 memory with 288.0 GB/s bandwidth ensures consistent memory performance without contention from other system components. The 28 RT cores give it a hardware advantage for ray tracing workloads.

The Intel Arc Pro B370 wins in power efficiency and integration. Its 25 W TDP allows deployment in systems where the AMD GPU's 110 W requirement would be impractical. The 3 nm process node represents a more advanced manufacturing technology, and the IGP form factor means no additional card slot or power wiring is needed. The FP16 ratio of 2:1 means the Intel part achieves 12.29 TFLOPS in half-precision workloads, which is closer to the AMD part's 17.56 TFLOPS than the FP32 comparison suggests. Applications using FP16 math narrow the performance gap.

The display output difference matters for use cases. The AMD GPU provides fixed outputs of 1x HDMI 2.1a and 1x DisplayPort 2.1, making it suitable for direct connection to displays. The Intel IGP's display outputs are portable device dependent, meaning the host device determines connectivity. The AMD GPU has a game clock specification, indicating it is designed for sustained gaming workloads. The Intel part has no game clock, consistent with its integrated, power-limited design.

The release timing differs by roughly five months, with Intel first and AMD later. The Intel part has a documented predecessor, HD Graphics-WM, while the AMD part does not. Both GPUs share identical DirectX, OpenGL, and Vulkan API support, so software compatibility is not a differentiator. The data positions the AMD Steam Machine GPU as the performance-oriented discrete option and the Intel Arc Pro B370 as the low-power integrated option, each serving distinct hardware environments.