AMD Radeon RX 9070 vs AMD Steam Machine GPU Comparison

Head-to-Head Benchmarks

The recorded database contains no direct head-to-head benchmark results between the AMD Radeon RX 9070 and the AMD Steam Machine GPU. The head-to-head benchmark array is empty, and there are no shared test entries where both products were measured under identical conditions. Consequently, a direct per-test comparison is not possible from the available data. Instead, the analysis must rely on the individual benchmark suites recorded for the RX 9070 and the architectural specifications of the Steam Machine GPU, which has no benchmark scores in the database.

The RX 9070 has an average benchmark score of 23877 across all recorded tests, placing it in the 69th percentile of all GPUs in the database. Its strongest recorded result is in Geekbench OpenCL, where it scores 131539 points. In Geekbench Vulkan, it delivers 58705 points. The 3DMark Steel Nomad DX12 test yields 6290 points. Passmark G3D shows 25381 points, while Passmark GPU Compute records 14737 points. DirectX 9 legacy performance in Passmark reaches 343 points, DirectX 11 scores 281 points, DirectX 10 scores 141 points, and DirectX 12 scores 74 points. The Passmark G2D score is 1280 points.

The Steam Machine GPU has no recorded benchmark scores, a zero average benchmark score, and no nearest rivals listed in the database. Its percentile versus all GPUs is 50, but this figure cannot be tied to any measured workload because no tests are registered. The data therefore supports only a specification-level comparison and a performance projection based on arithmetic differences in shading units, texture units, and clock rates. The RX 9070 carries 3584 shading units, 224 texture mapping units, and 128 render output units. The Steam Machine GPU carries 1792 shading units, 112 texture mapping units, and 64 render output units. These are exact 2:1 ratios in each category. The RX 9070 has a boost clock of 2520 MHz, while the Steam Machine GPU has a boost clock of 2450 MHz. The RX 9070 also has a higher game clock at 2070 MHz versus 2250 MHz for the Steam Machine GPU; the Steam Machine GPU actually runs a higher game clock by 180 MHz, although its boost ceiling is 70 MHz lower.

The FP32 throughput figures reflect these differences. The RX 9070 reaches 36.13 TFLOPS, exactly double the 17.56 TFLOPS of the Steam Machine GPU. Pixel rate for the RX 9070 is 322.6 GPixel/s versus 156.8 GPixel/s, a 2.06x advantage. Texture rate is 564.5 GTexel/s versus 274.4 GTexel/s, a 2.06x advantage. These ratios are consistent with the hardware resource counts: the RX 9070 has twice the shading units, twice the TMUs, and twice the ROPs, with only a small clock disadvantage in game clock. The memory subsystem amplifies the gap. The RX 9070 has 16 GB of GDDR6 on a 256-bit bus, delivering 644.6 GB/s of bandwidth. The Steam Machine GPU has 8 GB of GDDR6 on a 128-bit bus, delivering 288.0 GB/s. The RX 9070 offers 2.24x the memory bandwidth. In any bandwidth-sensitive workload, the RX 9070 holds a decisive advantage.

The nearest rivals for the RX 9070 in the database provide context for its average score. The NVIDIA GeForce GTX TITAN Z averages 23736 points, which is 0.6% lower than the RX 9070. The AMD Radeon RX 6800S averages 24063 points, which is 0.8% higher. The NVIDIA GeForce RTX 3080 Mobile averages 23628 points, which is 1.1% lower. The NVIDIA GeForce RTX 2080 SUPER averages 24170 points, which is 1.2% higher. These deltas are small, indicating that the RX 9070 sits in a tightly clustered performance band around these four products. The RX 9070 is effectively within 1.2% of all four nearest rivals, meaning its average benchmark score neither dominates nor trails by a meaningful margin. For the Steam Machine GPU, no such rival context exists in the database, so no comparable placement can be stated.

Architecture Differences

The RX 9070 uses the Navi 48 chip built on RDNA 4.0 architecture, fabricated on a 4 nm process at TSMC. The Steam Machine GPU uses the Navi 33 chip built on RDNA 3.0 architecture, fabricated on a 6 nm process at TSMC. The process node difference is significant: 4 nm versus 6 nm. The RX 9070 integrates 53,900 million transistors on a 357 mm² die, giving a transistor density of 151.0 million transistors per mm². The Steam Machine GPU integrates 13,300 million transistors on a 204 mm² die, giving a transistor density of 65.2 million transistors per mm². The RX 9070 has 4.05x the transistor count and 1.75x the die area, but 2.32x the transistor density. This density gap reflects both the newer manufacturing process and the different architectural generation.

The RX 9070 belongs to the Navi IV generation (RX 9000 series) and is listed as having the predecessor Navi III. The Steam Machine GPU is classified as a console GPU for Valve, with the codename Hotpink Bonefish. Its generation field reads "Console GPU (Valve)" and it has no predecessor or successor listed. The RX 9070 has no codename recorded, while the Steam Machine GPU has a codename. The RX 9070 has a launch MSRP of 549 USD, while the Steam Machine GPU has no launch MSRP recorded. The RX 9070 release date is 2025-03-05, while the Steam Machine GPU release date is 2026-06-28.

Both GPUs share the same API support: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither has tensor cores recorded. The ray tracing core counts differ: the RX 9070 has 56 RT cores, the Steam Machine GPU has 28 RT cores, again a 2:1 ratio. The RX 9070 has a TDP of 220 W and requires two 8-pin power connectors plus a 550 W suggested PSU. The Steam Machine GPU has a TDP of 110 W, no power connectors, and no suggested PSU recorded. The RX 9070 is dual-slot and uses a PCIe 5.0 x16 bus interface. The Steam Machine GPU has no slot width and no bus interface recorded, but its physical dimensions are listed: 156 mm length, 152 mm height, and 162 mm width. The RX 9070 has no dimensions recorded.

Display outputs differ in count and version. The RX 9070 provides one HDMI 2.1b port and three DisplayPort 2.1a ports. The Steam Machine GPU provides one HDMI 2.1a port and one DisplayPort 2.1 port. The RX 9070 offers more display outputs and newer port revisions. Memory clocks also differ. The RX 9070 memory clock is 2518 MHz with 20.1 Gbps effective data rate. The Steam Machine GPU memory clock is 2250 MHz with 18 Gbps effective. The RX 9070 uses GDDR6 memory, as does the Steam Machine GPU, but the bus width differs by 2x (256-bit versus 128-bit).

The FP16 performance is listed as 1:1 with FP32 for both parts: 36.13 TFLOPS for the RX 9070 and 17.56 TFLOPS for the Steam Machine GPU. Neither part has separate FP16 hardware acceleration beyond the 1:1 ratio. The RX 9070 base clock is 1330 MHz, which is 390 MHz lower than its boost clock of 2520 MHz. The Steam Machine GPU base clock is 1720 MHz, which is 730 MHz lower than its boost clock of 2450 MHz. The Steam Machine GPU has a higher base clock by 390 MHz, but a lower boost clock by 70 MHz. The game clock for the RX 9070 is 2070 MHz, while the Steam Machine GPU game clock is 2250 MHz, so the Steam Machine GPU runs 180 MHz faster in the game clock metric. Despite this, the RX 9070 still achieves double the FP32 throughput because it has exactly twice the shading units.

The Verdict

The data supports a clear performance hierarchy. The RX 9070 has double the shading units, double the texture mapping units, double the render output units, double the ray tracing cores, double the memory capacity, double the memory bus width, and more than double the memory bandwidth relative to the Steam Machine GPU. Its FP32 throughput is exactly 2.06x higher (36.13 TFLOPS versus 17.56 TFLOPS), its pixel rate is 2.06x higher, and its texture rate is 2.06x higher. The RX 9070 also uses a newer architecture (RDNA 4.0 versus RDNA 3.0), a newer process node (4 nm versus 6 nm), and a newer generation (Navi IV versus the console-class Navi 33). The RX 9070 has a recorded average benchmark score of 23877 with percentile 69, while the Steam Machine GPU has no recorded benchmark scores and a percentile of 50 that is not tied to any measured test.

For a user selecting between these two AMD parts strictly from the database, the RX 9070 is the stronger GPU for every measured category: compute throughput, pixel fill, texture fill, memory bandwidth, ray tracing core count, and API port revision. The Steam Machine GPU does hold advantages in base clock (1720 MHz versus 1330 MHz) and game clock (2250 MHz versus 2070 MHz), and it consumes half the power (110 W versus 220 W). It also has no power connectors, which suits compact console integration. The RX 9070 requires two 8-pin connectors and a 550 W suggested PSU, plus a dual-slot cooler and a PCIe 5.0 x16 slot.

The Steam Machine GPU is best understood as a lower-power, smaller-footprint part for a fixed console chassis. The RX 9070 is best understood as a full-size discrete graphics card. Neither product has head-to-head benchmark data, so the verdict rests on the specification deltas and the RX 9070's measured benchmark scores. The RX 9070 sits within 1.2% of four nearest rivals in the database, which places it in a competitive mid-range cluster. The Steam Machine GPU has no such cluster, so its performance cannot be anchored to any measured peer. The percentile field of 50 for the Steam Machine GPU should be treated cautiously, as it is not derived from any recorded benchmark score.

The RX 9070 is the choice when maximum throughput, higher memory bandwidth, and more display outputs are required. The Steam Machine GPU is the choice when power draw, physical size, and connector requirements dominate, because it draws 110 W versus 220 W, uses no power connectors, and measures 156 mm by 152 mm by 162 mm. The RX 9070 has a launch MSRP of 549 USD. The Steam Machine GPU has no launch MSRP recorded.

Specification Differences

The following fields differ between the two products in the database:

• Chip: Navi 48 versus Navi 33
• Architecture: RDNA 4.0 versus RDNA 3.0
• Codename: null versus Hotpink Bonefish
• Generation: Navi IV (RX 9000) versus Console GPU (Valve)
• Process node: 4 nm versus 6 nm
• Transistors: 53,900 million versus 13,300 million
• Die size: 357 mm² versus 204 mm²
• Transistor density: 151.0M / mm² versus 65.2M / mm²
• Base clock: 1330 MHz versus 1720 MHz
• Boost clock: 2520 MHz versus 2450 MHz
• Game clock: 2070 MHz versus 2250 MHz
• Memory clock: 2518 MHz 20.1 Gbps effective versus 2250 MHz 18 Gbps effective
• Memory size: 16 GB versus 8 GB
• Memory bus width: 256 bit versus 128 bit
• Memory bandwidth: 644.6 GB/s versus 288.0 GB/s
• Shading units: 3584 versus 1792
• TMUs: 224 versus 112
• ROPs: 128 versus 64
• RT cores: 56 versus 28
• Pixel rate: 322.6 GPixel/s versus 156.8 GPixel/s
• Texture rate: 564.5 GTexel/s versus 274.4 GTexel/s
• FP32: 36.13 TFLOPS versus 17.56 TFLOPS
• FP16: 36.13 TFLOPS (1:1) versus 17.56 TFLOPS (1:1)
• TDP: 220 W versus 110 W
• Slot width: Dual-slot versus null
• Power connectors: 2x 8-pin versus None
• Suggested PSU: 550 W versus null
• Bus interface: PCIe 5.0 x16 versus null
• Display outputs: 1x HDMI 2.1b, 3x DisplayPort 2.1a versus 1x HDMI 2.1a, 1x DisplayPort 2.1
• Dimensions: null versus 156 mm 6.1 inches length, 152 mm 6 inches height, 162 mm 6.4 inches width
• Release date: 2025-03-05 versus 2026-06-28
• Launch MSRP: 549 USD versus null
• Predecessor: Navi III versus null
• Successor: null versus null
• Benchmarks: 10 recorded tests versus none
• Percentile: 69 versus 50
• Average benchmark score: 23877 versus 0
• Nearest rivals: four entries versus none

Fields that are identical include manufacturer (AMD), memory type (GDDR6), API support (DirectX 12 Ultimate 12_2, OpenGL 4.6, Vulkan 1.4), tensor cores (null for both), and production status (Active).

FAQ

Q: Which GPU has higher FP32 compute throughput?

A: The AMD Radeon RX 9070 delivers 36.13 TFLOPS, exactly double the 17.56 TFLOPS of the AMD Steam Machine GPU.

Q: How much memory bandwidth does each GPU provide?

A: The RX 9070 provides 644.6 GB/s over a 256-bit bus with 16 GB of GDDR6. The Steam Machine GPU provides 288.0 GB/s over a 128-bit bus with 8 GB of GDDR6.

Q: What are the process nodes for the two GPUs?

A: The RX 9070 uses a 4 nm process at TSMC with 53,900 million transistors on a 357 mm² die. The Steam Machine GPU uses a 6 nm process at TSMC with 13,300 million transistors on a 204 mm² die.

Q: Does the Steam Machine GPU have any recorded benchmark scores?

A: No. The Steam Machine GPU has an empty benchmark array, an average benchmark score of 0, and no nearest rivals in the database. The RX 9070 has 10 recorded benchmark scores with an average of 23877 and a 69th percentile placement.

Q: What is the power draw difference between the two products?

A: The RX 9070 has a TDP of 220 W and requires two 8-pin power connectors with a 550 W suggested PSU. The Steam Machine GPU has a TDP of 110 W and uses no power connectors, with no suggested PSU recorded.

Q: Which GPU has more ray tracing cores?

A: The RX 9070 has 56 ray tracing cores. The Steam Machine GPU has 28 ray tracing cores, exactly half the count.