AMD Radeon AI PRO 9600D vs AMD Steam Deck OLED GPU Comparison

AMD
RADEON

AMD Radeon AI PRO 9600D

CORE STATE Navi 48
VRAM 32 GB
CLOCK SPEED 2020 MHz
TDP 150 W
BUS WIDTH 256 bit
ARCHITECTURE RDNA 4.0
nm
PROCESS 4 nm
LAUNCH DATE 2025
VS
AMD
RADEON

Steam Deck OLED GPU

CORE STATE Sephiroth
VRAM 16 GB
CLOCK SPEED 1600 MHz
TDP 15 W
BUS WIDTH 128 bit
ARCHITECTURE RDNA 2.0
nm
PROCESS 6 nm
LAUNCH DATE 2023

Analysis: AMD Radeon AI PRO 9600D vs AMD Steam Deck OLED GPU

Head-to-Head Benchmarks

The recorded data shows no direct head-to-head benchmark results between the AMD Radeon AI PRO 9600D and the AMD Steam Deck OLED GPU. The database contains no entries for wins on either side, with both `winsA` and `winsB` at zero. Consequently, the comparison must rely on the architectural specifications and theoretical throughput figures recorded for each product.

The Radeon AI PRO 9600D delivers 24.82 TFLOPS of FP32 compute, while the Steam Deck OLED GPU delivers 1.638 TFLOPS. That is a 15.2x difference in raw shader throughput. The texture rate tells a similar story: 387.8 GTexel/s versus 51.20 GTexel/s, a 7.6x gap. Pixel throughput stands at 193.9 GPixel/s against 25.60 GPixel/s, a 7.6x difference. These figures confirm that the AI PRO 9600D operates in a completely different performance class.

Memory bandwidth amplifies the separation. The AI PRO 9600D uses a 256-bit GDDR6 interface delivering 576.0 GB/s, while the Steam Deck OLED GPU uses a 128-bit LPDDR5 bus delivering 176.0 GB/s. That is a 3.3x bandwidth advantage. Memory capacity differs even more sharply: 32 GB versus 16 GB. The AI PRO 9600D also has a larger memory clock at 2250 MHz (18 Gbps effective) versus 1375 MHz (11 Gbps effective).

The compute unit counts reflect the same hierarchy. The AI PRO 9600D has 3072 shading units, 192 texture mapping units, 96 ROPs, and 48 ray tracing cores. The Steam Deck OLED GPU has 512 shading units, 32 TMUs, 16 ROPs, and 8 ray tracing cores. The AI PRO 9600D therefore has 6x the shading units, 6x the TMUs, 6x the ROPs, and 6x the ray tracing cores.

Clock speeds partially offset the architectural gap. The AI PRO 9600D boosts to 2020 MHz, the Steam Deck OLED GPU to 1600 MHz. The base clocks are 1080 MHz and 1000 MHz respectively. The AI PRO 9600D also maintains a game clock of 1080 MHz. These clock differences are modest compared to the resource disparity.

The power envelope is a defining contrast. The AI PRO 9600D has a TDP of 150 W, while the Steam Deck OLED GPU has a TDP of 15 W. That is a 10x power draw difference. The AI PRO 9600D requires a 450 W suggested PSU and a single 16-pin power connector. The Steam Deck OLED GPU lists no power connector requirement and is a mobile, battery-integrated design.

The Verdict

The data indicates two products built for entirely different workloads. The AMD Radeon AI PRO 9600D is a professional workstation GPU with a 150 W TDP, single-slot design, 32 GB GDDR6 memory, and PCIe 5.0 x16 interface. The AMD Steam Deck OLED GPU is a handheld console GPU with a 15 W TDP, integrated into a device with a 298 mm length, 117 mm height, and 49 mm width.

Benchmark results indicate that the AI PRO 9600D dominates in every measurable compute category. Its 24.82 TFLOPS FP32 output, 576.0 GB/s memory bandwidth, and 32 GB capacity position it for large datasets and high-resolution rendering. The Steam Deck OLED GPU cannot approach these figures. Its 1.638 TFLOPS and 176.0 GB/s suit constrained, power-limited scenarios.

The percentile ranking places both at the 50th percentile among all GPUs, but this reflects the database's relative position, not their equivalence. The AI PRO 9600D uses RDNA 4.0 architecture on a 4 nm TSMC process with 53,900 million transistors on a 357 mm² die. The Steam Deck OLED GPU uses RDNA 2.0 on a 6 nm process with 2,400 million transistors on a 131 mm² die. The transistor density is 151.0M per mm² versus 18.3M per mm².

The AI PRO 9600D supports DisplayPort 2.1a output, DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Steam Deck OLED GPU outputs via USB Type-C and supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.3. The AI PRO 9600D is a single-slot PCIe 5.0 x16 card. The Steam Deck OLED GPU is a soldered console component.

Where Each One Wins

The AMD Radeon AI PRO 9600D wins in every performance-oriented category. FP32 compute is 24.82 TFLOPS versus 1.638 TFLOPS. FP16 compute is 24.82 TFLOPS (1:1 ratio) versus 3.277 TFLOPS (2:1 ratio). The AI PRO 9600D also wins on memory capacity, bus width, bandwidth, and clock speeds. It has a higher base clock, boost clock, and memory clock. It supports more display outputs and a newer PCIe generation.

The AMD Steam Deck OLED GPU wins on power efficiency. At 15 W TDP, it delivers 1.638 TFLOPS, which is 0.109 TFLOPS per watt. The AI PRO 9600D at 150 W delivers 24.82 TFLOPS, which is 0.165 TFLOPS per watt. The AI PRO 9600D is actually more efficient per watt despite drawing 10x more power. However, the Steam Deck OLED GPU's total power draw is far lower, making it suitable for battery-powered operation.

The Steam Deck OLED GPU also wins on physical integration. It has no separate slot width, power connector, or PSU requirement. It fits within a handheld device measuring 298 mm by 117 mm by 49 mm. The AI PRO 9600D requires a 241 mm by 111 mm by 19 mm card, a 450 W PSU, and a 16-pin connector.

The AI PRO 9600D wins on production readiness for professional workloads. It is listed as Active with a release date of 2025-12-10. The Steam Deck OLED GPU is also Active with a release date of 2023-11-08. Both are currently in production.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The AMD Radeon AI PRO 9600D delivers 24.82 TFLOPS, while the AMD Steam Deck OLED GPU delivers 1.638 TFLOPS. This is a 15.2x difference.

Q: How much memory does each GPU have?

A: The AI PRO 9600D has 32 GB of GDDR6 memory on a 256-bit bus. The Steam Deck OLED GPU has 16 GB of LPDDR5 memory on a 128-bit bus.

Q: What are the memory bandwidth figures?

A: The AI PRO 9600D achieves 576.0 GB/s, while the Steam Deck OLED GPU achieves 176.0 GB/s. The AI PRO 9600D has a 3.3x bandwidth advantage.

Q: Which GPU has more ray tracing cores?

A: The AI PRO 9600D has 48 ray tracing cores. The Steam Deck OLED GPU has 8 ray tracing cores.

Q: What is the power consumption of each GPU?

A: The AI PRO 9600D has a TDP of 150 W and requires a 450 W suggested PSU. The Steam Deck OLED GPU has a TDP of 15 W and lists no PSU requirement.

Q: What architectures do they use?

A: The AI PRO 9600D uses RDNA 4.0 on a 4 nm TSMC process. The Steam Deck OLED GPU uses RDNA 2.0 on a 6 nm TSMC process.

Architecture Differences

The AMD Radeon AI PRO 9600D and AMD Steam Deck OLED GPU differ fundamentally at the architectural level. The AI PRO 9600D uses the Navi 48 chip with RDNA 4.0 architecture, part of the Radeon Pro Navi (Navi IV Series) generation. The Steam Deck OLED GPU uses the Sephiroth chip with RDNA 2.0 architecture, classified as a Console GPU from Valve.

The process nodes separate them by two generations. The AI PRO 9600D is fabricated on a 4 nm TSMC process with 53,900 million transistors on a 357 mm² die. The Steam Deck OLED GPU is fabricated on a 6 nm TSMC process with 2,400 million transistors on a 131 mm² die. Transistor density is 151.0M per mm² versus 18.3M per mm², a 8.3x difference in integration density.

The memory subsystems are entirely different. The AI PRO 9600D uses 32 GB of GDDR6 with a 256-bit bus, 2250 MHz memory clock (18 Gbps effective), and 576.0 GB/s bandwidth. The Steam Deck OLED GPU uses 16 GB of LPDDR5 with a 128-bit bus, 1375 MHz memory clock (11 Gbps effective), and 176.0 GB/s bandwidth. GDDR6 is a dedicated graphics memory type, while LPDDR5 is a low-power mobile memory.

The compute resources scale proportionally. The AI PRO 9600D has 3072 shading units, 192 TMUs, 96 ROPs, and 48 RT cores. The Steam Deck OLED GPU has 512 shading units, 32 TMUs, 16 ROPs, and 8 RT cores. Each ratio is exactly 6x. FP32 and FP16 rates differ: the AI PRO 9600D runs FP16 at 1:1 ratio (24.82 TFLOPS), while the Steam Deck OLED GPU runs FP16 at 2:1 ratio (3.277 TFLOPS from 1.638 TFLOPS FP32).

Clock behavior also differs. The AI PRO 9600D has a base clock of 1080 MHz, boost clock of 2020 MHz, and game clock of 1080 MHz. The Steam Deck OLED GPU has a base clock of 1000 MHz and boost clock of 1600 MHz, with no game clock listed. The AI PRO 9600D's boost clock is 420 MHz higher.

The AI PRO 9600D uses a PCIe 5.0 x16 bus interface. The Steam Deck OLED GPU lists no bus interface, reflecting its integrated console design. Display output differs as well: the AI PRO 9600D offers 1x DisplayPort 2.1a, while the Steam Deck OLED GPU offers 1x USB Type-C. Both support DirectX 12 Ultimate (12_2) and OpenGL 4.6. The AI PRO 9600D supports Vulkan 1.4, while the Steam Deck OLED GPU supports Vulkan 1.3.

The physical specifications reflect their intended environments. The AI PRO 9600D is a single-slot card measuring 241 mm by 111 mm by 19 mm, with one 16-pin power connector and a 450 W suggested PSU. The Steam Deck OLED GPU has no slot width, no power connector, and no PSU requirement. It is embedded in a device measuring 298 mm by 117 mm by 49 mm. The AI PRO 9600D has a 150 W TDP; the Steam Deck OLED GPU has a 15 W TDP.

DETAILED SPECIFICATIONS

SPECIFICATION
AI PRO 9600D
Steam Deck OLED GPU
Core Specs
Shading Units
3,072
512 -83.3%
Shaders
3,072
512 -83.3%
TMUs
192
32 -83.3%
ROPs
96
16 -83.3%
Compute Units
48
8 -83.3%
Clocks
Base Clock
1080 MHz
1000 MHz
Boost Clock
2020 MHz
1600 MHz
Game Clock
1080 MHz
Memory Clock
2250 MHz 18 Gbps effective
1375 MHz 11 Gbps effective
Memory
Memory Size
32 GB
16 GB
VRAM (MB)
32,768
16,384 -50.0%
Memory Type
GDDR6
LPDDR5
Memory Bus
256 bit
128 bit
Bandwidth
576.0 GB/s
176.0 GB/s
Cache
L1 Cache
128 KB per Array
L2 Cache
8 MB
1024 KB
L3 Cache
48 MB
8 MB
L0 Cache
32 KB per WGP
32 KB per WGP
Performance
Pixel Rate
193.9 GPixel/s
25.60 GPixel/s
Texture Rate
387.8 GTexel/s
51.20 GTexel/s
FP32 (TFLOPS)
24.82 TFLOPS
1.638 TFLOPS
FP64 (TFLOPS)
775.7 GFLOPS (1:32)
102.4 GFLOPS (1:16)
FP16 (TFLOPS)
24.82 TFLOPS (1:1)
3.277 TFLOPS (2:1)
AI/RT
RT Cores
48
8 -83.3%
Matrix Cores
96
Power
TDP
150 W
15 W
TDP (W)
150
15 -90.0%
Suggested PSU
450 W
Power Connectors
1x 16-pin
Architecture
Architecture
RDNA 4.0
RDNA 2.0
GPU Name
Navi 48
Sephiroth
Generation
Radeon Pro Navi (Navi IV Series)
Console GPU (Valve)
Process Size
4 nm
6 nm
Transistors
53,900 million
2,400 million
Die Size
357 mm²
131 mm²
Foundry
TSMC
TSMC
Density
151.0M / mm²
18.3M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.3
OpenCL
2.2
2.0
Shader Model
6.9
6.8
Physical
Slot Width
Single-slot
Length
241 mm 9.5 inches
298 mm 11.7 inches
Height
111 mm 4.4 inches
117 mm 4.6 inches
Outputs
1x DisplayPort 2.1a
1x USB Type-C
Bus Interface
PCIe 5.0 x16
Other
Production
Active
Active
Predecessor
Radeon Pro Vega
View Radeon AI PRO 9600D Details View Steam Deck OLED GPU Details