GeekRadar / The comparison desk
COMPARE.
GPU
AMD

AMD Radeon Pro W6900X

Graphics processor

Video memory
32 GB*
Memory bus
256 bit*
Rated memory bandwidth
512 GB/s*

Rendering performance

Median score · higher is better

Three-scene Blender rendering throughput. Versions, operating systems and rendering backends remain separate. This is not a gaming score.

Blender 4.0.0 / macOS / METAL
AMD Radeon Pro W6900X 1,987.74 samples/min 3 complete runs · Middle 50%: 1,968.57 - 1,993.07 · Latest: 2024-02-20

Community submissions; cooling, power limits and system configuration can affect results.

Graphics & compute benchmarks

Archived cohort medians

GFXBench, GLBenchmark and CompuBench results retain the tested system, API, operating system and resolution. Offscreen tests fix the render resolution; onscreen results also reflect the display workload. These historical submissions do not establish performance in current games.

1080p ALU 2 OffscreenOS X / Metal / 1920 x 1080 / offscreen · fps

Higher is better.

AMD Radeon Pro W6900X

5,195.46* fps

AMD Radeon PRO W6900X

Host configuration not reported.

1080p Driver Overhead 2 OffscreenOS X / Metal / 1920 x 1080 / offscreen · fps

Higher is better.

AMD Radeon Pro W6900X

408.48* fps

AMD Radeon PRO W6900X

Host configuration not reported.

1080p Manhattan 3.1 OffscreenOS X / Metal / 1920 x 1080 / offscreen · fps

Higher is better.

AMD Radeon Pro W6900X

1,578.51* fps

AMD Radeon PRO W6900X

Host configuration not reported.

1080p Manhattan OffscreenOS X / Metal / 1920 x 1080 / offscreen · fps

Higher is better.

AMD Radeon Pro W6900X

1,893.44* fps

AMD Radeon PRO W6900X

Host configuration not reported.

1080p T-Rex OffscreenOS X / Metal / 1920 x 1080 / offscreen · fps

Higher is better.

AMD Radeon Pro W6900X

2,801.23* fps

AMD Radeon PRO W6900X

Host configuration not reported.

1080p Texturing OffscreenOS X / Metal / 1920 x 1080 / offscreen · MTexels/s

Higher is better.

AMD Radeon Pro W6900X

446,831* MTexels/s

AMD Radeon PRO W6900X

Host configuration not reported.

1440p Manhattan 3.1.1 OffscreenOS X / Metal / 2560 x 1440 / offscreen · fps

Higher is better.

AMD Radeon Pro W6900X

1,011.50* fps

AMD Radeon PRO W6900X

Host configuration not reported.

4K Aztec Ruins High Tier OffscreenOS X / Metal / 3840 x 2160 / offscreen · fps

Higher is better.

AMD Radeon Pro W6900X

237.01* fps

AMD Radeon PRO W6900X

Host configuration not reported.

Aztec Ruins High Tier OffscreenOS X / Metal / 2560 x 1440 / offscreen · fps

Higher is better.

AMD Radeon Pro W6900X

470.99* fps

AMD Radeon PRO W6900X

Host configuration not reported.

Aztec Ruins Normal Tier OffscreenOS X / Metal / 1920 x 1080 / offscreen · fps

Higher is better.

AMD Radeon Pro W6900X

878.02* fps

AMD Radeon PRO W6900X

Host configuration not reported.

ALU 2OS X / Metal / 3840 x 2160 / onscreen · fps

Higher is better.

AMD Radeon Pro W6900X

61.66* fps

AMD Radeon PRO W6900X

Host configuration not reported.

Aztec Ruins High TierOS X / Metal / 5120 x 2880 / onscreen · fps

Higher is better.

AMD Radeon Pro W6900X

60.13* fps

AMD Radeon PRO W6900X

Host configuration not reported.

Aztec Ruins Normal TierOS X / Metal / 5120 x 2880 / onscreen · fps

Higher is better.

AMD Radeon Pro W6900X

60.17* fps

AMD Radeon PRO W6900X

Host configuration not reported.

Driver Overhead 2OS X / Metal / 5120 x 2880 / onscreen · fps

Higher is better.

AMD Radeon Pro W6900X

60.46* fps

AMD Radeon PRO W6900X

Host configuration not reported.

Manhattan 3.1OS X / Metal / 5120 x 2880 / onscreen · fps

Higher is better.

AMD Radeon Pro W6900X

60.01* fps

AMD Radeon PRO W6900X

Host configuration not reported.

ManhattanOS X / Metal / 5120 x 2880 / onscreen · fps

Higher is better.

AMD Radeon Pro W6900X

60.01* fps

AMD Radeon PRO W6900X

Host configuration not reported.

Median results, not the best run. Sample counts and individual test dates are unreported. Different systems using the same GPU remain separate configurations.

Benchmark results

  • G3D Mark · version unreported · g3d_mark17,413* pointsPlatform unreported Compare results
  • Geekbench GPU (reported compilation) · version unreported · Metal166,946*Platform unreported Compare results
  • Geekbench GPU (reported compilation) · version unreported · Metal166,946*Platform unreported Compare results
  • Geekbench GPU (reported compilation) · version unreported · Metal211,415*Platform unreported Compare results
  • Geekbench GPU (reported compilation) · version unreported · OpenCL151,211*Platform unreported Compare results
  • Geekbench GPU (reported compilation) · version unreported · OpenCL141,178*Platform unreported Compare results
  • Geekbench GPU (reported compilation) · version unreported · OpenCL141,178*Platform unreported Compare results
  • Geekbench GPU (reported compilation) · version unreported · Vulkan143,082*Platform unreported Compare results
  • Geekbench GPU (reported compilation) · version unreported · Vulkan105,424*Platform unreported Compare results
  • Geekbench GPU (reported compilation) · version unreported · Vulkan105,424*Platform unreported Compare results

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

Blender Cycles / classroomMETAL / Darwin · 4.0.0 · samples/min

AMD Radeon Pro W6900X

384.84 samples/min

3 submissions · median result

Latest submission:
Rendering backend
METAL
Operating system
Darwin
Statistic
Median
Evaluation type
community rendering submissions
Blender Cycles / Complete suiteMETAL / Darwin · 4.0.0 · samples/min

AMD Radeon Pro W6900X

1,987.74 samples/min

3 submissions · median result

Latest submission:
Rendering backend
METAL
Operating system
Darwin
Statistic
Median
Evaluation type
community rendering submissions
25th percentile
1968.5744815587
75th percentile
1993.0743193213
Blender Cycles / junkshopMETAL / Darwin · 4.0.0 · samples/min

AMD Radeon Pro W6900X

541.09 samples/min

3 submissions · median result

Latest submission:
Rendering backend
METAL
Operating system
Darwin
Statistic
Median
Evaluation type
community rendering submissions
Blender Cycles / monsterMETAL / Darwin · 4.0.0 · samples/min

AMD Radeon Pro W6900X

987.12 samples/min

3 submissions · median result

Latest submission:
Rendering backend
METAL
Operating system
Darwin
Statistic
Median
Evaluation type
community rendering submissions
Blender Cycles / classroomMETAL / Darwin · 3.1.0 · samples/min

AMD Radeon Pro W6900X

501.49 samples/min

3 submissions · median result

Latest submission:
Rendering backend
METAL
Operating system
Darwin
Statistic
Median
Evaluation type
community rendering submissions
Blender Cycles / Complete suiteMETAL / Darwin · 3.1.0 · samples/min

AMD Radeon Pro W6900X

2,133.37 samples/min

3 submissions · median result

Latest submission:
Rendering backend
METAL
Operating system
Darwin
Statistic
Median
Evaluation type
community rendering submissions
25th percentile
2112.5152867283
75th percentile
2181.215566634
Blender Cycles / junkshopMETAL / Darwin · 3.1.0 · samples/min

AMD Radeon Pro W6900X

546.74 samples/min

3 submissions · median result

Latest submission:
Rendering backend
METAL
Operating system
Darwin
Statistic
Median
Evaluation type
community rendering submissions
Blender Cycles / monsterMETAL / Darwin · 3.1.0 · samples/min

AMD Radeon Pro W6900X

1,085.14 samples/min

3 submissions · median result

Latest submission:
Rendering backend
METAL
Operating system
Darwin
Statistic
Median
Evaluation type
community rendering submissions

Processing & platform

Architecture
RDNA 2.0
Manufacturing process
7 nm
Process node
7 nm
GPU cores
5,120*
GPU base clock
1,825 MHz*
Graphics interface
Apple MPX*
Reported chip die area
520 mm²*
Transistors
26,800 million*
GPU boost clock
2,171 MHz*
GPU L2 cache
4 MB*
Graphics processor
Navi*
Texture mapping units
320*
Raster output units
128*
Ray tracing cores
80*
Chip foundry
TSMC*
Declared GPU TDP
300 W*
Suggested system PSU
700 W*
DirectX / feature level
12 Ultimate (12_2)*
OpenGL
4.6*
OpenCL
2.1*
Vulkan
1.4*
Shader model
6.8*

Memory & storage

Memory options
32*
Video memory
32 GB*
Video memory type
Reported value 1: GDDR5*
Reported value 2: GDDR6*
Reported value 3: GDDR6*
Reported value 4: GDDR6*
Reported value 5: GDDR6*
Reported value 6: GDDR6*
Memory bus
256 bit*
Rated memory bandwidth
512 GB/s*

Power specifications

Thermal design power
300 W*

System configuration

Detailed specifications

Specifications 1 · 8 fields
Product Name
Radeon PRO W6900X*
GPU Chip
Navi 21*
Released
Aug 3rd, 2021*
Bus
PCIe 4.0 x16*
Memory
32 GB, GDDR6, 256 bit*
GPU clock
1,825* MHz
Memory clock
2,000* MHz
Shaders TMUs ROPs
5120 / 320 / 128*
Specifications 2 · 26 fields
proc transistors
26,800 million*
proc die size
520* mm²
cores
5,120*
rt cores
80*
base clock
1,825*
boost clock
2,171*
memory clock
2,000*
ram
32*
name
AMD Radeon Pro W6900X*
ram type
GDDR6*
bus width
256*
bus interface
Apple MPX*
ram bandwidth
512* GB/s
fp16 performance
44.46*
fp32 performance
22.23*
fp64 performance
1.3894*
tdp
300* W
suggested psu
700* W
is retail board
No*
gpi value
13.64*
vendor
amd*
manufacturer
AMD*
arch name
RDNA 2.0*
card release date
2021-08-03*
proc foundry
TSMC*
proc process size
7* nm
Specifications 3 · 35 fields
DirectX / feature level
12 Ultimate (12_2)*
OpenCL
2.1*
OpenGL
4.6*
Shader model
6.8*
Vulkan
1.4*
Architecture
RDNA 2.0*
Graphics interface
Apple MPX*
Memory bus
256* bit
Reported shader processors
5,120*
Reported theoretical FP16 peak
44.46* TFLOPS
Scope and conditions
Arithmetic density
Not stated*
Reported theoretical FP32 peak
22.23* TFLOPS
Scope and conditions
Arithmetic density
Not stated*
Reported theoretical FP64 peak
1.3894* TFLOPS
Scope and conditions
Arithmetic density
Not stated*
Declared L1 cache configuration
128 KB per Array*
Declared GPU L2 cache
4 MB*
Reported display outputs
1x HDMI 2.1, 4x Thunderbolt*
Theoretical pixel fill rate
277.9 GPixel/s*
Reported chip die area
520 mm²*
Chip foundry
TSMC*
Process node declaration
7 nm*
Transistors
26,800 million*
Graphics processor
Navi*
Reported video memory
32* GB
Rated memory bandwidth
512.0 GB/s*
Video memory type
GDDR6*
Raster output units
128*
Ray tracing cores
80*
Reported shading units
5,120*
Suggested system PSU
700 W*
2:4 tensor sparsity
No*
Declared GPU TDP
300 W*
Theoretical texture fill rate
694.7 GTexel/s*
Texture mapping units
320*
GPU cores
5,120*
Video memory
32* GB
Video memory type
GDDR6*
Specifications 4 · 3 fields
G3D Mark
17,413* points
Scope and conditions
benchmark_context
suite
G3D Mark*
native_context
GPU model
Radeon Pro W6900X*
Snapshot date
2026-10-05*
Model
Radeon Pro W6900X*
Hardware class
Desktop*
Specifications 5 · 8 fields
Product Name
Radeon Pro W6900X*
GPU Chip
Navi 21*
Released
Aug 3rd, 2021*
Bus
Apple MPX*
Memory
32 GB, GDDR6, 256 bit*
GPU clock
1,825* MHz
Memory clock
2,000* MHz
Shaders / TMUs / ROPs
5120 / 320 / 128*
Specifications 6 · 57 fields
Brand
AMD*
Top BUS WIDTH
256* bit
Graphics Processor GPU Name
Navi 21*
Graphics Processor Architecture
RDNA 2.0*
Graphics Processor Process Size
7* nm
Graphics Processor Transistors
26,800 million*
Graphics Processor Die Size
520* mm²
Graphics Card Release Date
Aug 3rd, 2021*
Graphics Card Generation
Radeon Pro Mac (Navi II Series)*
Graphics Card Production
Active*
Name
Radeon Pro W6900X*
Graphics Card Bus Interface
Apple MPX*
Clock Speeds Memory Clock
2000 MHz 16 Gbps effective*
Memory Memory Size
32* GB
Memory Memory Type
GDDR6*
Memory Memory Bus
256* bit
Memory Bandwidth
512* GB/s
Render Config TMUs
320*
Top GRAPHICS PROCESSOR
Navi 21*
Render Config ROPs
128*
Theoretical Performance Pixel Rate
277.9 GPixel/s
Theoretical Performance Texture Rate
694.7 GTexel/s
Board Design TDP
300* W
Board Design Suggested PSU
700* W
Board Design Outputs
1x HDMI 2.1 4x Thunderbolt*
Board Design Board Number
MJ103AM/A*
Graphics Features DirectX
12 Ultimate (12_2)*
Graphics Features OpenGL
4.6*
Graphics Features OpenCL
2.1*
Graphics Features Vulkan
1.3*
source file
amd2021.json*
Graphics Processor Foundry
TSMC*
Recommended Resolutions
Good:1920x1080 | Good:2560x1440 | Good:3840x2160*
Graphics Card Launch Price
5,999 USD*
Board Design Length
267 mm 10.5 inches*
Graphics Processor GPU Variant
Navi 21 Pro-XTA*
Graphics Processor Density
51.5M / mm²*
Board Design Height
120 mm 4.7 inches*
Top TMUS
320*
Graphics Processor Chip Package
BGA-2425*
Top ROPS
128*
Clock Speeds Base Clock
1,825* MHz
Clock Speeds Boost Clock
2,171* MHz
Graphics Features Shader Model
6.8*
Top CORES
5,120*
Render Config Shading Units
5,120*
Render Config Compute Units
80*
Render Config L2 Cache
4* MB
Top MEMORY SIZE
32* GB
Theoretical Performance FP32 (float)
22.23* TFLOPS
Theoretical Performance FP64 (double)
1,389.4 GFLOPS (1:16)
Render Config L1 Cache
128 KB per Array*
Theoretical Performance FP16 (half)
44.46 TFLOPS (2:1)
Top MEMORY TYPE
GDDR6*
Render Config RT Cores
80*
Render Config L0 Cache
32 KB per WGP*
Render Config L3 Cache
128* MB
Specifications 7 · 9 fields
Device
AMD Radeon Pro W6900X*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Manufacturer
AMD*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Device
Radeon Pro W6900X*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
OpenCL
151,211*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Metal
211,415*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Metal
166,946*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Vulkan
143,082*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
OpenCL
141,178*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Vulkan
105,424*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Specifications 8 · 5 fields
Manufacturer
AMD*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Device
Radeon Pro W6900X*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Metal
166,946*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
OpenCL
141,178*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Vulkan
105,424*
Scope and conditions
benchmark_context
suite
Geekbench GPU (reported compilation)*
backend_columns
  • CUDA*
  • Metal*
  • OpenCL*
  • Vulkan*
test_configuration_verified
No*
comparable
No*
Specifications 9 · 27 fields
manufacturer
AMD*
manufacturer
AMD*
rop
128*
rop
32*
igp
No*
igp
No*
bus
PCIe 3.0 x16*
bus
PCIe 4.0 x16*
memType
GDDR5*
memType
GDDR6*
gpuChip
Navi 21*
gpuChip
Amethyst*
productName
Radeon Pro W6900X*
productName
Radeon Pro W6900X*
releaseYear
2,021*
releaseYear
2,021*
memSize
32*
memSize
32*
memBusWidth
256*
gpuClock
1,825*
gpuClock
723*
memClock
1,250*
memClock
2,000*
unifiedShader
2,048*
unifiedShader
5,120*
tmu
128*
tmu
320*

* Provisional data