60-Core M2 Ultra vs Radeon Pro 580X MPX

60-Core M2 Ultra
Radeon Pro 580X MPX
7680 Shaders
Shared Memory
1398MHz
2304 Shaders
8GB GDDR5
1220MHz
Peak AI Performance
21.47 TFLOPS
FP16
Peak AI Performance
11.24 TFLOPS
FP16
FP32
21.47 TFLOPS
FP32
5.62 TFLOPS
FP16
21.47 TFLOPS
FP16
11.24 TFLOPS
Form Factor
iGPU
-
Form Factor
MPX
2.0-Slots
TDP
Shared
TDP
185W
-
-
-
-
-
-
-
-
-
-
GB6 OpenCL N/A
0%
GB6 OpenCL N/A
0%
GB6 Metal N/A
0%
GB6 Metal N/A
0%
GB6 Vulkan N/A
0%
GB6 Vulkan N/A
0%
GB5 OpenCL 105,745
35%
GB5 OpenCL 41,950
14%
GB5 CUDA N/A
0%
GB5 CUDA N/A
0%
GB5 Metal 112,540
62%
GB5 Metal 38,490
21%
GB5 Vulkan N/A
0%
GB5 Vulkan N/A
0%
OCT 2020.1 N/A
0%
OCT 2020.1 N/A
0%
OCT Metal N/A
0%
OCT Metal N/A
0%
Manufacturer
Apple
Manufacturer
Apple
Chip Designer
Apple
Chip Designer
AMD
Architecture
Apple Gen 5
Architecture
GCN 4
Family
M Series
Family
Radeon Pro
Codename
G14D
M2 Ultra
-
-
Codename
Ellesmere
Polaris 20
Variant
Polaris 20 XTA
Market Segment
Desktop
Market Segment
Workstation
Release Date
6/5/2023
Release Date
12/11/2019
-
-
-
Foundry
GlobalFoundries
-
-
-
-
Fabrication Node
14LPP
-
-
-
-
Die Size
232 mm²
-
-
-
Transistor Count
5.7 Billion
-
-
-
-
Transistor Density
24.57M/mm²
-
Form
iGPU
Form
MPX
Shading Units
7680 Shaders
-
Shading Units
2304 Shaders
-
Texture Mapping Units
480 TMUs
Texture Mapping Units
144 TMUs
Render Output Units
240 ROPs
Render Output Units
32 ROPs
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Compute Units
36 CUs
Execution Units
960 EUs
-
-
Graphics Processing Clusters
60 GPCs
Graphics Processing Clusters
4 GPCs
-
-
-
1398MHz
-
-
1100MHz Base
1220MHz
Peak AI Performance
21.47 TFLOPS
FP16
Peak AI Performance
11.24 TFLOPS
FP16
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
FP16
21.47 TFLOPS
-
-
-
-
FP16
11.24 TFLOPS
-
-
-
-
FP32
21.47 TFLOPS
-
-
FP32
5.62 TFLOPS
-
-
-
-
-
FP64
350 GFLOPS
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Pixel Fillrate
335.52 GPixel/s
Pixel Fillrate
39.04 GPixel/s
-
-
-
-
Texture Fillrate
671.04 GTexel/s
Texture Fillrate
175.68 GTexel/s
-
-
-
-
L1
-
-
-
Unknown
L1
-
-
16KB/CU
-
L2
Unknown
L2
2MB Shared
-
-
-
-
-
-
Shared Memory
-
-
8GB
GDDR5
-
-
Bus Width
256Bit
-
-
-
-
-
-
Clock
1710MHz
Transfer Rate
6.8GT/s
Bandwidth
218.9GB/s
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
TDP
Shared
TDP
185W
-
-
Temp
100°C Max
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
2x HDMI 2.0
-
-
-
Max Resolution
7680x4320
Max Resolution
6016x3384
Max Resolution Refresh Rate
60Hz
Max Resolution Refresh Rate
60Hz
Variable Refresh Rate
-
-
-
Variable Refresh Rate
-
FreeSync
-
Display Stream Compression (DSC)
Supported
Display Stream Compression (DSC)
Not Supported
Multi Monitor Support
8
Multi Monitor Support
6
-
-
Content Protection
HDCP 2.2
Model
4x Apple Media Engine 2
Model
VCE 3.4
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
ProRes
ProRes Raw
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
Model
2x Apple Media Engine 2
Model
UVD 6.3
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
ProRes
ProRes Raw
Codec
MPEG-1
MPEG-2
MPEG-4
JPEG
VC-1
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
-
-
-
-
Direct X
12
Direct 3D
12_0
-
-
-
-
-
-
OpenGL
4.6
OpenCL
2.1
Vulkan
1.2
-
-
-
-
-
-
-
-
-
-
Shader Model
6.4
-
-
GFX
8
-
-
-
-
Not a Card
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Slots Required
2.0
PCIe Version
3.0
PCIe Lanes
16
-
-
-
-
Multi GPU Support
Supported
Type
Bridgeless
-
-
-
-
-
-
-
-
-
-
-
-
Change Comparison