M3 Ultra (32C80G) vs AMD A10-9700E

M3 Ultra (32C80G)
AMD A10-9700E
32 Cores
32 Threads
4.05GHz Clock
2 Cores
4 Threads
3GHz Clock
Socket
Unknown
Socket
AM4
iGPU
80-Core M3 Ultra
28.63 TFLOPS FP32
iGPU
Radeon R7 6CU 847MHz GCN 3
650 GFLOPS FP32

GB6 Single-Core

  • 5.11x faster vs A10-9700E
  • 80% slower vs M3 Ultra (32C80G)
M3 Ultra (32C80G) - GB6S 3,220
x5.11
A10-9700E - GB6S 630
x1

GB6 Multi-Core

  • 17.9x faster vs A10-9700E
  • 94% slower vs M3 Ultra (32C80G)
M3 Ultra (32C80G) - GB6M 27,750
x17.9
A10-9700E - GB6M 1,550
x1
  • 21.33x faster vs A10-9700E
  • 95% slower vs M3 Ultra (32C80G)
M3 Ultra (32C80G) - 819.2GB/s
x21.33
A10-9700E - 38.4GB/s
x1
  • 414% higher vs A10-9700E
  • 81% lower vs M3 Ultra (32C80G)
M3 Ultra (32C80G) - 180W
x5.14
A10-9700E - 35W
x1
  • 44.05x faster vs A10-9700E
  • 98% slower vs M3 Ultra (32C80G)
M3 Ultra (32C80G) - 28.63 TFLOPS FP32
x44.05
A10-9700E - 650 GFLOPS FP32
x1
GB6S 3,220
79%
GB6S 630
16%
GB6M 27,750
104%
Manufacturer
Apple
Manufacturer
AMD
Architecture
Apple Silicon 13
Architecture
Excavator+
Family
M Series
Family
A Series
Instruction Set (ISA)
AArch64
Instruction Set (ISA)
x86-64
Codename
Palma
M3 Ultra
P-Core
AS13P
E-Core
AS13E
Codename
Bristol Ridge
-
-
-
-
-
Market Segment
Desktop
Market Segment
Desktop
Release Date
3/5/2025
Release Date
9/5/2016
Foundry
TSMC
Foundry
GlobalFoundries
Node
N3B
Node
28nm
Die Size
1005 mm²
Die Size
250 mm²
Transistor Count
184 Billion
Transistor Count
3.1 Billion
Unknown
AM4
32 Cores
24 P-Cores
8 E-Cores
2 Cores
-
-
32 Threads
24 P-Threads
8 E-Threads
4 Threads
-
-
P-Core
4.05GHz
-
-
3GHz
3.5GHz Boost
E-Core
2.75GHz
-
-
Overclocking
Locked
Overclocking
Locked
Integrated Chipset
-
Type
PCIe
-
-
Transfer Rate
8GT/s
-
-
Lanes
4
-
-
Bandwidth
4GB/s
-
-
Bi-directional
8GB/s
L1 Instruction
192KB/P-Core
6-way set associative
128KB/E-Core
4-way set associative
L1 Instruction
96KB/Core
3-way set associative
-
-
L1 Data
128KB/P-Core
4-way set associative
64KB/E-Core
2-way set associative
L1 Data
32KB/Core
8-way set associative
-
-
L2
-
-
64MB Shared
16-way set associative
4MB Shared (E-Core)
16-way set associative
L2
1024KB/Core
16-way set associative
-
-
-
-
96MB SLC Cache
-
Channels
8
Channels
2
Max Memory
128GB
Max Memory
64GB
ECC
Not Supported
ECC
Not Supported
Bus Width/Channel
128Bit
Bus Width
1024Bit
Bus Width/Channel
64Bit
Bus Width
128Bit
Clock
3200MHz
Transfer Rate
6400MT/s
Clock
1200MHz
Transfer Rate
2400MT/s
Bandwidth/Channel
102.4GB/s
Bandwidth
819.2GB/s
Bandwidth/Channel
19.2GB/s
Bandwidth
38.4GB/s
TDP
180W
TDP
35W
Temp
100°C Max
-
-
No PCIe
PCIe 3.0 x8 Lanes
80-Core M3 Ultra
10240 Shaders
1398MHz
28.63 TFLOPS FP32
Radeon R7 6CU 847MHz GCN 3
384 Shaders
847MHz
650 GFLOPS FP32
No NPU
32 Cores
-
36 TOPS FP16
-
No Cellular
No Cellular
Change Comparison