GPUs

NVIDIA GA10B 32SM vs Apple 16-Core M4 Pro Full Specs

2,048 Shaders
1.3GHz
2,048 Shaders
1.58GHz
Shared Memory204.8GB/sLPDDR5
Shared Memory273.1GB/sLPDDR5X
··
5.33 TFLOPS
··
6.46 TFLOPS
Form Factor
iGPU
Form Factor
iGPU
TDP
Shared
TDP
Shared

GA10B 32SMGA10B 32SM340.8 TOPSINT4 Tensor Sparse
x52.72
16-Core M4 Pro16-Core M4 Pro6.46 TFLOPSFP16
x1

Clock Speed
··
Clock Speed
··
Peak OPS
340.8 TOPSINT4 Tensor Sparse
Peak OPS
6.46 TFLOPSFP16
Tensor FP16-16
42.6 TFLOPS
FP16-16 Tensor Sparse
85.2 TFLOPS
Tensor FP16-32
42.6 TFLOPS
FP16-32 Tensor Sparse
85.2 TFLOPS
-
Tensor BF16
42.6 TFLOPS
BF16 Tensor Sparse
85.2 TFLOPS
-
Tensor TF32
21.3 TFLOPS
Tensor TF32
-
FP32
5.33 TFLOPS
FP32
6.46 TFLOPS
Tensor INT4
170.4 TOPS
Tensor INT4
-
Tensor INT8
85.2 TOPS
-
Pixel Rate
41.6 GPixel/s
Pixel Rate
101 GPixel/s
Texture Rate
20.8 GTexel/s
Texture Rate
202 GTexel/s

Shaders
2,048 Shaders
Shaders
2,048 Shaders
TMUs
16 TMUs
TMUs
128 TMUs
ROPs
32 ROPs
ROPs
64 ROPs
Tensor Cores
64 T-Cores
Tensor Cores
-
RT Cores
-
RT Cores
16 RT-Cores
SMs
16 SMs
EUs
256 EUs

Boost Clock
1.3GHz
Boost Clock
1.58GHz

L2 Cache
4.1MB shared
L2 Cache
4.1MB shared

Shared MemoryLPDDR5
Shared MemoryLPDDR5X
Memory Bus
256-bit
Memory Bus
256-bit
Memory Speed
6.4GT/s
Memory Speed
8.5GT/s
Memory Bandwidth
204.8GB/s
Memory Bandwidth
273.1GB/s
ECC
No
ECC
No

Multi-Monitor
-
Multi-Monitor
3

Encoder Model
NVENC 7
Encoder Model
Apple Media Engine 3

Decoder Model
NVDEC 5
Decoder Model
Apple Media Engine 3

Form Factor
iGPU
Form Factor
iGPU

Manufacturer
Manufacturer
Chip Designer
Chip Designer
Architecture
Architecture
Family
Family
Branding
GA10B 32SM branding
Branding
M4 Pro branding
Codename
Hercules
Codename
G16S
Market Segment
Embedded
Market Segment
Laptop
Release Date
Jan 1, 2021
Release Date
Oct 29, 2024

Foundry
Samsung
Foundry
TSMC
Fabrication Node
8N
Fabrication Node
N3E
Die Size
-
Die Size
325mm²
Transistor Count
17 Billion
Transistor Count
55 Billion
Transistor Density
-
Transistor Density
169 MTr/mm²

No images available
No images available