NVIDIA GA10B 16SM 625MHz vs Intel Iris Xe 96EU 1100MHz Full Specs
1,024 Shaders 625MHz | 768 Shaders 1.1GHz |
Shared Memory68.3GB/sLPDDR5 | Shared Memory83.2GB/s ·· |
·· 1.28 TFLOPS | ·· 1.69 TFLOPS |
Form Factor iGPU | Form Factor iGPU |
TDP Shared | TDP Shared |
GA10B 16SM 625MHz81.92 TOPSINT4 Tensor Sparse
Iris Xe 96EU 1100MHz3.38 TFLOPSFP16Clock Speed ·· | Clock Speed ·· |
Peak OPS 81.92 TOPSINT4 Tensor Sparse | Peak OPS 3.38 TFLOPSFP16 |
Tensor FP16-16 10.24 TFLOPSFP16-16 Tensor Sparse 20.48 TFLOPSTensor FP16-32 10.24 TFLOPSFP16-32 Tensor Sparse 20.48 TFLOPS | - |
Tensor BF16 10.24 TFLOPSBF16 Tensor Sparse 20.48 TFLOPS | - |
Tensor TF32 5.12 TFLOPS | Tensor TF32 - |
FP32 1.28 TFLOPS | FP32 1.69 TFLOPS |
- | FP64 422.4 GFLOPS |
Tensor INT4 40.96 TOPS | Tensor INT4 - |
Tensor INT8 20.48 TOPS | - |
Pixel Rate 10 GPixel/s | Pixel Rate 26.4 GPixel/s |
Texture Rate 5 GTexel/s | Texture Rate 52.8 GTexel/s |
Shaders 1,024 Shaders | Shaders 768 Shaders |
TMUs 8 TMUs | TMUs 48 TMUs |
ROPs 16 ROPs | ROPs 24 ROPs |
Tensor Cores 32 T-Cores | Tensor Cores - |
SMs 8 SMs | EUs 96 EUs |
Boost Clock 625MHz | Boost Clock 1.1GHz |
L2 Cache 2MB shared | L2 Cache 1MB shared |
L3 Cache - | L3 Cache 4MB shared |
Shared MemoryLPDDR5 | Shared Memory ·· |
Memory Bus 128-bit | Memory Bus 128-bit |
Memory Speed 4.3GT/s | Memory Speed 5.2GT/s |
Memory Bandwidth 68.3GB/s | Memory Bandwidth 83.2GB/s |
ECC No | ECC No |
Multi-Monitor - | Multi-Monitor 4 |
Encoder Model NVENC 7 | Encoder Model Quick Sync Video 8 |
Decoder Model NVDEC 5 | Decoder Model Quick Sync Video 8 |
Form Factor iGPU | Form Factor iGPU |
Foundry Samsung | Foundry Intel |
Fabrication Node 8N | Fabrication Node Intel 7 |
Die Size - | Die Size 161mm² |
Transistor Count 17 Billion | Transistor Count 11.9 Billion |
Transistor Density - | Transistor Density 74 MTr/mm² |
No images available
No images available
CPUs | CPUs |