NVIDIA GB10 vs Intel GMA 900

NVIDIA GB10
Intel GMA 900
6144 Shaders
Shared Memory
2525MHz
4 Shaders
Shared Memory
333MHz
Peak AI Performance
992.87 TFLOPS
FP4 Tensor Sparse
Peak AI Performance
-
FP32
31.03 TFLOPS
-
-
FP16
62.05 TFLOPS
-
-
Form Factor
iGPU
-
Form Factor
iGPU
-
TDP
Unknown
TDP
Shared
-
-
-
-
-
-
-
-
-
-
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 N/A
0%
GB5 OpenCL N/A
0%
GB5 CUDA N/A
0%
GB5 CUDA N/A
0%
GB5 Metal N/A
0%
GB5 Metal N/A
0%
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%
Peak AI Performance
992.87 TFLOPS
FP4 Tensor Sparse
Peak AI Performance
-
FP4
496.44 TFLOPS Tensor
992.87 TFLOPS Tensor Sparse
-
-
-
FP8
-
248.22 TFLOPS Tensor (FP16 Accumulate)
496.44 TFLOPS Tensor (FP16 Accumulate) Sparse
124.11 TFLOPS Tensor (FP32 Accumulate)
248.22 TFLOPS Tensor (FP32 Accumulate) Sparse
-
-
-
-
-
-
FP16
62.05 TFLOPS
124.11 TFLOPS Tensor (FP16 Accumulate)
248.22 TFLOPS Tensor (FP16 Accumulate) Sparse
62.05 TFLOPS Tensor (FP32 Accumulate)
124.11 TFLOPS Tensor (FP32 Accumulate) Sparse
-
-
-
-
-
-
FP32
31.03 TFLOPS
-
-
-
-
-
-
FP64
15.51 TFLOPS
-
-
-
-
BF16
62.05 TFLOPS
62.05 TFLOPS Tensor
124.11 TFLOPS Tensor Sparse
-
-
-
-
TF32
31.03 TFLOPS Tensor
62.05 TFLOPS Tensor Sparse
-
-
-
-
-
-
-
-
-
INT8
-
248.22 TOPS Tensor
496.44 TOPS Tensor Sparse
-
-
-
-
INT32
31.03 TOPS
-
-
Ray Tracing
94.1 TOPS
-
-
Pixel Fillrate
80.8 GPixel/s
Pixel Fillrate
1.332 GPixel/s
-
-
-
-
Texture Fillrate
484.8 GTexel/s
Texture Fillrate
1.332 GTexel/s
Manufacturer
NVIDIA
Manufacturer
Intel
Chip Designer
NVIDIA
Chip Designer
Intel
Architecture
Blackwell
Architecture
Gen 3
Family
-
Family
Gen 3
Codename
Thor
GB10
-
-
Codename
Grantsdale
-
-
-
Market Segment
Mobile
Market Segment
Desktop
Release Date
1/6/2025
Release Date
6/1/2004
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Form
iGPU
Form
iGPU
Shading Units
6144 Shaders
-
Shading Units
4 Shaders
-
Texture Mapping Units
192 TMUs
Texture Mapping Units
4 TMUs
Render Output Units
32 ROPs
Render Output Units
4 ROPs
Tensor Cores
192 T-Cores
-
-
Ray-Tracing Cores
48 RT-Cores
-
-
Streaming Multiprocessors
48 SMs
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
2418MHz Base
2525MHz
-
-
133MHz Base
333MHz
-
-
-
-
L1
64KB/SM Tex
256KB/SM
-
-
L1
-
-
-
Unknown
L2
24MB Shared
L2
Unknown
-
-
-
-
-
-
Shared Memory
-
-
Shared Memory
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
TDP
Unknown
TDP
Shared
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
Max Resolution
7680x4320
Max Resolution
1920x1200
Max Resolution Refresh Rate
165Hz
Max Resolution Refresh Rate
60Hz
Variable Refresh Rate
-
-
-
Variable Refresh Rate
-
-
-
Display Stream Compression (DSC)
Supported
Display Stream Compression (DSC)
Not Supported
Multi Monitor Support
4
Multi Monitor Support
2
Content Protection
HDCP 2.3
-
-
Model
NVENC 9
No Encoders
-
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Model
NVDEC 6
No Decoders
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Direct X
12
Direct 3D
12_3
Direct X
9
Direct 3D
9_0
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
OpenGL
1.4
-
-
-
-
Shader Model
6.8
CUDA
13.0
-
-
PureVideo HD
VP13
VDPAU
Feature Set M
Shader Model
3.0
-
-
-
-
-
-
-
-
Not a Card
-
-
-
Not a Card
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Change Comparison