GPU Comparison
GPU
Matrox Parhelia 128 MB
CORE STATE
Parhelia-512
VRAM
128 MB
CLOCK SPEED
—
TDP
—
BUS WIDTH
256 bit
ARCHITECTURE
Parhelia
PROCESS
150 nm
LAUNCH DATE
2002
VS
GEFORCE
GeForce RTX 2070 Max-Q
CORE STATE
TU106
VRAM
8 GB
CLOCK SPEED
1185 MHz
TDP
90 W
BUS WIDTH
256 bit
ARCHITECTURE
Turing
PROCESS
12 nm
LAUNCH DATE
2019
DETAILED SPECIFICATIONS
SPECIFICATION
Matrox Parhelia 128 MB
RTX 2070 Max-Q
Core Specs
Shading Units
—
2,304
Shaders
—
2,304
TMUs
16
144
+800.0%
ROPs
4
64
+1500.0%
SM Count
—
36
Clocks
Base Clock
—
885 MHz
Boost Clock
—
1185 MHz
GPU Clock
220 MHz
—
Memory Clock
275 MHz
550 Mbps effective
1500 MHz
12 Gbps effective
Memory
Memory Size
128 MB
8 GB
VRAM (MB)
128
8,192
+6300.0%
Memory Type
DDR
GDDR6
Memory Bus
256 bit
256 bit
Bandwidth
17.60 GB/s
384.0 GB/s
Cache
L1 Cache
—
64 KB (per SM)
L2 Cache
—
4 MB
Performance
Pixel Rate
880.0 MPixel/s
75.84 GPixel/s
Texture Rate
3.520 GTexel/s
170.6 GTexel/s
FP32 (TFLOPS)
—
5.460 TFLOPS
FP64 (TFLOPS)
—
170.6 GFLOPS (1:32)
FP16 (TFLOPS)
—
10.92 TFLOPS (2:1)
AI/RT
RT Cores
—
36
Tensor Cores
—
288
Power
TDP
—
90 W
TDP (W)
—
90
Suggested PSU
200 W
—
Power Connectors
None
None
Architecture
Architecture
Parhelia
Turing
GPU Name
Parhelia-512
TU106
Generation
Parhelia
GeForce 20 Mobile
Process Size
150 nm
12 nm
Transistors
80 million
10,800 million
Die Size
174 mm²
445 mm²
Foundry
UMC
TSMC
Density
459.8K / mm²
24.3M / mm²
API Support
DirectX
8.1
12 Ultimate (12_2)
OpenGL
1.3
4.6
Vulkan
—
1.4
OpenCL
—
3.0
CUDA
—
7.5
Shader Model
—
6.8
Physical
Slot Width
Single-slot
MXM Module
Length
175 mm
6.9 inches
—
Outputs
2x DVI
Portable Device Dependent
Bus Interface
AGP 4x
PCIe 3.0 x16
Other
Production
End-of-life
End-of-life
Predecessor
—
GeForce 10 Mobile
Successor
—
GeForce 30 Mobile