GPU Comparison
RADEON
AMD AeroBox GPU
CORE STATE
Kryptos
VRAM
8 GB
CLOCK SPEED
985 MHz
TDP
100 W
BUS WIDTH
256 bit
ARCHITECTURE
GCN 1.0
PROCESS
16 nm
LAUNCH DATE
2020
VS
GEFORCE
GeForce GTX 1660 SUPER
CORE STATE
TU116
VRAM
6 GB
CLOCK SPEED
1785 MHz
TDP
125 W
BUS WIDTH
192 bit
ARCHITECTURE
Turing
PROCESS
12 nm
LAUNCH DATE
2019
PERFORMANCE BENCHMARKS
3dmark_3dmark_steel_nomad_dx12
geekbench_opencl
geekbench_vulkan
passmark_directx_10
passmark_directx_11
passmark_directx_12
passmark_directx_9
passmark_g2d
passmark_g3d
passmark_gpu_compute
DETAILED SPECIFICATIONS
SPECIFICATION
AeroBox GPU
GTX 1660 SUPER
Core Specs
Shading Units
896
1,408
+57.1%
Shaders
896
1,408
+57.1%
TMUs
56
88
+57.1%
ROPs
16
48
+200.0%
Compute Units
14
—
SM Count
—
22
Clocks
Base Clock
935 MHz
1530 MHz
Boost Clock
985 MHz
1785 MHz
Memory Clock
1066 MHz
2.1 Gbps effective
1750 MHz
14 Gbps effective
Memory
Memory Size
8 GB
6 GB
VRAM (MB)
8,192
6,144
-25.0%
Memory Type
DDR3
GDDR6
Memory Bus
256 bit
192 bit
Bandwidth
68.22 GB/s
336.0 GB/s
Cache
L1 Cache
—
64 KB (per SM)
L2 Cache
—
1536 KB
Performance
Pixel Rate
15.76 GPixel/s
85.68 GPixel/s
Texture Rate
55.16 GTexel/s
157.1 GTexel/s
FP32 (TFLOPS)
1.765 TFLOPS
5.027 TFLOPS
FP64 (TFLOPS)
110.3 GFLOPS (1:16)
157.1 GFLOPS (1:32)
FP16 (TFLOPS)
3.530 TFLOPS (2:1)
10.05 TFLOPS (2:1)
Power
TDP
100 W
125 W
TDP (W)
100
125
+25.0%
Suggested PSU
—
300 W
Power Connectors
—
1x 8-pin
Architecture
Architecture
GCN 1.0
Turing
GPU Name
Kryptos
TU116
Generation
Console GPU
(Chuwi)
GeForce 16
Process Size
16 nm
12 nm
Transistors
—
6,600 million
Die Size
—
284 mm²
Foundry
TSMC
TSMC
Density
—
23.2M / mm²
API Support
DirectX
12 (11_1)
12 (12_1)
OpenGL
—
4.6
Vulkan
1.2
1.4
OpenCL
1.2
3.0
CUDA
—
7.5
Shader Model
5.1
6.8
Physical
Slot Width
—
Dual-slot
Length
289 mm
11.4 inches
229 mm
9 inches
Height
238 mm
9.4 inches
111 mm
4.4 inches
Outputs
1x DVI
1x DVI1x HDMI 2.01x DisplayPort 1.4a
Bus Interface
—
PCIe 3.0 x16
Other
Launch Price
—
229 USD
Production
End-of-life
End-of-life
Predecessor
—
GeForce 10
Successor
—
GeForce 20