P102-100 and GeForce RTX 4070 Ti - video card comparison
P102-100
GeForce RTX 4070 Ti
Architecture
Pascal (2016−2021)
Ada Lovelace
GPU code name
GP102
AD104-400
Market segment
Desktop
Desktop
Release date
February 2018
January 2023
Interface
PCIe 3.0 x4
PCIe 4.0 x16
Supplementary power connectors
2x 8-pin
1x 12-pin
Boost clock speed
1683 MHz
2610 MHz
Core clock speed
1582 MHz
2310 MHz
Floating-point performance
10,125 gflops
-
Manufacturing process technology
16 nm
5 nm
Number of transistors
11,800 million
35,800 million
Pipelines / CUDA cores
3200
7680
Power consumption (TDP)
250 Watt
285 Watt
Texture fill rate
336.6
626.4
Memory type
GDDR5X
GDDR6X
Maximum RAM amount
5 GB
12 GB
Memory bus width
320 Bit
192 Bit
Memory clock speed
11008 MHz
21000 MHz
Memory bandwidth
440.3 GB/s
504.2 GB/s
Shared memory
-
-
Display Connectors
No outputs
1x HDMI 2.1, 3x DisplayPort 1.4a
HDMI
-
+
DirectX
12 (12_1)
12 Ultimate (12_2)
CUDA
6.1
8.9
OpenCL
1.2
3.0
OpenGL
4.6
4.6
Shader Model
6.4
6.7
Vulkan
1.2.131
1.3