Tesla D870 and GeForce RTX 4070 SUPER - video card comparison
Tesla D870
GeForce RTX 4070 SUPER
Architecture
Tesla (2006−2010)
Ada Lovelace
GPU code name
G80
AD104
Market segment
Workstation
Desktop
Release date
May 2007
January 2024
Interface
PCIe 1.0 x16
PCIe 4.0 x16
Supplementary power connectors
-
1x 16-pin
Boost clock speed
-
2475 MHz
Core clock speed
600 MHz
1980 MHz
Floating-point performance
2x 345.6 gflops
-
Manufacturing process technology
90 nm
5 nm
Number of transistors
681 million
35,800 million
Pipelines / CUDA cores
128
7168
Power consumption (TDP)
520 Watt
220 Watt
Texture fill rate
38.40
584.6
Memory type
GDDR3
GDDR6X
Maximum RAM amount
1536 MB
12 GB
Memory bus width
384 Bit
192 Bit
Memory clock speed
1600 MHz
21000 MHz
Memory bandwidth
76.8 GB/s
504.2 GB/s
Shared memory
-
-
Display Connectors
No outputs
1x HDMI 2.1, 3x DisplayPort 1.4a
HDMI
-
+
DirectX
11.1 (10_0)
12 Ultimate (12_2)
CUDA
+
8.9
OpenCL
1.1
3.0
OpenGL
3.3
4.6
Shader Model
4.0
6.7
Vulkan
N/A
1.3