GeForce GTX 1070 and GeForce 9800 GTX+ - video card comparison
Passmark - benchmark score GeForce GTX 1070 vs GeForce 9800 GTX+
GeForce GTX 1070
13916
GeForce 9800 GTX+
521
GeForce GTX 1070
GeForce 9800 GTX+
Architecture
Pascal (2016−2021)
Tesla (2006−2010)
GPU code name
Pascal GP104
G92B
Market segment
Desktop
Desktop
Release date
May 2016
January 2009
Bus support
PCIe 3.0
-
Interface
PCIe 3.0 x16
PCIe 2.0 x16
Recommended system power (PSU)
500 Watt
-
Supplementary power connectors
8-pin
2x 6-pin
SLI options
+
+
Boost clock speed
1683 MHz
-
Core clock speed
1506 MHz
738 MHz
CUDA cores
1920
128
Floating-point performance
6,463 gflops
470.0 gflops
Manufacturing process technology
16 nm
55 nm
Number of transistors
7,200 million
754 million
Pipelines / CUDA cores
1920
128
Power consumption (TDP)
150 Watt
141 Watt
Maximum GPU temperature
94 °C
105 °C
Texture fill rate
202.0
47.2 billion/sec
Memory type
GDDR5
GDDR3
Maximum RAM amount
8 GB
512 MB
Memory bus width
256 Bit
256 Bit
Memory clock speed
8 GB/s
1100 MHz
Memory bandwidth
256 GB/s
70.4 GB/s
Shared memory
-
-
Display Connectors
DP 1.42, HDMI 2.0b, Dual Link-DVI
HDTVDual Link DVI
Multi monitor support
+
+
HDMI
+
-
Maximum VGA resolution
-
2048x1536
G-SYNC support
+
-
Audio input for HDMI
-
S/PDIF
DirectX
12 (12_1)
11.1 (10_0)
CUDA
+
+
OpenCL
1.2
1.1
OpenGL
4.5
2.1
Shader Model
6.4
4.0
Vulkan
+
N/A
Ansel
+
-
GPU Boost
3.0
-
VR Ready
+
-