GeekRadar / The comparison desk
COMPARE.
GPUInno3D GTX 680 HerculeZ 3000
Graphics processor
- Video memory
- 2 GB*
- Memory bus
- 256 bit*
- Rated memory bandwidth
- 198.40 GB/s*
No matched benchmark results are available for this entry yet.
Processing & platform
- Architecture
- Kepler*
- Process node
- 28 nm*
- GPU cores
- 1,536*
- Graphics interface
- PCIe 3.0 x16*
- Reported chip die area
- 294 mm²*
- Transistors
- 3,540 million*
- Graphics processor
- GK104*
- Texture mapping units
- 128*
- Raster output units
- 32*
- Chip foundry
- TSMC*
- Card or module format
- Triple-slot*
- Declared GPU TDP
- 195 W*
- Suggested system PSU
- 450 W*
- External power connectors
- 2x 6-pin*
- DirectX / feature level
- 12 (11_0)*
- OpenGL
- 4.6*
- OpenCL
- 3.0*
- Vulkan
- 1.2.175*
- CUDA compute capability
- 3.0*
- Shader model
- 6.5 (5.1)*
- GPU L2 cache
- 512 KB*
Memory & storage
- Memory options
- 2*
- Video memory
- 2 GB*
- Video memory type
- GDDR5*
- Memory bus
- 256 bit*
- Rated memory bandwidth
- 198.40 GB/s*
Power specifications
- Thermal design power
- 195 W*
System configuration
Detailed specifications
Specifications 1 · 26 fields
- proc transistors
- 3,540 million*
- proc die size
- 294* mm²
- cores
- 1,536*
- base clock
- 1,098*
- boost clock
- 1,155*
- memory clock
- 1,550*
- ram
- 2*
- name
- Inno3D GTX 680 HerculeZ 3000*
- ram type
- GDDR5*
- bus width
- 256*
- bus interface
- PCIe 3.0 x16*
- ram bandwidth
- 198.4* GB/s
- fp32 performance
- 3.548*
- fp64 performance
- 0.1478*
- tdp
- 195* W
- suggested psu
- 450* W
- slot width
- Triple-slot*
- power connectors
- 2x 6-pin*
- api cuda
- 3*
- is retail board
- Yes*
- gpi value
- 2.71*
- manufacturer
- INNO3D*
- arch name
- Kepler*
- card release date
- 2012-03-22*
- proc foundry
- TSMC*
- proc process size
- 28* nm
Specifications 2 · 36 fields
- CUDA compute capability
- 3.0*
- DirectX / feature level
- 12 (11_0)*
- OpenCL
- 3.0*
- OpenGL
- 4.6*
- Shader model
- 6.5 (5.1)*
- Vulkan
- 1.2.175*
- Architecture
- Kepler*
- Graphics interface
- PCIe 3.0 x16*
- Memory bus
- 256* bit
- Reported shader processors
- 1,536*
- Reported theoretical FP32 peak
- 3.548* TFLOPS
Scope and conditions
- Arithmetic density
- Not stated*
- Reported theoretical FP64 peak
- 0.1478* TFLOPS
Scope and conditions
- Arithmetic density
- Not stated*
- Declared L1 cache configuration
- 16 KB (per SMX)*
- Declared GPU L2 cache
- 512 KB*
- Reported display outputs
- 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2*
- Theoretical pixel fill rate
- 36.96 GPixel/s*
- External power connectors
- 2x 6-pin*
- Reported chip die area
- 294 mm²*
- Chip foundry
- TSMC*
- Process node declaration
- 28 nm*
- Transistors
- 3,540 million*
- Graphics processor
- GK104*
- Reported video memory
- 2* GB
- Rated memory bandwidth
- 198.4 GB/s*
- Video memory type
- GDDR5*
- Raster output units
- 32*
- Reported shading units
- 1,536*
- Card or module format
- Triple-slot*
- Suggested system PSU
- 450 W*
- 2:4 tensor sparsity
- No*
- Declared GPU TDP
- 195 W*
- Theoretical texture fill rate
- 147.8 GTexel/s*
- Texture mapping units
- 128*
- GPU cores
- 1,536*
- Video memory
- 2* GB
- Video memory type
- GDDR5*
0 selected Compare 0
* Provisional data