GPU Comparison
GEFORCE
NVIDIA GeForce GTX 580
CORE STATE
GF110
VRAM
1536 MB
CLOCK SPEED
—
TDP
244 W
BUS WIDTH
384 bit
ARCHITECTURE
Fermi 2.0
PROCESS
40 nm
LAUNCH DATE
2010
VS
GEFORCE
GeForce RTX 3090 Ti
CORE STATE
GA102
VRAM
24 GB
CLOCK SPEED
1860 MHz
TDP
450 W
BUS WIDTH
384 bit
ARCHITECTURE
Ampere
PROCESS
8 nm
LAUNCH DATE
2022
PERFORMANCE BENCHMARKS
geekbench_opencl
3dmark_3dmark_steel_nomad_dx12
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
GTX 580
RTX 3090 Ti
Core Specs
Shading Units
512
10,752
+2000.0%
Shaders
512
10,752
+2000.0%
TMUs
64
336
+425.0%
ROPs
48
112
+133.3%
SM Count
16
84
+425.0%
Clocks
Base Clock
—
1560 MHz
Boost Clock
—
1860 MHz
GPU Clock
772 MHz
—
Shader Clock
1544 MHz
—
Memory Clock
1002 MHz
4 Gbps effective
1313 MHz
21 Gbps effective
Memory
Memory Size
1536 MB
24 GB
VRAM (MB)
1,536
24,576
+1500.0%
Memory Type
GDDR5
GDDR6X
Memory Bus
384 bit
384 bit
Bandwidth
192.4 GB/s
1.01 TB/s
Cache
L1 Cache
64 KB (per SM)
128 KB (per SM)
L2 Cache
768 KB
6 MB
Performance
Pixel Rate
24.70 GPixel/s
208.3 GPixel/s
Texture Rate
49.41 GTexel/s
625.0 GTexel/s
FP32 (TFLOPS)
1.581 TFLOPS
40.00 TFLOPS
FP64 (TFLOPS)
197.6 GFLOPS (1:8)
625.0 GFLOPS (1:64)
FP16 (TFLOPS)
—
40.00 TFLOPS (1:1)
AI/RT
RT Cores
—
84
Tensor Cores
—
336
Power
TDP
244 W
450 W
TDP (W)
244
450
+84.4%
Suggested PSU
550 W
850 W
Power Connectors
1x 6-pin + 1x 8-pin
1x 16-pin
Architecture
Architecture
Fermi 2.0
Ampere
GPU Name
GF110
GA102
Generation
GeForce 500
GeForce 30
Process Size
40 nm
8 nm
Transistors
3,000 million
28,300 million
Die Size
520 mm²
628 mm²
Foundry
TSMC
Samsung
Density
5.8M / mm²
45.1M / mm²
API Support
DirectX
12 (11_0)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
—
1.4
OpenCL
1.1
3.0
CUDA
2.0
8.6
Shader Model
5.1
6.8
Physical
Slot Width
Dual-slot
Triple-slot
Length
267 mm
10.5 inches
336 mm
13.2 inches
Height
111 mm
4.4 inches
140 mm
5.5 inches
Outputs
2x DVI1x mini-HDMI 1.3a
1x HDMI 2.13x DisplayPort 1.4a
Bus Interface
PCIe 2.0 x16
PCIe 4.0 x16
Other
Launch Price
499 USD
1,999 USD
Production
End-of-life
End-of-life
Predecessor
GeForce 400
GeForce 20
Successor
GeForce 600
GeForce 40