GPU Comparison
RADEON
ATI Radeon HD 3850 AGP
CORE STATE
RV670
VRAM
512 MB
CLOCK SPEED
—
TDP
75 W
BUS WIDTH
256 bit
ARCHITECTURE
TeraScale
PROCESS
55 nm
LAUNCH DATE
2008
VS
GEFORCE
GeForce RTX 2080 Max-Q
CORE STATE
TU104
VRAM
8 GB
CLOCK SPEED
1095 MHz
TDP
80 W
BUS WIDTH
256 bit
ARCHITECTURE
Turing
PROCESS
12 nm
LAUNCH DATE
2019
DETAILED SPECIFICATIONS
SPECIFICATION
ATI Radeon HD 3850 AGP
RTX 2080 Max-Q
Core Specs
Shading Units
320
2,944
+820.0%
Shaders
320
2,944
+820.0%
TMUs
16
184
+1050.0%
ROPs
16
64
+300.0%
Compute Units
4
—
SM Count
—
46
Clocks
Base Clock
—
735 MHz
Boost Clock
—
1095 MHz
GPU Clock
668 MHz
—
Memory Clock
828 MHz
1656 Mbps effective
1500 MHz
12 Gbps effective
Memory
Memory Size
512 MB
8 GB
VRAM (MB)
512
8,192
+1500.0%
Memory Type
GDDR3
GDDR6
Memory Bus
256 bit
256 bit
Bandwidth
52.99 GB/s
384.0 GB/s
Cache
L1 Cache
—
64 KB (per SM)
L2 Cache
256 KB
4 MB
Performance
Pixel Rate
10.69 GPixel/s
70.08 GPixel/s
Texture Rate
10.69 GTexel/s
201.5 GTexel/s
FP32 (TFLOPS)
427.5 GFLOPS
6.447 TFLOPS
FP64 (TFLOPS)
85.50 GFLOPS (1:5)
201.5 GFLOPS (1:32)
FP16 (TFLOPS)
—
12.89 TFLOPS (2:1)
AI/RT
RT Cores
—
46
Tensor Cores
—
368
Power
TDP
75 W
80 W
TDP (W)
75
80
+6.7%
Suggested PSU
250 W
—
Power Connectors
1x 8-pin
None
Architecture
Architecture
TeraScale
Turing
GPU Name
RV670
TU104
Generation
Radeon R600
(HD 3800)
GeForce 20 Mobile
Process Size
55 nm
12 nm
Transistors
666 million
13,600 million
Die Size
192 mm²
545 mm²
Foundry
TSMC
TSMC
Density
3.5M / mm²
25.0M / mm²
API Support
DirectX
10.1 (10_1)
12 Ultimate (12_2)
OpenGL
3.3 (full)
4.0 (partial)
4.6
Vulkan
—
1.4
OpenCL
—
3.0
CUDA
—
7.5
Shader Model
4.1
6.8
Physical
Slot Width
Single-slot
MXM Module
Outputs
2x DVI1x S-Video
Portable Device Dependent
Bus Interface
AGP 8x
PCIe 3.0 x16
Other
Production
End-of-life
End-of-life
Predecessor
Radeon R500 PCIe
GeForce 10 Mobile
Successor
Radeon R700
GeForce 30 Mobile