RTX 4000 SFF Ada Generation vs Tesla S870
We compared two Professional market GPUs: 16GB VRAM RTX 4000 SFF Ada Generation and 1536MB VRAM Tesla S870 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
RTX 4000 SFF Ada Generation 's Advantages
Released 11 years and 6 months late
Boost Clock1746MHz
More VRAM (16GB vs 1536GB)
Larger VRAM bandwidth (1.02TB/s vs 76.80GB/s)
3712 additional rendering cores
Lower TDP (300W vs 800W)
Score
Benchmark
FP32 (float)
RTX 4000 SFF Ada Generation
+3775%
13410
Tesla S870
346
Graphics Card
Nov 2018
Release Date
May 2007
Radeon Instinct
Generation
Tesla
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 1.0 x16
Clock Speeds
1200 MHz
Base Clock
-
1746 MHz
Boost Clock
-
1000 MHz
Memory Clock
800 MHz
Memory
16GB
Memory Size
1536MB
HBM2
Memory Type
GDDR3
4096bit
Memory Bus
384bit
1.02TB/s
Bandwidth
76.80GB/s
Render Config
60
Compute Units
-
48
SM Count
16
3840
Shading Units
128
240
TMUs
32
64
ROPs
24
192
Tensor Cores
-
48
RT Cores
-
16 KB (per CU)
L1 Cache
-
4 MB
L2 Cache
96 KB
Theoretical Performance
111.7 GPixel/s
Pixel Rate
14.40 GPixel/s
419.0 GTexel/s
Texture Rate
38.40 GTexel/s
26.82 TFLOPS
FP16 (half)
-
13.41 TFLOPS
FP32 (float)
345.6 GFLOPS
6.705 TFLOPS
FP64 (double)
-
Board Design
300W
TDP
800W
700 W
Suggested PSU
1200 W
1x mini-DisplayPort 1.4a
Outputs
No outputs
2x 8-pin
Power Connectors
-
Graphics Processor
Vega 20
GPU Name
G80
Vega 20 GLXT
GPU Variant
-
GCN 5.1
Architecture
Tesla
TSMC
Foundry
TSMC
7 nm
Process Size
90 nm
13.23 billion
Transistors
0.681 billion
331 mm²
Die Size
484 mm²
Graphics Features
12 (12_1)
DirectX
11.1 (10_0)
4.6
OpenGL
3.3
2.1
OpenCL
1.1 (1.0)
1.3
Vulkan
N/A
8.9
CUDA
1.0
6.7
Shader Model
4.0
