Report an Error

NVIDIA GRID A100B

Graphics Processor
GA100
Cores
6912
TMUs
432
ROPs
192
Memory Size
48 GB
Memory Type
HBM2e
Bus Width
6144 bit
The GRID A100B is a professional graphics card by NVIDIA, launched in May 2020. Built on the 7 nm process, and based on the GA100 graphics processor, the card supports DirectX 12 Ultimate. The GA100 graphics processor is a large chip with a die area of 826 mm² and 54,200 million transistors. It features 6912 shading units, 432 texture mapping units, and 192 ROPs. Also included are 432 tensor cores which help improve the speed of machine learning applications. NVIDIA has paired 48 GB HBM2e memory with the GRID A100B, which are connected using a 6144-bit memory interface. The GPU is operating at a frequency of 900 MHz, which can be boosted up to 1005 MHz, memory is running at 1215 MHz.
Its power draw is rated at 400 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. GRID A100B is connected to the rest of the system using a PCI-Express 4.0 x16 interface.

Graphics Processor

GPU Name
GA100
Architecture
Ampere
Foundry
TSMC
Process Size
7 nm
Transistors
54,200 million
Die Size
826 mm²

Graphics Card

Release Date
May 14th, 2020
Generation
GRID Ax
Production
Active
Bus Interface
PCIe 4.0 x16

Relative Performance

1%
GeForce 210
2%
GeForce 9400 GT
2%
Radeon HD 4550
3%
Radeon HD 5450
3%
Radeon HD 6450
3%
GeForce GT 520
4%
GeForce GT 220
5%
GeForce GT 430
5%
Radeon HD 5570
6%
Radeon HD 4670
6%
GeForce GT 440
6%
GeForce GT 240
7%
GeForce 9600 GT
8%
Radeon HD 5670
8%
GeForce GT 640
8%
GeForce 9800 GT
9%
Radeon HD 6670
9%
Radeon HD 4830
10%
Radeon HD 4770
10%
Radeon HD 4850
10%
GeForce GTS 250
11%
GeForce GTS 450
11%
Radeon HD 5750
13%
Radeon HD 7750
13%
Radeon HD 5770
13%
GeForce GTX 260
13%
Radeon HD 4870
13%
Radeon Vega 8
13%
GeForce GTX 550 Ti
14%
GeForce GTX 650
14%
Radeon HD 5830
15%
Radeon HD 6790
15%
Radeon HD 4890
16%
GeForce GTX 460
16%
Radeon RX Vega 11
16%
GeForce GT 1030
16%
GeForce GTX 275
16%
Radeon HD 7770 GHz Edition
16%
GeForce GTX 280
17%
GeForce GTX 465
18%
Radeon HD 6850
18%
GeForce GTX 285
19%
Radeon HD 5850
20%
GeForce GTX 650 Ti
20%
Radeon HD 7790
20%
Radeon RX 550
20%
GeForce GTX 470
21%
Radeon HD 6870
23%
Radeon HD 5870
23%
GeForce GTX 560 Ti
23%
Radeon HD 4870 X2
24%
GeForce GTX 750 Ti
24%
Radeon HD 6950
25%
GeForce GTX 295
25%
GeForce GTX 650 Ti Boost
25%
Radeon HD 7850
26%
GeForce GTX 480
27%
Radeon HD 6970
27%
Radeon R7 265
28%
Radeon RX 460
28%
Radeon R7 370
28%
GeForce GTX 660
28%
GeForce GTX 570
28%
Radeon RX 560
30%
GeForce GTX 950
31%
Radeon HD 7870 GHz Edition
31%
Radeon R9 270X
32%
GeForce GTX 660 Ti
32%
GeForce GTX 580
32%
GeForce GTX 1050
32%
Radeon HD 7950
33%
Radeon HD 5970
34%
GeForce GTX 760
36%
GeForce GTX 670
37%
GeForce GTX 960
38%
Radeon R9 380
38%
Radeon R9 285
39%
Radeon HD 7970
39%
GeForce GTX 680
40%
GeForce GTX 1050 Ti
41%
GeForce GTX 770
42%
Radeon HD 6990
43%
Radeon R9 280X
44%
GeForce GTX 590
44%
Radeon HD 7970 GHz Edition
48%
GeForce GTX 780
50%
GeForce GTX 1650
53%
Radeon RX 470
54%
Radeon R9 290
57%
Radeon RX 570
58%
Radeon R9 390
58%
Radeon R9 290X
59%
GeForce GTX 970
59%
GeForce GTX TITAN
60%
GeForce GTX 780 Ti
61%
Radeon R9 390X
61%
Radeon HD 7990
62%
Radeon RX 480
62%
GeForce GTX 690
64%
GeForce GTX 1060 6 GB
65%
Radeon RX 5500 OEM
66%
Radeon RX 580
66%
Radeon RX 5500 XT
67%
GeForce GTX 980
67%
GeForce GTX 1650 SUPER
68%
Radeon R9 FURY
71%
Radeon RX 590
75%
Radeon R9 295X2
75%
GeForce GTX 1660
76%
Radeon R9 FURY X
77%
GeForce GTX 980 Ti
79%
GeForce GTX TITAN X
84%
GeForce GTX 1660 SUPER
86%
GeForce GTX 1660 Ti
86%
GeForce GTX 1070
92%
Radeon RX Vega 56
97%
GeForce GTX 1070 Ti
100%
Radeon RX Vega 64
100%
GRID A100B
101%
GeForce RTX 2060
101%
GeForce GTX 1080
105%
Radeon RX 5700
114%
GeForce RTX 2060 SUPER
118%
GeForce RTX 2070
121%
Radeon RX 5700 XT
122%
Radeon VII
125%
TITAN X Pascal
129%
GeForce GTX 1080 Ti
130%
GeForce RTX 2070 SUPER
139%
GeForce RTX 2080
147%
GeForce RTX 2080 SUPER
162%
GeForce RTX 2080 Ti
214%
GeForce RTX 3080
236%
GeForce RTX 3090
Based on TPU review data: "Performance Summary" at 1920x1080, 4K for 2080 Ti and faster.
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

Base Clock
900 MHz
Boost Clock
1005 MHz
Memory Clock
1215 MHz
2.4 Gbps effective

Memory

Memory Size
48 GB
Memory Type
HBM2e
Memory Bus
6144 bit
Bandwidth
1,866 GB/s

Render Config

Shading Units
6912
TMUs
432
ROPs
192
SM Count
108
Tensor Cores
432
L1 Cache
192 KB (per SM)
L2 Cache
48 MB

Theoretical Performance

Pixel Rate
193.0 GPixel/s
Texture Rate
434.2 GTexel/s
FP16 (half) performance
55.57 TFLOPS (4:1)
FP32 (float) performance
13.89 TFLOPS
FP64 (double) performance
6.947 TFLOPS (1:2)

Board Design

Slot Width
IGP
TDP
400 W
Suggested PSU
800 W
Outputs
No outputs
Power Connectors
None

Graphics Features

DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
2.0
Vulkan
1.2
CUDA
8.0
Shader Model
6.5

Card Notes

specs unknown