Report an Error

NVIDIA Tesla P4

GP104
Graphics Processor
2560
Cores
160
TMUs
64
ROPs
8192 MB
Memory Size
GDDR5
Memory Type
256 bit
Bus Width
NVIDIA Tesla P4 Photo NVIDIA GP104 Photo
The Tesla P4 is a professional graphics card by NVIDIA, launched in September 2016. Built on the 16 nm process, and based on the GP104 graphics processor, in its GP104-895-A1 variant, the card supports DirectX 12.0. The GP104 graphics processor is a large chip with a die area of 314 mm² and 7,200 million transistors. It features 2560 shading units, 160 texture mapping units and 64 ROPs. NVIDIA has placed 8,192 MB GDDR5 memory on the card, which are connected using a 256-bit memory interface. The GPU is operating at a frequency of 810 MHz, which can be boosted up to 1063 MHz, memory is running at 1502 MHz.
Being a single-slot card, the NVIDIA Tesla P4 does not require any additional power connector, its power draw is rated at 75 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. Tesla P4 is connected to the rest of the system using a PCI-Express 3.0 x16 interface. The card measures 267 mm in length, and features a single-slot cooling solution.

Graphics Processor

GPU Name
GP104
GPU Variant
GP104-895-A1
Architecture
Pascal
Process Size
16 nm
Transistors
7,200 million
Die Size
314 mm²

Graphics Card

Release Date
Sep 13th, 2016
Generation
NVIDIA Tesla Pxx
Production Status
Active
Bus Interface
PCIe 3.0 x16

Relative Performance

2%
3%
3%
4%
5%
5%
6%
8%
8%
9%
9%
10%
11%
12%
13%
13%
13%
14%
15%
16%
16%
16%
17%
19%
20%
20%
20%
20%
20%
21%
22%
23%
23%
24%
24%
25%
25%
25%
25%
25%
27%
27%
29%
30%
30%
31%
31%
32%
34%
35%
36%
36%
36%
38%
38%
38%
39%
41%
42%
42%
42%
43%
43%
46%
46%
48%
48%
48%
49%
49%
50%
50%
51%
55%
57%
58%
58%
59%
60%
61%
63%
64%
66%
66%
66%
74%
81%
82%
88%
88%
89%
91%
91%
92%
92%
93%
93%
95%
100%
Tesla P4
102%
103%
103%
104%
114%
117%
122%
126%
139%
143%
157%
157%
163%
191%
196%
Based on TPU review data: "Performance Summary" at 1920x1080
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

GPU Clock
810 MHz
Boost Clock
1063 MHz
Memory Clock
1502 MHz
6008 MHz effective

Memory

Memory Size
8192 MB
Memory Type
GDDR5
Memory Bus
256 bit
Bandwidth
192.3 GB/s

Render Config

Shading Units
2560
TMUs
160
ROPs
64
SM Count
20

Theoretical Performance

Pixel Rate
68.03 GPixel/s
Texture Rate
170.1 GTexel/s
FP16 (half) performance
85.04 GFLOPS (1:64)
FP32 (float) performance
5,443 GFLOPS
FP64 (double) performance
170.1 GFLOPS (1:32)

Board Design

Slot Width
Single-slot
Length
10.5 inches
267 mm
TDP
75 W
Outputs
No outputs
Power Connectors
None

Graphics Features

DirectX
12.0 (12_1)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.1.73
CUDA
6.1
Shader Model
6.1

GP104 GPU Notes

PureVideo HD: VP8
VDPAU: Feature Set H