Understanding Display Interfaces

You are here

Display size, contrast, color, brightness, resolution, and power are key factors in choosing the right display technology for your application. However, making the right choice in how you feed the information to the display is just as vital, and there are many interface options available.

Understanding Display Technology

All displays work in a similar manner. In a very basic explanation, they all have many of rows and columns of pixels driven by a controller that communicates with each pixel to emit the brightness and color needed to make up the transmitted image. In some devices, the pixels are diodes that light up when current flows (PMOLEDs and AMOLEDs), and in other electronics, the pixel acts as a shutter to let some of the light from a backlight visible. In all cases, a memory array stores the image information that travels to the display through an interface.

Display Interface

According to Wikipedia, "an interface is a shared boundary across which two separate components of a computer system exchange information. The exchange can be between software, computer hardware, peripheral devices, humans, and combinations of these. Some computer hardware devices such as a touchscreen can both send and receive data through the interface, while others such as a mouse or microphone may only provide an interface to send data to a given system.” In other words, an interface is something that facilitates communication between two objects. Although display interfaces serve a similar purpose, how that communication occurs varies widely.

Serial Peripheral Interface (SPI)

Serial Peripheral Interface (SPI) is a synchronous serial communication interface best-suited for short distances. It was developed by Motorola for components to share data such as flash memory, sensors, Real Time Clocks, analog-to-digital converters, and more. Because there is no protocol overhead, the transmission runs at relatively high speeds. SPI runs on one master (the side that generates the clock) with one or more slaves, usually the devices outside the central processor. One drawback for SPI is the number of pins required between devices. Each slave added to the master/slave system needs an additional chip select I/O pin on the master. SPI is a great option for small, low-resolution displays including PMOLEDs and smaller LCDs.

SPI Benefits:

  • Ease of setup
  • Faster than I2C
  • Bandwidth capabilities up to ~10mb/sec

Inter-integrated Circuit (I2C)  

Philips Semiconductors invented I2C (Inter-integrated Circuit) or I-squared-C in 1982. It utilizes a multi-master, multi-slave, single-ended, serial computer bus system. Engineers developed I2C for simple peripherals on PCs, like keyboards and mice to then later apply it to displays. Like SPI, it only works for short distances within a device and uses an asynchronous serial port. What sets I2C apart from SPI is that it can support up to 1008 slaves and only requires two wires, serial clock (SCL), and serial data (SDA). Like SPI, I2C also works well with PMOLEDs and smaller LCDs. Many display systems transfer the touch sensor data through I2C.

I2C benefits:

  • Low energy consumption
  • Resistance to noise 
  • Handles a broad range of operating temperatures
  • Ease of use and troubleshooting
  • Bandwidth capabilities up to 1 MB/sec

RGB (Red Green Blue)

RGB is used to interface with large color displays. It sends 8 bits of data for each of the three colors, Red Green, and Blue every clock cycle. Since there are 24 bits of data transmitted every clock cycle, at clock rates up to 50 MHz, this interface can drive much larger displays at video frame rates of 60Hz and up. 

RGB benefits:

  • Low cost due to technology maturity
  • High performance
  • Works with medium to large displays
  • Bandwidth capabilities up to 1.2 GB/sec

RGB disadvantages:

  • Requires large pin real-estate (up to 29 pins) with expensive connectors
  • Fast edges on multiple wires can create electrical noise, which is particularly troubling in the many systems that include wireless modules

Low-voltage Differential Signaling (LVDS)

Low-Voltage Differential Signaling (LVDS) was developed in 1994 and is a popular choice for large LCDs and peripherals in need of high bandwidth, like high-definition graphics and fast frame rates. It is a great solution because of its high speed of data transmission while using low voltage. Two wires carry the signal,  with one wire carrying the exact inverse of its companion. The electric field generated by one wire is neatly concealed by the other, creating much less interference to nearby wireless systems. At the receiver end, a circuit reads the difference (hence the "differential" in the name) in voltage between the wires. As a result, this scheme doesn’t generate noise or gets its signals scrambled by external noise. The interface consists of four, six, or eight pairs of wires, plus a pair carrying the clock and some ground wires. 24-bit color information at the transmitter end is converted to serial information, transmitted quickly over these pairs of cables, then converted back to 24-bit parallel in the receiver, resulting in an interface that is very fast to handle large displays and is very immune to interference.

LVDS benefits:

  • Ideal for systems that include wireless transmitters, since it generates little interference
  • Works for larger displays 
  • Bandwidth capabilities up to 3.125 GB/sec

Mobile Industry Processor Interface (MIPI)

Mobile Industry Processor Interface (MIPI) is a newer technology that is managed by the MIPI Alliance and has become a popular choice among wearable and mobile developers. MIPI uses similar differential signaling to LVDS by using a clock pair and one to eight pairs of data called lanes. MIPI supports a complex protocol that allows high speed and low power modes, as well as the ability to read data back from the display at lower rates. There are several versions of MIPI for different applications, MIPI DSI being the one for displays.

MIPI disadvantages

  • Complex protocol and driver software 
  • Initially available mainly on cellphone-sized displays
  • Needs careful board layout to function at high speed

MIPI can be adapted to fit any industry’s needs such as mobile phones, PCs, or a variety of other consumer applications. 

How Do All The Interfaces Compare?

Below is a comparison table of the different interface technologies:

Technology # of Pins Max Bandwidth Applications Clock
SPI 5 ~10MB/sec Small displays on MCU’s 8 bits/10 clocks
I2C 4 400KB/sec or 1MB/sec Small displays on PC’s and PC peripherals N/A
RGB 29 for 8 bits/color 24 bits X 50MHz = 1.2Gb/Sec Larger displays Max clock 50MHz
LVDS 4, 6, or 8 pairs; 8, 12, or 16 + Clock Pair + grounds +3 = 25 6 x 300MHz x 2 = 2.4 GB/sec; Up to 3.125 GB/sec Larger displays, harsh environments Max clock at 300MHz
MIPI DSI 4 or 8 pairs plus clock pair, 10 or 18 4 x 1.5GB/sec = 6GB/sec Cellphones  

 

Bandwidth Matters

Display components stretch the limitations of bandwidth. For perspective, the most common internet bandwidth in a residential home runs on average at around 20 megabits per second or 20 billion 1s and 0s per second. Even small displays can require 4MB per second, which is a lot of data in what is often a tightly constrained physical space.  

Computing Bandwidth

To give an example, a small monochrome PMOLED with a resolution of 128 x 128 contains 16,384 individual diodes. A still image of various diodes carrying current represents a frame. A frame rate is the number of times that a picture needs refreshing. Most videos have a frame rate of 60 fps (frames per second), which means that it is updated 60 times every second.

This number is important because it is the rate that the average human eye does not notice the flicker caused by the transition between frames. 

Take the same PMOLED display with the 128 x 128 resolution and 16,384 separate diodes; it requires information as to when and how brightly to illuminate each pixel. For a display with only 16 shades, it takes 4 bits of data. 128 x 128 x 4 = 65,536 bits for one frame. Now multiply it by the 60Hz, and you get a bandwidth of 4 megabits/second for a small monochrome display. 

Below is simplified version of how to compute the bandwidth for a display.

You will need:

  • Resolution (pixel²)
  • Brightness/Pixel (Pixel Ram)
  • Frame Rate (Hz)
  • Number of colors

Steps to compute:

  1. Resolution x Pixel Ram (Brightness/Pixel) = Single Frame
  2. Frame Rate x Single Frame = Bits/Second
  3. Bits/Second x Colors (RGB) = Bandwidth
Size (inches) Resolution (bits) Pixel Ram Frame Rate (Hz) Colors Bandwidth per channel
1.5" PMOLED 128 x 128 = 16,384 4 60 1 3.9 (MB/sec)
12.3" TFT LCD 1920 x 720 = 1,382,400 8 60 3 1,991 (2 gigabits/sec)

 

Conclusion

Choosing the right display for your application requires more than knowing what it needs to look like and what the optical environment is. 

Crucial factors:

  • Knowing how to connect the display to the electronic system
  • Choosing the right interface to the display

Key parameters:

  • The amount of data
  • The refresh rate needed
  • The number of colors of the GUI

Don't forget to future proof the interface for when your marketing department wants a new GUI for the next software update.

If you have more questions or need help finding the right display interface, send us an email to sales@usmicroproducts.com.

 

 

[1] https://en.wikipedia.org/wiki/Interface_(computing)

MIPI
LVDS
SPI
I2C
RGB
display interfaces

Request Additional Info

US Micro on Twitter

Follow Us on Social Media

Contact Us

North America HQ

6207 Bee Caves Rd., Ste 330
Austin, TX 78746

Tel: (512) 385 9000
Fax: (512) 385 9002