DisplayPort

DisplayPort
DisplayPort entered the market in 2008. HDMI has been in use since 2002 so already had a big market share before DisplayPort was released. DisplayPort is different from HDMI in that it uses packets to transmit data. Using packets is also used by other communication systems like Ethernet, USB and PCI Express.

DisplayPort does not use a clock signal. This means it does not require any dedicated pins for transmitting a clock and thus reduces the number of pins required. DisplayPort uses differential signaling to transmit data, which allows the clock and data to be transferred using the same pins. More on that later in the video.

Since DisplayPort uses packets to send data, it can be extended to include new features. DisplayPort is primarily used for video, however it also supports audio transmissions. Audio is usually transferred with video, but this is not required. You could, for example, use DisplayPort to transfer audio only, but as DisplayPort was designed with video in mind you probably won’t ever find it used for audio without video also being transferred.

DisplayPort has an auxiliary channel which allows traffic to be transferred in the opposite direction. Therefore, using this auxiliary channel allows for bi-directional communication. This is essential for receiving device management information and can carry data for USB signals. However, you will find that DisplayPort is generally not used for USB traffic, although it is possible to do so.

There have been a number of different versions of DisplayPort. I will now have a look at some of these different versions.

Version 1.1a
DisplayPort has been through a number of different versions and revisions to some of those versions. Rather than talking about them all, I will combine some of them together, as some are not widely used, and others include only minor changes.

Version 1.1a was released in 2008. There was version 1 and version 1.1 before this, but they were not adopted by the market. Version 1.1a transfers data at 10.8 Gigabits per second. There is some transmission overhead so the actual amount of data that is transferred is 8.64 Gigabits per second.

In order to be considered to support these standards, this speed has to be maintained over a cable length of up to two meters. Keep this in mind if you purchase a cable that is longer than two meters. The cable speed may be reduced and still considered to be compliant with the standard.

This version has four data lanes and one auxiliary lane. All versions of DisplayPort have four lanes and one auxiliary lane. The four data lanes allow four streams of data to be sent down the cable to the device on the other end. The auxiliary lane is the one lane that runs in the opposite direction. This is what gives DisplayPort its ability to transfer data in both directions.

The next feature is it supports High-bandwidth Digital Content Protection or HDCP. This is copy protection that was originally designed by Intel. It also supports DisplayPort Content Protection or DPCP developed by Philips.

Both are copy protection systems that are designed to prevent the signal being duplicated and copied once it leaves the computer via the connector. The idea being that only a supported device would be able to connect to the DisplayPort connector, thus preventing splitting devices and other recording equipment from being used. With the increase of live streaming and other services it becomes difficult to support copy protection as live streaming by its nature won’t work with copy protection. You can understand why technology like this is not widely used. Using it would mean only supported devices could be used and certain setups would not work.

The basics of DisplayPort have not changed for newer versions; the differences are the speeds they run at and also the addition of new features. I will now have a look at the next version of DisplayPort.

Version 1.2a
The next version I will look at is version 1.2a. Version 1.2 was released in 2010 and revision 1.2a in 2013. There is not too much difference between the two, so I will combine them together and, in a moment, explain the difference.

Version 1.2a, after removing the overhead, transfers data at 17.28 Gigabits per second. This effectively doubles the data rate from version 1.1. The speed of the auxiliary channel has increased to 720 Megabits per second. You can see the auxiliary channel is very slow compared to the data channels. Remember that the primary purpose of DisplayPort is to transfer video data which can be quite large and does not require much data to be transferred back.

The next feature added is multiple display streams. This allows more streams to be transferred over a single cable. This essentially means the one monitor could be used to display multiple monitor video streams. This is useful for applications like CCTV systems where you may have multiple camera signals that you want to display on the same screen.

The next feature added is the ability to daisy chain monitors together. This means that multiple monitors can be connected together rather than running separate cables from the computer to each monitor. This is a feature that Thunderbolt also supports.

The next feature is support for stereoscopic 3D. This is particularly important for virtual reality and 3D screens. Version 1.2 also added additional color spaces. The color space determines the number of colors that can be transmitted to the monitor. Some color spaces include larger color ranges; it is useful to have this extra color range when applying extra effects. For example, if a monitor was changing how bright or dark an image was displayed on the monitor, having the extra color range would give a better result.

DisplayPort Connector
Before I look at the next feature version 1.2 adds, I will first look at the connector that DisplayPort uses. DisplayPort uses the 20-pin connector and plug as shown. The same connector and plug is used for all versions of DisplayPort.

DisplayPort supports latches, so you will find most connectors will have latches on them, and these latches hold the connector in the plug. When removing the connector, make sure you disengage the latches before attempting to remove it. Usually, the latches are disengaged by pressing down on the connector.

With version 1.2, a smaller connector was added which is essentially the same as the larger connector. The smaller connector uses 20 pins just like the larger connector and so essentially works the same way. Using a Mini DisplayPort connector does not affect performance or change any of the features that are supported, it is just smaller.

The Mini DisplayPort was first used by Apple in 2008. In 2010 with the release of version 1.2 of DisplayPort it was added to the standard. You will primarily find that the Mini DisplayPort is used by Apple products, however in newer products Apple has started using the USB-C connection rather than the Mini DisplayPort. In PC computers you will mainly find the larger DisplayPort connection, but you may come across the smaller one from time to time.

Before I move on to the next version of DisplayPort, I will have a look at the features that revision 1.2a adds.

Version 1.2a (Differences)
Version 1.2a adds one feature over 1.2 and this feature is VESA Adaptative-Sync. This feature allows the computer to change the refresh rate of the monitor to match the speed it can update the screen. This prevents what is referred to as screen tear. Screen tear is when the screen is updated while it is being drawn. When screen tear occurs, you will see the upper part of the screen is disjointed compared to what is shown on the lower part of the screen and is most noticeable when there is a lot of motion on the screen. When it occurs, it looks like the upper and bottom part of the screen are offset.

Adaptive-Sync is optional. So essentially this means that to be version 1.2a compliant the device does not have to have Adaptive-Sync. Therefore, since this is optional, it is possible for a 1.2 device to be retrospectively updated to 1.2a, so you can see why I have combined 1.2 and 1.2a into one. Since the only difference is optional, do not assume if it is labeled 1.2a that it has VESA Adaptive-Sync. Also keep in mind, there is no difference between version 1.2 and 1.2a cables.

VESA Adaptive-Sync is used by AMD’s Freesync technology. As the name suggests this technology is free, unlike other technology such as G-Sync which requires proprietary hardware modules supplied by Nvidia.

So, you can see that 1.2 and 1.2a are essentially the same. If you purchase a version 1.2a device, and if you need VESA Adaptive-Sync, you need to confirm the device supports it – do not assume it does.

Version 1.3
The next version that I will look at is version 1.3. This version was released in 2014. It increases the data transfer rate to almost 26 Gigabits per second. This is almost 70% faster than the previous version.

Version 1.3 adds High-bandwidth Digital Content Protection version 2.2. Previously, version 1.3 was available in DisplayPort. Version 2.2 is designed to be completely different from the previous version. Thus, all devices connected will need to support the same version in order for them to work together.

The last new feature is that it includes mandatory support for Dual Mode. This allows for simple passive adapters to be used with DVI and HDMI. Before this, this support was optional. With DVI, it is able to send one or two signals along the same cable. The second signal effectively doubles the amount of data that can be sent; however, it increases the number of wires in the cable.

Being able to transfer more data essentially means higher resolutions and frame rates. DVI calls one signal single mode and two signals dual mode. With DisplayPort, being able to support DVI by remapping the wires and using limited electronics in the cable is called Dual Mode. This is where it gets confusing.

DisplayPort Dual Mode will only support one signal in DVI. This is because DisplayPort only does basic signal processing rather than signal conversion. DisplayPort has enough wires to do this with a single signal, but not enough for two signals.

Remapping the wires requires a passive adapter which are cheap to make as they do not require much electronics. If you wish to use higher resolutions with a DVI signal you will need an active adapter. Later in the video I will look at Dual Mode in more detail. The important take away here is that version 1.3 will natively support passive DVI to DisplayPort adaptation, whereas before this it was optional. However, with DVI, it won’t support higher resolutions, due to it being limited by how much data it can send.

Version 1.4a
The next version, version 1.4 was released in 2016 and version 1.4a in 2018. These versions have the same speed as the previous versions. The first feature it adds is Display Stream Compression. This is a technology that uses lossless compression to reduce the amount of data that is transferred over the cable. Lossless essentially means that data, when decompressed, is identical to before it was compressed. Using some data compression will reduce the amount of data that is sent over the cable, allowing for higher resolutions.

With DisplayPort version 1.4a, there were added improvements to Display Stream Compression. The improvements add some extra color spaces and some bug fixes. So, you can see there is not too much of a difference between the two versions.

The next feature added is High Dynamic Range or HDR. In video and photography, HDR captures the same image multiple times using different exposures. The end result is a range of images at different brightnesses. The example shows two images, one dark and one bright; however, more than two could be used.

Capturing multiple images allows the camera to capture what the scene looks like at different light levels. This gives a higher range of bit depths, luminance and color. Essentially, the high range means better colors; for example, whites look brighter and blacks looks darker.

HDR also requires more bandwidth to transmit, so using it may reduce the resolution that the screen can display.

Version 2.0
The newest version of DisplayPort is 2.0 released in 2019. This increased the speed to just over 77 Gigabits per second, which is three times faster than the previous version. The cable does run at a faster speed; however, this is only one of the improvements to get that extra speed. Extra speed is also achieved by making the encoding used in the data signal more efficient. This encoding reduces the amount of overhead required to send data. I will look more at the encoding later in the video.

Unlike the other versions, version 2.0 has three different speeds the cable can run at. The lowest speed is just over 38 Gigabits per second and the mid-range speed is just over 52 Gigabits per second. The cable you use and the hardware in the video adapter and the screen used will determine which speed is used. As we will see later in the video, even the lowest speed gives us some impressive results.

It would seem logical that, with this version of DisplayPort, three different speeds were included to future proof the design. The manufacturer could simply decide to support a lower speed if that is all they needed. If they need more, they would support a faster speed. Faster generally means the electronics that run it cost more, so it is unlikely they would include a faster speed unless it is needed. The speeds are currently represented by UHBR followed by the speed of the signal data lane. Remember that there are four data lanes. Since this version is new technology, it would be difficult to find a device that supports it, but keep in mind, by the time it becomes commonplace there will probably be a marketing term that is used to describe each version.

Version 2.0 has a number of improvements to DisplayPort. For example, it has a higher refresh rate, improvements to augmented reality and virtual reality (just to name a few).

Version 2.0 essentially improves on the previous versions; the high bandwidth allows for higher resolutions all the way up to 8K. Later in the video, I will compare the different versions and what resolutions you can expect to achieve.

One new feature to this version is Panel Replay. This allows the screen to only update those parts of the screen that have changed. Updating less of the screen reduces power consumption. This is particularly useful if the screen displays a static image for a long time, for example, if the screen is left open on a static web page. With small portable devices, being able to update only the parts of the screen that are changing potentially reduces the amount of power required to run the device, thus making it able to run longer between recharges.

At the time of making of this video, version 2.0 was the latest version of DisplayPort. You can see they are some feature differences between the versions; however, the most important difference I have not looked at yet is the resolution each version can support, which ultimately determines which version you will need to use.

Maximum Resolutions
Shown here is the upper resolution you can expect from different versions. The resolution is determined by a number of factors – The first one being the speed of the cable and the second is the amount of data that is going through the cable.

Different versions that have the same speed have been grouped together since they will give similar results. Keep in mind that, in this example, I am only considering 24bit uncompressed RGB data. Essentially, each color uses 8 bits of data, making a total of 24 bits for each pixel.

If using high dynamic range or a different color space, this may increase or decrease the resolution you can get. This table is a basic guide of what sort of resolutions you can achieve, but if you are using different color spaces, you may get a higher or lower resolution.

The green bars indicate what sort of maximum resolution and frame rate you can achieve. Keep in mind that in most cases this will only give you a refresh rate of 30Hz. This may not be too bad for office work; however, if you are playing games or editing video this will look very sluggish.

For this reason, I have also included the 60Hz maximum resolution which is a reasonable refresh rate for most applications. In the case of version 2.0, the 60Hz resolution is the lowest speed the version runs at, whereas the other resolution is using the fastest speed.

You can see that starting with version 1.2, you get a good refresh rate at 4K resolution. For most people, for the moment at least, version 1.2 would be a good starting point since 4K screens are becoming more common. You won’t see as many devices that support version 1.3 or 1.4 and are extremely rare for version 2.0 at the time this video was made.

Now that we understand what resolutions you can expect to achieve, I will now have a look at how the data itself is transferred.

Differential Signalling
DisplayPort uses differential signalling to transfer data and the clock signal, thus requiring no separate wires for the clock signal. In order for this system to work, the difference between the two signals is measured to determine if the data is a one or a zero.

To better understand how this works, consider that you have the source of the transmission. This transmits two signals at once. Transmitting across the wires is either a positive or negative voltage or, to think of it another way, the voltage is either high or low.

If I consider the first signal, we see that it has a high or positive voltage. When this occurs, the second signal will be low. When this is received at the other end, the difference between the two signals is measured. In this case, the difference between the two voltages is positive and thus is considered to be one bit.

To send a zero bit, the voltages that are transmitted over the two wires are reversed. Now when we compare the two signals, this time we get a negative difference rather than a positive difference, so we know the data being transferred is a zero bit.

You can also see that, between each of the data transmissions, both wires’ signals go back to zero. This is how the device knows when a data signal starts and stops. By using this method, you can see there is no need for a clock signal. The device simply looks for both wires to go to zero and thus knows that the previous data transmission has ended and the next one is about to start.

You can see this process is simply repeated for each transmission of data. Since the signal goes to zero between each transmission, you can see that even if the data signal is not perfectly timed, that is the high and low go on a little longer or end a little earlier, it does not matter. As long as both wires go back to zero voltage it is possible to detect when data starts and stops.

Differential signaling allows data to be transmitted at a lower voltage, which has its advantages but also causes problems. Let us have a look at how differential signaling overcomes these problems.

Interference
The biggest problem with using low voltage is that it is more sensitive to interference. Using differential signaling helps to get around the problems of interference; let’s have a look how.

Consider that you have two wires that are used to transfer data. Without any interference, you will notice that the signals are well defined. In order to work out if the data being sent is a zero or a one all you need to do is calculate the difference between the two signals. The result will either be a positive or negative depending on what data is being transmitted.

Now let’s consider what happens when you have some interference. With differential signaling, the two wires run alongside each other. This means that when there is some interference it will generally affect both wires fairly equally, since they are next to each other.

Now consider what happens when the same signal is sent through these wires. You will notice that the interference has affected both signals equally. Although the signals look different from the original signals, you will notice that the difference between the two signals has remained the same.

You can see that differential signaling handles interference quiet well. So, we now have a system for transferring data with an embedded clock signal that can handle interference. However, there are still problems with this system, depending on what data is sent.

8b/10b Encoding
The problem when using differential signaling is that, when you transmit too many zeros or ones the wires become unbalanced. To prevent this happening, the encoding system attempts to achieve DC-balance, which is essentially attempting to get the number of zeros and ones transmitted to be equal.

There are a number of steps in the encoding process, including changing the order of bits before encoding. I won’t go through the whole process here as it can change depending on the implementation. However, the basics of how it works is that two extra bits are set to attempt to achieve DC-balance.

You can see in this example that five bits are transmitted, then an extra bit, three bits and then an extra bit. Depending on how many ones and zero are transmitted, these two extra bits will be set, in order to attempt to even out the number of zeros and ones transmitted.

If you consider you had a container that you fill with water. Essentially almost full is a high signal, half full is used for the clock timing and close to empty is a low signal. If, for example, you have a lot of high signals in a row, you are attempting to fill the bucket, then empty the bucket so it is half full then refill it. As time goes on, generally when attempting to get the bucket back to half full, some water will be left resulting in the average water level going up.

This is referred to as DC offset, as essentially the halfway mark keeps increasing or getting offset from where it should be. As the offset gets bigger, it gets harder to determine the difference between the halfway mark and where the high signal is.

So, essentially what is happening is you are getting an accumulation effect or a memory if you will. To get around this, you essentially need to empty the tank or put some opposite signals through. This will help get the offset back around the halfway mark. By keeping a nice even flow of zeros and ones through the wires, this helps the DC offset from getting too skewed from where it should be, which will improve the reliability of the system.

However, all this comes at a cost. By adding these two extra bits you essentially increase the overhead of transmission by 20%. This method is used for all versions of DisplayPort except for version 2.0. Let’s have a look at how version 2.0 encodes data.

128b/132b Encoding
Version 2.0 of DisplayPort uses a different type of encoding. This encoding is essentially 64b/66b encoding done twice. What occurs is that a 64-bit block is randomized before transmission and then derandomized when it is received.

By randomizing the data, this effectively balances out the ones and zeros in transmission. This helps keep the signal, travelling down the wires, balanced. The extra two bits are two preamble bits that are added before transmission. These two bits are used as control and error signals.

Using 64-bit blocks reduces the overhead, but by combining two blocks together reduces the overhead some more. You can see how using this encoding reduces the overhead while also keeping the signal more balanced. This improves the reliability of the signal and allows more data to be transmitted faster.

Differential signaling is becoming more common for high speed data transmissions. You will see it used in USB 3 and PCI Express to transmit data. It has the advantages of using low power and having good reliability.

DisplayPort Dual Mode (DP++)
To end this video, I will have a look at DisplayPort’s ability to support DVI and HDMI through the use of passive adapters. An active adapter contains additional electronics which convert one signal to another, whereas passive adapters contain minimal components if any at all. Passive adapters are cheaper than active adapters, as they generally just re-wire one connector to another or do some basic conversion like changing voltages. DisplayPort version 1.3 and upwards has native support for passive adapters; however, before that, support was optional.

To understand how this works, consider that you have a device that only has a DVI output port. The DVI port has 24 pins with some extra pins used for analog signals. DisplayPort only supports digital signals so if you have a DVI that only supports analog you won’t be able to use it without a converter or active adapter.

A passive adapter can be as little as a cable with a different connector on the other end. You will find that some passive adapters will have some very basic components. Essentially one side will have DVI and the other side will be DisplayPort. The adapter or the cable will remap the pins on the DVI side to the required pins on the DisplayPort side. The small chip you can see in the passive adapter changes the voltages in the signal and does some minor signaling changing. The original DVI signal is not converted fully, just some of the voltages have changed. This is why passive adapters may not always work; because essentially, they only do part of the job.

The problem with this is the DVI plug has four extra pins. Since the DisplayPort only has 20 pins, these extra pins can’t be mapped to pins on the DisplayPort.

DVI supports dual signals to double the amount of data it can transmit at once. Since DisplayPort does not have enough pins to map to all the DVI pins, it will only be able to support one signal using a passive adapter. This limits the resolution that can be used to 1920×1200 at 60Hz. If you need a higher resolution, you will need to use an active adapter or a converter. These devices will be able to take the data and clock rate transmitted by the DVI connector and combine them into the one signal. Combining these signals reduces the number of pins required and, thus, dual DVI signals can be sent to a DisplayPort connector.

If you are using HDMI, the HDMI plug has 19 pins. So, you can see that since DisplayPort has 20 pins, it is not a problem to map all the pins from one to the other, as there are enough pins. The problem that occurs is this, even though both use differential signaling, the differential signaling they are using is different.

The solution to this is to have a tiny little chip on the passive adapter. This chip changes the voltage of the signal and performs some basic functions like correcting timing skews. The chip is considered to be a repeater. A repeater, as the name suggests, repeats a signal rather than doing any processing of the signal. Since HDMI and DisplayPort both use differential signaling, it is not too hard to convert between the two using a minimal amount of electronics.

Since this adapter is considered to be a passive adapter, as it does very basic signal conversion and rewiring, it may not work with DisplayPort before version 1.3. After 1.3, the standard required passive adapters like this to be supported in order to be considered to be compliant with the standard. If you are having problems, you may need to pay more for an active adapter which does a full conversion of the signal rather than just adjusting a few basic things and changing the voltages.

That concludes this video on DisplayPort from ITFreeTraining. I hope you have found this video useful and look forward to seeing you in the next video from us. Until the next video, thanks for watching.

References
“The Official CompTIA A+ Core Study Guide (Exam 220-1001)” Chapter 5 Position 81 – 88
“CompTIA A+ Certification exam guide. Tenth edition” Pages 755 – 756
“DisplayPort” https://en.wikipedia.org/wiki/DisplayPort
“64b/66b encoding” https://en.wikipedia.org/wiki/64b/66b_encoding
“Picture: Person holding package” https://pixabay.com/vectors/toy-monster-fantasy-strange-576520/
“Picture: Loud speaker” https://pixabay.com/illustrations/megaphone-speaker-speak-loud-2374502/
“Picture: cat 1” https://unsplash.com/photos/zBvVuRJ71vU
“Picture: cat 2” https://unsplash.com/photos/IFxjDdqK_0U
“Picture: cat 3” https://unsplash.com/photos/gKXKBY-C-Dk
“Picture: cat 4” https://unsplash.com/photos/NodtnCsLdTE
“Picture: SubtractiveColor” https://commons.wikimedia.org/wiki/File:SubtractiveColor.svg
“Picture: DisplayPort Connector” https://en.wikipedia.org/wiki/File:DisplayPort_Connector.svg
“Picture: DisplayPort Mini Adapter” https://en.wikipedia.org/wiki/Mini_DisplayPort#/media/File:Mini_DisplayPort_(connector).PNG
“Picture: Logo for AMD’s FreeSync” https://en.wikipedia.org/wiki/File:Logo_for_AMD%27s_FreeSync_technology.png
“Picture: Man running” https://unsplash.com/photos/Xcmfp0gBukI
“Picture: Copyright” https://pixabay.com/vectors/copyright-symbol-intellectual-30343/
“Picture: Binary code” https://pixabay.com/illustrations/binary-code-binary-binary-system-475664/
“Picture: HDR National Expansion Memorial” https://en.wikipedia.org/wiki/File:StLouisArchMultExpEV-4.72.JPG
“Picture: Archive compress” https://pixabay.com/vectors/compression-archiver-compress-149782/
“Picture: Tools” https://pixabay.com/photos/tools-hammer-nail-nails-drill-nut-15539/
“Picture: Differential signaling” https://commons.wikimedia.org/wiki/File:DiffSignaling.png
“Video Random numbers” https://pixabay.com/videos/random-numbers-led-digital-clock-33218/
“Picture: Display Port” https://commons.wikimedia.org/wiki/File:DisplayPort_plus_plus.svg
“Picture: DVI pinout” https://commons.wikimedia.org/wiki/File:DVI_pinout.svg
“Picture: HDMI connector” https://en.wikipedia.org/wiki/HDMI#/media/File:HDMI_Connector_Pinout.svg

Credits
Trainer: Austin Mason http://ITFreeTraining.com
Voice Talent: HP Lewis http://hplewis.com
Quality Assurance: Brett Batson http://www.pbb-proofreading.uk