What is DASH?
DASH – short for “Dynamic Adaptive Streaming over HTTP” – is a streaming protocol that uses adaptive bitrate streaming for distributing high-quality media content via the Internet. Similar to HLS, DASH splits a large video file into many small sections. These usually are slightly shorter than the HLS standard of 10 seconds, namely only 2 – 4 seconds. So a fast change is possible when the internet connection gets lower or better. DASH is more modern than HLS but is not compatible with iOS devices.
HLS vs. DASH
HLS and DASH have a few commonalities. They are both entirely based on HTTP, and both split the video data into segments, which enables adaptive bitrate streaming. But there are a few differences:
Encoding: MPEG-DASH allows the use of any encoding standard. HLS, on the other hand, requires the use of a specific one.
Device support: HLS is the only format supported by Apple devices. iPhones, MacBooks, and other Apple products cannot play videos delivered via DASH.
Segmentation: DASH typically segments video for delivery into smaller segments than HLS. The default segment length for HLS is 10 seconds, DASH segments generally are 2 to 4 seconds long. This means that DASH allows faster switching between quality levels to adapt more quickly to network conditions.
Standardization: DASH is an international standard. HLS was developed by Apple and was not published as an international standard, although it has broad support.
HTML5 support: HLS is automatically supported by HTML5, but DASH is not. This means that some browsers or apps cannot play MPEG-DASH video streams, even on non-Apple devices.
What is LL-DASH?
Low Latency-Dash (LL-Dash) is an extension of DASH and was first introduced in 2017. It is often referred to as CMAF-CTE (Chunked Transfer Encoding) in the streaming industry. This designation is misleading, as CMAF-CTE only describes DASH’s method to add low latency capabilities.
This method describes a procedure in which the video segments are divided into even smaller units, so-called chunks. As these chunks are independent of each other, the player does not have to load the entire segment but can also load individual chunks. This results in much higher flexibility and lower latency. In addition, the player no longer has to cache entire segments but can store the much smaller chunks and delete them as soon as they are no longer needed.