TL;DR: The minimum detectable stalling duration in video streaming is between approximately 80 and 200 milliseconds, for most viewers.

One of our customers recently asked about some possible video measurement artifacts: streams that ran into a lot of minor stalling events. That is, video rebuffering shortly, as indicated by the HTML5 video player’s event timeline. Their question was: do users notice these stallings at all? In particular when they’re as short as 100 ms, or less? Initially, we couldn’t provide an immediate answer, but we settled on a default filter to eliminate stallings of <100 ms from our statistics.

We wanted to go deeper and understand what really happens on a psychophysical level — do viewers even notice 100 ms of stalling? Are they more sensitive than that? Or does it really not matter?

To find out, we searched for academic studies, and we found some interesting results. Read on to discover what we learned about stalling detection thresholds, and the importance of knowing the context (content motion, scene context, and viewer attention) to decide whether to consider them or not.

Read more

The ISO/OSI model is the invisible backbone of our daily internet use. It’s an abstract framework that organizes the complex process of digital communication into seven distinct layers: Physical, Data Link, Network, Transport, Session, Presentation, and Application. However, the complexity of this stack leads to two important issues when it comes to measuring end-user experience:

1. Many network measurement tools focus on metrics from a single layer, providing only a partial view – measuring or optimizing one layer in isolation doesn’t guarantee a better overall user experience. In other words: problems experienced by users often stem from complex interactions between these layers.
2. The highest ISO/OSI layer still does not tell you what the user really experienced. While it gives you important information about what was transmitted on the network, knowing what happened on the screen is the most crucial aspect for the final Quality of Experience.

At AVEQ, our Surfmeter platform is built on a cross-layer measurement methodology. This means Surfmeter doesn’t just look at one piece of the puzzle; it analyzes how the different layers work together to deliver services to your end-users

And while over the course of time some layers have become less important, there is something new on the horizon: Layer 8 – the end-user layer. This is where the actual customer experience happens. In the current research discussions, Layer 8 is becoming increasingly talked about, even though there is no standardized definition of that yet.

Stakeholders across the entertainment and networking industries realize that this holistic approach is crucial for truly understanding and improving Quality of Experience (QoE). Let’s explore this top-most layer in conjunction with the ISO/OSI layers to see why this matters.

Read more

Networks can behave like humans: Sometimes they have a shiny great day and everything works like a charm – and sometimes they feel tired and are soooo slow. Usually when humans get tired, they need a rest from the hard work they did. Networks don’t have that luxury. Under stress, they greet us with loading spinners, rotating hourglasses or just a simple “the service you are calling is not available” message in the browser. How can you, as a network operator, make the right decisions and make networks more resilient against packet loss or jitter, inefficient routing and transmission delays. And how do you pre-emptively solve customer complaints? The answer is: continuous quality monitoring. For video, web browsing, or other OTT apps like conferencing services.

But all the best data is not useful if you can’t make sense of the numbers you’re seeing. So in this article, we want to go deeper and explain how you can understand the data from the monitoring itself. In particular, the Mean Opinion Score.

Read more