Frequently asked questions
Why Buy Testvids?
Q: I'm a broadcaster / user of encoders - why would I buy Tvids and not just use video I have?
A: There are several reasons:
TestVid clips have been carefully designed and compiled by video codec experts to be an exhaustive and really tough test of encoders but provided in a relatively short set of videos (each set takes less than an hour to run). To get the equivalent coverage with 'normal' programs you would have to watch each program to document all the features, then find the programs that gave the equivalent coverage, then play them all - all of which would take way longer.
If you use compressed video as test material for an encoder, it has to be decoded first - and the decoder can of course affect the video quality. TestVid clips are provided uncompressed, so they go straight into your encoder.
Supposing you want to check if the latest 'upgrade' to your current encoders and decoders really is an upgrade - or in fact works fine for most video but is a real dog with some video types - then you need a clip set that gives great test coverage and that you can plug in and run where the decoder cannot affect what happens (as is the case with the uncompressed TestVid clips)
Likewise to compare encoders before buying them; the TestVid clips sets provide the best test set for checking out the encoder.
And as clips are provided with audio, you can check to make sure that the encoder keeps the audio in sync with the video.
Or supposing you normally encode SD or 720p but want to try 1080i or 1080p or 2K or higher bit depths? Or similarly you want to see how your encoder does when generating parallel low resolution content for your website? TestVid offers all these different video types.
Perhaps you think that your encoder has a problem with say grainy night-time video where there are highlights. How quickly could you find video with these features (or any other specific features)? It only takes a few seconds with TestVid clips to search the user manual PDF for these features, to find the relevant clips.
Q: I'm a codec developer - why would I buy TestVid clips and not just use the occasional free test clips that are available?
A: There are several reasons:
Test coverage 1: all the free clips available are pretty short and relatively few in number. It is also vital to ensure that you are testing the full range of video types, lighting types, subject types, movement types etc. at all the different resolutions that your encoder is designed to handle. Using just the free test clips will not give the same test coverage. (And generally do not have audio.)
Test coverage 2: it is possible to use a bunch of standard DVDs (and we suggest you do this as an additional test) but these will not be documented for test purposes and are unlikely to cover all the resolutions, frame rates, and bit depths you require. Likewise these will have to be decoded first, and take a long time to play through to try all the different subject types etc - running the TestVid clip set through your encoder will be much quicker.
During development, for example you want to check that the motion estimation works fine, and it is much easier to start with slow-moving defined motion as in the T2V006 Synthetic clip set to try this out.
You need to be sure that your encoder does not slow down when the video is complex, to be sure that the encoder keeps the audio in sync with the video. TestVid clips are provided with associated audio, allowing you to do this easily.
Or supposing you need to test SD and lower resolutions as well as 720p and 1080i and 1080p and 2K, and perhaps higher bit depths (such as up to 14-bit)? TestVid offers all these different video types.
Q: Why not just shoot my own video?
A: Time and money are the short answers.
Yes, you could go out and do all the filming we did, then the editing and documentation. daytime, evening, night time, spring, summer, winter, New York, San Francisco, Las Vegas, London, Barcelona, Munich, Bruges...
Spend some days planning where to film, then planning each shot to get good coverage of lots of different subjects, global and subject motion, colors, features, etc.
Then hire the cameras, and travel to the locations.
Then spend days and days editing the clips to get a good representative set and range of clips and features as per our list (e.g. in T2V001 USA East ). Then re-size them to the sizes you require, then write some software to make letterboxed versions..
Then document them all so you can easily select them.
In all, it took 6-8 man weeks per clip set, plus the time to write & test the software applications provided. So without doubt buying TestVid clips is a lot more cost-effective - and you can get them right now.
Q: What test coverage do TestVid clips offer?
A: How well does your encoder cope with grainy night-time video? Or video where there is a lot of fast movement / scene changes? Or water, or moire patterns, or perhaps monochromatic blue sky? Or will your encoder have a problem where video goes rapidly in and out of focus? Or where there is lots of text, or very bright colors, or low contrast, or many areas that look very similar - will your encoder pick the right ones to track the motion of? TestVid allows you to instantly have a comprehensive test set for your encoder - so you know it will cope with anything, efficiently and robustly. (And the PDF documentation allows you to quickly find particular clips which have features that are 'difficult' for your encoder.)
Typically each set comprises a range of subjects, motion, colors, light levels designed to test and stress video encoders by providing a varied set of conditions:
Subject types such as people, traffic, buildings, sky, water, trees, text
Movement types such as panning, zooming in/out, tracking, hand-held camera
Subject motion such as into, out of or across the picture, in front of and partially behind objects
Lighting conditions, from bright sunlight, dull daylight, shaded areas, indoors
Varying camera properties such as depth of field, in/out-of-focus
Hard to encode items such as reflections, fine lines, patterns, round objects
And with sound associated with some clips
In many cases the video is harder to encode than might normally be expected, as the lighting conditions are not ideal or there is significant camera movement, or the focus varies. These features are deliberately used as they often cause the most difficulty to video encoders and represent the worst case that the encoder should encounter in 'normal / real' use.
Q: Why buy the TestVid clips instead of other video clips available, including free ones?
A: In short, TestVid clips give you far better and wider test coverage than anything else available - at a better price. Some things to consider:
Quantity: not always a good indicator, but in this case it is - how many minutes of video and different clips do you get? TestVid clip sets include a large number of videos - typically ~180 or more - one set even has 840 clips! (How long would it take you to make these?)
Length: how long is each clip? It's fine to have some clips that just last 10 seconds or less, but for a proper test you need longer clips (e.g. to see if the encoder memory suffers overflow)
Quality: were the clips filmed in HD or SD?
Features: do the clips cover the range of brightness, contrast, colors, content, global and local movement. And do the clips include things such as transitions, fade-ins, fade-outs - which your encoder will have to deal with?
Audio: is there any audio provided? OK, you are not buying these specifically for the audio (probably), but how often do you get video with no sound? After all, you want to be able to check that when your encoder finishes with the video, that the audio will still be in sync. And if you are not provided with the right audio, how can you do this? (Other than spend time making your own...)
Q: What about T2V006 Synthetic / T2V016 Synth3060? What's the point in 'synthetic' video clips?
A: Certainly, 'real' video is the ultimate test, but when doing encoder comparison or development it can be very useful to have a set of clips where there is defined and regular movement, e.g. to check out motion vectors / motion estimation. Or defined shapes, or encoding of colors such as on a test chart / color bars, or text at multiple / varying sizes. The T2V006 Synthetic product page has more information.