Analogue Video Capture Details - Terms and Concepts Explained

Resolution

Computer images are made from picture elements called pixels. Each pixel is a rectangle (usually square) which has its own colour. Generally, the more pixels the image has, the better quality it looks to the eye but the more space it takes up in the computer's memory.

Resolution is the width and height of an image expressed in the number of pixels from left-right and top-bottom. For example, a resolution of 320 x 240 means the image has 320 pixels from left to right and 240 pixels from top to bottom.

Resolution is particularly important when it comes to movie files because it has the most influence on how big the movie file is. Also when capturing video, depending on the power of the PC used, we may be limited to capturing below certain resolutions.

To see what effect resolution has on image quality and file size click here.

Colour Format

For each pixel within each frame, the colour format determines how much memory is required to store the pixel's brightness and colour. Similar to resolution, the colour format of an image determines its quality and also determines how much hard disk space is required to store it. 

The following table shows some common colour formats and their typical uses. 

To see some example images stored in different colour formats click here.

Frame Rate

The frame rate determines the number of images that are stored and displayed per second. The greater the frame rate, the smoother the motion. Generally, you will want to match the capture frame rate to that of the source. You may however wish to reduce the frame rate to:

• reduce the captured file size (since the file size is approximately proportional to the frame rate)
• reduce the load on the computer's processor to stop dropped frames

The table below summarises some common frame rates:

Why 29.97 Frames Per Second?

The NTSC video standard actually runs at 29.97 frames per second - not 30 frames per second as is commonly quoted. For short captures of a couple of minutes or so, using 30 frames per second will be satisfactory. For longer captures however, capturing at 30 frames per second from an NTSC will cause audio / video synchronization problems.

We recommend that when capturing long clips from an NTSC source, that the frame rate is set to 29.97 frame per second.

Video Compression

Video compression is important for two main reasons:

• It can vastly reduce the captured movie file size.
• It reduces the data rate to the hard disk.

As with resolution, colour format and frame rate, choosing the right compression settings is a trade off between image quality and file size. Click here to see some examples of a compressor used with different quality settings and the resultant file size.

Note that some video compressors are not intended for real time compression. This means that if you try and capture video and compress using one of these compressors, the computer will not be able to keep up with the frames coming from the capture card and dropped frames will occur.

For general use and good compatibility we recommend using an Indeo version 5 compressor. Other compressors however may be more suited to high quality video or high data rates.

Audio Settings

If you wish to capture audio along with your video, you have a choice of audio settings to use. Click here to see what options are available and what implications these have.

Data Rate

Data rate is a very important concept when it comes to video capture. Understanding data rate helps us understand how the speed of our PC and hard disk limits the quality the we can capture at.

The diagram below illustrates the different data rates at different stages of the capture process.

The highest data rate occurs between the video capture card and the video compressor. This is where the full, uncompressed frames are captured by the hardware and send to the video compressor. The video compressor takes each frame and applies its compression algorithm to reduce the image size in bytes and hence reduce the output data rate. This video compression all occurs in the computer's memory and does not depend on the hard disk. It is therefore the speed of your PCs processor and memory, and the video compressor used that dictates the maximum data rate at this stage that can be used. Remember that increasing the image resolution, increasing the frame rate and changing the colour format can all increase the data rate. If the PC cannot cope with the data rate, dropped frames will occur.

Note that some video capture cards have the video compressor built in and use hardware to perform the compression. In this case, the above paragraph is not applicable.

The output from the video compressor is added to the audio stream to form the resultant data stream which is written to a file on the hard disk. This final data rate is therefore dependent on:

• the audio format used, and
• the video data rate, which is dependent on the video compressor type and settings, image resolution, colour format and frame rate.

Note that the video data rate usually dominates the final data rate.

The video and audio data is then written to the hard disk at the final data rate. If the hard disk cannot write the data fast enough, something has to give and dropped frames occur.

Dropped Frames

Dropped frames are missing or skipped frames the can occur when capturing video or playing back video. They are visible in a video when playing back a movie file and the motion stops or jerks for a moment and then continues.

Dropped frames can occur on playback when the computer cannot read the data off the hard disk fast enough or if the PC cannot decompress the video fast enough. They are however, much more common and troublesome when they occur during capture.

If dropped frames occur during capture it is usually due to:

• The data rate from the capture card being too great for the video compressor, or
• The final data rate is too high for the hard disk to sustain.

There could be other reasons too, like for example if some other program (virus checker or screensaver) is running at the same time as capturing.

To solve problem 1, there are several steps that can be taken:

• Reduce the capture resolution
• Reduce the frame rate
• Choose a capture image format that uses less bits per pixel (see colour format table)
• Choose a different compressor

To solve problem 2, you can:

• Reduce the quality setting of the compressor to reduce the output data rate
• Reduce the audio quality settings
• Make sure your hard disk is optimized for capture. Click here to see how to do this.

The 2GB AVI File Limit

The 2GB AVI file limit applies to AVI 1.0 format files (the most common AVI format). It occurs because the file format uses 32 bit integers to store offsets within the file and this limits the file to about 2 gigabytes. To overcome this problem, the AVI 2.0 format was created which uses 64 bit integer offsets. AVI 2.0 files can therefore be virtually unlimited in size, there are however two points to bear in mind when using AVI 2.0 format files:

• Not all editing software supports AVI 2.0 files. 
• There is a 4Gb limit on all files with the FAT32 file system on Windows 95 and Windows 98. So unless you are using NTFS on Windows NT or Windows 2000, you are limited to 4Gb files anyway.

So some advice would be to stick to using AVI 1.0 files and capture in sections shorter than 2GB and only use AVI 2.0 if:

• You absolutely have to capture more than 2GB in one continuous section, and
• You have an editing program which supports the AVI 2.0 standard (such as Pure Motion's EditStudio).