How Burn-in works

How Burn-in works

You will probably find it helpful if you have a little understanding of how the Burn-in Time Stamp and Burn-in Metadata tools work. In fact it is quite straightforward and uses the same techniques as most video editing software.


To keep things simple we will take a quick look at how photo editing software works when you edit the still photo images that were taken by your digital camera. DVMP Pro does not work with still photo images, so this is just an analogy - please bear with us, it will become clear later.


Please note that we have over-simplified some of the details to make it easier for beginners to understand.


Photo Editing - an Analogy


As you'll probably know, a photographic image that you view on your computer screen is comprised of lots of pixels, each with its own numeric color and brightness values. If each of the pixels is stored separately (with their separate color and brightness values) in a file, then we say that the photographic image is stored "uncompressed" within the file. A Microsoft Bitmap (BMP) file is an example of an uncompressed image file where all of the pixels values are stored separately as a long list inside the file. This means that an uncompressed file can be very large. An image that contains twice as many pixels as another image will be approximately twice the size.


When image editing software opens and reads a digital image file, it must store all of the pixel color and brightness values separately in its working memory. This then makes it very easy for the image to be modified within the editing software. For example if you told the software to brighten the image then it would simply go through each of the pixel brightness values held in its memory and change them to a greater brightness value.


An uncompressed file can be very easily loaded into the image editing software because each of the pixel values is already stored separately, so each pixel value just needs to be "copied" from the file into the software's working memory.


However, uncompressed files are very large, and camera storage media needs to hold lots of digital photograph files. Therefore, the images are usually stored in a compressed format which takes up less space than uncompressed, so the files are smaller.


Compression involves looking through all of the pixel values in the image for repeating occurrences of values - for example a large area of similar color or a checkerboard floor will contain lots of repeated pixel values. The pixels in these repeating areas are grouped together and a special code value is used to represent the group of pixels. Then each time the same group of pixels occurs in the image, the group can be represented by just the special code value instead of all of the color and brightness values for every pixel in the group. Consequently this encoded image will take up a lot less space and the image file will therefore be a lot smaller than the original uncompressed image. This means that our camera can store more of these compressed image files on its media.


However, these compressed files can not be modified directly in image editing software because we have lost sight of the color and brightness values for each pixel - many of them have been replaced by the special codes. Remember that image editing software needs the image in its working memory to be in the uncompressed state, with all of the pixel values stored separately. So when you open a compressed file, the software must decompress (or uncompress) the image before it can work with it in its working memory. This involves replacing each of the special codes with all the original pixel color and brightness values - this is the opposite procedure to when we created the compressed files.


Hence we have image compression to produce smaller files, and decompression to restore the original image in its uncompressed state.


Of course there are lots of different methods of searching and grouping repeating pixel values, and of creating the special codes. Each method is referred to as a "compression scheme".


The compression of the image is usually carried out by a "compressor" (or "encoder") - this is simply a chunk of software that contains the method used to compress the image, and it actually performs the compression. Similarly the decompression is carried out by a "decompressor" (or "decoder"). The image editing software simply runs the compressor or decompressor whenever it needs to compress the image (when saving the file) or decompress the image (when opening/reading the file).


Obviously, the decompressor must be using the same compression scheme (method) as the compressor otherwise the special codes will not be correctly converted back to the original pixel values. Thus the compressor (or encoder) and decompressor (or decoder) are usually grouped together as a matching pair, and are referred to as a "codec" (i.e. encoder-decoder). In the case of photo editing software these codecs are usually built into the software, but it is also sometimes possible to add extra codecs, perhaps provided by a completely different software supplier.


The camera itself will have a built-in compressor that allows it to store lots of image files on its media. When one of these compressed image files is opened in image editing software on your computer then it follows that the editing software must have a matched decompressor (that uses the same compression scheme) to recreate the uncompressed image in its working memory.


OK, you may say, I can see that a codec is very useful - it can make image files smaller. But why do we need lots of different codecs?


Well, each codec contains a different method of compressing the image (the method is usually referred to as the "compression scheme" or the "algorithm"). Some algorithms produce better image quality, some produce smaller files, and some take less time to compress or decompress the image. You can therefore choose whichever codec best suits your own requirements.


There are two types of compression - "lossless" and "lossy". So a compression scheme may be either lossless or lossy.


In lossless compression, all the detail of the original image is preserved exactly. So when the image is compressed and then decompressed, you end up with pixel values that are exactly the same as in the original image. TIFF is an example of a lossless codec - the compressed image is stored in files with the file name extension .tif.


In lossy compression, some of the very fine image detail is thrown-away. The thinking behind this is that the eye may not miss very fine detail in the image. Because lots of very fine detail does not have to be stored, then the compressed image can be very much smaller than for lossless compression. Often a lossy codec may have a quality setting which allows you to tell it how much detail you are happy to lose - the more detail you are willing to sacrifice the smaller the compressed image (and therefore the file) will be. JPEG is an example of lossy compression - the compressed image is stored in files with the file name extension .jpg.


Burning-in Video


Having taken the time to read the above analogy, you'll be pleased to know that the world of video is very similar. Video is compressed and decompressed using video codecs, and video effects can only be added to uncompressed video frames. So video frames are decompressed, then effects are added to them, then they are compressed again. Most video editing software works this way, not just DVMP Pro.


Because video is effectively a rapid succession of images (i.e. frames), it takes up a lot more space than a single photo image. Uncompressed video can therefore take up a huge amount of disk space, so it is more common to use compression for video.


To burn-in text onto video, the video data is decompressed by being passed through a video "Decoder" to produce uncompressed video frames in working memory. The text is then added to the uncompressed frames which are then compressed using a "Compressor". The compressed data is then stored in the output file. If the AVI output type is selected in DVMP Pro's options, it will be an AVI file; or if you choose the WMV output type it will be a WMV file. If you choose the AVI output type, you can also choose what compression scheme is used by selecting an AVI Video Compressor from any of those that are already present on your computer or any that you choose to install from other (third-party) software suppliers. The WMV output type always uses the VC-1 compressor.


This "decompression-then-compression" cycle is not just something that DVMP Pro does - all video editing and processing software works this way (and photo-editing software too). Any type of video effect can only be applied to video frames after they have been decompressed in memory by a decoder. Some camcorder formats require a lot of processing power to do the decompression - especially AVCHD which requires a fairly powerful PC.


The speed of the Burn-in tool can vary widely, depending on your computer hardware and which Decoder and AVI Video Compressor is selected in DVMP Pro options. Some factors that affect the speed of burn-in are:


How powerful the computer/processor is
The speed of the Video Decoder that you are using which is set in Tools > Options > Video Decoders
If you have selected the AVI output type, the speed of the AVI Video Compressor that you selected (and perhaps any of the compressor's own options that you may have set - e.g. multithreading) which is set in Tools > Options > Burn-in > Output
If the input files or the burnt-in output files are on an external hard drive, the slow USB2 interface is a bottleneck which might slow everything down. So make sure that the external hard drive and the computer's USB port it is connected to are both rated at USB3. If just one of these is USB2 then the whole connection will run at the slow USB2 speed.
If you have chosen an AVI Video Compressor that produces very large output files, the speed of the hard drive where the output files are created may be a bottleneck due to the large amount of data that must be written to it


For a particular computer, the factor that most affects the speed of burn-in is usually the choice of AVI Video Compressor; the Video Decoder has less influence.


There is a three way trade-off between picture quality, file size, and time taken to compress - it's an unavoidable principle of any type of video processing. DVMP Pro does not tie you into a particular AVI video compressor, so you are free to choose whatever AVI video compressor works best for your own particular requirements.


For example, someone working in video production wanting utmost quality would typically use "Uncompressed" (the default when DVMP Pro is installed) or more likely "Lagarith" to produce lossless quality burnt-in AVI files - this produces quality effectively the same as the original but with very large files (Lagarith less so). These large files are then used as inputs to their video editing program (giving top quality to begin editing with) or DVD authoring program. As these files are very large it is difficult or impossible to play them in real time on a PC without skipping or freezing due to the sheer amount of data that must pass through the system - but that's OK because that is not what they are intended for; they are intended as top quality inputs to editing programs, transcoders or DVD authoring programs.


When you choose "Uncompressed", you are effectively telling DVMP Pro to skip the compression phase of the "decompression-then-compression" cycle mentioned above. The decompressed video frames are written directly to the output AVI file so you avoid the processing effort of compressing them first. This reduces the load on the PC's processor but the files are extremely large because they contain uncompressed data, and writing this large amount of data to the output hard drive is effectively a bottleneck which slows the speed of burn-in. The "Lagarith" compressor produces almost identical quality to "Uncompressed" because it is a "lossless" compressor, and the files are significantly smaller (though still quite big); but unlike "Uncompressed", Lagarith requires some processing power to do the compression, but because the output files are smaller than Uncompressed it is less of a bottleneck.


On the other hand, if you wanted smaller files that would play directly comfortably on a PC you may want instead to try as an AVI video compressor "x264 vfw", DivX, Xvid, or an MJPEG-based compressor. You could also try out the WMV (VC-1) output type which you can select in DVMP Pro's Tools > Options > Burn-in > Output options.


It's also worth trying the free "Cineform" AVI video compressor. This gives near-lossless quality but the compression techniques are so good that it is very fast and produces relatively small files. You will understand from the above discussion that this is an ideal combination of properties that few other AVI video compressors can equal, and for this reason is used a lot in professional video production. The Cineform codec is free, but it is only available if you install the free "GoPro Cineform Studio Software" - the Cineform codec then becomes available for use in other software, including as a selectable AVI video compressor in DVMP Pro.


You can find more recommendations of AVI Video Compressors to choose from in the AVI Video Compressors topic.


You can also select "MPG" or "MPG for DVD" as the output file type. If either of these is chosen, the video frames are passed through an MPEG-2 video compressor that is built into some versions of Windows, and stored in a .mpg output file.


The "MPG for DVD" output type is the same as "MPG" but it constrains the video dimensions and frame rate (and other low-level data) to conform to the video DVD specification - in theory this should make it quicker to use in DVD authoring software because it is already in a suitable compressed format for video DVDs. Unfortunately, a lot of DVD authoring software insists on always re-compressing (i.e. decompressing then compressing again) all video files when it makes a DVD even if the file is already DVD compliant. Consequently, if your DVD authoring software always re-compresses video files then there is little point in using "MPG for DVD" as the output type in DVMP Pro when the AVI output type may be faster to burn-in.



IMPORTANT: The "MPG" and "MPG for DVD" output types are only available if you are using Windows Vista or Windows 7. It is also available if you are using Windows 8 or 8.1 but ONLY if you also have installed the "Windows Media Center" feature pack. Otherwise these two output types can not be selected. For more details please see Options - Burn-in Output


All of the above choices will affect the time taken to burn-in a file.


For example, if you just want your files to be inputs to a DVD authoring program, and your DVD authoring software always re-compresses its input files (see above), then you could opt for the AVI output type with the "Lagarith" AVI video compressor. The DVD authoring program will be doing some compression to the DVD native format anyway, so there is not much point in using a time consuming compressor in DVMP Pro just to produce smaller files for input to the DVD authoring program. This may of course mean that you need lots of disk space but with multiple-TB disks now very affordable (and getting cheaper), this is probably a better option than waiting for a heavily processor intensive compressor to complete. If you decide to go for a large external hard drive, make sure that both the hard drive and your computer have USB3 ports which are faster than USB2. A better alternative to Lagarith that you might want to consider is Cineform which will produce smaller files and consequently is likely to be faster to complete.


Also, remember that although the AVI file format is one of the most versatile and ubiquitous video file formats, its method of storing the Display Aspect Ratio (16:9 or 4:3) of the video frames is not supported by some video editing software. DVMP Pro does store the correct display aspect ratio in the burned-in AVI files. However, if these files are imported into video editing or DVD authoring software that does not support display aspect ratio in AVI files, the software may just make a guess at it - and it might get it wrong! So in some cases a burned-in AVI file may be incorrectly interpreted as 4:3 (or square pixels) by video editing packages and other programs - this should not happen if your camcorder shoots in "square pixel" resolutions such as 1920x1080 or 1280x720. But if your camcorder shoots in 1440x1080 (non-square pixels) you may have to manually set the display aspect ratio to 16:9 within your editing/authoring program - for example, in Adobe Premiere the "Interpret footage" command does just that. Alternatively the editing program may allow you to set "rules" so that imported 1440x1080 files are always interpreted as 16:9.


The display aspect ratio of burned-in WMV files should always be interpreted correctly, but the file type is less versatile, less well supported by other programs, and is quite processor intensive and hence tends to be slower to encode/compress.


See Also:

Time Stamping Video

Burning-in Metadata

Burn-in Time Stamp

Burn-in Metadata

Options - Burn-in