Talk:Transcoding

This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Computing Low‑importance This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles Low This article has been rated as Low-importance on the project's importance scale.

With regards to audio/video:Transcoding usually refers to compressing files(such as MPEG) to a lower bitrate without changing formats,Encoding(also called Rendering) usually refers to changing formats(such as AVI to MPEG).

In a strict sense, the definition of this word has absolutely nothing to do with the lossy/losslessness of the conversion, and if anything the 'literal' definition would be more likely to imply a lossless conversion than a lossy one. Transcoding is, quite simply, transfering the same information from one encoding to another. It matters not whether the conversion induces loss, as long as it represents the same information (ie. it doesn't matter if it's bit-perfect, as long as it represents the same song). I don't think we should be perpetuating the current definition. Only the first line is even close to technically accurate. --Ktims 10:00, 28 May 2006 (UTC)[reply]

This should some how be merge with the article in video tracoding. There should also be more examples of software transcoder, and maybe link to the mpeg formats in some way. —The preceding

The process of converting CD audio, or DVD audio, for example, into a digital format (lossly or not), is generally called encoding. I've also seen it called transcoding. Any thoughts on whether a difference between "encode" and "transcode" actually exists as I've described, or am I just mixing up expressions for the same thing? Bernard S. Jansen (talk) 08:49, 4 January 2008 (UTC)[reply]

The article makes a distinction that "true transcoding" means "the bitstream format of one file is changed from one to another without its undergoing another complete decoding and encoding process" but there's no citation for this and -- if so -- then the rest of the article needs to be clarified I think to distinguish between "true transcoding" and "transcoding" vs "decoding/encoding" —Preceding unsigned comment added by 202.78.240.7 (talk) 01:51, 31 July 2009 (UTC)[reply]

as far as I understood it: the difference between transcoding and reencoding is that transcoding converts from one compressed format into another compressed format without decompressing the input, reencoding does the same but with decoding the material before encoding it. Since most tools that reencode like tranzcode/mencoder/ffmpeg do the decompressing internally (in RAM) most people confuse them with transcoders.

short:
transcoding = compressed -> compressed with no decompression in-between
encoding = uncompressed -> compressed
reencoding = compressed -> uncompressed -> compressed

Neither transcoding nor encoding require any form of compression. Encoding is simply the process of representing raw data using a code that conforms to some scheme that is agreed upon in advance, often in the form of a standard file format, although two people could easily devise their only encoding scheme (a simple substitution cipher would be an example of this).

One of the simplest examples is ASCII. Software encodes the information I type using the keyboard by representing the characters of the alphabet as numbers in a form that has been agreed in advance. The information is decoded by converting the number into some representation I can use, perhaps a pattern of pixels on a display. Software might transcode this data into another format such as UTF-16. Usually this requires decoding and then encoding in the new format although this may not always be the case, as in ASCII to UTF-16 where it is not necessary to know what each letter means because you can transcode it by simply taking the lower 7bits of the ASCII value and copying that into a 16bit area of memory. For image and video files things are usually not so simple and decoding and encoding would be necessary. Whether decoding and encoding are involved or not, it is still transcoding.

Notice how in the above example there was no compression used. Similarly for image files, the camera or scanner will take the raw sensor data and encode it in a format of some sort. It may be compressed or uncompressed, lossy or lossless it does not matter it still needs to be encoded in some way that is agreed upon in advance by anyone wanting to use the data. It could even be a proprietary format used internally in the camera, in which case the company that developed the proprietary format is the only one who has agreed to use it. If that proprietary format then gets converted into a standard format in order to export the photos, then it has been transcoded. Trevorw (talk) 05:04, 17 January 2015 (UTC)[reply]

Hello, techies out there....

May we please simplify the introductory paragraph and put some plain, easily understandable English up front and save the technical jargon (and concepts beyond a lay encyclopedia-user's grasp) for a second sentence or short follow-up paragraph there? *Please*.

I just ran into the word reading a Google article. I popped it into Wikipedia and still did not have a remote grasp of what "transcoding" was until I had read the entire thing (and initially was wildly confused and nonplused just reading the introduction, going backwards in understanding rather than forward: lay expression is only introduced later when discussing consumer product applications of the concept like music players and cell phones).

May we not begin the article by saying something to the effect that "transcoding" is the process of utilizing a universally understood intermediate code to allow devices employing disparite ones to communicate (or something to that effect, technically accurate in expression), then, once the reader understands what the concept represents, get into the tech talk of "lossy codecs", "bitstreams" PCM, and YUC, later.

I realize the intro may sound simple to one who is technically familiar with the concepts and terminology, and may even appear to do what is being suggested to the tech savvy, but as a layman I can assure you it does not.

Perhaps something like:

"Transcoding is the process of transforming one body of data into a format readable by a device employing another by utilizing a universally understood intermediate format. Digital media players and mobile phones often rely on such transcoding to display content generated by otherwise incompatible devices.

"Transcoding often but does not necessarily result in diminished file size and resolution or audio quality, which may or may not be problematic depending on the type of data being transcoded and the nature of the destination device. Diminished audio quality on a home audio system may not be desirable, for example, while a reduction in file size and resolution to the level of a mobile phone's capabilities may not result in a diminished experience for the end-user."

"[Get more technical here]...."

Anyone out there capable of tackling this? I'm just passing through, trying to get enlightened by Wikipedia while reading the morning news.

PS: I've bracketed "Transrating" as a concept that deserves but as yet does not have its own Wikipedia article. Maybe someone who knows enough about Transcoding to make the above suggested changes in its introduction can bat out a paragraph or two for a new Transrating page, that at the least could reference a reader back to the Transcoding page for a fuller fleshing out of a related technology. That would be great.Wikiuser100 (talk) 12:31, 28 August 2008 (UTC)[reply]

The article

have been removed.Dogru144 (talk) 00:05, 19 February 2010 (UTC)[reply]

The explanation of generational loss given in introduction is flawed in that whether the input is lossy or lossless has no bearing on whether generational loss occurs. The loss of data only ever occurs during the encoding into a lossy format - the decoding of the input produces no loss because that data was already lost when it was encoded originally.

If I take a lossy encoded JPEG and transcode it to a lossless format such as BMP I will lose no additional information that wasn't already lost when the raw source data was converted to the JPEG format. When viewed, the JPEG and BMP will be indistinguishable by any means and there will be no new compression artifacts.

I have therefore deleted reference to whether the input is lossy or lossless from the discussion of transcoding.

Trevorw (talk) 02:11, 16 January 2015 (UTC)[reply]

Recoding should not be merged with transcoding. We need a page "Recoding" that also lead to the biological (and synthetic biology) process of Recoding (translation at the ribosome). I don not know how to do that split techically. 82squaremetres (talk) 11:21, 27 November 2017 (UTC)[reply]

I agree and I changed several things to fix this: I redirected

Recoding (biology). Lambtron (talk) 21:16, 29 November 2017 (UTC)[reply

]

While I'm no expert in this, lossy to lossless transcoding will just copy the data with no quality loss, am I wrong? The quality loss should only happen when transcoding into a lossy format. Why would there be a quality loss in the audio? Wouldn't it just increase the size? No source is cited about this in the article and most of the discussion I've found in the Internet argues against it, so I'm doubting its accuracy.--31.223.101.97 (talk) 15:10, 2 November 2018 (UTC)[reply]

Indeed, that's not really a good explanation. I think what the author of the edit really meant was that the decoded audio from the lossy format loses quality compared to the original file. But the wording is just unfortunate. It should probably be changed.—J. M. (talk) 17:06, 2 November 2018 (UTC)[reply]