RLE - Revision history

imported>Ikskoks at 13:07, 25 July 2021

2021-07-25T13:07:22Z

← Older revision	Revision as of 13:07, 25 July 2021
(No difference)

imported>Ikskoks at 13:07, 25 July 2021

2021-07-25T13:07:22Z

← Older revision		Revision as of 13:07, 25 July 2021
Line 45:		Line 45:

	* The Riddle of Master Lu ( ''' [[The Riddle of Master Lu_SS\|*.ss]] ''' )		* The Riddle of Master Lu ( ''' [[The Riddle of Master Lu_SS\|*.ss]] ''' )


			<br><br>
			[[Category:Articles]]

92.50.144.206 at 07:13, 13 September 2011

2011-09-13T07:13:26Z

← Older revision	Revision as of 07:13, 13 September 2011
(No difference)

92.50.144.206 at 07:13, 13 September 2011

2011-09-13T07:13:26Z

← Older revision		Revision as of 07:13, 13 September 2011
Line 1:		Line 1:
	== General Description ==		== General Description ==

	RLE (Run Length Encoding) describes a compression technique which tries to take advantage of long "runs" of a single symbol in the input. Instead of then storing all repetitions, the symbol is just stored once, together with an information on the length of this run. Similary, long runs of symbols without repetitions are stored literally.~~<br>~~		RLE (Run Length Encoding) describes a compression technique which tries to take advantage of long "runs" of a single symbol in the input. Instead of then storing all repetitions, the symbol is just stored once, together with an information on the length of this run. Similary, long runs of symbols without repetitions are stored literally.

	The advantage of RLE are obviously an easy an fast compression/decompression technique and good compression ratios on uniform inputs (e. g. images with a single background colour in large areas). However, on more complicated inputs RLE will often produce outputs that are even larger than the input.		The advantage of RLE are obviously an easy an fast compression/decompression technique and good compression ratios on uniform inputs (e. g. images with a single background colour in large areas). However, on more complicated inputs RLE will often produce outputs that are even larger than the input.

Line 9:		Line 10:
	A file processed a "standard implementation" of RLE can best be recognized when an uncompressed output file is known:		A file processed a "standard implementation" of RLE can best be recognized when an uncompressed output file is known:

	* If the output file is quite uniform, look out for larger areas of<tt> 0xff </tt> with one other byte in between.		* If the output file is quite uniform, look out for larger areas of <tt>0xff</tt> with one other byte in between.
	* If the output file is very non-uniform, look out for frequent occurrences of<tt> 0x00 </tt> followed by a byte whose value matches the distance to the next<tt> 0x00 </tt>.		* If the output file is very non-uniform, look out for frequent occurrences of <tt>0x00</tt> followed by a byte whose value matches the distance to the next <tt>0x00</tt>.


Line 17:		Line 18:
	When decompressing data, the basic task is to differentiate between literal and compressed data and reproduce the correct number of compressed characters. The pseudo code below shows a possible RLE decompression routine. Please note that the RLE method is quite widespread and has often been altered by developers to best match the type of data they want to compress. Thus, the below example just illustrates the general method of decompression, but has been used quite a few times in this variation:		When decompressing data, the basic task is to differentiate between literal and compressed data and reproduce the correct number of compressed characters. The pseudo code below shows a possible RLE decompression routine. Please note that the RLE method is quite widespread and has often been altered by developers to best match the type of data they want to compress. Thus, the below example just illustrates the general method of decompression, but has been used quite a few times in this variation:

	<br>		<syntaxhighlight lang=pascal>
	~~<tt>~~		while InputPos < InputSize do
	~~<b>~~while~~</b>~~ InputPos < InputSize ~~<b>~~do~~</b><br>~~		begin
	~~<b>~~begin~~</b><br>~~		RunLength := Input[InputPos++]
	: RunLength := Input[InputPos++]~~<br>~~		if RunLength = 0 then
	~~: <b>~~if~~</b>~~ RunLength = 0 ~~<b>~~then~~</b><br>~~		begin
	~~: <b>~~begin~~</b><br>~~		Count := Input[InputPos++]
	:: Count := Input[InputPos++]~~<br>~~		for i := 1 to Count do
	~~:: <b>~~for~~</b>~~ i := 1 ~~<b>~~to~~</b>~~ Count ~~<b>~~do~~</b><br>~~		Output[OutPos++] := Input[InPos++]
	~~:::~~ Output[OutPos++] := Input[InPos++]~~<br>~~		end
	~~: <b>~~end~~</b><br>~~		else
	~~: <b>~~else~~</b><br>~~		begin
	~~: <b>~~begin~~</b><br>~~		Symbol := Input[InputPos++]
	:: Symbol := Input[InputPos++]~~<br>~~		for i := 1 to RunLength do
	~~:: <b>~~for~~</b>~~ i := 1 ~~<b>~~to~~</b>~~ RunLength ~~<b>~~do~~</b><br>~~		Output[OutPos++] := Symbol
	~~:::~~ Output[OutPos++] := Symbol~~<br>~~		end
	~~: <b>~~end~~</b><br>~~		end
	~~<b>~~end~~</b><br>~~		</syntaxhighlight>
	</~~tt>~~
	~~<br~~>

	== Further Information ==		== Further Information ==

imported>Dinoguy1000: /* Games */

2006-08-21T02:24:37Z

Games

← Older revision	Revision as of 02:24, 21 August 2006
(No difference)

imported>Dinoguy1000: /* Games */

2006-08-21T02:24:37Z

Games

← Older revision		Revision as of 02:24, 21 August 2006
Line 45:		Line 45:
	== Games ==		== Games ==

	* [[The Riddle of Master Lu\|The Riddle of Master Lu]]		* The Riddle of Master Lu ( ''' [[The Riddle of Master Lu_SS\|*.ss]] ''' )

imported>DenizOezmen at 18:15, 11 June 2005

2005-06-11T18:15:49Z

← Older revision	Revision as of 18:15, 11 June 2005
(No difference)

imported>DenizOezmen at 18:15, 11 June 2005

2005-06-11T18:15:49Z

New page

== General Description ==

RLE (Run Length Encoding) describes a compression technique which tries to take advantage of long "runs" of a single symbol in the input. Instead of then storing all repetitions, the symbol is just stored once, together with an information on the length of this run. Similary, long runs of symbols without repetitions are stored literally. 
The advantage of RLE are obviously an easy an fast compression/decompression technique and good compression ratios on uniform inputs (e. g. images with a single background colour in large areas). However, on more complicated inputs RLE will often produce outputs that are even larger than the input.

== How to Recognize an RLE-compressed File ==

A file processed a "standard implementation" of RLE can best be recognized when an uncompressed output file is known:

* If the output file is quite uniform, look out for larger areas of<tt> 0xff </tt> with one other byte in between.
* If the output file is very non-uniform, look out for frequent occurrences of<tt> 0x00 </tt> followed by a byte whose value matches the distance to the next<tt> 0x00 </tt>.

== Decompression ==

When decompressing data, the basic task is to differentiate between literal and compressed data and reproduce the correct number of compressed characters. The pseudo code below shows a possible RLE decompression routine. Please note that the RLE method is quite widespread and has often been altered by developers to best match the type of data they want to compress. Thus, the below example just illustrates the general method of decompression, but has been used quite a few times in this variation:

 
<tt>
while InputPos < InputSize do 
begin 
: RunLength := Input[InputPos++] 
: if RunLength = 0 then 
: begin 
:: Count := Input[InputPos++] 
:: for i := 1 to Count do 
::: Output[OutPos++] := Input[InPos++] 
: end 
: else 
: begin 
:: Symbol := Input[InputPos++] 
:: for i := 1 to RunLength do 
::: Output[OutPos++] := Symbol 
: end 
end 
</tt>
 

== Further Information ==

More extensive information on this technique can easily be found on the web. Starting points might be [http://www.prepressure.com/techno/compressionrle.htm this page] showing a graphical explanation of the compression method using a more theoretically generalized approach than the sample implementation above. There are hundreds of further pages (too much to list them all here) describing a multitude of RLE variants.

== Games ==

* [[The Riddle of Master Lu|The Riddle of Master Lu]]