CBR File Documentation

Summary

A Comic Book RAR Archive file is a digital comic: page images (JPG or PNG) packed into a RAR archive and renamed .cbr, which a comic reader shows one page at a time in filename order. It has no format of its own — it is literally a RAR file, so it starts with 52 61 72 21 1A 07 (“Rar!”). Read it with YACReader; the common goal is converting it to PDF or the more open CBZ.

Technical details

FeatureValue
Full nameComic Book RAR Archive
File extension.cbr
MIME typeapplication/vnd.comicbook-rar
Format typeRAR archive of ordered page images, renamed
DeveloperCommunity convention (CDisplay, ~2003); RAR by Eugene Roshal / RARLAB
Introduced~2003 (the CDisplay reader popularised .cbr/.cbz)
Open standardNo — RAR is a proprietary compression format
Container / base formatRAR archive (extract with any RAR-capable tool)
Byte orderLittle-endian (RAR header fields)
Magic number (RAR4)52 61 72 21 1A 07 00 (7 bytes, “Rar!\x1A\x07\x00”)
Magic number (RAR5)52 61 72 21 1A 07 01 00 (8 bytes)
Page imagesJPG / PNG / WebP, sorted by filename = page order
Optional metadataComicInfo.xml (series, issue, writer, page count)
Open siblingCBZ = the same idea using a ZIP archive
Other siblings.cb7 (7-Zip), .cbt (TAR), .cba (ACE)
Related extensions.cbz, .rar, .zip, .epub, .pdf
Referenceen.wikipedia.org/wiki/Comic_book_archive
File signature (magic bytes)
52 61 72 21 1A 07 00

Offset 0. A .cbr is a RAR archive with the extension changed, so it carries the RAR signature Rar!\x1A\x07. The seventh byte gives the version: ...1A 07 00 is a 7-byte RAR4 signature, while RAR5 uses an 8-byte marker 52 61 72 21 1A 07 01 00. This is why a plain ZIP tool cannot open a .cbr — it is not ZIP. The sibling CBZ is a ZIP archive and starts with 50 4B 03 04 (“PK”) instead. The extension letter (R, Z, 7, T) tells you which archive type is underneath.

What is a CBR file?

CBR stands for Comic Book RAR, and it is the most common member of the comic book archive family. There is no special comic format under the hood: a .cbr is an ordinary RAR archive containing a set of sequentially named image files — page01.jpg, page02.jpg, and so on — with the extension changed from .rar to .cbr so that a comic reader knows to treat it as a paged comic rather than a generic archive. The convention was popularised around 2003 by the CDisplay reader on Windows.

The reader’s job is simple: open the archive, list the image entries, sort them by filename, and display them one page at a time with page-fitting, double-page spreads, bookmarks and (for manga) right-to-left reading. The comic itself is just pictures; the archive is just a wrapper. Everything distinctive about CBR comes from that wrapper being RAR specifically, which is what the rest of this article is about — the RAR signature and header, why only some tools can read it, and why the ZIP-based CBZ is the format the community actually recommends creating.

The comic archive family: same idea, five containers

Every “CBx” format works identically and differs only in which archive format supplies the container. The last letter of the extension is a direct label for the compression underneath:

ExtensionArchive underneathSignature at offset 0
.cbzZIP50 4B 03 04 (“PK”)
.cbrRAR52 61 72 21 1A 07 (“Rar!”)
.cb77-Zip37 7A BC AF 27 1C
.cbtTAR(no magic; ustar at offset 257)
.cbaACE(ACE header)

Because they are simply renamed archives, you can extract any of them with the matching archiver and get the page images back unchanged. This also means “converting” between .cbr and .cbz is not a re-encoding of the pictures — the images are already final JPG/PNG files — it is only a repackaging: unpack the RAR, repack the same images as ZIP.

The RAR signature and why the version byte matters

Rename a .cbr to .rar and a hex viewer shows the RAR marker block at offset 0. There are two forms in the wild, and telling them apart is a matter of one byte:

RAR 4.x:  52 61 72 21 1A 07 00           "Rar!" 1A 07 00      (7 bytes)
RAR 5.0:  52 61 72 21 1A 07 01 00        "Rar!" 1A 07 01 00   (8 bytes)

The first six bytes, 52 61 72 21 1A 07, spell Rar! followed by a 0x1A (Ctrl-Z, the old DOS end-of-text byte) and 0x07. The next byte is the format-version discriminator: 0x00 closes the classic RAR 4 marker, whereas RAR 5 inserts 0x01 0x00, extending the signature to eight bytes. This is not cosmetic. RAR 5 redesigned the archive-header layout and compression engine, so an extractor built only for RAR 4 will refuse a RAR 5-based .cbr. After the marker comes an archive header and then, for each page image, a file header giving the entry name, its uncompressed and compressed sizes, and a CRC, followed by the (usually stored, not heavily compressed) image data. The stored image order is irrelevant; the reader sorts by the file names, which is why consistent zero-padded names like p001.jpg matter for correct page sequence.

The RAR catch: extract freely, but only WinRAR creates it

RAR is a proprietary format, and this is the single most important practical fact about CBR. RARLAB publishes a free decompressor, so almost any archiver — 7-Zip, WinRAR, PeaZip, Keka, The Unarchiver — can extract a .cbr. But the RAR compressor is closed: only RARLAB’s own WinRAR (and its command-line rar) can legally create a genuine RAR archive. No free tool can produce one.

That asymmetry is exactly why the community prefers the ZIP-based CBZ for making comics: any tool on any platform can both read and write ZIP, so a comic you build as CBZ is reproducible by anyone, whereas a CBR you receive can be read but not cleanly re-created without WinRAR. It is also why so many readers and library managers standardise on CBZ, and why “convert CBR to CBZ” is a routine request. The conversion is loss-free because it never touches the images: extract the RAR, then ZIP the same JPG/PNG files and rename to .cbz.

ComicInfo.xml: series and issue metadata

Beyond the page images, a modern comic archive often carries one extra file at the archive root: ComicInfo.xml. This is a small XML sidecar, originating with the ComicRack reader, that records the series title, issue number, volume, writer, publisher, page count and reading direction. Library servers such as Komga and taggers such as ComicTagger read and write it so a shelf of hundreds of issues can be organised without opening each file. It is optional — a bare folder of images zipped up is still a valid comic — but it is what turns a loose archive into a catalogued library entry. Because it is just another entry inside the RAR, readers that do not understand it simply skip it and show the images.

Reading a CBR, and why ZIP tools fail on it

To read a CBR as a comic rather than a heap of files, use a dedicated reader: YACReader or CDisplayEx on Windows, YACReader or Simple Comic on macOS, Okular or MComix on Linux (both need an unrar/unar backend for the RAR decoding), and Panels or Chunky on an iPad. These add the page-fitting, spreads and bookmarks that make a comic pleasant to read. To pull out just the raw page images, extract the .cbr with any RAR-capable archiver — the pictures are already JPG/PNG inside, so no conversion is needed.

One failure is common enough to call out: Windows’ built-in ZIP extractor and other ZIP-only tools cannot open a .cbr, because it is RAR, not ZIP. The fix is to use a RAR-capable archiver (7-Zip, WinRAR, Keka) or, better, a comic reader that bundles RAR support. On the safety side, a .cbr is a RAR archive, so the usual archive cautions apply: a maliciously crafted archive could target a parser bug in the reader, and a file could in theory hide entries other than page images, so open comics from untrusted sources in an up-to-date reader. The page images themselves are inert. Avoid sketchy “CBR reader” download portals that bundle adware, and get YACReader, CDisplayEx or Calibre from their official sites.

Frequently asked questions

What is the difference between CBR and CBZ?

Both are comic archives of ordered page images; the only difference is the container. A .cbr is a RAR archive, a CBZ is a ZIP archive. CBZ is preferred because any tool can both read and create ZIP, while only WinRAR can create RAR — so CBZ is the open, reproducible choice for making comics.

Why won’t my CBR open in Windows’ built-in ZIP extractor?

Because a .cbr is a RAR archive, not a ZIP one. Its bytes start with Rar! (52 61 72 21 1A 07), which the ZIP-only extractor does not recognise. Use a RAR-capable archiver such as 7-Zip, WinRAR or Keka, or open the file in a comic reader that handles RAR.

Do I lose quality converting CBR to CBZ or extracting the pages?

No. The pages are stored as finished JPG or PNG files inside the archive, and repackaging or extraction copies those bytes unchanged. Converting CBR to CBZ only swaps the RAR wrapper for a ZIP wrapper; the images are never re-encoded, so there is no generation loss.

References