Comic book archive formats are the simplest ebook formats you'll encounter. A CBR file is a RAR archive of images. A CBZ file is a ZIP archive of images. That's it. There's no markup language, no style sheets, no metadata schema baked into the spec. You rename a .cbz to .zip, extract it, and you're looking at a folder of page images.

This simplicity is intentional. Comics are visual media where every pixel matters. The layout, panel arrangement, lettering, and art are all baked into the page image. There's nothing to reflow, nothing to re-render, nothing for a reading app to interpret beyond "show this image, then the next one." The format's job is to deliver those images in order, compressed for storage, and get out of the way.

The Archive Formats: CBR, CBZ, CB7, CBT

Each comic book archive format is just a standard archive format with a different extension:

ExtensionArchive FormatCompressionLicenseTool Support
.cbrRARRAR (proprietary)ProprietaryGood (WinRAR, 7-Zip, unrar)
.cbzZIPDEFLATE (open)Open standardExcellent (universal)
.cb77zLZMA/LZMA2 (open)LGPLGood (7-Zip, p7zip)
.cbtTARNone (uncompressed)Open standardExcellent (universal)

CBZ is the recommended format for new archives. ZIP is universally supported on every operating system and programming language. CBR persists because many older scans were distributed as RAR. CB7 offers better compression ratios (10-20% smaller than CBZ) but less tool support. CBT is rare — TAR provides no compression, so it only makes sense for already-compressed image formats like JPEG.

File Naming and Page Order

Comic readers sort images within the archive alphabetically to determine page order. The convention is zero-padded sequential numbering:

page_001.jpg
page_002.jpg
page_003.jpg
...
page_024.jpg

The actual naming pattern doesn't matter as long as alphabetical sorting produces the correct page order. Common patterns:

  • 001.jpg, 002.jpg — simplest, works fine
  • Batman_001_page_01.jpg — descriptive, still sorts correctly
  • cover.jpg, page_001.jpg — cover sorts before pages alphabetically

Pitfalls: page_1.jpg, page_2.jpg, ..., page_10.jpg will sort as 1, 10, 11, ..., 19, 2, 20 — alphabetically, "10" comes before "2" because "1" < "2". Always zero-pad to at least three digits. For archives with 100+ pages, use four digits. Some readers implement "natural sorting" (treating numeric parts as numbers), but don't rely on it.

Subdirectories within the archive are handled inconsistently across readers. Some readers flatten the directory structure; others display subdirectories as separate sections. For maximum compatibility, put all images in the archive root with no subdirectories.

Image Resolution and Format

Comic page images need to be high enough resolution that text is legible and art detail is preserved, but not so large that file sizes become unmanageable.

Resolution Guidelines

Standard US comic dimensions are 6.625" x 10.25". At different resolutions:

  • 150 DPI (994 x 1538) — Minimum readable on tablets. Text is legible but fuzzy. Fine for quick browsing, not for reading
  • 200 DPI (1325 x 2050) — Acceptable quality on 10" tablets. Standard for many fan scans from the 2000s-2010s
  • 300 DPI (1988 x 3075) — Professional quality. This is what publishers use for digital distribution. Sharp text, visible line art detail, and good readability at any zoom level
  • 600 DPI (3975 x 6150) — Archival quality. Overkill for reading but preserves every detail. File sizes are 4x the 300 DPI version

A practical minimum for readable comics is 1800 x 2700 pixels per page (roughly 270 DPI). Below this, fine lettering becomes blurry on high-DPI screens. For manga (smaller trim size, denser line art), aim for at least 1200 x 1800.

JPEG vs PNG for Comic Pages

Most comic archives use JPEG at quality 85-95. JPEG compression is well-suited to comic art — large areas of flat color compress efficiently. A 2000x3000 color comic page at JPEG quality 90 is typically 1-3MB. A 24-page issue runs 25-70MB as a CBZ.

PNG is better for: black-and-white comics (LZW/DEFLATE handles 1-bit and limited palette images very efficiently), comics with large flat color areas, and archival purposes where any lossy compression is unacceptable. The file size penalty is significant though — the same 2000x3000 page can be 5-15MB as PNG versus 1-3MB as JPEG.

WebP would theoretically give the best of both worlds (lossy with better compression than JPEG, lossless with better compression than PNG), but reader support for WebP in comic archives is inconsistent. Stick with JPEG for color, PNG for B&W or archival.

ComicInfo.xml: Metadata for Comics

ComicInfo.xml is a metadata file included in the archive root, originally created by the ComicRack application. It's become the de facto metadata standard for digital comics, supported by ComicRack, Kavita, Komga, Mylar, Codex, and most modern comic library managers.

<?xml version="1.0"?>
<ComicInfo>
  <Title>The Dark Knight Returns</Title>
  <Series>Batman: The Dark Knight Returns</Series>
  <Number>1</Number>
  <Count>4</Count>
  <Year>1986</Year>
  <Month>2</Month>
  <Writer>Frank Miller</Writer>
  <Penciller>Frank Miller</Penciller>
  <Inker>Klaus Janson</Inker>
  <Colorist>Lynn Varley</Colorist>
  <Publisher>DC Comics</Publisher>
  <Genre>Superhero</Genre>
  <PageCount>48</PageCount>
  <LanguageISO>en</LanguageISO>
  <Manga>No</Manga>
  <Pages>
    <Page Image="0" Type="FrontCover"/>
    <Page Image="1" Type="Story"/>
  </Pages>
</ComicInfo>

Notable fields: <Manga> signals right-to-left reading order. <Pages> can mark individual pages as FrontCover, Story, Advertisement, Editorial, etc. Credit fields are granular: Writer, Penciller, Inker, Colorist, Letterer, CoverArtist, Editor — reflecting comic industry roles accurately.

Comic Reader Apps

The comic reader app ecosystem is mature across every platform:

Desktop

CDisplayEx (Windows) — The spiritual successor to CDisplay, the original comic reader from 1999. Fast, lightweight, handles CBR/CBZ/CB7/PDF. The reference implementation for desktop comic reading. YACReader (Windows/Mac/Linux) — Open source, includes a library manager with series tracking, supports ComicInfo.xml, and has a companion iOS/Android app. MComix (Linux) — GTK-based, lightweight, handles CBR/CBZ/CB7/CBT and image directories.

Mobile and Tablet

Panels (iOS/iPadOS) — The best iOS comic reader. Clean UI, ComicInfo.xml support, guided reading (panel-by-panel zoom), OPDS catalog support for server integration. Chunky (iPad) — Designed specifically for iPad's larger screen. Handles CBR/CBZ/PDF with smooth page transitions. Perfect Viewer (Android) — Highly configurable, supports CBR/CBZ/CB7/EPUB, ComicInfo.xml metadata, and automatic page splitting for double-page spreads. Tachiyomi/Mihon (Android) — Open source manga reader with extension system for online sources, also handles local CBR/CBZ files.

Self-Hosted Servers

Kavita — Open source comic/manga/ebook server. Reads ComicInfo.xml for metadata, supports CBR/CBZ/CB7/EPUB/PDF, has a web reader and OPDS feed. Komga — Similar to Kavita, with strong ComicInfo.xml support and OPDS integration. Both solutions let you host a comic library on a server and read from any device via web browser or OPDS-compatible app.

CBR to CBZ Conversion: Just Re-Archive

Converting CBR to CBZ is the simplest conversion in all of digital publishing: extract the RAR archive, re-archive as ZIP, change the extension. The images are unchanged. No transcoding, no quality loss, no metadata transformation. It's a container swap.

Convert CBR to CBZ here — our backend extracts the RAR and re-archives as ZIP. The process preserves every image file, ComicInfo.xml if present, and the file naming order. File size may change slightly (RAR and ZIP have different compression ratios), but the images are bit-for-bit identical.

Why convert? RAR is proprietary — the compression algorithm is owned by Alexander Roshal, and distributing RAR-compatible software requires a license. ZIP is an open standard with native support in every operating system. Some comic readers (especially on iOS) handle CBZ more reliably than CBR. For long-term archival, open standards are always safer.

Other conversions: CBR to EPUB and CBZ to EPUB wrap the images in an EPUB fixed-layout container. CBR to PDF and CBZ to PDF render images as PDF pages. These conversions don't add reflowable text — the comic pages remain images.

Creating Comic Archives

To create a CBZ file from page images:

# Ensure images are named for correct sort order
ls pages/
page_001.jpg  page_002.jpg  page_003.jpg  ...

# Create CBZ (just a ZIP with .cbz extension)
cd pages/
zip -r ../my_comic.cbz *.jpg ComicInfo.xml

For scanned physical comics: scan at 300+ DPI, crop to the art area (remove scanner bed edges), save as JPEG quality 90-95 for color or PNG for B&W, number sequentially with zero-padding, optionally create a ComicInfo.xml, and archive as CBZ.

For digital art: export pages at your working resolution (typically 300 DPI at print size), flatten layers, save as JPEG or PNG depending on art style, and archive. If you're creating a webtoon-style vertical scroll comic, use very tall images (one per episode) or split into scroll segments and consider EPUB fixed layout instead of CBZ.

Comic book archives are the most honest file format in digital publishing. There's no abstraction layer pretending the content is something it's not. Pages are images. The archive holds them in order. The reader shows them. That transparency makes the format easy to work with, easy to convert, and easy to debug.

If you're managing a comic library, standardize on CBZ for the open format guarantee and better tool support. Convert your CBR collection to CBZ — it's a lossless operation that takes seconds per file. For metadata, add ComicInfo.xml to enable proper library management in Kavita, Komga, or YACReader.