How to use the binary file diff tool
- Choose the first file (A) and the second file (B). Each can be any kind of file, up to 25 MB.
- Optionally set a gap in bytes. Differences closer together than the gap are joined into one range.
- Read the summary: whether the files are identical, how many bytes differ, how many ranges, any extra bytes at the end and where the first difference is.
- Look at the table of ranges and the hex view of the bytes around the first difference, and download the list as CSV.
Formula
The files are compared from the first byte up to the length of the shorter one
A byte differs when the value at the same offset is not equal in A and B
Neighboring differing bytes form a range, written as a start offset and a length
Bytes past the end of the shorter file are counted as extra bytes and are not part of any range
Offsets are counted from 0 and shown in hexadecimal and decimal. A range with a length of 1 is a single changed byte.
One flipped byte
Take a 64-byte file whose bytes are 0, 1, 2 and so on, and copy it with the byte at offset 0x10 changed from 10 (hex 10) to hex FF. The tool reports exactly one range, at offset 0x00000010 with a length of 1, and shows 10 in file A and FF in file B. One byte differs, there are no extra bytes and the first difference is at offset 16.
If the second file is the same but has 3 more bytes at the end, no range is reported, the bytes in common are identical, and the page shows 3 extra bytes only in B. With bytes 5, 6, 7 and 10 changed, a gap of 0 gives two ranges, and a gap of 3 joins them into one range from offset 5 to 10.
What a byte diff is good for
A text diff compares lines, which is no use for data that is not text. A byte diff shows exactly what changed between two builds of a program, two exports of a file, two versions of a document, or a file and its copy after a transfer. Seeing a few changed bytes in a header often tells you what the change was, such as a timestamp, a checksum or a version number.
To read the raw bytes of one file, use the hex viewer. To compare text line by line, use the diff checker, and to compare two pictures use the image diff checker.
Reading the results
Many small ranges that are close together usually mean the same kind of change repeated, for example a shifted block, and a large gap setting joins them into a readable summary. If the files are the same length and differ everywhere, they are probably different data or one is compressed or encrypted. If one file is longer and the rest matches, the extra bytes were appended or the other file was cut short.
A file that has shifted by even a single byte at the start shows differences from that point to the end, because the comparison is by position and does not search for moved blocks. To check what type of file you have, see the file type identifier.
Assumptions and limits
- Files are compared byte for byte at the same offsets. Insertions and deletions are not aligned.
- Each file is up to 25 MB, and both are read into memory in your browser.
- The table shows the first 500 ranges. The CSV lists all of them.
- Bytes are shown in hexadecimal, with up to 16 bytes of each range.
Frequently asked questions
How do I compare two binary files?
Choose both files. The page lists every range of bytes that differ, with the offset, the length and the bytes on each side.
What does the gap setting do?
It joins ranges that are closer together than the gap, so a scattered set of changes becomes a shorter list.
What if the files have different lengths?
The shared part is compared byte by byte, and the extra bytes of the longer file are counted separately as extra bytes at the end.
Are my files uploaded?
No. Both files are read and compared in your browser and never leave your device.