They are described in detail in the API for DataOutput, an interface which RandomAccessFile implements. Basically, writeChars() uses Unicode values, writing two bytes per character. While writeUTF() uses a slightly-modified version of UTF-8, which uses 1 to 3 bytes per character. Also, writeUTF() writes two bytes at the beginning which identify the number of characters which will be in the UTF-8 representation, so the readUTF() method will know when to stop. The writeChars() method does not do this, and also does not write a delimiter (like 0) at the end of the String - so I'm not sure what it's really useful for. The DataInput interface does not define a readChars() method, because writeChars() does not provide enough information to know reliably where the string should terminate. There are ways to fix this, but in general it's too much trouble - I'd usually just ignore writeChars() and use writeUTF() instead, whenever I have String data to write to a binary file. Of course, if you're writing a text file rather than a binary file, you probably shouldn't be using any of the DataOutput / DataInput methods anyway. They have too many methods which generate unprintable chars if you don't know exactly what you're doing. Use Readers and Writers; they're much better.