I expected a text file that I can understand.

You've explicitly asked perl convert each byte to a number, so you're expectations were naive.

A quick look on the web turns up this page which identifies:

1. The first four bytes are the signature: '.sff';

   Which you have.

2. The next four byte form an unsigned 32-nit integer that is the version: Apparently always 1.

   Which you have.

3. The "index offset" and "index length", which it doesn't state the size, but from the example, the offset must be at least 4 bytes and the index at least two. Either or both could also easily be 8 bytes. Especially the offset.

   You have 14 zeros; which perhaps means your file doesn't contain an index.

4. Lots more fields.

   You haven't shown us the data.

The point of that was to make it clear that you cannot "convert binary to text" unless you understand what the binary is.

That means you need to know:

• what fields the file contains;
• in what order;
• and in what binary format they are stored;

And that is true regardless of whether you read the file using Perl, C, or any other language.

As an aside, I'd bet that -- given the information required -- a Perl script can decode one of these files just as quickly (within a few percent) as a program written in C.

And it would be a whole heap faster to write the Perl version. (Admittedly, this is the type of thing I've been doing week in, week out for a decade or more. :)

(Also, if you post a link to a file in this format, I might have a go at decoding it. I've been sitting here waiting for a simulation to complete for 3 days now, and I'm a little bored.)