This problem seems to occur with non-UTF-8 encoded log files. I'm not sure if EAC defaults to that (the log files I looked at in Notepad++ were in UCS-2 LE BOM), or what, but trying to open a log file with a non-UTF-8 will produce this encode read error. Switching the encoding to UTF-8 and saving works, but going through and re-saving every log file is obviously not feasible. A quick google search says that the script needs to forcibly read/add UTF-8 to fix the issue.

I uploaded a sample log file. It should be in UCS-2 LE BOM originally (at least I don't think it'll change it's unicode enoding if you open it in a different computer, etc.).