To ease the burden of typing, what I do is this: I have a file containing chess-related commands I have run. I keep this file in a directory I can easily find. Most times I want to do a chess-related command, it's a slight variant of one I ran earlier, e.g. to run Popeye on some other input file. So I'd replace the old input file name with the new one, resulting in something like this:
cd "D:\games\by game\chess\problems\Popeye"
C:\Progra~1\Popeye\pywin64 is just where I installed the Popeye executable on my Windows PC. Replace this with whatever is the file name (including drive-letter and full path) of where you've installed the Popeye executable on yours.
Smith_h#3.pyin you see there is the name of the file which Popeye is to read. As it stands, that command would run Popeye, read that file and write its output to the command-prompt window. This is what that Popeye documentation was referring to when it said "To solve problems recorded in file XX". By XX they mean the name of the file in which you have written the Popeye commands that describe the problems(s) you want Popeye to solve. That Popeye documentation describes the Popeye commands which you need to put into the input file to describe a problem to Popeye.
If you want to capture this output in a file, here are two ways to do it. One is to specify the output file in the command, e.g.
C:\Progra~1\Popeye\pywin64 Smith_h#3.pyin > Smith_h#3.pyout
This would redirect output to
Smith_h#3.pyout but would not write anything to the window while Popeye is running. Another is to put the Popeye command
in Popeye's input. This would redirect output to the specified output file and write it to the window as well.