A relatively minor update to the Windows 10/8.1/8 version of Joystick-To-Mouse is now available, tagged as Version 2.80 Release 5. Changes were made to the Joystick-To-Mouse executables, along with date and release text updates. This update is an example of the maintenance that is required of software - here is the background and details:
A new customer was using Joystick-To-Mouse on their Home Theater PC, using an Xbox 360 wireless controller, but ran into an issue when going into Start->Settings->System->Display->Change text size (or Display | Advanced Settings, set Custom Text size) to make the text easier to see on their flat screen TV. Basically whenever the setting was something other than 100%, the mouse cursor ran to the bottom-right of the screen, and made the software unusable. A Tech Support case was created, and it bubbled up to the development staff, since there was no quick solution available. The first step in resolving issues is to be able to duplicate the problem, and this was quickly done, but trying to understand why the DPI change was the cause of the problem seemed odd, as the software already had a SetProcessDPIAware call that was added back in the updates for Windows 8. DPI stands for Dots Per Inch, which is used in Windows to map things on to the display. In digging into the code and documentation, it turns out the preferred approach for a process is to use now is the program manifest vs. the API call, and it must be noted that is not uncommon for the preferred solution to change when dealing with Windows API based software. In testing and working with rebuilt/remanifested executables, it was discovered that the existing software did work just fine with custom text sizes in Windows 8, but not in Windows 8.1 and Windows 10. Also it should be noted that testing in a normal system would never expose this issue - only doing custom text sizing would make the problem occur. So the only required change was to remove the deprecated SetProcessDPIAware call, and update the manifests for the required executable files. Then comes the testing, and rolling into an updated release.

In an ideal world, with infinite resources, complete regression testing would have caught this issue. But customers simply won't pay thousands of dollars for a little utility program. Plus why should a problem that was found, and fixed, and in the software need testing on what was a mostly a cosmetic update to Windows 8? It turns out the API used had some issues, so it was deprecated (most likely around the Windows 8.1 release), but the thought that every Windows API used in a Windows software program needs to be reviewed for every Windows release just to see if there have been changes makes little business sense, so customers need to be understanding of the realities of modern software. Maybe someday non-IMG software will be engineered and maintained in a disciplined manner, but that is not the world we live in today.

Joystick-To-Mouse continues to be in demand, and provides a unique capability that IMG updates and maintains. Now that only secure, code-signed programs can access the user interface capabilities needed to perform the function that Joystick-To-Mouse provides, it may be a good idea to "Tell a Friend", and let your gamer friends know that they can "play" their internet browser, or "drive" Windows, or use Windows on a large screen TV with large text! Just as many eyes make bugs shallow, many customers make resources available for more development, more testing, and more features...