The reason for software having volume controls is so that a user can control the volume of each individual application, ie. having a youtube video play very loudly, and having a game you're playing be muted. The reason for the OS having volume controls is strictly for immersion, the user can control their experience (volume) without having to break out of their comfort zone/immersion in the UI (take their hands off of the keyboard & mouse, and reach forward to rotate a slider), there is also a practicality reasoning for this, what if this is a computer in a theater/auditorium, with huge speakers, and the speakers and their sliders are out of reach? How will the user adjust the volume if the OS slider is gone? The other practicality case would be for a user who has two computers, connected to the same speaker(s), how would they adjust the volume (assuming the speaker(s) are not designed for such a use case)? The only real redundancy I see in audio sliders, is in the fact that most major OS's (Linux-based OS's and Windows, not sure about OSX) have a "mixer" panel, where you can control the volume of each application, this becomes redundant when an application has it's own volume applications, there are now 4 sliders deciding the volume of a single application. But, the redundancy can be overlooked because most users are not aware of the "mixer" menu's, because some applications can have multiple outputs (e.g. a browser with two-audio outputting tabs, would be seen as the same process by windows, and you wouldn't be able to mute/make one quieter or louder than the other, so you would need each audio-output to have it's individual control), and because some applications refuse to even offer the user a slider to mute them.