Upsampling and oversampling are optional.
A DAC that handles 24/96 means it can handle that signal natively. This is something you would get by buying high resolution music or listening to a high res station like Toronto Jazz 91, or by upsampling your own music via software such as ffmpeg or sox (Linux/MacOS).
It does not mean that all input signals are up or over sampled. It just means if your data is that dense it can still read it. Vendors make different choices and some use very closely guarded algorithms to accomplish up sampling, while some just use simple linear interpolation. For instance, some convert everything to quad rate DSD. That's their shtick. Others insist on never over or up sampling. That's Audio Note's shtick.
Lots of makers claim upgrade-ability, but never actually offer it.
Honestly from a personal perspective, I find the DSD and it's derivatives over-hyped. I've not personally not been blown away by any of those standards. The current MQA hype is in there as well. Usually any differences in sound quality are really engineering differences and not really related to the underlying technology.
What I have found that seems to matter a lot more is the quality of the output stage. I love Ayre for this reason. Whether it's a CD player, preamp or Pono, I just love love the sound, no matter the original format. Same thing for portable players. I can play normal and high def music, and the difference isn't as big a deal to my ears as the amplifier and headphones. I'd much rather listen to redbook CD via an Ayre QB-9 than high rez or SACD through a lot of other players.
I would strongly suggest you start listening with this in mind and see what you are sensitive to, and what kind of music production seems to appeal to you the most.
A DAC that handles 24/96 means it can handle that signal natively. This is something you would get by buying high resolution music or listening to a high res station like Toronto Jazz 91, or by upsampling your own music via software such as ffmpeg or sox (Linux/MacOS).
It does not mean that all input signals are up or over sampled. It just means if your data is that dense it can still read it. Vendors make different choices and some use very closely guarded algorithms to accomplish up sampling, while some just use simple linear interpolation. For instance, some convert everything to quad rate DSD. That's their shtick. Others insist on never over or up sampling. That's Audio Note's shtick.
Lots of makers claim upgrade-ability, but never actually offer it.
Honestly from a personal perspective, I find the DSD and it's derivatives over-hyped. I've not personally not been blown away by any of those standards. The current MQA hype is in there as well. Usually any differences in sound quality are really engineering differences and not really related to the underlying technology.
What I have found that seems to matter a lot more is the quality of the output stage. I love Ayre for this reason. Whether it's a CD player, preamp or Pono, I just love love the sound, no matter the original format. Same thing for portable players. I can play normal and high def music, and the difference isn't as big a deal to my ears as the amplifier and headphones. I'd much rather listen to redbook CD via an Ayre QB-9 than high rez or SACD through a lot of other players.
I would strongly suggest you start listening with this in mind and see what you are sensitive to, and what kind of music production seems to appeal to you the most.