When it comes to any shielded cables with metal plugs and shield attached to the ground, you most likely will hear a difference. You are breaking the galvanic isolation. It can sound better initially. Things will sound smoother, less grainy, darker and more detailed. It's bringing in noise to the player.
You can do this test with a Cat6 S/STP or any Cat7 or Cat8 as they are similar and double shielded designs.
I did this for multiple pieces. Buy 3-4 versions of the same cable in the same exact length. The only thing you change is whether the ground is still attached to the shield or not. You can break and remove the metal plugs on the cables. That way you can the same cable with grounds attached on both ends, one end, or no end.
As for hearing differences between similar designs, please always pay attention to the length being used. IME the differences are more exaggerated at short lengths (less than 3m).
Some of these cables work, but don't pass or exceed the spec, like a Belden bonded would. That would be one thing to look at. Hook each cable to a Fluke DTX-1800 and see how it does.
If you have a good cable that measures well on a Fluke, then try multiple versions of it and just vary the length.
You never see that type of control with these reviews. Instead, it's bunch of cables at different lengths.
Provided the cables are well made, such as Belden bonded Cat6 UTP, I don't think they are making things worse/better themselves, but they are rather influencing the behaviour of the PHY (especially in most devices as they aren't using TOTL ethernet packages). So the problem might be with the PHY itself, and the cables are being used to fix/reduce the problem.
I've done similar testing with SPDIF and AES cables. Same wire, same connectors, only vary the length systematically. They all work, but sound slightly different from one another. I tend to prefer longer AES and SPDIF cables with my particular DAC/transport as well.
I think there should be more systematic control in these type of observations.
