In our market, we have many customers who are having problems with cables that they've purchased elsewhere. They have data integrity problems (ie, the cameras aren't as fast as they should be due to dropped frames), or during device testing they have ESD or EMI problems.
After so many years in this industry, we have come to learn a lot about the intricacies of building a proper USB cable - one that is not just to spec, but one that is properly made for a specific industrial application, like machine vision, which can require tighter specs than what the USB spec requires - especially if you need to go outside of the spec, such as needing an extension cable (which is not permissible in the spec) or if you have a high static environment and you need to drain the ESD (no fixes listed in the spec), nor using angled connectors (only a couple are even listed in the spec, and those are solely for USB 2).
Because there are so many things that we could discuss, I would like to welcome you to contact me directly for a specific question. When I find some that are noteworthy enough, I'll publish the details here for the common benefit and the advancement of proper techniques within our industry.
To be frank, for an advanced level of usage of USB 3.0 / USB 3.1, the scope that we need to operate at is very high; unfortunately, many cables that we've tested, even cables that have been certified, are so loose within the spec (although they do appear to be within spec) that they cannot fit into an MV application. They cause strange errors and differ from cable to cable. It's unfortunate, but the reality.
So, as I get your questions, I'll publish more. Contact me at our website, USBFireWire.com