The level of detail that you can get from the real world is infinite. It is only limited by the devices we use to capture it. Today’s advanced perception systems are hungry for data to help them understand the real world and make informed decisions. In the last couple of years there’s been a surge in the use of synthetic data to train computer vision-based perception systems. And rightfully so, I think that the value of synthetic data to reduce cost and improve efficiency is proven but, can synthetic data capture those infinite details when reproducing the real world? Can we reduce the domain gap? What are the limits? Synthetic Hyperspectral Data may be the answer. We will explore the requirements to generate hyperspectral data, what’s its value, how to use it and some results.