Flight controllers have to implement an inverter to process the SBUS signal. Why is SBUS not just uninverted?
SBUS and IBUS are both digital protocols and for a digital protocol, you define two voltage values to represent an active state and one inactive state. This is done for every digital protocol.
You can specify that 0V is inactive and 1V is active (this is called active high). But you can also define that 0V is active and 1V is inactive (active low).
Many bus protocols are active low and SBUS is also such a protocol. The reason for this choice is that active low signals can handle noise better. This is probably the reason why Futaba designed SBUS this way.
The ARM chips used on the flight controllers have the hardware to decode serial signals (both SUBS and IBUS are serial protocols). This decoder also is built for active high or active low signals. If you want to decode an active low signal (SUBS) with a decoder that is built for active high you have to add an inverter between so that both "talk the same language". This is the case for the F4 chips since ARM did not include an inverter on the chip. For F3/F7 this problem does not exist since ARM included the inverter in the chip and the software can enable/disable the inverter.
You get similar performance from SBUS and IBUS since both are a serial protocol with comparable frame times.
There isn't actually an advantage for SBUS over IBUS in digital serial protocols, and with STM32 F4 processors needing inverter hardware. The frame performance is basically identical, although IBUS does enable an extra 2 channels compared to D16 SBUS (and 10 more compared to D8).
There are small budget SBUS receivers that outperform their IBUS equivalents, otherwise, I consider IBUS to be a superior protocol because it doesn't require inversion... but I personally run mostly a mixture of XM+ compact receivers, and X6B FlySky receivers for larger quads.
The inversion is due to the electrical design of the receiver. Usually, you will find an uninverted pad on the receiver too.