Choosing a hydrogen engine (which only really makes sense for upper stages) also means you need a hydrogen tank... a cryogenic hydrogen tank. They chose to make such a tank with a weird shape that fit the unnecessarily weird shape of their prototype. I think the major cost of getting their craft repaired (or more realistically, having a new one built) was to build a new custom hydrogen tank.

Crazy, when they didn't need anything optimized/complicated for testing launch and landing.

Also wild that they went with an F/A-18 accelerometer/gyro package. The first commercial mass market MEMS accelerometer was introduced in 1991 and was in volume production in 1993. I mean, they had to pick something and the ADXL50 wasn't ready yet (and they would still have had to design a 3-axis solution around it if it were), so I don't blame them for that (expensive) decision.

https://qringtech.com/TryMe/wp-content/uploads/2014/01/Histo... https://www.analog.com/media/en/technical-documentation/obso...

I do very much blame them for the choice of the RL-10, though. And for going immediately to such a large size -- that's the same thing as saying "we plan on not making any mistakes".