>>12197768Aerospikes to have slightly lower TWR and a slightly larger diameter footprint for the same thrust, however I'd take this with a grain of salt. While the RS2200 was cutting edge at the time, it was still built using RS-25 guts and the old fassioned hand-designed way. Frankly I think there's still substantial room for maturation of the aerospike design which won't be done the oldspace way.
Firstly, the RS2200 is a ramp/wedge type aerospike with only two sides, the least efficient way to do an aerospike, there's lost surface area which could be put to good use. J2T does it better, but it's still running off the turbomachinery of the anemic LOX/LH2 J2 rocket, ideally you'd want a truncated cone aerospike's surface to have "scoops" or corrugation, this would probably give it more surface area than an equivalent bell engine, and thus allow it to become at least equally space efficient.
The best version of an aerospike though would be the rotating pulse detonation aerospike, RPD rockets are still a very immature technology but there have been successful test firings that indicate they could yield huge TWR and ISP enhancements to bipropellant rockets, not quite as great an ISP boost as Nuclear Thermal, but still quite significant, 25% improvement.
