Re: Not sure what they used...
The standard reference for old NASA space computers is
Computers In Spaceflight: The NASA Experience where you'll find out all sorts of quite detailed stuff about how NASA built and ran those missions, right back to the days when the state of the are was a "cam timer," essentially the device used in old washing machines.
TTL has a rep for being power hungry, but it wasn't too bad if you kept the clock frequency down. The standard 16 and 18 pin packages used made the packing density quite good (for the time). Today we'd go surface mount and increase it 4x at a stroke.
Quite a few of these processors were bit serial, with "word length" set by width of registers (which might also be serial, being a string of ultrasound pulses in a delay line memory).
The availability of a 4 bit ALU (LS74181 and it's CMOS equivalent) made new processors easier, if you could take the clock speed limits and you can operate in chunks of 4 bits, which was OK for a lot of people.
When you control the hardware if it''s not fast enough not only can you hack the code, you can hack the instruction set as well. :-)
Then you hack the assembler to support the new instructions (no HLL, no YACC or Lex to write one)
From that era it seems only the RCA 1802 was available early enough and rugged enough for space use. It's sort of like the SPARC, a big register set and on chip DMA, DMA is very handy for space probes.
A very different world.
Today you hand BAe Systems $150K for a rad hard POWERPC board instead.