When Gimbal Lock Is A Bad Thing.

[NeoMorph]

How do you have four axis of control? Are we counting rotation as an axis here?

Nope... the four axis are in the gyro... Basically when you have three gyro wheels and two align up it causes a problem that cannot be fixed without spending hours taking star sightings to work out the true alignment of the spacecraft. If you have four gyro wheels though you can have two gyro wheels accidentally align and you can use the fourth wheel to unstick the gimbal lock.

Sounds complicated unless you played with gyros as a kid (made my own from a toy gyro spaceship and some bits and bobs that dad got from his works). I had a seat that was on castors that I used to control the direction by spinning up the gyro and then tried to turn it and intead the seat turned around.

Regarding the Noun and Verb... there was Prog too. Program was the program loaded into the memory. Verb was the type of thing that the program will run and noun was the area of memory that will be worked on. It's mind blowing how the system worked but here is a simple example... if you hit Verb 16 noun 36 and then hit enter it will display the time. Bit long winded when you compare it to todays systems but remember that the tech was such that the programs were actually WOVEN into ropes in the main ROM. I kid you not.

You can play with a virtuall Apollo Guidance Computer by downloading a program from http://www.ibiblio.org/apollo/ which has something called VirtualAGC. Playing with a real DSKY opens the mind to how amazing it was to get to the moon and back.

Oh and if you ever see me put "P00 and Accept" it's a DSKY joke (DSKY stands for "Display and Keyboard"). P00 is program zero which means you have finished with the computer and have put it into idle mode (which isn't really idle because it is still running the spaceship). It's the equivelant to the End of Line joke.

Here they are, weaving the Apollo computer memory lol.

Edited July 19, 2013 by NeoMorph

[DerekL1963]

Did you know the Gemini spaceship had a FOUR axis version that apparently didn't get into Gimbal Lock like the Apollo did?

That's correct, Gemini had a four axis guidance platform that could not get into gimbal lock, the why is because of quirk in space history... most people don't realize that Apollo was designed *before* Gemini.

The Apollo CSM was originally born as a general purpose earth orbiter, sort of a supersupersized Mercury. When Kennedy made the call to go to the Moon, it was uprated for the lunar mission and eventually the LM added to the package. (Originally the CSM itself was going to land on the moon.) That's why there were Block I and Block II command modules - Block I was the original version, and Block II the upgraded version. Because it was born as a general purpose orbiter, not only did they not think to install a 4-axis gimbal system (virtually a requirement for docking), the Block I _didn't even have a docking system_. Like the oversized propulsion system on the Service module, the gimbal was one of the things left 'as-is' because changing it was a hassle and at the time (this is still the early 60's) they didn't realize it (the gimbals) would be a problem.

Gemini came about as a bit of a kludge... NASA officialdom realized that not only would Apollo take years to develop, but that it's expense and complexity meant fewer flights would be possible. Together this meant that not only would their be a politically unacceptable gap in flights between Mercury and Apollo, but also that it would be hard to get the flight experience and testing needed before heading off to the moon while also meeting the "end of the decade" deadline. So they flailed about a bit until they discovered an unsolicited proposal from McDonnell (it wasn't yet McDonnell Douglas) for a Mercury MK II - a two man Mercury orbiter. This proposal was accepted, and rather quickly became Gemini. Since rendezvous and docking was part of the program goals, a four axis gimbal system was part of the specifications...

(voice = "Paul Harvey") And now you know... the rest of the story. (/voice)

[Deadweasel]

That's correct, Gemini had a four axis guidance platform that could not get into gimbal lock, the why is because of quirk in space history... most people don't realize that Apollo was designed *before* Gemini.

The Apollo CSM was originally born as a general purpose earth orbiter, sort of a supersupersized Mercury. When Kennedy made the call to go to the Moon, it was uprated for the lunar mission and eventually the LM added to the package. (Originally the CSM itself was going to land on the moon.) That's why there were Block I and Block II command modules - Block I was the original version, and Block II the upgraded version. Because it was born as a general purpose orbiter, not only did they not think to install a 4-axis gimbal system (virtually a requirement for docking), the Block I _didn't even have a docking system_. Like the oversized propulsion system on the Service module, the gimbal was one of the things left 'as-is' because changing it was a hassle and at the time (this is still the early 60's) they didn't realize it (the gimbals) would be a problem.

Gemini came about as a bit of a kludge... NASA officialdom realized that not only would Apollo take years to develop, but that it's expense and complexity meant fewer flights would be possible. Together this meant that not only would their be a politically unacceptable gap in flights between Mercury and Apollo, but also that it would be hard to get the flight experience and testing needed before heading off to the moon while also meeting the "end of the decade" deadline. So they flailed about a bit until they discovered an unsolicited proposal from McDonnell (it wasn't yet McDonnell Douglas) for a Mercury MK II - a two man Mercury orbiter. This proposal was accepted, and rather quickly became Gemini. Since rendezvous and docking was part of the program goals, a four axis gimbal system was part of the specifications...

(voice = "Paul Harvey") And now you know... the rest of the story. (/voice)

You are the newest addition to the "My Favorite People" list. Love reading about these little details I didn't know about before. Paul Harvey reference and everything! <3

/Paul Harvey was the only excuse for AM radio's existence during my youth, in my opinion.

[NeoMorph]

Regarding that gyro toy I mentioned... I actually found it with the power of the net...

Was that exact same model. Was what got me interested in gyros in later life.

Edited July 20, 2013 by NeoMorph

[JDobs]

The Apollo missions considered using a 4 axis backup gimbal but decided against it due to weight and size factors within the CSM. If you watch Apollo 13 there is a point just after the explosion that you hear Jim Lovel (Tom Hanks) yelling "We are about to hit gimbal lock!"

Also, if you have an iPhone, well essentially any modern smart phone, you have about 2 to 3 times the computing power in your phone than the entire Apollo computer had on board!

http://www.hq.nasa.gov/alsj/e-1344.htm

[Deadweasel]

The Apollo missions considered using a 4 axis backup gimbal but decided against it due to weight and size factors within the CSM. If you watch Apollo 13 there is a point just after the explosion that you hear Jim Lovel (Tom Hanks) yelling "We are about to hit gimbal lock!"

Also, if you have an iPhone, well essentially any modern smart phone, you have about 2 to 3 times the computing power in your phone than the entire Apollo computer had on board!

http://www.hq.nasa.gov/alsj/e-1344.htm

Actually, keep going. Your estimate describes the old IBM PC XT (8086), which ran around 4MHz. A modern USB thumb drive has more processing ability than the Apollo systems had. iPhone isn't even comparable. I'd say many tens of thousands of times more powerful by comparison.

[Akinesis]

There's a few things you can do to gyros to stop gimbal lock. But nowadays, we use laser and fibre-optic gyros. So they have no moving parts at all now so gimbal lock is a thing if the past.

Maybe Kerbals went straight to this?

[Deadweasel]

There's a few things you can do to gyros to stop gimbal lock. But nowadays, we use laser and fibre-optic gyros. So they have no moving parts at all now so gimbal lock is a thing if the past.

Maybe Kerbals went straight to this?

They do seem to have a tendency to jump straight for the stuff they understand the least, don't they? I mean, the current tech tree reflects that well. They have rocket technology long before they have solar power, batteries or even the power of basic atmospheric flight!

[magnemoe]

Happens in 3D programs too. Can be annoying:

(Z and Y are almost locked here)

Gimbal lock is very common and very annoying then working with 3d, Most irritating then you import something like an gun turret you have to roll it 90 degree to place it and this lock out the rotation or elevation axis, Many programs has the option to change the axis orientation, this removes the gimbal, yes it might come back but the gun turret would be safe as you never roll it.

[Akinesis]

They do seem to have a tendency to jump straight for the stuff they understand the least, don't they? I mean, the current tech tree reflects that well. They have rocket technology long before they have solar power, batteries or even the power of basic atmospheric flight!

Indeed! It's almost like the machines are in control, and they are using Kerbals to test their vehicles before they send their own hardware up!

[KSK]

That's correct, Gemini had a four axis guidance platform that could not get into gimbal lock, the why is because of quirk in space history... most people don't realize that Apollo was designed *before* Gemini.

*lots of cool information*

I'm sure I read somewhere that the Gemini capsule was more advanced than the Apollo CSM in many ways. The capsule systems were far more modular, so components were much easier to access and swap in and out during assembly and testing (and presumably in flight if required). Whereas the wiring on Apollo was a real cats-cradle that was much harder to work with.

[annallia]

So what you're saying is that Kerbals are actually pretty realistic?

I laughed entirely too hard at this...

[Luckless]

Actually, keep going. Your estimate describes the old IBM PC XT (8086), which ran around 4MHz. A modern USB thumb drive has more processing ability than the Apollo systems had. iPhone isn't even comparable. I'd say many tens of thousands of times more powerful by comparison.

Yeah, your camera embedded in any modern phone can crunch far more numbers than all the computers on the Apollo capsules.

The phone itself has can crunch more numbers than all the computing NASA had access to at the time.

[Wallace]

Gimbal lock is a problem for programming too, if you use Euler rotations x,y,z , you can reach situation where they will not combine properly due to gimbal lock, to avoid it mostly Quaternions s,x,y,z (tho confusingly some programming languages have them x,y,z,s ) are used which have very nice ways of combining and also of rotating vectors, but I digress, the problem is, humans have a hard time intuitively understanding quaternions, if you see the Euler rotation, 90,0,0 you can probably tell it's 90 degrees off in the X axis, so you can easily visualise that, and tell that the top of the object would now be facing right, but with quaternions, it is pretty much incomprehensible, so you have to convert from quaternion rotations to Euler rotations for the stupid humans to understand it and input their controls and that's where the problems occur.

Also I heard that washing machines that have more computing power than the Apollo capsule, and I heard that one in the 90's before they started adding the silly little dot matrix screens on appliances.

Edited November 3, 2013 by Wallace

[jebisanaggie]

Its amazing to think of what a modern day capsule would look like and how (relatively) easy it would be to fly it.

[hobbsyoyo]

Did you know the Gemini spaceship had a FOUR axis version that apparently didn't get into Gimbal Lock like the Apollo did?

In some ways the Gemini capsule was more advanced than Apollo as work on it started after work had already begun on Apollo by a year or two. Also, the company that built the Mercury capsule built the Gemini capsule, but not the Apollo capsule. This meant that the company already had experience in building them and put that to good use making Gemini whereas the contractor building Apollo didn't have that.

Edit: Crap I didn't see there was already this post:

I'm sure I read somewhere that the Gemini capsule was more advanced than the Apollo CSM in many ways. The capsule systems were far more modular, so components were much easier to access and swap in and out during assembly and testing (and presumably in flight if required). Whereas the wiring on Apollo was a real cats-cradle that was much harder to work with.

Edit: Yet another post I should have read, lol:

That's correct, Gemini had a four axis guidance platform that could not get into gimbal lock, the why is because of quirk in space history... most people don't realize that Apollo was designed *before* Gemini.

Edited November 3, 2013 by hobbsyoyo

[Creideiki]

The flight control computers for the Apollo and Gemini spacecraft are very "neat" machines. While true that the Gemini capsule was more advanced than the Apollo capsule for several reasons, least of which was actually asking the astronauts what they think the capsule should have, its flight computer was not as advanced as the main flight computers of the Apollo capsules.

Of minor interest, the Apollo missions -- that carried a LEM -- actually had four flight computers. Two Apollo Guidance Computers (the AGC), one Abort Guidance System (AGS), and one Launch Vehicle Digital Computer (LVDC). One AGC was located in the CM, the second AGC and the AGS were located in the LEM; the two AGC's had different sets of program in their ROM, of course. The LVDC is often ignored, but it was the computer that actually flew the Saturn V (and Saturn IB) rockets. The AGS and LVDC were quite similar to the Gemini OBC actually.

While many people assert that the Apollo missions were done with less computing power than a modern calculator... I would like to state, categorically, that saying that is very wrong. While true that the spacecraft does have less computing power than a graphing calculator (take my TI-89 Titanium for instance; uses a Moto 68K processor, with 256KB of RAM and 4MB of flash memory for firmware), when you add the ground computers to the mix, you'd probably get away with something like an IBM 5150 (the PC) or an IBM 5160 (the PC/XT), without any of the reliability.

It's interesting comparing and contrasting the architecture's of NASA's Apollo and Gemini computers with the computers that were available at the time. At the time of the Apollo and Gemini programs in the 1960s, there were roughly two "types" of computer system one could get a hold of: large systems (large as in terms of size, and usually capabilities; please note also, "large" does not necessarily equal "mainframe"), and minicomputers (which means a computer which instead of taking up a room with all of its parts, can usually fit into one or more 19" equipment racks; minicomputers are not, and never will be, considered mainframes). Large systems were machines like the IBM 709x series (1959), Burroughs B5000 (1961), IBM System/360 (1964; the defining example of a mainframe, period), and Digital Equipment Corporation PDP-10 (1966). Minicomputers were machines like the Digital Equipment Corporation PDP-8 (1965), Varian Data Systems 620i (1966), Hewlett-Packard HP2100 (1966), MIT's Lincoln Labs LINC (1961; the design is public domain, most were built by DEC), and Honeywell DDP-516 (1966; used in 1969 as the central machine of the "Interface Message Processor", the first (internet) router).

Note, the two lists of computers I made were machines either contemporaneous to the Apollo and Gemini programs, or were used in the Apollo and Gemini programs; NASA had several IBM 7094s, and eventually purchased IBM System/360 machines, NASA actually kept a 7094 up and running at least until the completion of Apollo 11, since not all of their software had been ported to their 360s.

The AGC has an instruction set that is superficially similar to that of the DEC PDP-8, although in depth the instruction set architecture of the AGC is quite different, mostly in regards to memory addressing and I/O device interface. The PDP-8 has a 12-bit word length with 3-bit instruction field and 9-bit addressing field (1-bit direct/indirect reference flag, 1-bit page 0/current page reference flag, 7-bit page address) for the six memory reference instructions, the I/O instruction is 3-bit instruction, 6-bit device address, 3-bit device flags, and the last instruction is 12-bits of "microcoded" instruction; the AGC has a 16-bit word length, with 3-bit instruction field, and 12-bit addressing field (2-bit bank selection field, 10-bit bank address). For I/O the PDP-8 has a specific I/O instruction to make use of its input/output devices (and certain internal processor options); while the AGC on the other hand uses memory mapped I/O space, like the later DEC PDP-11 system, however the way the AGC works it does provide specific I/O instructions because the I/O channel memory area overlaps with the permanently resident writable core.

The astute, assembler literate, programmer will notice I have yet to mention general registers for either the AGC or PDP-8 instruction sets. This is because they don't have them. Both machines are "accumulator machines" with one register to which functions are applied (the accumulator); that is an instruction which adds two values would start with the first value in the accumulator register, and then reference the memory location of the second value with the summed value being located in the accumulator at the end of it ("AD VALUE" on the AGC, "TAD VALUE" on the PDP-8, where AD and TAD are the instruction mnemonics, and VALUE is an assembler symbol referring to the address of the value). For those who do not know what the difference between a general register machine and an accumulator machine is, what follows is a code snippet of what it would take to multiply two numbers and store the high and low results, on a PDP-8 and a PDP-11.

On a PDP-8 (accumulator based) machine the instructions would look like:

[table=width: 480, class: grid]

[tr]

[td]

        CLA CLL         /CLEAR AC AND LINK
        TAD    VEL      /LOAD AC WITH VELOCITY
        MQL             /LOAD MQ WITH VELOCITY, CLEAR AC
        MUY             /MULTIPLY VELOCITY BY TIME
TIM,    (0              /TIME VALUE, DEFINED AS ZERO, MODIFIED DURING EXEC
        DCA    HDIST    /DEPOSIT HIGH-ORDER BITS OF DISTANCE INTO MEMORY
        CLA MQA         /PUT LOW-ORDER BITS OF MULTIPLICATION INTO AC
        DCA    LDIST    /DEPOSIT LOW-ORDER BITS OF DISTANCE INTO MEMORY

Of note; the PDP-8 multiply instruction (MUY) works by multiplying the MQ register by the value in the next memory word after it; resuming program instruction the word after that. So it would look like: MUY --> {Data word... TIM in this case.} --> DCA... when disassembled.[/td]

[td]

        MOV     #VEL,R0  ;Put the velocity into register 0
        MUL     R0,#TIM  ;Multiply the velocity (in R0) by the time
        MOV     R0,#LDIST ;Move low-order bits of result to low-order value of distance
        MOV     R1,#HDIST ;Move high-order bits of result to high-order value of distance

[/td]

[/tr]

[/table]

The AGC, OBC, AGS, and LVDC were programmed similar to the PDP-8 example on the left, while the general register concept on the right only applied to the NASA's System/360 machines, the IBM 7094s that NASA had were accumulator machines.

I mentioned earlier that the AGS, OBC, and LVDC were accumulator machines, they had their own programming paradigms. The AGS was an 18-bit machine whose programming method was very close to that of a PDP-8. The OBC conversely, had a much stranger architectural paradigm where each word in memory was 39-bits, broken into three 13-bit "syllables." During flight syllables 0 and 1 were writable, while syllable 2 was read only, with the result that never changing code would be stored in syllable 2, while data and loadable code would get put in syllables 0 and 1. Instruction format was 4-bit opcode followed by 9-bit operand field which depended on the instruction; the normal OBC data format was 26-bits wide, so one piece of data and one instruction could be stored in one 39-bit memory word. The LVDC was similar to the OBC in its architectural paradigm, with 28-bit wide memory words holding 14-bit "syllables" (13-bits of each being used for information, the last being a parity bit); instructions were 13-bits wide with a 4-bit instruction field and 9-bit operand; data was a full 26-bits wide. Both LVDC and OBC had paged memory like the PDP-8, with those instructions whose operands referred to memory locations having one bit dedicated to selecting whether the eight-bit address was on the current memory page, or on the "residual" (always accessible) page.

Addenda:

One set of computer architectures which the Gemini OBC and the LVDC were quite similar to were the Autonetics D-17B, D-37C and RECOMP computers. The D-17 and D-37 were -- and in the case of the D-37, I believe it still is -- used as guidance computers for Minuteman ICBMs. The RECOMP and RECOMP II were deskside computer systems which had a similar architecture.

Something which some of you may find mildly comical, at the end-of-life of the D-17B Minuteman I computers, instead of scrapping them, the military put them up for surplus. Which resulted in several universities getting their hands on one and putting them to use as slightly odd laboratory computer systems.

I could probably ramble on for longer, but I don't think one would find the nitty-gritty details of computer architecture to be all that interesting.

Edited November 13, 2013 by Creideiki
Added extra content.

[3_bit]

Ah, the good old Apollo Guidance Computer! That thing did a lot of stuff, from moving telescopes to controlling engines... I know there's a video from the 60's about it's production and use floating around on YouTube.

[Geschosskopf]

It's interesting comparing and contrasting the architecture's of NASA's Apollo and Gemini computers with the computers that were available at the time. At the time of the Apollo and Gemini programs in the 1960s, there were roughly two "types" of computer system one could get a hold of: large systems (large as in terms of size, and usually capabilities; please note also, "large" does not necessarily equal "mainframe"), and minicomputers (which means a computer which instead of taking up a room with all of its parts, can usually fit into one or more 19" equipment racks.....

Thank you for all the great info an nostalgia. I wrote my 1st programs on punch cards in IBM System/360 Assembler using macho EBCDIC instead of lame ASCII .

But I must offer 1 very slight correction to the above statement. There was at that time a 3rd type of computer: analog. Geez, I love analog computers. The sheer genius required, in what now seems a totally alien field to modern understanding...... Know any good stories about them, too?

[3_bit]

Thank you for all the great info an nostalgia. I wrote my 1st programs on punch cards in IBM System/360 Assembler using macho EBCDIC instead of lame ASCII .

But I must offer 1 very slight correction to the above statement. There was at that time a 3rd type of computer: analog. Geez, I love analog computers. The sheer genius required, in what now seems a totally alien field to modern understanding...... Know any good stories about them, too?

I hate to say it, but I'm jealous you were around when that kind of stuff was big. The best I can do right now is programming a COSMAC Elf.

[doggie015]

Apollo 13 came dangerously close to it after their not-so-minor event. Thankfully they were able to avoid it as they had no way to tell legitimate stars from debris from the event and as such they would be flying blind and unable to tell which way they are pointing. You can review the 765-page air-to-ground transcript as well as the 930 page mission commentary transcript at http://www.jsc.nasa.gov/history/mission_trans/apollo13.htm

[Creideiki]

Ah, the good old Apollo Guidance Computer! That thing did a lot of stuff, from moving telescopes to controlling engines... I know there's a video from the 60's about it's production and use floating around on YouTube.

There are a few videos, I think. There are also videos from the rather... annoying... people who claim the moon landing was a hoax, and use their total non-comprehension of the technologies involved to say the AGC never worked. (It did work, quite well. And even if they don't believe it flew a man to the moon, they can't say it didn't work since the AGC was also used as a fly-by-wire computer in a US military test air plane.)

Thank you for all the great info an nostalgia. I wrote my 1st programs on punch cards in IBM System/360 Assembler using macho EBCDIC instead of lame ASCII .

Fun fact, the IBM System/360 actually supports both ASCII and EBCDIC. The bit in the PSW which controlled ASCII/EBCDIC mode would, in the System/370 and later, end up becoming the extended PSW mode bit (you trade ASCII mode for 31-bit memory addressing; I think that's a good trade).

The '360 did a lot of nice things. Count-Key-Data DASD devices were, and are, very interesting. Plus, you just gotta love how a big System/360 setup looks (yes, their blinkenlights are a part of it, but it's amazing when you realize "I'm not sitting AT the computer, I'm sitting IN the computer!"

Since you mentioned being a mainframe programmer, there's actually a software package out there that'll let you emulate any IBM mainframe from System/370 on upwards (yep, even a z/10). OS wise, OS/360 is free, so is the 3.8J version of MVS, and there's a few distributions of VM/370 as well. The more modern OSes, and the program products (CICS, IMS, ISPF, DB2, the language compilers from MVS and upwards except the assembler, et cetera) are not available... well except by special means (you can run z\OS and z\VM if you can get the CDs, though technically you shouldn't because they are tied to hardware and blah...).

But I must offer 1 very slight correction to the above statement. There was at that time a 3rd type of computer: analog. Geez, I love analog computers. The sheer genius required, in what now seems a totally alien field to modern understanding...... Know any good stories about them, too?

Ah yes, analog computers. They're quite neat, even if somewhat forgotten. I didn't really count them in my post as I was referring to digital computers.

I don't know if you'll count it as an analog computer, but my physics professor -- I'm still a university student in his 20s -- still makes use of a slide rule, in class, instead of the pocket calculators everyone else has. (I use a TI-89 Titanium as my calculator... Mmm do I love its CAS (CAS = Computer Algebra System).)

But yes, if you include analog computers, and slide rules, you'd probably see an analog computer or two in the backroom with the engineers, slide rules for all of the engineers, and I'm going to guess that there was at least two slide rules in the Apollo spacecraft themselves.

[lazarus1024]

Thank you for all the great info an nostalgia. I wrote my 1st programs on punch cards in IBM System/360 Assembler using macho EBCDIC instead of lame ASCII .

But I must offer 1 very slight correction to the above statement. There was at that time a 3rd type of computer: analog. Geez, I love analog computers. The sheer genius required, in what now seems a totally alien field to modern understanding...... Know any good stories about them, too?

What is more amussing is our agency STILL uses EBCDIC on our mainframes.

Good old z/OS.

So when we take stuff in we have to convert from ASCII to EBCDIC.

Sigh.

[csiler2]

That's correct, Gemini had a four axis guidance platform that could not get into gimbal lock, the why is because of quirk in space history... most people don't realize that Apollo was designed *before* Gemini.

The Apollo CSM was originally born as a general purpose earth orbiter, sort of a supersupersized Mercury. When Kennedy made the call to go to the Moon, it was uprated for the lunar mission and eventually the LM added to the package. (Originally the CSM itself was going to land on the moon.) That's why there were Block I and Block II command modules - Block I was the original version, and Block II the upgraded version. Because it was born as a general purpose orbiter, not only did they not think to install a 4-axis gimbal system (virtually a requirement for docking), the Block I _didn't even have a docking system_. Like the oversized propulsion system on the Service module, the gimbal was one of the things left 'as-is' because changing it was a hassle and at the time (this is still the early 60's) they didn't realize it (the gimbals) would be a problem.

Gemini came about as a bit of a kludge... NASA officialdom realized that not only would Apollo take years to develop, but that it's expense and complexity meant fewer flights would be possible. Together this meant that not only would their be a politically unacceptable gap in flights between Mercury and Apollo, but also that it would be hard to get the flight experience and testing needed before heading off to the moon while also meeting the "end of the decade" deadline. So they flailed about a bit until they discovered an unsolicited proposal from McDonnell (it wasn't yet McDonnell Douglas) for a Mercury MK II - a two man Mercury orbiter. This proposal was accepted, and rather quickly became Gemini. Since rendezvous and docking was part of the program goals, a four axis gimbal system was part of the specifications...

(voice = "Paul Harvey") And now you know... the rest of the story. (/voice)

That's not entirely accurate. The Apollo CSM was always designed for lunar missions, the major debate was in regards to the mission profile; Direct Ascent, Earth Rendezvous, or Lunar Rendezvous. The command module design started in 1961 with an assumed mission mode of direct ascent and was too far along in development to stop Block 1 production before NASA settled on LOR in mid-1962. Instead of wasting Block 1, NASA decided to use it for unmanned test flights (after the Apollo 1/AS-204 fire). Once it became clear that NASA's budget would be rolled back, the designers began considering alternate uses for the Apollo CSM. Google Apollo applications program for more info. Despite rollbacks, NASA had high enough hopes that a manned Venus flyby was considered, but eventually Skylab was the result, and NASA started considering alternatives to expendable vehicles. By the time we got to the Apollo/Soyuz flight, NASA had decided to ditch expendable launch systems and pinned their hopes on the reusable space shuttle.

What's most exciting (for me) is that the US got very lucky when it came to the space race. The initial specifications for what would eventually become the Saturn family of launchers were discussed as early as 1958, over a decade before we landed on the moon! The military mentioned that it might want a super-heavy lifter in its arsenal, and Von Braun's team started making initial proposals, but the program was on the rocks, kept alive through Von Braun's force of will until Kennedy made a moon landing NASA's top priority. He even threatened to quit if Saturn development was cancelled. Check out "Stages to Saturn" if you want to learn more about the Saturn family than you ever knew you didn't know.

[LethalDose]

so what you're saying is that kerbals are actually pretty realistic?

bwahahahahahahahaha!!!! Love it!