Re: Jeff, why waste the time?

From: Earl Colby Pottinger (earlcp_at_idirect.com)
Date: 03/11/05 

Date: Fri, 11 Mar 2005 11:55:02 -0600

"Jeff Findley" <jeff.findley@ugs.nojunk.com> :

> I'm not talking about SS1 here. With a better vehicle design, you
shouldn't
> need VAB's, crawlers, and huge launch pads. Reducing the size and cost of
> the ground infrastructure will reduce launch costs. Your "old school"
> launch vehicle designer is so interested in reducing the wet mass of the
> vehicle itself to the barest minimum that the ground infrastructure costs
> are ignored.
>
> A vehicle proposal like the Delta Clipper didn't need all this ground
> infrastructure. DC-X proved that this concept ought to work. Certinaly
> DC-X was small, but even for its size, the infrastructure and ground crews
> needed to launch it were absolutely *tiny* by traditional aerospace
> standards. That's the way you reduce launch costs, by looking at all your
> costs and not worrying so much about the wet mass of your vehicle.

As an Armadillo fan I would like to point out the same thing happens when
thier designs are debugged. The crew needed to setup/fuel/launch/land/and
return the craft to storage totals less than ten people. The smallest
lanuches by NASA still seem to need a team up to 100 people from start to
finish.

NASA seems to have to major problems in thier organizations. (1) Paper,
paperwork and paperwork. Tons of money is wasted filling out papers that
have nothing to do with flying the crafts. (2) Not Invented Here, NASA seems
to hate farming out work to small outside companies. Result they keep on
staff people who do not work most of the time, they ae just there for when
they are really needed, in some case this could be less than10% of their
working time.

> And at the time, the "usual suspects" in the computer industry would say
> that "that isn't a real computer" because it could never do as much as a
> mainframe computer. This is *exactly* like SS1 is today. The "usual
> suspects" in the aerospace industry think SS1 is a toy and say it's "not a
> real spaceship", because it can't get into orbit.
>
> The really important thing to note is that the customers for these "toys"
> won't necessarily be the existing customers who buy the older "real"
> products. The customers for these "toys" will largely come from completely
> new markets. In the case of personal computers (like the IBM PCs, Apples,
> Commodores, Ataris, and etc.), it was small businesses and individuals who
> could never afford a mainframe computer. In the case of manned launch
> vehicles (like the follow-on to SS1) the customers will be individuals
> looking for "the ultimate thrill", not big companies interested in
launching
> comsats. Completely different market for a completely different vehicle.
>
> Eventually though, personal computers became so powerful (in numbers) that
> they began to replace mainframes for many tasks. As these small,
> inexpensive, manned space vehicles grow in capabilities (i.e. when they can
> get into orbit), they too will begin to replace ELV's for many tasks (i.e.
> small LEO satellite launches).
>
> I'm old enough to remember the change from mainframes to peronal computers.
> I played text games on terminals attached to a minframe as a kid in the
> early 80's. As a teenager, I bought a C-64 and learned basic programming.
> At college in the late 80's, I was again using terminals to write my
> engineering programs (Fortran 77 and C) that ran on huge Unix mainframes
> (with hundreds of students sharing the same mainframe at the same time).
> But by the time I neared graduation in the early 90's, I was working on my
> senior design project on an 80386 class PC, Macintosh computers, and little
> Sun "pizza boxes" (which ran Unix).
>
> Today, I'm working on a Pentium III computer that's at least 10 times more
> powerful and costs less than 1/10 the money of my first computer cost at
> work (it was a "high end" SGI workstation that cost the company over $20k).
>
> We'll eventually see the same changes happen with manned launch vehicles,
> but it will take time since the existing markets are very small, so the
> amount of money which will be invested in the startups is also very small
> (by big aerospace company standards).

Does that take me back, by 1980 I came out of college trained to program
mainframes but entered into the microcomputer field instead. Back then if I
went to major computer conferences like the CCC I would get comments about
working on 'toys'.

One thing people use to doing things the old way have problems with is
learning to think about solving a problem diffirently. I remember one person
coming into the store to complain about his C64 'only' having 38K of useable
ram and that his pipeline program ran out of memory when he wanted the full
size array.

For some reason he could not get out of the 'floating point' mode of thinking
and realize that the integer arrays can be treated as 'fixed point decimal'
if he wrapped the right code around the array access. I had to write him the
sample code and point out since his expected range of temperture was 35c at
the input end and he wanted to find where it dropped to 25c to locate the
next heater/pump that 10 degrees in a variable that had a value range of 65
made recording the sections to one thousand of a degree was a piece of cake.

He had been programing for years on mainframes but could not by himself
(atleast in the short term) make the jump to fixed decimal on a C64 just
because that was not a directly supported variable type.

And as for the number of UNIX people who put down the NeXT machine and Linux,
I have lost count.

Same for the number of people who told me VT terminals was all that were
needed and that GUIs were a waste.

The present Aerospace companies biggest mistakes will come when they try to
treat the new spacecrafts coming online by the standards of the old rocket
designs. Perfect example is Derek Lyons who thinks the SS1 is worthless
because it does no go into orbit. Give him a SS1 for free and he still could
not make money off it because he will try to use it wrongly, or will not even
try because it is sub-orbital and thus not a real space craft.

                  Earl Colby Pottinger 

--

Relevant Pages

  • Re: Death Sentence for the Hubble?
    ... to your vehicle because it's cheaper. ... We have to show investors through many baby steps that as launch costs drop, ... NASA may want to keep ...
    (sci.astro)
  • Re: Death Sentence for the Hubble?
    ... to your vehicle because it's cheaper. ... We have to show investors through many baby steps that as launch costs drop, ... NASA may want to keep ...
    (sci.space.policy)
  • Re: Modest Proposal - Common Interplanetary Booster
    ... sized vehicle can do much to reduce costs as well. ... The smallest vehicle I have analyzed involves building an annular ... It is attached to a flight element that has 15,384 pounds ... per launch. ...
    (sci.space.policy)
  • Re: Jeff, why waste the time?
    ... > need VAB's, crawlers, and huge launch pads. ... > vehicle itself to the barest minimum that the ground infrastructure costs ... That's the way you reduce launch costs, ... In the case of personal computers (like the IBM PCs, Apples, ...
    (sci.space.policy)
  • Re: Modest Proposal - Common Interplanetary Booster
    ... Recall, that high specific impulse on orbit, translates to more ... sized vehicle can do much to reduce costs as well. ... It is attached to a flight element that has 15,384 pounds ... per launch. ...
    (sci.space.policy)
Quantcast