[OSRS-PROJ] da, df

[Clifford J Mugnier]

Don't blame the GPS manufacturers; just remember recent history (or
ancient, depending on how old you are), and recognize the fact that GPS is
a military system, and GPS has ALWAYS been a military system.  (Lots of
opinions on the future.)

Decades ago (3) when I was a Engineer Company Commander, the first
electronic satellite system was the Army SECOR.  It compactly fit into two
semi-trailer trucks!  All processing was done back at Army Map Service in
D.C.  That was done on 300 bpi 9-track tapes and punch cards.

That was followed in the early 70's with the Navy TRANSIT system, and the
first portable receivers were the Magnavox 1502 units.  These units fit
into a large suitcase (plus automobile batteries, 1 per 8-hour session),
and had a single-board computer.  With some VERY skillful programming, the
1502's could compute geocentric X,Y,Z coordinates from observations and
store differences in the ellipsoid parameters from the TRANSIT system's
ellipsoid.  Any other ellipsoid used would be a difference from that basic
defining ellipsoid.  This was sufficient to get local Datum Latitude and
Longitude with accuracy to support coordinated (no pun intended) artillery
fire and air to ground operations.  The accuracy was good enough for the
blast radius of large ordinance.  That system was later de-classified in
1976, and everybody and his brother picked up on the "defacto standard" of
the Magnavox 1502.

That "standard," or convention, or custom continues today because that's
the way it started.  It may seem silly now, but that's why it exists and
will probably continue to exist.

All of it from way back in the early 1970's.

Prof. Clifford J. Mugnier (cjmce at LSU.edu)
Surveying, Geodesy, & Photogrammetry
LOUISIANA STATE UNIVERSITY
2408 CEBA Building
Baton Rouge, LA  70803
Voice & Facsimile: (225) 578-8536
========================================
SEE: http://www.ASPRS.org/resources.html
         http://www.ce.LSU.edu/~mugnier/
========================================

I didn't mean to be obscure and was only commenting on the d f/a parameters
and the factor that it appears that the *reason* that they used this novel
form
was that fewer bytes of storage were required to save the list of ellipsoid
parameters
in this form as compared to the full specification of each f and a.

20-30 years ago, when computer memory was a scarce resource we would go
to great lengths such as this differential method but I would not bother
now.

If there is any criticism, it is solely directed against the coders of the
GPS recievers.

"Felipe G. Nievinski" wrote:

> On 04/08/01 at 20:19 Gerald I. Evenden wrote:
> >Remarkable.  The lengths some will go to to save such a trivial number
of bytes.
> >
> >I used to do this back in the 60's when memory was dear, but now???
> >
>
> Some days ago there was a doubt in this list about what the da and df
parameters mean. I found an explanation (from the U.S. NIMA) and sent it
hoping it could help someone.
>
> Gerald, I'm sorry, English is not my mother tongue, and I just did not
understand what you meant. :((
>
> Felipe.
>
> >"Felipe G. Nievinski" wrote:
> >
> >> FYI (about the DA and DF parameters)
> >>
> >> The DA and DF are the delta values between the ellipsoid used by
WGS-84 and another given ellipsoid:
> >> da = [a(WGS 84 ellipsoid) - a(second ellipsoid)] | a = semi-major
(equatorial) axis
> >> df = [f(WGS 84 ellipsoid) - f(second ellisoid)*10000] | f = flattening
> >>
> >> They are commonly used by NAVSTAR-GPS receivers, which shall already
know the WGS-84 model, and therefore needs less space to store just the
delta values for the others Earth models than the a and f values
themselves.
> >> For example, the Finnish ellipsoid can be defined by:
> >> da = -251 m
> >> df = -0.14192702
> >
> >    ...
> >
> >----------------------------------------
> >PROJ.4 Discussion List
> >See http://www.remotesensing.org/proj for subscription, unsubscription
> >and other information.
>
> ----------------------------------------
> PROJ.4 Discussion List
> See http://www.remotesensing.org/proj for subscription, unsubscription
> and other information.

----------------------------------------
PROJ.4 Discussion List
See http://www.remotesensing.org/proj for subscription, unsubscription
and other information.


----------------------------------------
PROJ.4 Discussion List
See http://www.remotesensing.org/proj for subscription, unsubscription
and other information.