coordinate transformations formulas ... ?

[Manfred Meier]

I had written a C program and if I remember right, I typed the constants 
as integers instead of floats, that means something like
...(24*pow(cos(...
instead of
...(24.0*pow... .
And therefore the compiler did some type conversion magic.

Manfred



Janeks Kamerovskis schrieb:
> It is Rebol scripting language and there are only integers and decimals that
> is based on 64-bit standard IEEE floating point numbers. Is it enought for
> these formulas?
> Could you plese send more information about how you find solution un your
> case.
> 
> Janeks
> 
> 
>>-----Original Message-----
>>From: UMN MapServer Users List [mailto:MAPSERVER-USERS at LISTS.UMN.EDU] On
>>Behalf Of Manfred Meier
>>Sent: Saturday, February 11, 2006 6:19 AM
>>To: MAPSERVER-USERS at LISTS.UMN.EDU
>>Subject: Re: [UMN_MAPSERVER-USERS] coordinate transformations formulas ...
>>?
>>
>>Hello,
>>
>>I had a similar problem. What programming language do you use? I had
>>trouble with some double to float or double to long conversion (done by
>>the compiler) because i declared some variables not appropriate.
>>
>>Manfred
>>
>>
>>Janeks Kamerovskis schrieb:
>>
>>>Yes and I tested, that I can safely ignore those ..., because they have
>>>realy small influence on result.
>>>But the next problem is that I can get those example value for M an M0.
>>
>>They
>>
>>>are sligtly different for the given example (from EPSG quidance
>>>http://www.epsg.org/guides/docs/G7-2.pdf) it differs by some cople of
>>>thousands (5596050.46000000 5602234.40780535 and 5429228.60000000
>>>5435961.03820551)
>>>I checked many of the sources of Transverse Mercator formulas
>>
>>description
>>
>>>(EPSG of course and other) they all looked the same.
>>>I checked my implementations of formulas not only twice, but the result
>>
>>is
>>
>>>the same.
>>>
>>>Janeks
>>>P.S. Sorry about looking like offtopic - you could ask why I need to
>>>implement things that already has in proj4. I just have a case, when it
>>
>>is
>>
>>>difficult to use, but finaly data any way will come to Mapserver.
>>>
>>>
>>>
>>>>-----Original Message-----
>>>>From: UMN MapServer Users List [mailto:MAPSERVER-USERS at LISTS.UMN.EDU] On
>>>>Behalf Of Stephen Woodbridge
>>>>Sent: Wednesday, February 08, 2006 5:00 PM
>>>>To: MAPSERVER-USERS at LISTS.UMN.EDU
>>>>Subject: Re: [UMN_MAPSERVER-USERS] coordinate transformations formulas
>>
>>...
>>
>>>>?
>>>>
>>>>The ... means "and additional terms following the sequence of the prior
>>>>terms". It is basically a short hand notation for a summation of n to
>>>>infinity. So you if you want more terms you need to first infer the
>>>>sequence from the prior terms like [2, 4, 6, ...] is a sequence and the
>>>>next terms are  8, 16, 32, ...
>>>>
>>>>Each additional term adds some additional fractional accuracy to the
>>>>computation, at some point the fractional accuracy is beyond the point
>>>>that it is significant to to our uses.
>>>>
>>>>If you calculate just the 3rd term and see how big its value is it will
>>>>give you the magnitude of the change it is impacting on the equation and
>>>>the next term in the sequence will tern to be significantly smaller.
>>>>
>>>>-Steve W.
>>>>
>>>>Janeks Kamerovskis wrote:
>>>>
>>>>
>>>>>Hi list,
>>>>>
>>>>>
>>>>>
>>>>>What means those ... in the formula, that is part of coordinate
>>>>>transformations formulas?
>>>>>
>>>>>
>>>>>
>>>>>M = a[(1 - e2/4 - 3e4/64 - 5e6/256 -....)j - (3e2/8 + 3e4/32 +
>>>>>45e6/1024+....)sin2j + (15e4/256 + 45e6/1024 +.....)sin4j - (35e6/3072
>>
>>+
>>
>>>>>....)sin6j + .....]
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>brgds
>>>>>
>>>>>Janeks
>>>>>
>>>>>
>>>>>
>>>
>>>
> 
>