[LSC] File Format Restrictions for Official Parts

[Michael Heidemann]

-Precision
I think the first attempt with 3 (4 for hi-res) decimal will be ok. If 
you calculate you have much more decimals and if I look into most parts 
it seems to be a consens in the past. So I like to keep this.

-Coplanarity
Philo is right, the algorithm is not needed here if specified what we 
want exactly. So the suggestion from Travis is fine.

- identical lines
There should be no identical lines at all that starts with another value 
than 0.
Also the hint from Travis for linetype 5 and 2 is important.

mikeheide


Philo schrieb:

> William Howard wrote:

>

>> Another version uploaded to the scratch pad.

>>

>> Still a work in progress, so feel free to rip it to bits

>>

>> W

>

>

> - Precision "In general, three decimal places are sufficient for parts"

> My rule of thumb for pieces is to use only 2 decimal place for 

> position in linetype 1 or for apex coordinates (other line types), and 

> 3 decimal places for rotation/scaling matrixes. I have not seen any 

> ill effect so far...

>

> - coplanarity tolerance [[Need an algorithm here]]

> I don't think an algorithm is needed in specification (should be in a 

> separate article). I would simply say that "this angle is the angle 

> between normals to these triangles" which is the mathematical 

> definition... You may improve my approximative english ;o)

>

> - identical lines: "Ordering of points is not considered when checking 

> for equality." This is not perfectly clear for me. Are identical 

> linetype1 allowed? why?

>

>

> Philo

> _______________________________________________

> LSC mailing list

> LSC at ldraw.org

> http://five.pairlist.net/mailman/listinfo/lsc

>