Subject:
|
Quality of authored parts
|
Newsgroups:
|
lugnet.cad.dev
|
Date:
|
Fri, 26 Feb 1999 18:02:00 GMT
|
Viewed:
|
945 times
|
| |
|
|
Perhaps this would be a good time to have a discussion of standards for
authoring new parts--separate from any particular parts or authors.
How realistic do you think parts ought to be in order to be considered for
inclusion in the official L-CAD updates?
Here are some of the guidelines I have picked up from the list over the last
9 months:
-Details at least 1 ldu in size should be modelled. Measurements should be
accurate to 1 ldu, but most importantly all connection points on a part
should line up with standard parts.
-There are some standard exceptions to this rule: undersides of studs, the
gripping protrusions on the inside of bricks, raised "LEGO" design on studs,
raised part numbers, groove along bottom of tiles (though this is currently
under review), "lips" on the end of Technic pegs (though IMO this should be
fixed).
-Looks are what is important. When the part is used in Ldraw, Ldlite,
POV-Ray, it should look as close as possible to the real thing. Details
like coincident surfaces are not a problem.
-Exceptions to this rule include the filling of the inter-brick space of
real Lego bricks, and areas of a part which are very unlikely to ever see
the light of day, such as the insides of compound parts like the battery
boxes, inflatable boat, rack winder.
-Type 2 and 5 lines are important to add wherever an edge exists. Type 5
lines may be omitted on concave surfaces, but watch out for this on
transparent parts.
-When creating a patterned part, the author should remove the patterned
face, add in the pattern, then fill in the space around the pattern.
Stickers present a problem in this regard, as they do lie on top of the
part, so they should be modeled as separate parts which could be placed on
other parts.
Some problems with the current "standards":
-Current parts do not always meet these standards.
-Some compound curves and intersections of curves are extremely difficult to
model to these standards, so best approximations can be used when the
"looks" are good enough.
-A tension will always exist between those who want to uphold standard of
"as good as we can possibly make them" -v- a standard of "good enough to
make it obvious what part this is". Under the first standard, we get
better, but probably fewer parts than under the second.
What do others think? (Try not to get personal.)
-John Van
|
|
Message has 3 Replies:
 | | Re: Quality of authored parts
|
| John VanZwieten wrote in message ... (...) Thank you, John, you have boiled it all down to its component parts. I can think of no reason not to follow any of your points. That does include the "don't get personal" aside. Well done. Bravo Zulu Roy (25 years ago, 27-Feb-99, to lugnet.cad.dev)
| | | Re: Quality of authored parts
|
| (...) <snip> (...) Rightly or wrongly, I am definitely an "as good as we can possibly make them" person. I'm sorry if this offends others. Chris (25 years ago, 27-Feb-99, to lugnet.cad.dev)
| | | Re: Quality of authored parts
|
| (...) Sometimes 'gripping extrusions' are drawn, sometimes not. Depends on the size of the extrusion. I agree about the lips. Why don't you try one, and see how it looks? Steve (25 years ago, 1-Mar-99, to lugnet.cad.dev)
|
7 Messages in This Thread:
  
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
