RE: [cgmo-webcgm] second NURBS test

[Lofton Henderson <lofton@rockynet.com>]

Rob,

I tend to agree with you about the niceness of having some real-world 
tests.  But I also think that the ratio of nurbs/other content is too low 
-- MetaCheck shows 74 NURBS in the 348K file, and 2048 Polylines of varying 
lengths.

If people want this one in the test suite, I'll try to tidy it up and make 
it NURBS03.  But personally I'd rather see a real-world test that has 
higher proportion of NURBS content.

-Lofton.




At 04:26 PM 3/3/2006 -0700, Robert Orosz wrote:
>Lofton,
>
>The NURBS02.cgm file is a good test.  I would definitely keep it.
>
>Regarding the Catia file, I have some thoughts.  First, I believe that it is
>useful to have some "real-world" examples in a test suite.  Second, the fact
>that CAD data often contain NURBS was the justification for including NURBS
>in WebCGM 2.0 in the first place, so why not have a file produced by a CAD
>system in the test suite?
>
>Having said that however, it would be nice if the NURBS/other ratio were a
>little higher.  I guess that I could go either way on that one.
>
>Rob
>
>-----Original Message-----
>From: Lofton Henderson [mailto:lofton@rockynet.com]
>Sent: Thursday, March 02, 2006 9:00 AM
>To: cgmo-webcgm@lists.oasis-open.org
>Subject: [cgmo-webcgm] second NURBS test
>
>
>I have integrated and uploaded the proposed second NURBS test (from
>Forrest) to FTP.  It consists of the three files:
>
>NURBS02.cgm
>NURBS02.htm
>NURBS02.png
>
>Personally, I like this as our second NURBS test.  As the attached
>MetaCheck trace shows, it contains 28 NURBS, all of have a significant
>amount of data.  It also contains filled NURBS (in a Closed Figure).  I
>think the only potential criticism would be "you can do that with
>Beziers".   But we're not assessing and arguing the requirement here (reqt.
>is agreed), we're demonstrating and testing it.
>
>By comparison, I have also uploaded the earlier proposal, that we discussed
>yesterday -- a Catia output (from Dave).  In 340K of CGM data, it contains
>only a half-dozen or so NURBS.
>
>NURBS02-dc.cgm
>NURBS02-dc.htm
>NURBS02-dc.png
>
>Btw, I know that two viewers pass the proposed new NURBS02 test.
>
>Thoughts?
>
>-Lofton.