Old BUG, new version?

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 06/01/2003 at 10:09, xxxxxxxx wrote:

User Information:
Cinema 4D Version:   8.012 
Platform:      
Language(s) :     C++  ;

---------
I have followed the SDK documentation. I have spent days trying to figure this one out. The same problem was in v7.303 and I believe it was Philip Losch who said it was a bug and it would be addressed. It has seemingly not been addressed and remains a roadblock to us finishing our renderer. This is a very serious issue and should be addressed ( as well as the NULL objects being removed from various SendModelingCommands ). 
I would like the source code to be compiled and the issue to be acknowledged ( again ).
http://www.bhodinut.com/forum/viewtopic.php?p=2464#2464
David Farmer - bhodiNUT

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 08/01/2003 at 02:16, xxxxxxxx wrote:

"[mime type text/plain was deactivated by an board admin, therefore this attachment is not displayed]" - It's a good thing you reported this in three places...
I've tried the code and can confirm that it seems hard to create hard edges when any kind of antialiasing is on. I couldn't locate the previous correspondence you refer to. (If possible, try to give a date and or excerpt from such old mails.)
My guess is that the reason that your square isn't perfectly hard is that the filters look on more than the current pixel. Try for example the Cone filter at 0% for hard edges. (But that'll destroy the rest of the AA.)
I'll post an internal question if it's possible to use ExecuteLine() to apply effects after the antialiasing pass.

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 08/01/2003 at 02:28, xxxxxxxx wrote:

Oh, and as to your hidden comments in the .cpp-file:
* Again I couldn't find any correspondence that indicate that the NULL behaviour would change. As back then the recommendation still is to use an empty polygon object (PolygonObject::Alloc(0,0)) instead for these structural NULLs. (Too much depended on this collapsing of NULL hierarchies internally.)
* As to CalcHardShadows() : yes, please do try it out! All accounts I have say that it should have been fixed. (I don't know the issues well enough to confirm it myself.)

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 08/01/2003 at 03:10, xxxxxxxx wrote:

I just got the reply that there's no direct way to write non-AAed data in ExecuteLine(). However, a suggested workaround is to do this: store the line values into a buffer and write it in Execute() using your own filter functions into the image.
Hope this helps!

THE POST BELOW IS MORE THAN 5 YEARS OLD. RELATED SUPPORT INFORMATION MIGHT BE OUTDATED OR DEPRECATED

On 08/01/2003 at 09:19, xxxxxxxx wrote:

I could have done that all along. It just makes the result look unprofessional. Aack, seems like a hack. Not something I would push upon paying users. VELMA shelved... 
>
> **** 
>
> Quote: Originally posted by Mikael Sterner on 08  January 2003
>
> * * *
>
> I just got the reply that there's no direct way to write non-AAed data in ExecuteLine(). However, a suggested workaround is to do this: store the line values into a buffer and write it in Execute() using your own filter functions into the image.
>
> Hope this helps!
>
> * * *