Jaime Vives Piqueres schrieb:
>   Very good modeling... can I guess you did come up with the rad-proximity
> hack just for this? ;)

Indeed. You gotta have dirt on these things - *lots* of dirt - and it 
must match the detail level of it, so it's probably a safe bet that a 
df3-based approach is prohibitive. Memory consumption should be ok (some 
300 MB of raw data should give me a resolution of 0.5 inch, which should 
suffice), but generating that data with a trace()-based SDL script... 
*DO NOT WANT!* >_<

>   I'm trying to remember the name of the guy that years ago modeled some
> very detailed and accurate trains with POV-Ray CSG... well, I give up... my
> memory is usually good for names, but this time I had to google it: it was
> David S. Merchant :
> 
>   http://www.nesys.com/images/pov/3d_berk.html

Duh. Not bad, just about what I'm going for (except that it's a 2-8-4, 
not a 2-8-2, but my choice was just due to what blueprints I could find 
on the net for free).

Though this work of his appears to be actually even closer to what I 
/really/ want:

http://www.irtc.org/irtc/irtc?_n&pg=ViewSubmission&id=StillImages_May-June1999_giants.jpg

Unfortunately I don't have blueprints of such a giant. And even more 
unfortunately, Dave didn't publish the SDL code... *DO WANT!* >_<

Post a reply to this message