🚀 go-pugleaf

In article <C1CI80.Iq@taligent.com> Jack Palevich <Jack_Palevich@Taligent.com> writes:
>I bought the Macintosh verson of Playmation, and I'm not entirely
>satisfied with it.
>
>Playmation is available on the Amiga, the Macintosh, and Windows. It's
>originally an Amiga program, and it's kept its Amiga style user interface
>as it's moved from machine to machine.

   The Windows version is not exactly Windows standard either, but that
does allow easier support.

>Pros:
> * Cheap! $300, plus another $100 if you want the not-yet-released 24-bit
>renderer.
> * Superb character animation features. You can use animated motions as
>subroutines. You can edit the individual surface definition points. You
>can use a "spine" to quickly animate complex objects.
> * Nice incremental redraw in modeler.
> * Spline surfaces. No more jaggy polygons.
> * Supports QuickTime
> * Comes with lots of sample models and animations.
> * Comes with a tutorial video.
> * Files are ASCII, and they provide limited documentation of the file
>format on their BBS system.

   This one is a bigger pro than you know. ASCII files will allow
movement from system to system. Model on a Mac, Animate on an Amiga
and render on a number of PCs.

>Cons:
> * Weird user interface. Doesn't work like a Mac program. Limited Undo.

    Different user interface. Most parts of the Mac paradime don't
apply. Most 3D programs have *no* undo at all.

> * Comes in five stand-alone pieces. (Model objects, apply surface
>attributes, combine objects into actors and design animation subroutines,
>combine actors, render.)

  This is a good thing. Interfaces are optimized for function, and the 
size of the executables is smaller. Really, I am quite happy with a
modular program as it forces one to adopt good work habits, saving your
work when you exit each module. Also, most modules allow "updating"
i.e. create the animation, fine-tune the objects later, and select
update to load the new objects from disk or the running modeler.

> * Crashes all the time. (This is the first release; quality may improve
>in time.)

   The Windows version is fairly robust.

> * Slow renderer. Only allows ray-tracing.

   A very fast ray-tracer, but I do want a scan-line option.

> * Only renders 320 x 200 x 8, which is pretty crude by Mac standards.

    The 24-bit renderer should not be considered an option.

    In my experience the 24-bit renderer is pixel-per-pixel faster than
the screen renderer. In the DOS version, everything is done in Windows
except 24-bit rendering which is done at the command line. It's uses
a DOS extender. As near as I can tell, all rendering is done 24-bit
and quantized to 8-bit for display. Actually it's a true 32-bit
renderer, and I've been able to alpha-channel matte images over TARGA
backgrounds.

                                Chris Williams
                               chrisw@fciad2.bsd.uchicago.edu

In article <1993Jan24.093443.6267@midway.uchicago.edu> chrisw@fciad2.bsd.uchicago.edu (chris williams) writes:
|In article <C1CI80.Iq@taligent.com> Jack Palevich <Jack_Palevich@Taligent.com> writes:
|>Cons:
|> * Weird user interface. Doesn't work like a Mac program. Limited Undo.
|
|    Different user interface. Most parts of the Mac paradime don't
|apply. Most 3D programs have *no* undo at all.

Weird or different, it's still not as good.  Node selection is not a natural
extension of typical point selection techniques, but a completely different and
nonintuitive offshoot.  Spline curvatures are edited by nonintuitive curvature
control parameters attached to the nodes, and when nodes are used as control
points for multiple splines, many of the curvature parameters are lost to the
editor.  (Sure, you can get at them by editing the ascii model file, but what a
pain.)  There is pointer based control of these curvature parameters, but it's a
crude second order effect (e.g., you move the cursor back and forth, and the
curvature parameter is modified over SOME range, which shows up as some one or
two axis flex of the curve.
________________________________________________________________________________
Robert Reed			Home Animation Ltd.		503-656-8414
robert@slipknot.rain.com	5686 First Court, West Linn, OR 97068

When I'm really bored, I hope to drive downtown, get a great parking spot,
then sit in my car and count how many people asked me if I'm leaving.
--Steve Wright
________________________________________________________________________________

 
Robert Reed writes:
>chris williams writes:
>|Jack Palevich writes:
>|>Cons:
>|> * Weird user interface. Doesn't work like a Mac program. Limited Undo.
>|
>|    Different user interface. Most parts of the Mac paradime don't
>|apply. Most 3D programs have *no* undo at all.
 
>Weird or different, it's still not as good.  Node selection is not a natural
>extension of typical point selection techniques, but a completely different
>and nonintuitive offshoot. 
 
   Intuitive in the real sense, or "intuitive" is the Mac-buzzword sense?
Press the "g" key for group. Press "Return" to un-group all. About two minutes
to get it, tops.
 
>Spline curvatures are edited by nonintuitive curvature control parameters
>attached to the nodes, 
 
   Complain to Ed Catmull. Catmull-Rom splines behave differently than the
Bezier splines that you are used to. The advantages of these splines for
easy objects creation are fairly obvious. Bezier patches are difficult to
control when you want to create a mesh, while continuity between Catmull
patches is simple. B-Spline patches have continuity, but require "ghost
points" outside the object.
 
>There is pointer based control of these
>curvature parameters, but it's a crude second order effect (e.g., you 
>move the cursor back and forth, and the curvature parameter is modified
>over SOME range, which shows up as some one or two axis flex of the curve.
 
   I have built *many* objects, and have *yet* to use the curve controls.
I suspect that you are trying to force the Catmull-Rom splines to behave as
Bezier patches, and are getting frustrated. Read up on Catmull-Rom patches.
(...and why you shouldn't put <tab>'s in an ASCII forum)
 
>________________________________________________________________________________
>Robert Reed                    Home Animation Ltd.             503-656-8414
>robert@slipknot.rain.com       5686 First Court, West Linn, OR 97068
>
>When I'm really bored, I hope to drive downtown, get a great parking spot,
>then sit in my car and count how many people asked me if I'm leaving.
>--Steve Wright
>________________________________________________________________________________
 
                             Chris Williams
                              chrisw@fciad2.bsd.uchicago.edu

 
In article Robert Reed writes:
>In article chrisw@fciad2.bsd.uchicago.edu (chris williams) writes:
>|In article  Jack Palevich <Jack_Palevich@Taligent.com> writes:
>|>Cons:
>|> * Weird user interface. Doesn't work like a Mac program. Limited Undo.
>|
>|    Different user interface. Most parts of the Mac paradime don't
>|apply. Most 3D programs have *no* undo at all.
 
>Weird or different, it's still not as good.  Node selection is not a natural
>extension of typical point selection techniques, but a completely different
>and nonintuitive offshoot. 
 
   Intuitive in the real sense, or "intuitive" is the Mac-buzzword sense?
Press the "g" key for group. Press "Return" to un-group all. About two minutes
to get it, tops.
 
>Spline curvatures are edited by nonintuitive curvature control parameters
>attached to the nodes, 
 
   Complain to Ed Catmull. Catmull-Rom splines behave differently than the
Bezier splines that you are used to. The advantages of these splines for
easy objects creation are fairly obvious. Bezier patches are difficult to
control when you want to create a mesh, while continuity between Catmull
patches is simple. B-Spline patches have continuity, but require "ghost
points" outside the object.
 
>There is pointer based control of these
>curvature parameters, but it's a crude second order effect (e.g., you 
>move the cursor back and forth, and the curvature parameter is modified
>over SOME range, which shows up as some one or two axis flex of the curve.
 
   I have built *many* objects, and have *yet* to use the curve controls.
I suspect that you are trying to force the Catmull-Rom splines to behave as
Bezier patches, and are getting frustrated. Read up on Catmull-Rom patches.
(...and why you shouldn't put <tab>'s in an ASCII forum)
 
>________________________________________________________________________________
>Robert Reed                    Home Animation Ltd.             503-656-8414
>robert@slipknot.rain.com       5686 First Court, West Linn, OR 97068
>
>When I'm really bored, I hope to drive downtown, get a great parking spot,
>then sit in my car and count how many people asked me if I'm leaving.
>--Steve Wright
>________________________________________________________________________________
 
                             Chris Williams
                              chrisw@fciad2.bsd.uchicago.edu

In article <1993Jan26.024907.1777@midway.uchicago.edu> chris williams,
chrisw@fciad2.bsd.uchicago.edu writes:
>   Intuitive in the real sense, or "intuitive" is the Mac-buzzword sense?
>Press the "g" key for group. Press "Return" to un-group all. About two
minutes
>to get it, tops.

No offence, but we just aren't going to agree on this issue -- you think
intuitive means "easy to learn, and powerful once learned", while I think
it means "doesn't require learning". Try this test: Get five novices,
hand them the program and a list of tasks to do, and sit back and see if
they can do the tasks without _any_ prompting from you or from the
manual. Some of the best Mac user interfaces (Finder, ClarisWorks)
actually pass this test. I somehow doubt Playmation would. Often the
changes required to make an interface pass my definition of intuitive are
quite small, and do not require much recoding (nor much relearning on the
part of the veteran users). The trick is to be really honest about the
test -- it's no fair prompting the novices to "help" them over the rough
spots.

> 
>>Spline curvatures are edited by nonintuitive curvature control
parameters
>>attached to the nodes, 
> 
>   Complain to Ed Catmull. Catmull-Rom splines behave differently than
the
>Bezier splines that you are used to. The advantages of these splines
for...

I beg your pardon. I don't deal with Bezier curves -- I'm used to
genuine, grade-a NURBs, thank you very much! :-)