🚀 go-pugleaf

I posted this to comp.sys.sgi a couple weeks ago, but there were no replies,
either from SGI folks or from knowledgeable Explorer users.  Now that this
newsgroup is up and running, I'll repost it here.  Briefly stated, Explorer
seems to misbehave when rendering on an Indigo/XS without the z-buffer.  Is
this a limitation (if so, where is it documented?), or is it a bug (in other
words, should Explorer be smart enough to recognize that there is no z-buffer
and do its rendering differently)?

---------------

A couple months ago I posted the following:

| We've started playing with Explorer on a couple of our systems, and have
| noticed some differences between behaviour on the Indigo/XS compared with
| the 4D/xxG and 4D/xxTG.
|
| If I drag the "simple" map from the modules list to the map editor window,
| the object which appears in the "render" window (for lack of a better word,
| I call it "the flesh-coloured curvaceous cube-like thing with holes through
| it") undergoes various types of weird distortions when rotated in different
| directions.  For the most part, it looks OK.  But there are some positions
| where the object appears to have been attacked by an egg slicer (there are
| grey planar cut-like marks through part of it, parallel to one of the faces).
| Other positions produce a distorted perspective, in that the closer face (or,
| at least, the one which appears to be closer) seems smaller, as if it ought
| to be the more distant face.
|
| But my favourite distortion is an Escher-like object which consists of what
| seems to be two of the parallel faces, connected at one corner only, and each
| having one of its edges passing through the other face's hole!  (I could send
| a compressed+uuencoded copy of this to anyone who might consider it useful to
| help track down the problem.  I considered appending it here, but decided it
| would be a rather anti-social thing to do, in view of the fact that it's a bit
| over 2000 lines.  Nonetheless, it is an interesting image.  It also exhibits
| the egg-slicer problem.)  None of these oddities are seen on our PIs.

Peter Daifuku replied:

| I'm guessing that you're running 4.0.2. There were a number of z-buffering
| bugs in that first release which exhibit the egg slicer (an apt description,
| BTW) behaviour that you describe. Contact your local sales rep for 4.0.4 or
| later software. Many other bugs were fixed after 4.0.2, so there's a good
| chance that your other problems are fixed also.

Well, I just upgraded from 4.0.2 to 4.0.4, and the problem is still there.
Recently, Terry Fong posted the following:

| I don't know how many times this has come up over the past few
| months, but if you did not purchase the hardware z-buffer option for
| your XS or XS24, you do not have ANY z-buffering. Neither the XS or
| XS24 does software z-buffering!!! Only the Entry level Indigo does. The
| Indigo Elan, however, comes standard with hardware z-buffering.
|
| I wish SGI sales people would make it clear that if you "upgrade" from
| the Entry Indigo to either an XS or XS24 without the hardware
| z-buffer, then you are actually losing some capability (i.e.,
| z-buffering).

Yeah, it certainly wasn't clear to *us* when we bought the XS upgrade (or
perhaps it should be called a "sidewaysgrade", since you both gain and lose
functionality), that the Indigo would no longer do software z-buffering.

So, basically, my question is this:  Given that we have an Indigo/XS without
the z-buffer option, is Explorer still supposed to render things correctly?
I.e. should it recognize that there is no z-buffer and act accordingly?  Or
are people with "upgraded" Indigos sans z-buffer just out of luck?

If this is considered a bug in Explorer, will it be fixed in 4.0.5?  (It's
certainly not fixed in 4.0.4.)  If it's considered a restriction, perhaps it
could be documented.  (Or is it documented already?  If so, where?)

Mark Bartelt                                                    416/978-5619
Canadian Institute for                                 mark@cita.toronto.edu
Theoretical Astrophysics                               mark@cita.utoronto.ca

"Clothes not busy being worn are busy drying."    -    Dylan, on laundry day