Re: ANNOUNCE:GLView 2.02 a new WIN VRML OpenGL Browser

Mitra (mitra@earth.path.net)
Sat, 25 Nov 1995 16:52:18 -0800


At 6:15 AM 11/25/95, Holger Grahn wrote:
>
>> I tested it against my vrml test site -
>>http://earth.path.net/mitra/vrmltests.
>I lost the url to your page, I have tried some of Eric's tests.
>
>> It did pretty good on some of the tough network tests that fail many
>> browsers. Only failing one of these - Test 3 - which results in a
>> Redirected URL.
>Can you give me s mint how to detect this, without going to Netscape

Sure, if you are using HTTP V0.9 then just ...
telnet earth.path.net
GET /mitra/vrmltests/redirectfrom.wrl

You should see the precise format of the document you get back.
If you are using HTTP V1.0 then you'll have to figure out what the sequence
is yourself, I can't remember the exact protocol for 1.0, it generates a
return code with a specific number.
>?
>
>>
>> It failed all the texture tests (8, 9, 10, 10a) , are you supporting
>> textures in files yet?
>
>You have to explicitly enable Texture mapping and loading for *each*
>scene using the "Render->Textured" menu command.
>Reason: I`am using a 14 K Modem and don't want to wait until all
>textures have been loaded from the net.
>So I have tried the texture tests, and for me it seems to work.
>Should I provide a poup like
> "This scene contains textures, do you want to load them ?"
>?

No - I think you should always load textures if that is what the author
specified, the author has to make
the trade-off as to whether to use a lot of polygons, or to use fewer
polygons and use textures. However having a preference that a user could
chose (Always, Ask, Never) would be a reasonable compromise. You shouldn't
have to wait anyway because you can always display the untextured scene
while waiting for the textures to load, and of course the textures can be
cached as well.
>
>>
>> It failed the WWWAnchor with POINT test.
>Ok,
>>
>> It also appears to have a nasty in that almost anything sets the object
>> spinning - including it appears mouse activity in some other window.
>
>This is for the "new" user to see something is happening
>Please disable this "Autospin" in Dialog Options->Config->Autospin at
>startup, and also important is to specify a cache directory in
>"WWW->Options" dialog.

These sound like odd defaults. In fact a major grumble about this product
(though not anything which makes it non-conforming) is the number of
options, that will confuse most users substantially, and the likelihood of
them setting it up to be a "correct" VRML browser, i.e. one which presents
the intentions of the author, is fairly low.
>
>
>Thanks for testing.

Your welcome, could you send me whatever info I need to get rid of the
"unregistered Evaluation Version", I'll keep testing it as things progress.

By the way, do you want to be added to the vrml-developers mailing list,
its currently an extremely low volume mailing list - just the VAG and
developers, but it might hot up a bit if some organisational changes I
expect happen. Also I'd be interested in what your intentions are with
this product, and VRML in general. There may be room for collaboration in
various forms.

- Mitra

=================================================================
Mitra
mitra@earth.path.net voice: (415)826-2499 fax: (415)826-4423
<http://earth.path.net/mitra>

Always remember you're unique, just like everyone else.


  • Next message: Scatt: "FAQ?"
  • Previous message: Len Bullard: "PHIL: RE: Impatience"
  • Maybe in reply to: Holger Grahn: "ANNOUNCE:GLView 2.02 a new WIN VRML OpenGL Browser"
  • Next in thesad: Holger Grahn: "Re: ANNOUNCE:GLView 2.02 a new WIN VRML OpenGL Browser"