Rear vew mirror - how to?

> No, i don't think so, i mean that it wouldn't be a MIRROR image, ie
> not reversed, as a "rear view mirror" should be.

This could be done with a browser, that would support a negative
Transform.scale as extension for mirroring. The only browser with this
extension i know is the covise/cover VRML browser 8-(

so long
MUFTI
--
Chatetiquette
.... Wenn für einen Moment Sie lassen müssen, Sie müssen "BRB" eingeben
(Sie sein rechte Rückseite)
Micro$oft Knowledge Base Artikel-ID: 178294 Version: 3.1

On 22 Jul, 10:47, Joerg Scheurich aka MUFTI
<rusmu...@helpdesk.bera.rus.uni-stuttgart.de> wrote:
> > Any idea how to go to implement a "rear view mirror"?
> > I mean, something that simply looks and works like the usual rear vew
> > mirror in the average cars game.
> > Ideally, I am looking for a pure VRML97 solution that does not involve
> > embedding the same world twice in my web page.
>
> The immersive VRML97 browser cover/covise has such a node as unportable
> Browser extension.
>
> http://www.hlrs.de/organization/vis/covise/


That's cool stuff, but the software engineer in me has to cry: when
will you guys get the whole point of standardization?

It's just a mess, even worse than 10 years ago. At least at that time
we had Cosmo Player to do basic (but both standard and reliable!)
things, and for the rest we simply knew the technology was not yet
ready.

Today the situation is simply ridiculous.


>
> You need to ask for permission to use it...


I won't, but thanks anyway (speaking a bit rethorically here). I'm
busy trying to build web based 3d applications where *any user* can
just browse to my page and it will work, because I am a bloody
brilliant programmer, knowing all the relevant standards and writing
bloody first class code always... so no, I won't ask for permissions:
what for?

-LV


>
> so long
> MUFTI
> --
> Chatetiquette
> ... Wenn für einen Moment Sie lassen müssen, Sie müssen "BRB" eingeben
> (Sie sein rechte Rückseite)
> * * * * * * Micro$oft Knowledge Base *Artikel-ID: 178294 Version: 3.1

>
> This could be done with a browser, that would support a negative
> Transform.scale as extension for mirroring. The only browser with this
> extension i know is the covise/cover VRML browser 8-(
>
> so long
> MUFTI

tried that years ago, and was really surprised to find the standard
specifically precludes such a useful feature, but i did find from
experimentation that Cortona does it OK, but that Contact gets the
lighting angles reversed on the mirrored geometry, which looks very
strange

BTW these were both in VRML, i suppose X3D might be more picky.

> > > Any idea how to go to implement a "rear view mirror"?

> > The immersive VRML97 browser cover/covise has such a node as unportable
> > Browser extension.

> That's cool stuff, but the software engineer in me has to cry: when
> will you guys get the whole point of standardization?

AFAIK, there is no similar function in the standards. The guys of cover/covise
simply need this feature (for a car simulation made from a scrap Porsche
car running in a 4 wall cave).
Should they wait 10 years, till someone set a standard for this, before they
may implement this ?

Please read

http://www.web3d.org/x3d/specificati...xtensions.html

| Annex F(informative)

| Recommendations for non-normative extensions

The VRML97 Standard advices browser builder how to build nonstandard
extensions, so why should this annex not to be uses ?

> It's just a mess, even worse than 10 years ago.

There are simply more features, that can be done now.
Some have been simply ignored by the X3D specifications for years.

A good example are realtime shadows.
In 1997, this was too hard for the 100000 Dollar 200 MHz Risc desktop crown
machines of this time.
5 years later, the power of this 100000 Dollar machine was beaten by a
100 Dollar Nvidia gforce graphicscard.

Today, we have at least 2 implementations of simple real time shadows in
VRML related players: kambi and blax(x)un cc3d (in cc3d since 5 years ?)

http://vrmlengine.sourceforge.net/im...een_demo_4.png
http://www.bitmanagement.de/develope...irectional.wrl

but AFAIK nothing in the standards 8-(
At least i did not heared anything about shadows in the next version of X3D 8-(

> > You need to ask for permission to use it...

> ... so no, I won't ask for permissions:
> what for?

cover/covise is also a commercial product, the builtin VRML browser is only
a minor part of it.

so long
MUFTI
--
ar -r ../lib/libswt.a swt.o browser.o tree.o header.o groupbox.o
statusbar.o toolbar.o util.o resload.o
ar: ../lib/libswt.a is a fat file (use libtool(1) or lipo(1) and ar(1) on it)
MacOSX errormessage

On 24 Jul, 15:29, Joerg Scheurich aka MUFTI
<rusmu...@helpdesk.bera.rus.uni-stuttgart.de> wrote:
> > > > Any idea how to go to implement a "rear view mirror"?
> > > The immersive VRML97 browser cover/covise has such a node as unportable
> > > Browser extension.
> > That's cool stuff, but the software engineer in me has to cry: when
> > will you guys get the whole point of standardization?
>
> AFAIK, there is no similar function in the standards. The guys of cover/covise
> simply need this feature (for a car simulation made from a scrap Porsche
> car running in a 4 wall cave).


I was not at all complaining with the guys at cover/covise, that
should have been clear enough, although of course it wasn't.

The point maybe some of you guys maybe have not yet realized is: there
is a ***super-massive*** potential market share for VRML/X3D on the
***World Wide Web***, just waiting to be exploted. Along with that
there is also the whole business and financial software markets! And
then actually most anything one can think of. It just should work!

So, you are elaborating about non-normative extensions and real-time
shadows, and I am digging the web for a browser plugin that works
enough and is enough available to the end-user.

How many "billions" are we missing? Can you see my point now? Working
applications! Software!! This thing, VRML, has got potentials way
beyond design. Billion ways beyond.

-LV


> Should they wait 10 years, till someone set a standard for this, before they
> may implement this ?
>
> Please read
>
> http://www.web3d.org/x3d/specificati...772-VRML97/par...
>
> | Annex F(informative)
>
> | Recommendations for non-normative extensions
>
> The VRML97 Standard advices browser builder how to build nonstandard
> extensions, so why should this annex not to be uses ?
>
> > It's just a mess, even worse than 10 years ago.
>
> There are simply more features, that can be done now.
> Some have been simply ignored by the X3D specifications for years.
>
> A good example are realtime shadows.
> In 1997, this was too hard for the 100000 Dollar 200 MHz Risc desktop crown
> machines of this time.
> 5 years later, the power of this 100000 Dollar machine was beaten by a
> 100 Dollar Nvidia gforce graphicscard.
>
> Today, we have at least 2 implementations of simple real time shadows in
> VRML related players: kambi and blax(x)un cc3d (in cc3d since 5 years ?)
>
> http://vrmlengine.sourceforge.net/im...shadow/lili_sh...
>
> but AFAIK nothing in the standards 8-(
> At least i did not heared anything about shadows in the next version of X3D 8-(
>
> > > You need to ask for permission to use it...
> > ... so no, I won't ask for permissions:
> > what for?
>
> cover/covise is also a commercial product, the builtin VRML browser is only
> a minor part of it.
>
> so long
> MUFTI
> --
> ar -r ../lib/libswt.a swt.o browser.o tree.o header.o groupbox.o
> statusbar.o toolbar.o util.o resload.o
> ar: ../lib/libswt.a is a fat file (use libtool(1) or lipo(1) and ar(1) onit)
> * * * * * * * * * * * * * MacOSX errormessage