Sign in with
Sign up | Sign in
Your question

Advice Please: Creating And Displaying a 135 Degree Scene

Last response: in Graphics & Displays
Share
Anonymous
a b U Graphics card
a b C Monitor
February 6, 2005 11:34:03 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Can anyone tell me if it is possible to create and display a panoramic
video that encompasses a 135 degree view across three monitors?

As for creating it, if there exists any kind of wide angle lense for
something like this, I would assume that it would cause noticeable
distortion in the image.

So I'm guessing that it is more plausible to just pull back and video
the part of the scene that would have corresponded to 135 degrees, and
then crop the video by zooming in to what would correspond to 1/3rd(45
degrees-horizontal) of the scene.

This brings me to the display part.

The idea is to be able to run said video on three monitors, meaning
each monitor together would display the entire scene.

Can this be done with video,(or static images for that matter)?

Thanks a lot.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 7, 2005 7:57:32 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Searcher7 wrote ...
> Can anyone tell me if it is possible to create and display a panoramic
> video that encompasses a 135 degree view across three monitors?

It was done in your (grand)parent's era on 35mm. Look up "Cinerama"
Anonymous
a b U Graphics card
a b C Monitor
February 7, 2005 8:49:28 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Richard Crowley wrote:
> Searcher7 wrote ...
> > Can anyone tell me if it is possible to create and display a
panoramic
> > video that encompasses a 135 degree view across three monitors?
>
> It was done in your (grand)parent's era on 35mm. Look up "Cinerama"

Are you saying that the video I want can't be created with modern
equipment?

Thanks.

Darren Harris
Staten Island, New York.
Related resources
Anonymous
a b U Graphics card
a b C Monitor
February 8, 2005 12:56:42 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

In article <1107827368.212880.274730@o13g2000cwo.googlegroups.com>,
Searcher7@mail.con2.com wrote:

> Richard Crowley wrote:
> > Searcher7 wrote ...
> > > Can anyone tell me if it is possible to create and display a
> panoramic
> > > video that encompasses a 135 degree view across three monitors?
> >
> > It was done in your (grand)parent's era on 35mm. Look up "Cinerama"
>
> Are you saying that the video I want can't be created with modern
> equipment?
>
> Thanks.
>
> Darren Harris
> Staten Island, New York.

Heavens.

Of course it can.

Take 3 standard camcorders. Put them on tripods. Point and/or zoom them
to fit your 135 degree requirement. Play the resulting tapes on 3
monitors placed side by side.

Voila.

Precisely the same technique used in Cinerama which had 3
cameras/projectors sync'd to create the ultra-wide screen effect.

Not particularly cheap, but it sounds like precisely what you're trying
to accomplish.

What's the problem?
Anonymous
a b U Graphics card
a b C Monitor
February 8, 2005 10:49:22 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Searcher7 wrote ...
>
> Richard Crowley wrote:
>> Searcher7 wrote ...
>> > Can anyone tell me if it is possible to create and display a
> panoramic
>> > video that encompasses a 135 degree view across three monitors?
>>
>> It was done in your (grand)parent's era on 35mm. Look up "Cinerama"
>
> Are you saying that the video I want can't be created with modern
> equipment?

What I am saying is that it was done 53 years ago and is pretty
well documented. Easy for anyone with sufficient curiosity to
gather the learnings and BKMs (best known methods) and do it
again (or better) with modern equipment. Sounds like a neat
idea, actually. Or even in the round with 9 cameras like the
Disney people do.
Anonymous
a b U Graphics card
a b C Monitor
February 8, 2005 1:55:37 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Scala software will do just that

<Searcher7@mail.con2.com> wrote in message
news:1107750842.988838.54450@l41g2000cwc.googlegroups.com...
> Can anyone tell me if it is possible to create and display a panoramic
> video that encompasses a 135 degree view across three monitors?
>
> As for creating it, if there exists any kind of wide angle lense for
> something like this, I would assume that it would cause noticeable
> distortion in the image.
>
> So I'm guessing that it is more plausible to just pull back and video
> the part of the scene that would have corresponded to 135 degrees, and
> then crop the video by zooming in to what would correspond to 1/3rd(45
> degrees-horizontal) of the scene.
>
> This brings me to the display part.
>
> The idea is to be able to run said video on three monitors, meaning
> each monitor together would display the entire scene.
>
> Can this be done with video,(or static images for that matter)?
>
> Thanks a lot.
>
> Darren Harris
> Staten Island, New York.
>
Anonymous
a b U Graphics card
a b C Monitor
February 8, 2005 1:55:38 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

"Digital Video Solutions" wrote ...
> Scala software will do just that

Presumably you mean editing three timelines in sync?
Anonymous
a b U Graphics card
a b C Monitor
February 8, 2005 6:53:37 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

I had already thought about the three camera idea, but since the three
videos have to be seemless as a single video, attempting to get a
perfect set-up is not within my ability.

I'd like to mention that when I run it as a single video(on a single
monitor), the intent is to use the horizontal scroll to see either the
left, right, or center.

So basically, I'd have to be able to store a panoramic video, and have
the bandwidth to display it all. Bandwidth that is at least three times
the normal bandwidth needed for displaying a regular full screen video.

Thanks.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 8, 2005 9:36:53 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Searcher7 wrote ...
>I had already thought about the three camera idea, but since the three
> videos have to be seemless as a single video, attempting to get a
> perfect set-up is not within my ability.

Not clear why you were asking, then?

> I'd like to mention that when I run it as a single video(on a single
> monitor), the intent is to use the horizontal scroll to see either the
> left, right, or center.

You could just shoot "extreme letterbox" depending on how much
resolution you need at the end of the process.
Anonymous
a b U Graphics card
a b C Monitor
February 8, 2005 11:40:59 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Richard Crowley wrote:
> Searcher7 wrote ...
> >I had already thought about the three camera idea, but since the
three
> > videos have to be seemless as a single video, attempting to get a
> > perfect set-up is not within my ability.
>
> Not clear why you were asking, then?

Well, I guess this whole thing is about what hardware I'd need for the
project.

Three cameras cannot be in the same space at the same time, so that is
one reason a seemless video would be difficult to make. Another is
making sure the verticl is *perfect* between the three frames.
Horizontal can be overlapped and then edited when putting the three
frames together.

> > I'd like to mention that when I run it as a single video(on a
single
> > monitor), the intent is to use the horizontal scroll to see either
the
> > left, right, or center.
>
> You could just shoot "extreme letterbox" depending on how much
> resolution you need at the end of the process.

"Extreme Letterbox"? I assume this is an in-camera function. Does it
cover 135 degrees?

Thanks.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 9, 2005 12:29:07 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

>> I'd like to mention that when I run it as a single video
>> (on a single monitor), the intent is to use the horizontal
>> scroll to see either the left, right, or center.

> Richard Crowley <rcrowley7@xprt.net> wrote:
> You could just shoot "extreme letterbox" depending on how
> much resolution you need at the end of the process.

Or mount a 2:1 anamorphic fisheye lens on the HD cam.

My reference books are in storage, but my recollection
is that 3-strip Cinerama (35mm 6-perf x 3) was far from
a routine process, with severe limitations on original
photography (e.g. focal length), expense in post (e.g.
color matching panels), and complications in exhibition
(extracting 35mm 4-perf and 70mm 5-perf prints for
standard theatres). Many of those issues would have
congruents in a 3-cam SDTV setup.

In its latter days the Cinerama brand switched to single-strip
65/70mm anamorphic, largely, I'd guess, due to the challenges.

--
Regards, Bob Niland mailto:name@ispname.tld
http://www.access-one.com/rjn email4rjn AT yahoo DOT com
NOT speaking for any employer, client or Internet Service Provider.
Anonymous
a b U Graphics card
a b C Monitor
February 9, 2005 2:56:37 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

>> You could just shoot "extreme letterbox" depending on how
>> much resolution you need at the end of the process.

Or extreme anamorphic, assuming that 3:1 lenses even exist.

> <Searcher7@mail.con2.com> wrote:
> "Extreme Letterbox"? I assume this is an in-camera function.

No. It means throwing away vertical information in order to
obtain a wider aspect ratio (a/r). Irrespective of the angle
of coverage you're seeking, your desire to display it on three
standard 4:3 (1.33:1) displays computes to an overall image
a/r of 12:3 (4:1).

Matte off the top and bottom third of your single-lens SDTV
image, and voila, 12:3 image. Blow it up on a 3x3 video wall
that only has the middle three displays horizontally.
Resolution will be horrible.

The movie industry (esp. in Italy) did something similar in
the 1960s with Techniscope. Instead of shooting anamorphic
onto a 4-perf 35mm frame, they'd shoot spherical onto 2-perf.

Cut raw neg cost in half. Also cut the resolution delivered
to the eventual 4-perf anamorhpic theatrical prints. Film
had enough res to get away with this. SDTV doesn't, really.

Going anamorphic would at least not waste any of the basic
SDTV res. Anamorphic to HD (assuming 16:9 native) would
require less squeeze (an additional 2.25:1 vs. 3:1) and would
deliver more res into the post production workflow vs SDTV.

> Does it cover 135 degrees?

Coverage angle for a single-lens capture is a function of
the lens focal length. On a 35mm film camera, that view
would be at the borderline of "fisheye" ultra wide angle
lenses, I'm guessing around 12mm, which are very expensive.

Caveat: the above is actually more than I know about the topic.

--
Regards, Bob Niland mailto:name@ispname.tld
http://www.access-one.com/rjn email4rjn AT yahoo DOT com
NOT speaking for any employer, client or Internet Service Provider.
Anonymous
a b U Graphics card
a b C Monitor
February 9, 2005 7:02:35 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Bob Niland wrote:
> >> You could just shoot "extreme letterbox" depending on how
> >> much resolution you need at the end of the process.
>
> Or extreme anamorphic, assuming that 3:1 lenses even exist.

Obviously there would be too much distortion invlolved, so a single
lense is not an option. :-(

> > <Searcher7@mail.con2.com> wrote:
> > "Extreme Letterbox"? I assume this is an in-camera function.
>
> No. It means throwing away vertical information in order to
> obtain a wider aspect ratio (a/r). Irrespective of the angle
> of coverage you're seeking, your desire to display it on three
> standard 4:3 (1.33:1) displays computes to an overall image
> a/r of 12:3 (4:1).
>
> Matte off the top and bottom third of your single-lens SDTV
> image, and voila, 12:3 image. Blow it up on a 3x3 video wall
> that only has the middle three displays horizontally.
> Resolution will be horrible.

Yes I already mentioned the possibilty oif soing this. But if the
resolution is poor, this wouldn't be an option either.

> The movie industry (esp. in Italy) did something similar in
> the 1960s with Techniscope. Instead of shooting anamorphic
> onto a 4-perf 35mm frame, they'd shoot spherical onto 2-perf.
>
> Cut raw neg cost in half. Also cut the resolution delivered
> to the eventual 4-perf anamorhpic theatrical prints. Film
> had enough res to get away with this. SDTV doesn't, really.
>
> Going anamorphic would at least not waste any of the basic
> SDTV res. Anamorphic to HD (assuming 16:9 native) would
> require less squeeze (an additional 2.25:1 vs. 3:1) and would
> deliver more res into the post production workflow vs SDTV.

Non-linear editing on undistorted source material(which at least *good*
resolution) is my goal. So this is a brick wall.

> > Does it cover 135 degrees?
>
> Coverage angle for a single-lens capture is a function of
> the lens focal length. On a 35mm film camera, that view
> would be at the borderline of "fisheye" ultra wide angle
> lenses, I'm guessing around 12mm, which are very expensive.
>
> Caveat: the above is actually more than I know about the topic.

Okay, so here are my conclusions...

1) I'll have to find a single lens(with the least distortion). Since my
hobby is 35mm(film) photography, I wouldn't know what the optimal
lens(focal length) for a video camera would be.

2) The video camera would have to be set up to pivot perfectly at the
CCD plane.

This would probably require an extremely(custom) tripod head and a lot
of trial and error.

Now if I'm successful, all that would be left is merging the three or
more video together to get the three "horizontal-view" "screen-lengths"
I need.

In the mean time, what about bandwisdth considerations? And idea what
kind of card would handle a full screen video(X 3)?

I assume that the display buffer would be important, and a large amount
of ram on the video card.(But correct me if I'm wrong).

Thanks a lot.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 10, 2005 10:07:58 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Scala software can, from a single source computer, display three different
scenes to three different monitors simultaneously. http://www.scala.com/ I
used their software to display 24 different works of art to 24 different
screens in random to each screen in time and art piece displayed. It works
with video sources as well.

"Richard Crowley" <rcrowley7@xprt.net> wrote in message
news:110hnv41je2f9d2@corp.supernews.com...
> "Digital Video Solutions" wrote ...
> > Scala software will do just that
>
> Presumably you mean editing three timelines in sync?
Anonymous
a b U Graphics card
a b C Monitor
February 12, 2005 5:37:55 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Digital Video Solutions wrote:
> Scala software can, from a single source computer, display three
different
> scenes to three different monitors simultaneously.
http://www.scala.com/ I
> used their software to display 24 different works of art to 24
different
> screens in random to each screen in time and art piece displayed. It
works
> with video sources as well.

Third party software is not an option, since it will have to be written
from the ground up.

But any advice on what kind of video card I'd need would be
appreciated.

Thanks a lot.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 8:50:56 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Digital Video Solutions wrote:
> Write what from the ground up. Scala can do it, and yet you say the
thing
> has to be written from the ground up. It's like trying to teach a pig
to
> sing with some people.

I have no idea what you are talking about.

But this project involves having not only the ability to create
seemless, unseen, and and immediate changes at the press of a button
with nothing else but the video on the screen, it also involves some
on-the-fly editing changes that won't be doable in existing software.
And it will have to be run out of ram disks(with no visible GUI).

So this really comes down to hardware(Speed and bandwidth) that will
allow me to do what I want.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 9:49:49 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

<Searcher7@mail.con2.com> wrote ....
> I have no idea what you are talking about.
>
> But this project involves having not only the ability to create
> seemless, unseen, and and immediate changes at the press of a button
> with nothing else but the video on the screen, it also involves some
> on-the-fly editing changes that won't be doable in existing software.
> And it will have to be run out of ram disks(with no visible GUI).
>
> So this really comes down to hardware(Speed and bandwidth) that will
> allow me to do what I want.

The limitations that you have revealed (only after being offered
good suggestions), and the seemingly conflicting requirements
leave many of us dubious that you can do what you are seeking.
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 10:18:59 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Richard Crowley wrote:
> <Searcher7@mail.con2.com> wrote ....
> > I have no idea what you are talking about.
> >
> > But this project involves having not only the ability to create
> > seemless, unseen, and and immediate changes at the press of a
button
> > with nothing else but the video on the screen, it also involves
some
> > on-the-fly editing changes that won't be doable in existing
software.
> > And it will have to be run out of ram disks(with no visible GUI).
> >
> > So this really comes down to hardware(Speed and bandwidth) that
will
> > allow me to do what I want.
>
> The limitations that you have revealed (only after being offered
> good suggestions), and the seemingly conflicting requirements
> leave many of us dubious that you can do what you are seeking.

I have no idea what you are talking about.

The original goal was a request for info on how to create a
distortion-free video scene large enough to be displayed across three
monitors. The feedback I got help me determine that the only plausble
way to do this would involve a single camera as opposed to three, that
must be set up to pivot at it's CCD plane.

The poster who kept pushing "Scala software" force me to give a reason
why that wasn't plausible, but that had nothing to do with the original
post.

So what "conflicting requirements" are you referring to.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 11:50:37 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Searcher7 wrote ...
> The original goal was a request for info on how to create a
> distortion-free video scene large enough to be displayed
> across three monitors. The feedback I got help me determine
> that the only plausble way to do this would involve a single
> camera as opposed to three, that must be set up to pivot at
> it's CCD plane.

By "video" we are assuming you mean moving images (not
snapshots). We also don't know what your concept of "quality"
is. Doing what you say with a single camera, even a very very
expensive one, would give very limited quality not worth viewing
on three screens, even small ones. At least IMHO. Perhaps not
in yours?

I don't see anything in the record that indicates that three cameras
cannot be used. It is likely done several times a month by scores
or even hundreds of people. If you are saying that you can't
afford to use three cameras, then your project seems to be
stillborn.

How would you propose shooting a 3-panel video with a
single camera by "pivoting at the CCD plane"? That is a valid
solution for shooting still pictures of stationary things: landscapes,
interiors, etc., but not suitable for anything moving, and certainly
not for shooting video.

> So what "conflicting requirements" are you referring to.

Maybe you can explain this statement of yours...

> "Three cameras cannot be in the same space at the same
> time, so that is one reason a seemless video would be
> difficult to make."

It appears that you summarily rejected the most commonly-
used solution without even understanding it. What do YOU
mean by "same space at the same time"?

And then you said...

> "Another is making sure the verticl is *perfect* between
> the three frames. Horizontal can be overlapped and then edited
> when putting the three frames together."

Perhaps you will have to explain your definition of "*perfect*"
If you have some sort of requirements that exceed those of
all the previous implementations of this method, you have not
revealed them here.
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 1:02:10 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Write what from the ground up. Scala can do it, and yet you say the thing
has to be written from the ground up. It's like trying to teach a pig to
sing with some people.


<Searcher7@mail.con2.com> wrote in message
news:1108247875.594634.32150@l41g2000cwc.googlegroups.com...
>
> Digital Video Solutions wrote:
> > Scala software can, from a single source computer, display three
> different
> > scenes to three different monitors simultaneously.
> http://www.scala.com/ I
> > used their software to display 24 different works of art to 24
> different
> > screens in random to each screen in time and art piece displayed. It
> works
> > with video sources as well.
>
> Third party software is not an option, since it will have to be written
> from the ground up.
>
> But any advice on what kind of video card I'd need would be
> appreciated.
>
> Thanks a lot.
>
> Darren Harris
> Staten Island, New York.
>
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 3:04:04 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Searcher7 wrote ...
> Three cameras cannot be used because it would make the
> three videos almost if not impossible to splice together
> *seemlessly*, due to the fact that you cannot get the exact
> same perspective, because the cameras are in three
> *different* positions.

If I understand your requirements, you don't want "exactly
the same perspective". You want three different perspectives,
each matched to the adjacent one. One to show on each of your
three screens.

Have you actually seen any multi-camera, multi-screen
productions? Cinerama was one of the first, some 50 years ago.
But there are many in use today. My favorite is the travelouge
of France at Epcot. The images are eye-popping and the music
exquisite. I have only seen it 4 or 5 times, but I never noticed
any problems at the junction between the three images which
are projected on screens several metres tall and something like
20 metres wide.

> Have you tried it?

Hundreds of people have shot dozens of successfull productions
with it.

> There will be very little in the way of movement because the
> video will be of a landscape.

If the landscape is so stationary that you can shoot your three
perspectives sequentialy (as contrasted with concurrently),
without temporal discontinuities, then likely all you need is a
series of still images.

> And what I mean by the "same space at the same time" is
> just that. One cannot position three camera in the same space
> at the same time).

Well, first of all the cameras don't occupy "the same space at
the same time". This is obviously impossible. If you have seen
the description of Cinerama, you would know that the three
cameras shoot across each other.

> This is why I will need to pivot a single camera(at the CCD
> plane) in at least three directions to get the three videos that
> have to be spliced together(seemlessly).

To paraphrase you: "you cant shoot three perspectives at the
same time with a single camera". The "fourth dimension"
(time) prevents you from doing what you propose "seamlesly".
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 7:43:07 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Richard Crowley wrote:
> Searcher7 wrote ...
> > Three cameras cannot be used because it would make the
> > three videos almost if not impossible to splice together
> > *seemlessly*, due to the fact that you cannot get the exact
> > same perspective, because the cameras are in three
> > *different* positions.
>
> If I understand your requirements, you don't want "exactly
> the same perspective". You want three different perspectives,
> each matched to the adjacent one. One to show on each of your
> three screens.

Or three different views from one perspective, depending on how you
want to say it. Spliced together all three views would cover a 135
degree perspective.

> Have you actually seen any multi-camera, multi-screen
> productions? Cinerama was one of the first, some 50 years ago.
> But there are many in use today. My favorite is the travelouge
> of France at Epcot. The images are eye-popping and the music
> exquisite. I have only seen it 4 or 5 times, but I never noticed
> any problems at the junction between the three images which
> are projected on screens several metres tall and something like
> 20 metres wide.

Yes, but this is not Cinerama. The closer the foreground objects, even
if stationary, the more difficult it is to keep everything in their
proper perspective. I have been saying that a single camera will have
to pivot "at the CCD plane". If this can be accomplished, then it
automatically keeps everything in their proper perspective. The camera
of course will not move when each of the three(or more) video frames
are being recorded.

> > Have you tried it?
>
> Hundreds of people have shot dozens of successfull productions
> with it.

Then why did you say the following?

"How would you propose shooting a 3-panel video with a single camera by
"pivoting at the CCD plane"? That is a valid solution for shooting
still pictures of stationary things: landscapes, interiors, etc., but
not suitable for anything moving, and certainly not for shooting
video."

> > There will be very little in the way of movement because the
> > video will be of a landscape.
>
> If the landscape is so stationary that you can shoot your three
> perspectives sequentialy (as contrasted with concurrently),
> without temporal discontinuities, then likely all you need is a
> series of still images.

Exactly, that is what I've been attempting to convey.

> > And what I mean by the "same space at the same time" is
> > just that. One cannot position three camera in the same space
> > at the same time).
>
> Well, first of all the cameras don't occupy "the same space at
> the same time". This is obviously impossible. If you have seen
> the description of Cinerama, you would know that the three
> cameras shoot across each other.

I saw that site. But that was the only way to do it since there is so
much movement in the scenes. As a result, there is a limit to how close
to the three cameras objects/people in the scene can be.

> > This is why I will need to pivot a single camera(at the CCD
> > plane) in at least three directions to get the three videos that
> > have to be spliced together(seemlessly).
>
> To paraphrase you: "you cant shoot three perspectives at the
> same time with a single camera". The "fourth dimension"
> (time) prevents you from doing what you propose "seamlesly".

?!? You didn't paraphrased me. I never said that.(And we've both
misspelled "Seamlessly"). :-)

Anyway, I've decided that for now, I'll just do this project for
displaying a single monitor like I mentioned. I'll just have to
incorporate a way to use the horizontal scroll to see the entire
video.(Of course, only 1/3rd of the video at any time can be seen this
way).

But I still need to find a video card that can handle this.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 7:57:35 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

> > There will be very little in the way of movement because the
> > video will be of a landscape.
>
> If the landscape is so stationary that you can shoot your three
> perspectives sequentialy (as contrasted with concurrently),
> without temporal discontinuities, then likely all you need is a
> series of still images.

Exactly, that is what I've been attempting to convey.

Actually, I wasn't attempting to convey that.(It's difficult keeping up
with all of your misunderstandings).

I am looking for a realistic effect that cannot be achieved using
"still images".

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 8:05:17 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Larry J. wrote:
> Waiving the right to remain silent, Searcher7@mail.con2.com said:
>
> > Three cameras cannot be used because it would make the three
> > videos almost if not impossible to splice together *seemlessly*,
> > due to the fact that you cannot get the exact same perspective,
> > because the cameras are in three *different* positions.
>
> I've described briefly how this is done earlier in the thread. by
> carefully controlling angle and focal length, it's been accomplished
> many, many times with good results.

What thread did you describe this in?

And does it involve scenes with close foreground objects?

Thanks.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 10:19:30 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

On 2/13/2005, Searcher7@mail.con2.com managed to type:
> Larry J. wrote:
>> Waiving the right to remain silent, Searcher7@mail.con2.com said:
>>
>>> Three cameras cannot be used because it would make the three
>>> videos almost if not impossible to splice together *seemlessly*,
>>> due to the fact that you cannot get the exact same perspective,
>>> because the cameras are in three *different* positions.
>>
>> I've described briefly how this is done earlier in the thread. by
>> carefully controlling angle and focal length, it's been accomplished
>> many, many times with good results.
>
> What thread did you describe this in?
>
> And does it involve scenes with close foreground objects?
>
> Thanks.
>
> Darren Harris
> Staten Island, New York.

Just for reference - if you are indeed going to pivot a camera to
accomplish your task, better rotate around the optical center of the
lens, not around the CCD.

Gino

--
Gene E. Bloch (Gino)
letters617blochg3251
(replace the numbers by "at" and "dotcom")
Anonymous
a b U Graphics card
a b C Monitor
February 13, 2005 11:11:33 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Gene E. Bloch wrote:
> On 2/13/2005, Searcher7@mail.con2.com managed to type:
> > Larry J. wrote:
> >> Waiving the right to remain silent, Searcher7@mail.con2.com said:
> >>
> >>> Three cameras cannot be used because it would make the three
> >>> videos almost if not impossible to splice together *seemlessly*,
> >>> due to the fact that you cannot get the exact same perspective,
> >>> because the cameras are in three *different* positions.
> >>
> >> I've described briefly how this is done earlier in the thread. by
> >> carefully controlling angle and focal length, it's been
accomplished
> >> many, many times with good results.
> >
> > What thread did you describe this in?
> >
> > And does it involve scenes with close foreground objects?
> >
> > Thanks.
> >
> > Darren Harris
> > Staten Island, New York.
>
> Just for reference - if you are indeed going to pivot a camera to
> accomplish your task, better rotate around the optical center of the
> lens, not around the CCD.

Thanks.

That saved me some trial and error time. :-)

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
March 5, 2005 1:37:35 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Searcher7@mail.con2.com wrote:
> Gene E. Bloch wrote:
> > On 2/13/2005, Searcher7@mail.con2.com managed to type:
> > > Larry J. wrote:
> > >> Waiving the right to remain silent, Searcher7@mail.con2.com
said:
> > >>
> > >>> Three cameras cannot be used because it would make the three
> > >>> videos almost if not impossible to splice together
*seemlessly*,
> > >>> due to the fact that you cannot get the exact same perspective,
> > >>> because the cameras are in three *different* positions.
> > >>
> > >> I've described briefly how this is done earlier in the thread.
by
> > >> carefully controlling angle and focal length, it's been
> accomplished
> > >> many, many times with good results.
> > >
> > > What thread did you describe this in?
> > >
> > > And does it involve scenes with close foreground objects?
> > >
> > > Thanks.
> > >
> > > Darren Harris
> > > Staten Island, New York.
> >
> > Just for reference - if you are indeed going to pivot a camera to
> > accomplish your task, better rotate around the optical center of
the
> > lens, not around the CCD.

Actually, would it be better to set it at a popint in between the
optical center and the CCD?

Thanks.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
March 5, 2005 8:05:21 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

On 3/5/2005, Searcher7@mail.con2.com managed to type:
> Searcher7@mail.con2.com wrote:
>> Gene E. Bloch wrote:
>>> On 2/13/2005, Searcher7@mail.con2.com managed to type:
>>>> Larry J. wrote:
>>>>> Waiving the right to remain silent, Searcher7@mail.con2.com said:
>>>>>
>>>>>> Three cameras cannot be used because it would make the three
>>>>>> videos almost if not impossible to splice together *seemlessly*,
>>>>>> due to the fact that you cannot get the exact same perspective,
>>>>>> because the cameras are in three *different* positions.
>>>>>
>>>>> I've described briefly how this is done earlier in the thread. by
>>>>> carefully controlling angle and focal length, it's been accomplished
>>>>> many, many times with good results.
>>>>
>>>> What thread did you describe this in?
>>>>
>>>> And does it involve scenes with close foreground objects?
>>>>
>>>> Thanks.
>>>>
>>>> Darren Harris
>>>> Staten Island, New York.
>>>
>>> Just for reference - if you are indeed going to pivot a camera to
>>> accomplish your task, better rotate around the optical center of the
>>> lens, not around the CCD.
>
> Actually, would it be better to set it at a popint in between the
> optical center and the CCD?
>
> Thanks.
>
> Darren Harris
> Staten Island, New York.

No.

Draw pictures and it should clarify this.

HTH,
Gino

--
Gene E. Bloch (Gino)
letters617blochg3251
(replace the numbers by "at" and "dotcom")
Anonymous
a b U Graphics card
a b C Monitor
March 6, 2005 8:19:29 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Gene E. Bloch wrote:
> On 3/5/2005, Searcher7@mail.con2.com managed to type:
> > Searcher7@mail.con2.com wrote:
> >> Gene E. Bloch wrote:
> >>> On 2/13/2005, Searcher7@mail.con2.com managed to type:
> >>>> Larry J. wrote:
> >>>>> Waiving the right to remain silent, Searcher7@mail.con2.com
said:
> >>>>>
> >>>>>> Three cameras cannot be used because it would make the three
> >>>>>> videos almost if not impossible to splice together
*seemlessly*,
> >>>>>> due to the fact that you cannot get the exact same
perspective,
> >>>>>> because the cameras are in three *different* positions.
> >>>>>
> >>>>> I've described briefly how this is done earlier in the thread.
by
> >>>>> carefully controlling angle and focal length, it's been
accomplished
> >>>>> many, many times with good results.
> >>>>
> >>>> What thread did you describe this in?
> >>>>
> >>>> And does it involve scenes with close foreground objects?
> >>>>
> >>>> Thanks.
> >>>>
> >>>> Darren Harris
> >>>> Staten Island, New York.
> >>>
> >>> Just for reference - if you are indeed going to pivot a camera to
> >>> accomplish your task, better rotate around the optical center of
the
> >>> lens, not around the CCD.
> >
> > Actually, would it be better to set it at a popint in between the
> > optical center and the CCD?
> >
> > Thanks.
> >
> > Darren Harris
> > Staten Island, New York.
>
> No.
>
> Draw pictures and it should clarify this.

Okay.

The reason I asked is because the light stream is narrowest and the
image inverts at that point between the lens and the CCD.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
March 6, 2005 10:30:51 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Searcher7 wrote...
> Okay.
>
> The reason I asked is because the light stream is narrowest and the
> image inverts at that point between the lens and the CCD.

This was all worked out a couple generations ago. Studing the
history of the craft may prove valuable.
Anonymous
a b U Graphics card
a b C Monitor
March 6, 2005 10:50:02 AM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Richard Crowley wrote:
> Searcher7 wrote...
> > Okay.
> >
> > The reason I asked is because the light stream is narrowest and the
> > image inverts at that point between the lens and the CCD.
>
> This was all worked out a couple generations ago. Studing the
> history of the craft may prove valuable.

I know enough history.

What I'm doing hasn't be done before. And if it was, it isn't well
documented.

Darren Harris
Staten Island, New York.
Anonymous
a b U Graphics card
a b C Monitor
March 6, 2005 9:08:33 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

On 3/6/2005, Searcher7@mail.con2.com managed to type:
> Gene E. Bloch wrote:
>> On 3/5/2005, Searcher7@mail.con2.com managed to type:
>>> Searcher7@mail.con2.com wrote:
>>>> Gene E. Bloch wrote:
>>>>> On 2/13/2005, Searcher7@mail.con2.com managed to type:
>>>>>> Larry J. wrote:
>>>>>>> Waiving the right to remain silent, Searcher7@mail.con2.com said:
>>>>>>>
>>>>>>>> Three cameras cannot be used because it would make the three
>>>>>>>> videos almost if not impossible to splice together *seemlessly*,
>>>>>>>> due to the fact that you cannot get the exact same perspective,
>>>>>>>> because the cameras are in three *different* positions.
>>>>>>>
>>>>>>> I've described briefly how this is done earlier in the thread. by
>>>>>>> carefully controlling angle and focal length, it's been accomplished
>>>>>>> many, many times with good results.
>>>>>>
>>>>>> What thread did you describe this in?
>>>>>>
>>>>>> And does it involve scenes with close foreground objects?
>>>>>>
>>>>>> Thanks.
>>>>>>
>>>>>> Darren Harris
>>>>>> Staten Island, New York.
>>>>>
>>>>> Just for reference - if you are indeed going to pivot a camera to
>>>>> accomplish your task, better rotate around the optical center of the
>>>>> lens, not around the CCD.
>>>
>>> Actually, would it be better to set it at a popint in between the
>>> optical center and the CCD?
>>>
>>> Thanks.
>>>
>>> Darren Harris
>>> Staten Island, New York.
>>
>> No.
>>
>> Draw pictures and it should clarify this.
>
> Okay.
>
> The reason I asked is because the light stream is narrowest and the
> image inverts at that point between the lens and the CCD.
>
> Darren Harris
> Staten Island, New York.

What you say is either false or doesn't make sense - I'm not quite sure
:-)

The light stream is narrowest at the optical center, and that is where
the inversion happens.

However, the above is a bit of an abstraction for today's multi-element
lenses. So the optical center is often pictured as if there were a
simple (one-element) lens substituting for the actual lens.

It even sometimes ends up being defined as the point around which you
can pivot the lens without shifting the image oddly ;-)

Once more: try drawing the conventional picture where the rays of light
from various points on the object pass through the center of a double
convex lens on the way to the corresponding points of the image on the
film and you might actually come to understand this.

Bye-bye,
Gino

--
Gene E. Bloch (Gino)
letters617blochg3251
(replace the numbers by "at" and "dotcom")
Anonymous
a b U Graphics card
a b C Monitor
March 8, 2005 2:38:42 PM

Archived from groups: comp.sys.ibm.pc.hardware.video,rec.video.desktop,rec.video.production (More info?)

Darren,

What is it you want to do? Is this just a fun project or are you trying to
develop something?

The technology exists in projection systems, it's called Jackman Media(tm)
and probably goes far beyond your expectations.

Jackman Media(tm) is to video what Dolby is to sound. It's not past history,
it is current and available technology. It is eight years of development. A
lot of it is in the public domain but much of it is still proprietary. Using
three monitors is not feasible and screen blending would be impossible
however, It is intended to be used with a single video projector and single
video source. and in is viewed seamlessly on our SharkScreen front/rear
projection screens are available in several types and configurations.
Multiple scaleable screens used are in stock aspect ratios of 32:9, 16:9 and
5:9. other ratios are also available.

Content and production is just as important as the hardware. It can be
simple static .jpg images to MPEG3, miniDV, and all the way up to Sony
CineAlta 1080i/24p high definition digital video at 40mbps with Dolby 5.1

One layout similar to your 135 degree inquiry is our 180 degree virtual
view, the main screen is normally 40" high by 144" wide with 2 flanking
right and left screens 40X72". Three camera productions provide new creative
tools for the videographer. and three views per frame really challenge the
animators.

Tell me more about your project! and where you want to go with it.

Paul Jackman
Jackman Media Group
--
Jackman Media(tm) http://www.jacktech.net

<Searcher7@mail.con2.com> wrote in message
news:1110124202.923625.205030@o13g2000cwo.googlegroups.com...
>
> Richard Crowley wrote:
> > Searcher7 wrote...
> > > Okay.
> > >
> > > The reason I asked is because the light stream is narrowest and the
> > > image inverts at that point between the lens and the CCD.
> >
> > This was all worked out a couple generations ago. Studing the
> > history of the craft may prove valuable.
>
> I know enough history.
>
> What I'm doing hasn't be done before. And if it was, it isn't well
> documented.
>
> Darren Harris
> Staten Island, New York.
>
!