3-D generic question

Ralph · 07-12-2003, 11:10 PM

Sitting and watching the GA when it occurred to me to wonder just what is it that makes this a 3-D and Jim's SS2 not. Both have fish swimming in front of and behind things. The GA fish spend more time swimming away or towards the screen but they do not get smaller / bigger or fade into the background. I can not move my vantage point to see behind a fish, plant or rock.
The image is on a flat 2 dimensional surface..... how can one expect to get a 3 dimensional image? Does it not ultimately boil down to illusions and tricks... How will Jim's much talked about 3-D version handle the laws of physics?
Just curious.....

Jim Sachs · 07-12-2003, 11:28 PM

Each background object in the Goldfish tank is a separate element with its own 3D coordinates. The Marine Aquarium uses a 2-dimensional background. If you ever see the Aquarium with 3D glasses, this will be instantly obvious.

Let me take a second to clear up another point. I've seen many posts from people who seem to believe that the Marine Aquarium fish do not swim front-to-back or get larger or smaller. The Aquarium is only 12" thick, so when the fish swim from the front of the tank to the rear or vice-versa, it doesn't take them very long. They would do it a lot more often if you just had one or two fish in the tank, because they need a clear shot to make it between the rocks or around the edges.

The camera is 10 feet away, so fish in the back of the tank should be 10% smaller than those in the front. I accentuated this effect by doubling it (they are 20% smaller). So lets not hear any more about the fish not changing size when they go to the rear of the tank.

As far as fading into the background, this is the much-used "fog" effect. It is much-used because it is so easy in Direct3D -- you just turn it on. It was my decision to go for crystal clear water, and not use fog. How dirty would the water have to be for the fish to fade into the background after only going 12 inches into the distance?

cjmaddy · 07-13-2003, 04:15 AM

Ralph, - Have you tried filling the tank with the max number of the same fish?
It shows up the smaller / bigger effect more clearly, ....... but they all do it!

You have a point there though, Jim's SSA IS 3D in that sense!
I think sharpness of detail has a lot to do with the illusion, compare the very 2D rock in GA, with the coral in SSA!

There used to be a good illustration that Morgan did of how the tank works ...... but it's not there now!

https://www.feldoncentral.com/forums...6901#post16901

feldon34 · 07-13-2003, 06:38 AM

Here

Ralph · 07-13-2003, 08:08 AM

Thanks.... I can visualize Morgans image but still have trouble imagining anything else......... To me its like layers in Photoshop, one can have any number of them but they have a thickness of next to nothing. What would a comparable image of the GA look like?
Obviously I am hung up on the concept....... sorry I brought it up. Sometimes it is better just to enjoy it and not wonder how it works

Jim Sachs · 07-13-2003, 09:59 AM

In the Marine Aquarium, the layers do have a thickness of zero (except the fish, which are 3D). In the Goldfish tank, the floor, rocks, etc. are all 3D models.

feldon34 · 07-13-2003, 04:54 PM

Originally posted by Ralph
Thanks.... I can visualize Morgans image but still have trouble imagining anything else......... To me its like layers in Photoshop, one can have any number of them but they have a thickness of next to nothing. What would a comparable image of the GA look like?

The fish are not flat. Just look at them. There's no way they're flat.

Just realize that they will only swim in two possible depths--Near and Far. And they can only swim between the two depths in "free" areas of the tank without coral in the way.

cjmaddy · 07-14-2003, 03:45 AM

Morgan, - May I just add a note to that last comment, - your picture illustrates my point perfectly! - The sharpness of detail, and shading, in that coral is what makes all the difference for me.
To me, the illusion of 3D in the coral image is complete, - even though it may be only 2D in reality.
The rock in GA does not 'look' 3D, - even though apparently it is! ..... Strange that! .... I can only put in down to sharpness of detail, and the shading applied!

I think perfectionists call it 'attention to detail' ... After all, all we can have on a 2D monitor screen, is illusion!

Ranger · 07-14-2003, 07:30 AM

If you press "w" while the screensaver is running you will see whats 3D and what isn't...."w" shows the wireframe on the fish, rocks, and everything just to show they are not 2D....

feldon34 · 07-14-2003, 09:25 AM

Ranger said:
If you press "w" while the screensaver is running you will see whats 3D and what isn't...."w" shows the wireframe on the fish, rocks, and everything just to show they are not 2D....

You can't go by that. There are wireframes draped over the rocks in MA and GA to generate the shadow/light effects.

The actual rocks in GA are flat, just like the coral in MA.

cjmaddy said:
The sharpness of detail, and shading, in that coral is what makes all the difference for me.
To me, the illusion of 3D in the coral image is complete, - even though it may be only 2D in reality.

It's funny you say all this since this is probably the worst example of how the coral can look. The snapshot was taken at 1280x1024 with the software scaling the background up from 1024x768 rather harshly. I wish the background could be scaled up to 1280x1024 thru a Bicubic method, sharpened, then saved in RAM until the Aquarium closes. It would add a trivial 2-3 seconds to Aquarium startup times in exchange for a smoother background.

P.S. Anyone really annoyed at the fake Z-buffering being done in MA2 now? The fins change color and the tail loses its perfectly alpha-blended look when the fish is turned away from us. I wish they'd just put it back and put a checkbox "Economize Video Memory".

cjmaddy · 07-14-2003, 11:08 AM

My fault Morgan, for not being pedantic enough. - I meant the original coral in SSA, not your picture of it! - Though I see little wrong with your 'snapshot'

That particular piece of coral, the one to the right of your picture, has always looked superb to me!

The nearest to 3D a 2D image can get! IMHO. - I am not questioning that it couldn't look better, just that it 'looks' so real, compared to the GA rock.

There are parts of the coral that I think aren't too good, - the (orangey) highest point to the left, is one that I am thinking of, to me that looks very two dimensional. - But I think the one behind the Yellow Tang (the one we lost in the Time version), and the one below it to the right, - are masterpieces!

ps, Fake Z-buffering! ..... I still use 1.1wide, - I will check it out!

FishyBusiness · 07-14-2003, 06:04 PM

The camera is 10 feet away, so fish in the back of the tank should be 10% smaller than those in the front. I accentuated this effect by doubling it (they are 20% smaller). So lets not hear any more about the fish not changing size when they go to the rear of the tank.

I think it is very obvious that the fish get smaller. I just think that they get too small. In fact, I have always thought that.

Here a good example of Jim's 2D background and a 3D fish:

Jim Sachs · 07-14-2003, 06:08 PM

Ha, ha. They look so REAL!

feldon34 · 07-14-2003, 06:36 PM

If it weren't for the glare, I could make it look quite real.

That "look" is a common mistake in 3D design. People need to properly light any 2D textures they are photographing to match the background it will be on. Once you shoot something bright, you can't darken it very much, and the reverse is true.

FishyBusiness · 07-14-2003, 07:22 PM

Unfortunately, that glare was from my flash. I wished I broguht my external flash, then I could have just lite up Elena.

cjmaddy · 07-15-2003, 03:16 AM

Nice one Frank, - took me by surprise for a moment! - thought you'd posted the wrong pic!!!

Morgan is right of course, about lighting, but I still think it's also down to shading and sharpness. - It's the shading and lack of sharpness on the space suits that doesn't match that on Elena or the background, (or the astronauts faces), that makes it jump out at you. IMO

Mixing low res enlargements with real life isn't ever going to work.

............ Now if we had really really really sharp enlargements !

cjmaddy · 08-02-2003, 12:33 PM

I'm going to stick my neck out here, and take issue with both Jim and Morgan! ......... So I'd better start praying!

This point has been going around in my mind for some time now, and when Ralph started this thread, I thought we may get the situation clarified.

Originally posted by Jim
Each background object in the Goldfish tank is a separate element with its own 3D coordinates. The Marine Aquarium uses a 2-dimensional background. If you ever see the Aquarium with 3D glasses, this will be instantly obvious.

I think it is fair to say that the vast majority of people will never see the aquarium with 3D glasses, so Ralph's original point is still valid. IMHO!
Unless the background is traversing across the front of us, allowing a 3D image to give the illusion of being able to see around the sides, then ALL images, wether 2D or 3D, will alway be relying on sharpness, shading and size difference, to complete the illusion of depth.

The descriptions given of the pending 3D background will, (I assume!), satisfy all these requirements, because it will be moving!

.... But a static 3D background will always look 2D ...... unless viewed through 3D glasses!

Originally posted by Jim
In the Marine Aquarium, the layers do have a thickness of zero (except the fish, which are 3D). In the Goldfish tank, the floor, rocks, etc. are all 3D models.

The reason why we can appreciate the 3D 'depth' to the fish, is because they are moving, and that allows us to see around the edges of the fish as they swim.

Originally posted by Morgan
The fish are not flat. Just look at them. There's no way they're flat.

In that static image, the fish ARE flat! - only when they move, will they appear to be 3D.

Originally posted by Ralph
The image is on a flat 2 dimensional surface..... how can one expect to get a 3 dimensional image? Does it not ultimately boil down to illusions and tricks... How will Jim's much talked about 3-D version handle the laws of physics?

IMO, unless it's moving, it wont! - Yes, it does ultimately boil down to illusion.
I don't think it is fair to assume that we will all sit using our computers whilst wearing 3D glasses!

Originally posted by Ranger
If you press "w" while the screensaver is running you will see whats 3D and what isn't...."w" shows the wireframe on the fish, rocks, and everything just to show they are not 2D....

Ranger, - I think not! ......
Even so, that would be implying that wire frame mode is needed to establish which is 3D! .... that can't be right!

If we need a perfect example of how a 2D image, can, with the help of lighting and shade only, still look three dimensional ..... check out Frank's pick above! ... (No, I don't mean the astronauts!)

I do hope this post is received in the way in which it is intended. ie. Not as criticism, but as a point for discussion, and to clarify the degree of 3D illusion obtainable, - and also when and where we will see it!

Jim Sachs · 08-02-2003, 01:03 PM

Cliff - You are absolutely right about being unable to tell what is 2D and what is 3D when the 2D images are not moving and 3D glasses are not used. That is exactly why I chose to make the background an economical 2D object. At the time I had not planned on ever moving the camera, and had never seen the Aquarium through 3D glasses.

Prolific developed the Goldfish tank with these things in mind, so they needed a 3D environment.

Once you have seen the Aquarium using 3D glasses, there's no turning back -- you can never really think of the background as 3D again.

Ralph · 08-03-2003, 08:10 AM

I guess I need to find some 3-D glasses. Where does one go for such a device? Are there different qualities to look for.... good ones bad ones?
The last 3-D glasses I think I saw were from a box of corn flakes.... forgot the brand

Tiny Turtle · 08-03-2003, 08:12 AM

Probably a pair of iKellogg's(tm)

07-12-2003, 11:10 PM	#1
Ralph Retired Join Date: Jan 2002 Location: Edmonton Alberta Ca Posts: 2,443	3-D generic question Sitting and watching the GA when it occurred to me to wonder just what is it that makes this a 3-D and Jim's SS2 not. Both have fish swimming in front of and behind things. The GA fish spend more time swimming away or towards the screen but they do not get smaller / bigger or fade into the background. I can not move my vantage point to see behind a fish, plant or rock. The image is on a flat 2 dimensional surface..... how can one expect to get a 3 dimensional image? Does it not ultimately boil down to illusions and tricks... How will Jim's much talked about 3-D version handle the laws of physics? Just curious.....

07-12-2003, 11:28 PM	#2
Jim Sachs Developer Join Date: Dec 2000 Location: Southern Oregon Posts: 9,791	Each background object in the Goldfish tank is a separate element with its own 3D coordinates. The Marine Aquarium uses a 2-dimensional background. If you ever see the Aquarium with 3D glasses, this will be instantly obvious. Let me take a second to clear up another point. I've seen many posts from people who seem to believe that the Marine Aquarium fish do not swim front-to-back or get larger or smaller. The Aquarium is only 12" thick, so when the fish swim from the front of the tank to the rear or vice-versa, it doesn't take them very long. They would do it a lot more often if you just had one or two fish in the tank, because they need a clear shot to make it between the rocks or around the edges. The camera is 10 feet away, so fish in the back of the tank should be 10% smaller than those in the front. I accentuated this effect by doubling it (they are 20% smaller). So lets not hear any more about the fish not changing size when they go to the rear of the tank. As far as fading into the background, this is the much-used "fog" effect. It is much-used because it is so easy in Direct3D -- you just turn it on. It was my decision to go for crystal clear water, and not use fog. How dirty would the water have to be for the fish to fade into the background after only going 12 inches into the distance? Jim Sachs Creator of SereneScreen Aquarium

07-13-2003, 08:08 AM	#5
Ralph Retired Join Date: Jan 2002 Location: Edmonton Alberta Ca Posts: 2,443	Thanks.... I can visualize Morgans image but still have trouble imagining anything else......... To me its like layers in Photoshop, one can have any number of them but they have a thickness of next to nothing. What would a comparable image of the GA look like? Obviously I am hung up on the concept....... sorry I brought it up. Sometimes it is better just to enjoy it and not wonder how it works Ludwig Wittgenstein (1889 – 1951) The limits of my language are the limits of my mind. All I know is what I have words for.

07-13-2003, 09:59 AM	#6
Jim Sachs Developer Join Date: Dec 2000 Location: Southern Oregon Posts: 9,791	In the Marine Aquarium, the layers do have a thickness of zero (except the fish, which are 3D). In the Goldfish tank, the floor, rocks, etc. are all 3D models. Jim Sachs Creator of SereneScreen Aquarium

07-14-2003, 03:45 AM	#8
cjmaddy Registered Join Date: Nov 2001 Location: Lancashire, UK Posts: 7,854	Illusions of 3D ! Morgan, - May I just add a note to that last comment, - your picture illustrates my point perfectly! - The sharpness of detail, and shading, in that coral is what makes all the difference for me. To me, the illusion of 3D in the coral image is complete, - even though it may be only 2D in reality. The rock in GA does not 'look' 3D, - even though apparently it is! ..... Strange that! .... I can only put in down to sharpness of detail, and the shading applied! I think perfectionists call it 'attention to detail' ... After all, all we can have on a 2D monitor screen, is illusion!

07-13-2003, 04:15 AM	#3
cjmaddy Registered Join Date: Nov 2001 Location: Lancashire, UK Posts: 7,854	Ralph, - Have you tried filling the tank with the max number of the same fish? It shows up the smaller / bigger effect more clearly, ....... but they all do it! You have a point there though, Jim's SSA IS 3D in that sense! I think sharpness of detail has a lot to do with the illusion, compare the very 2D rock in GA, with the coral in SSA! There used to be a good illustration that Morgan did of how the tank works ...... but it's not there now! https://www.feldoncentral.com/forums...6901#post16901

07-14-2003, 07:30 AM	#9
Ranger 1337 Gamer Join Date: Jul 2003 Location: SanAntonio Posts: 24	If you press "w" while the screensaver is running you will see whats 3D and what isn't...."w" shows the wireframe on the fish, rocks, and everything just to show they are not 2D.... "I am a man of a plan, and I will do what I can"

07-14-2003, 09:25 AM	#10
feldon34 Forum Administrator Join Date: Dec 2000 Location: Rock Hill, SC Posts: 10,939	Ranger said: If you press "w" while the screensaver is running you will see whats 3D and what isn't...."w" shows the wireframe on the fish, rocks, and everything just to show they are not 2D.... You can't go by that. There are wireframes draped over the rocks in MA and GA to generate the shadow/light effects. The actual rocks in GA are flat, just like the coral in MA. cjmaddy said: The sharpness of detail, and shading, in that coral is what makes all the difference for me. To me, the illusion of 3D in the coral image is complete, - even though it may be only 2D in reality. It's funny you say all this since this is probably the worst example of how the coral can look. The snapshot was taken at 1280x1024 with the software scaling the background up from 1024x768 rather harshly. I wish the background could be scaled up to 1280x1024 thru a Bicubic method, sharpened, then saved in RAM until the Aquarium closes. It would add a trivial 2-3 seconds to Aquarium startup times in exchange for a smoother background. P.S. Anyone really annoyed at the fake Z-buffering being done in MA2 now? The fins change color and the tail loses its perfectly alpha-blended look when the fish is turned away from us. I wish they'd just put it back and put a checkbox "Economize Video Memory". "Journalism is printing what someone else does not want printed. Everything else is public relations." - George Orwell "If voting changed anything, they'd make it illegal." - Emma Goldman Last edited by feldon34; 07-14-2003 at 10:55 AM.

07-14-2003, 11:08 AM	#11
cjmaddy Registered Join Date: Nov 2001 Location: Lancashire, UK Posts: 7,854	My fault Morgan, for not being pedantic enough. - I meant the original coral in SSA, not your picture of it! - Though I see little wrong with your 'snapshot' That particular piece of coral, the one to the right of your picture, has always looked superb to me! The nearest to 3D a 2D image can get! IMHO. - I am not questioning that it couldn't look better, just that it 'looks' so real, compared to the GA rock. There are parts of the coral that I think aren't too good, - the (orangey) highest point to the left, is one that I am thinking of, to me that looks very two dimensional. - But I think the one behind the Yellow Tang (the one we lost in the Time version), and the one below it to the right, - are masterpieces! ps, Fake Z-buffering! ..... I still use 1.1wide, - I will check it out!

07-14-2003, 06:04 PM	#12
FishyBusiness Registered Join Date: Jan 2001 Posts: 2,206	The camera is 10 feet away, so fish in the back of the tank should be 10% smaller than those in the front. I accentuated this effect by doubling it (they are 20% smaller). So lets not hear any more about the fish not changing size when they go to the rear of the tank. I think it is very obvious that the fish get smaller. I just think that they get too small. In fact, I have always thought that. Here a good example of Jim's 2D background and a 3D fish: Frank

07-14-2003, 06:08 PM	#13
Jim Sachs Developer Join Date: Dec 2000 Location: Southern Oregon Posts: 9,791	Ha, ha. They look so REAL! Jim Sachs Creator of SereneScreen Aquarium

07-14-2003, 06:36 PM	#14
feldon34 Forum Administrator Join Date: Dec 2000 Location: Rock Hill, SC Posts: 10,939	If it weren't for the glare, I could make it look quite real. That "look" is a common mistake in 3D design. People need to properly light any 2D textures they are photographing to match the background it will be on. Once you shoot something bright, you can't darken it very much, and the reverse is true. "Journalism is printing what someone else does not want printed. Everything else is public relations." - George Orwell "If voting changed anything, they'd make it illegal." - Emma Goldman

07-14-2003, 07:22 PM	#15
FishyBusiness Registered Join Date: Jan 2001 Posts: 2,206	Unfortunately, that glare was from my flash. I wished I broguht my external flash, then I could have just lite up Elena. Frank

07-15-2003, 03:16 AM	#16
cjmaddy Registered Join Date: Nov 2001 Location: Lancashire, UK Posts: 7,854	Nice one Frank, - took me by surprise for a moment! - thought you'd posted the wrong pic!!! Morgan is right of course, about lighting, but I still think it's also down to shading and sharpness. - It's the shading and lack of sharpness on the space suits that doesn't match that on Elena or the background, (or the astronauts faces), that makes it jump out at you. IMO Mixing low res enlargements with real life isn't ever going to work. ............ Now if we had really really really sharp enlargements !

08-02-2003, 12:33 PM	#17
cjmaddy Registered Join Date: Nov 2001 Location: Lancashire, UK Posts: 7,854	I'm going to stick my neck out here, and take issue with both Jim and Morgan! ......... So I'd better start praying! This point has been going around in my mind for some time now, and when Ralph started this thread, I thought we may get the situation clarified. Originally posted by Jim Each background object in the Goldfish tank is a separate element with its own 3D coordinates. The Marine Aquarium uses a 2-dimensional background. If you ever see the Aquarium with 3D glasses, this will be instantly obvious. I think it is fair to say that the vast majority of people will never see the aquarium with 3D glasses, so Ralph's original point is still valid. IMHO! Unless the background is traversing across the front of us, allowing a 3D image to give the illusion of being able to see around the sides, then ALL images, wether 2D or 3D, will alway be relying on sharpness, shading and size difference, to complete the illusion of depth. The descriptions given of the pending 3D background will, (I assume!), satisfy all these requirements, because it will be moving! .... But a static 3D background will always look 2D ...... unless viewed through 3D glasses! Originally posted by Jim In the Marine Aquarium, the layers do have a thickness of zero (except the fish, which are 3D). In the Goldfish tank, the floor, rocks, etc. are all 3D models. The reason why we can appreciate the 3D 'depth' to the fish, is because they are moving, and that allows us to see around the edges of the fish as they swim. Originally posted by Morgan The fish are not flat. Just look at them. There's no way they're flat. In that static image, the fish ARE flat! - only when they move, will they appear to be 3D. Originally posted by Ralph The image is on a flat 2 dimensional surface..... how can one expect to get a 3 dimensional image? Does it not ultimately boil down to illusions and tricks... How will Jim's much talked about 3-D version handle the laws of physics? IMO, unless it's moving, it wont! - Yes, it does ultimately boil down to illusion. I don't think it is fair to assume that we will all sit using our computers whilst wearing 3D glasses! Originally posted by Ranger If you press "w" while the screensaver is running you will see whats 3D and what isn't...."w" shows the wireframe on the fish, rocks, and everything just to show they are not 2D.... Ranger, - I think not! ...... Even so, that would be implying that wire frame mode is needed to establish which is 3D! .... that can't be right! If we need a perfect example of how a 2D image, can, with the help of lighting and shade only, still look three dimensional ..... check out Frank's pick above! ... (No, I don't mean the astronauts!) I do hope this post is received in the way in which it is intended. ie. Not as criticism, but as a point for discussion, and to clarify the degree of 3D illusion obtainable, - and also when and where we will see it!

08-02-2003, 01:03 PM	#18
Jim Sachs Developer Join Date: Dec 2000 Location: Southern Oregon Posts: 9,791	Cliff - You are absolutely right about being unable to tell what is 2D and what is 3D when the 2D images are not moving and 3D glasses are not used. That is exactly why I chose to make the background an economical 2D object. At the time I had not planned on ever moving the camera, and had never seen the Aquarium through 3D glasses. Prolific developed the Goldfish tank with these things in mind, so they needed a 3D environment. Once you have seen the Aquarium using 3D glasses, there's no turning back -- you can never really think of the background as 3D again. Jim Sachs Creator of SereneScreen Aquarium

08-03-2003, 08:10 AM	#19
Ralph Retired Join Date: Jan 2002 Location: Edmonton Alberta Ca Posts: 2,443	I guess I need to find some 3-D glasses. Where does one go for such a device? Are there different qualities to look for.... good ones bad ones? The last 3-D glasses I think I saw were from a box of corn flakes.... forgot the brand Ludwig Wittgenstein (1889 – 1951) The limits of my language are the limits of my mind. All I know is what I have words for.

08-03-2003, 08:12 AM	#20
Tiny Turtle is pleased Join Date: Mar 2001 Location: Stockholm, Sweden Posts: 7,365	Probably a pair of iKellogg's(tm) Thanks to Morgan, Tiny Snapshots is up and running again with "Tiny Järvafält" as the latest addition – Go have a look and tell me what you think. "We knew the world would not be the same. A few people laughed, a few people cried, most people were silent. I remembered the line from the Hindu scripture, the Bhagavad-Gita. Vishnu is trying to persuade the Prince that he should do his duty and to impress him takes on his multi-armed form and says, 'Now, I am become Death, the destroyer of worlds.' I suppose we all thought that, one way or another." /Robert Oppenheimer on witnessing the first thermonuclear detonation in history.