United3dArtists Archive

Since few months, a number that often appears is "millions of polygons" . Whouaa! A number that dream many people. The real feat is to do better but simpler. but this is not the target of marketing today. About all these render engine, they make you think you can do everything. You must know that the scenes in the demos require much work and talent. we do not become a virtuoso in buying a Stradivarius. It is urgent to wait whether a standard will appear in a few years. In the meantime, enjoy your tools and use your money for the holidays.

Millions of polygons of garbage is still garbage.
I just don't see the fire so I am in no rush to bale out.
I'm still having fun!

we just see to live in a "must have" kind of world, it doesn't really matter if you need to buy a new graphics card, the software which is as risky as playing on the stock market, then they'll blame the computer coz it just isn't working right and buy another one just in case.. at the end of the day if you can't get a pretty picture with your present software spending 3 grand+ aint gonna help!
Its kinda like the folks who did the complaining over at the cali forums because one bug totally disabled TS for them for some strange reason and were jumping off the cliff like lemmings when it never came, I'm having a good week if TS doesn't crash at least twice lol, its part of its character

I think is better I enjoy my vacations instead producing scenes even that while in school my productivity goes down...
Another think I want to clarify is that a good render can make a sphere or cube looks real when a normal render make a super model looks fake or not good at all, that's the difference.

Leon:
A curious thing with photorealistic is the ability to reproduce the default (or artistic effect) of a real photography. for example the reality have no dof. We used to see dof in movie then a picture with dof seem more realistic whereas it should seem false. Our vision is still influenced by the ancient techniques and their shortcomings. a person who has never seen a picture would not understand the significance of this dof. but for us it is more real and we are wrong. Before the invention of photography, this notion existed only for those who had poor eyesight and now, it's a quality!

There was a guy over at the SCream forum (SCream is the panoramic game engine for Scratches) an he was trying to fake "true" sight and it was very difficult, very involved and I don't think he ever pulled it off, Its another thing I guess the human eye can't quite get to grips with even if in theory the guy got it right

The dof isn't wrong at all, we can see all things at the same time the human's eyes was made to focus something to then we be able to hunt and follow it. The landscape view is when we focus the minimum is possible, but even with the ground in sight you see the landscape not the near ground and for it those guys thought about blur it, also because cameras do it.

leonregis wrote:The dof isn't wrong at all, we can see all things at the same time the human's eyes was made to focus something to then we be able to hunt and follow it. The landscape view is when we focus the minimum is possible, but even with the ground in sight you see the landscape not the near ground and for it those guys thought about blur it, also because cameras do it.

No. the focus (for eyes) is a problem of convergence of two superimposed clear images, traduced in 3d by a blur. when you see a near object, the background is not blur, you have two superimposed clear images (or one clear image for one eye). the only blur you can have (for a good eye) is when the object is too near the eye (it is the limit of the capacity of the eye). when we do a blur on the background, it is for artistic effect to show an object, not the reality of a human vision. Also, a blur front and behind an object at the same time is not possible for a human eye. when we do a clear object and a blur behind, we are wrong. the background can seem blur because we have two clear surimposed images.
Just test this:
put your finger near your nose, what do you see? a blur finger with a clear background (due to the limit of the eye) . now try to have a clear finger (move it) with a blur background. it is not possible with a normal eye (the background is always clear). it is possible with camera or glasses. As i said, we are influenced by photography.
It is why why speak about photorealistic, not reality.

Pwned!

Sorry but when I move my finger away from my nose I then can focus on the finger
and the background is blurred.
When I look around a room everything is blurred except the area my eye focuses on. Behind, in front and to the sides are all blurred.
I won't bring up the fact we actually see upside down too but our brain flips correctly...

I think Steinie is right, I did the test and the background looked a bit blurred just if I look at it it looks good, so now that Steinie confirmed I can say it without worry about eye problems.

you are right...and wrong: the dof of the eye is only 25cm! but the mental image have no dof.
It is not easy to explain certain subtlety with bad English. my example was perhaps not good. to understand, imagine in "4 dimensions". forget the distance too short which exceeds the capacity of the eye. if you take a photo in the street with a depth of field so that the clearest point is 5 meters. before and after it becomes blurry. we can emulate it with a 3d image and dof. Yet this is not what we actually see because everything we see on the street is clear. Why?
you must imagine the focus point as a pixel of the image. to see the whole image, the eye moves and the focus changes constantly and instantaneously. the brain assembles all these points of focus to create a clear image regardless of the distance of objects. which cancels the effect of dof. when you look at an image created with a dof, it can be accurate when you look at the focus point but if the eye moves on the image, the focus point can not change and we fix a point blur when in reality it should be clear (because the focus would have changed in the reality). image represents only a fraction of a second and a step of the mental image. it is incomplete because a party is unclear (because not analyzed). this is not what the brain remembers. It have a global view with all clear point. it is like if each focus point was differents images then it assemble only the clear points to create the image without dof (it is my "4th dimension"). the camera record only one image and we see only a step of the construction of a mental image. we are so accustomed to this, it seems more real than reality.
hope you understand what i mean. maybe too intellectual?

Steinie is correct about the actual eye view.

Now it may be that some dof methods used in TS/CG graphics aren't a realistic simulation of what is happening with real light and real lenses.

Our bodies are marvelous machines built by nanotechnology (dna etc.) Our eyes are cameras. Of course the technology isn't exactly that of film or digital. A typical human sees at approximately 12mp color and 120mp brightness per eye, based on the number of sensor cells, and 60fps. Our sensor cells are concentrated more to the center of view. Our lenses are flexible.

An eye focuses on a particular distance per the brain's commands. Things not at that distance are out of focus. The farther from that distance the more out of focus.

The two images, one for each eye, is how we see 3D. While that 3D ability is a perception of depth is isn't related to focus depth.

Marcel makes a good point about the mental image and constant refocusing.

Our mental image is built from much information collected over time and is not the same as the image an eye, or both eyes, sees. Our mental image or mental surroundings map is in 4D too (with time) and includes concepts and assumptions which we apply from memory.

The constant refocusing as we pan our view around gives the illusion of everything being in focus when in any one frame it isn't.

So it is important to know if you are trying to mimic the mental image or the eye image in a TS/CG image. Simulating a camera is good for simulating the image from an eye. That is probably not good for simulating a mental image.

About the constant refocusing: it would be nice to have a screen who can know the zone you see to adapt the dof! combined with stereo it would be the perfection of the reality. maybe after hdtvhd, after stereo hdtv, the mental tv.
since a few years we can see many new render engine. that is possible to do many thing due to the progress of the hardware. but many new parameters can appears soon (as the merge of different image to create a "mental image" (in 20 years?) or the modification of the reality due to warm air, humidity who can disturb ray of light, etc..). what i said in this thread was the doubt to have a real standard soon due the numbers of uncompatible render engine. With a standard, we could work on a render engine then save the setting and material optimized for the hardware and load this on another render engine (with another card) in the same way we can import/export a 3d file from a 3d soft to another. it is difficult to have a standard when things change every year.

I see it more lmited on the other side. We here are limited to what exports and what we can use with the technology we have here at the moment.
Based upon files that we create. What we use as render engines depends on what they can read.Those render engines arent that much different from each other.
All those new technologies are fine but as long as we dont have an interface or exporter that someone takes the time to write we dont have the option.
there is a handfullof options and we all are using them ..... as realistic or unrealistic as they can get....
So we see that gpu rendering will be the future and we gain a bit of time for each frame... doesnt realy change what we are rendering does it ....

Agree with Rayman. The gpu renderer's good thing is speed. That's great for gigapixel Omnimax film frames and quick feedback, like workspace, for working. The speed will help some with some things through "rapid prototyping". But it is still a raytrace or GI or what ever and it doesn't make us better modelers or more creative.

Yes. My real problem is to find time to model and imagine my animation. I can render the animation when i sleep!
but the render engine can help you for the creation if you can try more rapid test to choose the best way. After 36 years of work with computer i saw how the life is short and what i could do with 36 years and modern hardware. but i was wrong. Even with the biggest computer of the world, i need time to create as the fruit ripens slowly. Hardware saves us time only for the duration of production. It may even become a problem for the professional creation when creation must keep pace with production. the result is that the creation seems increasingly formatted. it's just a feeling.

Life is good the only secret is how much time you need to put in work and how much in fun to then you in the end say Oh I worked too much and hadn't too much fun! or Oh I had fun too much and did nothing, futile existence of mine!

I worked too much with much fun. Now I have even more fun to work less.

Seems you had the right combination!

The inevitable thing about life is that every day you get out of bed is a good day.....

Today i have a new internet connexion. I can see my email 10 times faster. The problem is i can't read 10 times faster!

When a 2 Mpbs is fully running you can watch videos without stop it to wait to see, so 25 Mpbs makes videos load super fast like a 10 minute video load in just 30 seconds or less but you can't accelerate time...

i tested a download at 86 megabits and watch full hd tv at the same time.

one minute to download an entire long movie but it is impossible to understand the story with this speed. Someone can help me?

Marcel I even feel that you are typing faster now ..... one post right after the other... I see you got a faster internet !

Aww, I'm on 1 megabit/s. Well, it is 1.5mb/s but local phone lines will only support 1. I wouldn't want to live in a big city but it does have some advantages like getting the high tech stuff first.

Yeah and my father uses most of it also the VoiP line takes 400 kbps so we are in a similar situation and I'm not on USA so you are winning.