Image 'DEPTH' , the next hurdle.

This is total speculation but my guess is that Sony will take that crown with its Ruby replacement (whenever that may come). If other LCoS pjs like the JVC RS1 can now do 15,000:1 native, then I have to figure Sony's Ruby 2 will offer similar native CR but that they'll likely add a DI on top of that - and perhaps be able to obtain upwards of 30,000:1 with minimal brightness compression or related artifacts. Should be interesting.

Image depth is an interesting one.

image depth is also HD source + 2K-4K pj + total black room and not too big a screen....

HD gives already a ton of extra depth compared to dvd

but it depends also on the film maker ! take for instance CAPTAIN CORRELLI's MANDOLIN
or RUSH HOUR 1 or 2: lots of depth there (depth+ contrast).
KK looks often flat compared (so CGI processed).
HD video vs HD film also gives lots more depth.

When you say 'not too big a screen', do you mean one so large that it is not bright enough? Otherwise, I thought a large screen was good in 'drawing one into' the picture?

has to do with brightness also: a pj fit for a 10ft screen will be flat no punch at 15ft

I haven't seen anything that matches sim2's 3 chip offerings for depth.

I have been following this thread, but I have to confess that I don't know what image depth is.

What, exactly, is image depth?

Is is measurable?

Combination of brightness,MTF, ANSI contrast and Sequential contrast capabilites.Interesting to debate which of these are most important in achieving that.

Art

Have you seen the Sharp XV-Z20000?

At the Stockholm shootout a few weeks ago I saw the ProjectionDesign Model 3 1080. That PJ has the best image depth/3D picture I have ever seen. I think it wins hands down over Sharp z21000. Although the Sharp together with Projectiondesign model 3 are the best PJs out there at the moment. (Samsung 710 is also a great PJ with it's color and great gammacurve)

I think it's in image depth DLP wins over Lcos and LCD. If they could get higher OnOff CR and Brightness it would offcourse be even better. The Sharp z21000 is in this area great with awsome blacks and great image depth.


All the best

Shovven

Is the z20000 your new reference? better than Marantz and Digital projection1080p units?

For me this was the most important consideration in buying a projector and I'm afraid I don't have any real explanation of what gives the most image depth. After demoing most of the newest projectors, I honestly can't match up any one particular attribute that governs depth, all I can say is 'newer and more of' seems to be better. After taking all the numbers and technology into consideration, real world performance often differs from what you would expect.
For instance, I find the C3X produces a lot more depth than the Infocus 777 but the technology and numbers are more or less the same.
I think image depth is perhaps one of the most subjective areas in PJ appraisal but FWIW, IMO quality of lens seems to play a large part with optic path, ANSI, brightness and processing. More importantly, how these attributes are configured together. I haven't seen the Marantz 11s1 or the JVC RS1 but my current favourites are: sim2 HT5000, C3X, D80, Projectiondesign model 3 and you will have to make up your own mind on the Sharp Z20000! My advice is: demo in 2 setups and never buy blind

"Image depth" depends on the projector and the images projected. Some images have no/little depth on any projector that has not On-Off contrast of several 10000:1. Others have excellent depth with On-Off of 1000:1 if you have ANSI contrast of 500:1 at the same time. Basically the higher both are the less material will lack depth assuming sharpness and detail are adequate too. In absolute numbers ANSI will reach a plateau much quicker than On-Off ( 500-1000:1 versus several 10000:1-several 100000:1).

No it's not measurable, it's how we perceive a sense of 'depth' in an image. This relies on many things, including the nature of the image itself and your distance from it. Some images that have a great sense of depth may not have a similar depth if you move a different ratio away. Some displays like DLPs can have a good bit of depth especially because of high ANSI CR, and others like CRTs can also have incredible depth despite the fact that they often have low ANSI. So depth is really a whole lot of attributes and it's kind of just a perceived result. Great source content (clean HD) really helps out too. I feel that things like grain also seriously negatively impact a sense of depth, but I like grain so go figure!

I think Chris put that really well. There are many factors that contribute to our perception of depth, and the ones that are most important depend on the image content. It is basically our ability to differentiate the "depth clues" in the image that improves our perception of depth. For instance, in a dark image, the full-field (on-off) contrast ratio may be the dominant factor if it eliminates haze because we can more clearly differentiate the dark objects from one another and the background. That's why CRT projectors can have such excellent depth in dark scenes. But in bright scenes, intra-image contrast (as measured by ANSI contrast ratios) is more important in differentiating changes within the image. Sharpness is also important because it helps to differentiate edges. Resolution is important if the image contains a lot of detail that helps us distinguish layers of depth in the image. Color resolution is a big factor that gives some images much better depth in high-definition than the same image in standard-definition. So in general, if the projector excels in multiple performance parameters, it is also likely to produce better depth in more types of images.

Well, if this isn't a hint Do I smell a new "Greg Rogers approved reference PJ choice" here... We will know next week. Everybody get a WSR subscription!

Oh well, I'm getting too excited again...

And oddly enough, I think we see better depth in many high end home systems (especially with HD content) than in an average theater. I think a significant exception to this would be IMAX presentations, but I think as we sit usually farther away in our home systems, for some reason I always am much more impressed with a sense of depth in my own system and others than I usually do in a regular theater presentation.

If you ever really want to play around with getting an image to look really 3-d without using glasses or real 3d systems, just try covering one of your eyes. You might be surprised just how much 'depth' the image can convey just based on focus difference across the image that you lose when you open both of your eyes! That's one of the main things in the structure of the image itself that really helps make an image "pop", if you're in a similar kind of viewpoint as the camera might be. Move too close and you lose that depth, move too far away and you lose it too!

It has always been my experience, that ANSI CR contributes the most to the perception of image depth. That, plus sharpness, I think are key. Moving from my C3X, to my Sharp 20000, I find the Sharp is slightly better in this aspect. This is one of my biggest concerns with the RS1, but hopefully, I will be pleasantly surprised, when I get it... or it will be for sale shortly thereafter.

Phil - as we've been discussing a bit in other threads, one can likely make improvements to their HT envrionment to bring ANSI CR way up. This is a good place to start because otherwise the ANSI CR of the pj doesn't make as much difference as it could.

For instance I get 80:1 at the screen with the Ruby and 100:1 with the Sharp 10K. These are far lower than what the pjs are capable of. At the pj I measured 104:1 and 279:1 respectively. So its interesting to note that a higher native CR from the pj still results in a higher ANSI CR at the screen despite room reflections. At the same time, however, much of the benefit is lost. In this case a pj that has over 2x ANSI CR is yielding just 25% more at the screen.

Certainly the sharpness of the RS1 will be beyond our control, but with the help of some HT "dressings" perhaps we can make overall improvements to really bring the ANSI CR at the screen up considerably.

BTW, I too share your concern about the RS1 and its sharpness as compared to the Sharp 20K. It was very encouraging to hear some folks say that the RS1 looked equally as sharp side by side with the 20K. Then again, others reported that the Pearl and the RS1 looked almost identical in certain scenes ,and my impression is that the Pearl is a bit soft so it seems there is somewhat of a disconnect in this regard between those reports IMO...

You guys are leaving the SCREEN out of the equation. That is wrong. Nothing kills image depth as a hotspot does (at one end) or as the softening of the image by a perf screen (at the other end). White field uniformity is paramount to image depth, but so is gain, and high MTF (no visible holes or particles).

I don't think uniformity is that important. CRTs have fantastic depth and terrible uniformity.

Brightness uniformity. That is easily achieved with OpticaL SCREENS LIKE THE torus (IN CRT CASE). The people that have compared how the TORUS brings more depth out of a CRT image are saying so because of the added gain coupled with good brightness uniformity. Plasmas also are great in that area.

Here are my thoughts after reading these posts on how to define and quantify a sense of depth:

We want to feel the detachment of the subject from the background. It is some sense of separation between two things that make us feel the depth. Now the question is what visual detaching or separating tool creates this sense or feeling?

In graphics designing when one wants to draw a two-dimensional image which generates a sense of depth, one uses a gradient that is a gradual variation of intensity from dark to light. This gradual variation in shade intensity makes the image felt 3-dimensional. The move can be from black to white or for any color, from its darker tone to its lighter tone. So factors in the science of image that seem relevant to me and are already very well quantified are the following:
1- Contrast: as a general conecpt in an image; it is the available various shades of dark and light colors. In a display device, it is its ability to depict various intensity levels of light.
How many degrees or levels of shade intensity can be covered by a certain projector or display device when moving from black to white? The more steps are there in between, means the more image depth. Display devices certainly vary in this regard.
2- Another relevant concept is dynamic range: in photography it is measured by the number of steps between the lowest level of light to highest level that can be recorded or displayed. One step up or down means twice more light or twice less light. Human eye has more dynamic range than any light recording or displaying device he has made so far. The next best is negative film. Negative film has a very thick latitude which enables it to capture a wide range of light intensities. Positive film (slide) how lower dynamic range which is only 5 steps, i think. Digital recorders such as digital cameras and video cameras have 4-5 steps of dynamic range. These digital displays have less ability than negative film to distinguish between two tones that are very close in intensity. That is one reason the image they generate looks flat comparing to an image made using film if considering other factors being the same (such as lens blurring which is another method to create the sense of separation).

Another tool for creating a visual sense of separation is the amount of details. In this regard i think observing more details can contribute to sense of depth one gets from an image). So the ability of a projector to depict more details, contributes to its image being felt 3-dimensional. So projectors with higher resolution give the impression of more depth in the image they show. In photography, medium format or large format comparing to 35mm has the same effect because one can see more details in the pictures made by them. This coupled with better contrast is what makes a projector's image felt even more 3-dimensional. That is another reason why images made by digital cameras have less depth into them: as of now, digital media still has lower resolution than what film has (they also have smaller sensor, which reduces the optical effect of the variation in aperture which contributes to blurriness of the background).

So if a low resolution image is projected on a large screen, its depth is reduced since the details are distributed in a wider area hence , less details per each unit of area is available.

There are other tools for observing the sense of detachment that is feeling the separation of two objects. Blurriness of the background where the subject is sharp. But that comes from source and the displaying device can not do anything about it (or as of now they are not designed to do such things as using AI to decide which parts of the image should be blurred by the brain of the digital processing unit).

You are confusing contrast ratios with bit-depth. Contrast ratios quantify a ratio between only two shades: black and white. CR has no bearing on bit depth or any shades of gray in between, though CR performance is important in being able to resolve those shades when present. This is, unfortunately, something that is commonly confused.

Thanks Chris for the clarification. If you could also clarify this terminology to me? I thought what you are referring to as contrast is what that is called 'on/off contrast' (or 'contrast ratio'?). One question: What is the term 'ANSI contrast' referring to?

Taken that as on/off contrast, i thought there is also a somehow broader concept of 'contrast' that refers to the bit depth: the ability to show more different shades of gray when moving from black to white. Am i using a wrong vocabulary here when using the word 'contrast'? Is there any right word for that?

Thanks, Ramin

gray scale resolution.