Image 'DEPTH' , the next hurdle. - Page 2

Ramin,

No.. in fact your post above is spot on. Several people on this forum seem to have difficulty comprehending perceived image intensity or bit depth, coupled with any information one might glean from measuring sequential 'on/off' or ANSI checkerboard test patterns.
http://www.infocomm.org/index.cfm?ob...BBD7812103C7A5

Peter H. Putman, CTS, ISF
Pete Putman is a contributing editor for Pro AV magazine and president of ROAM Consulting, Doylestown, Pa. An industry speaker and educator, Putman is well known for the product testing/development services he provides manufacturers of projectors, monitors, integrated TVs, and display interfaces. Over the past two decades, he has authored hundreds of technical articles, reviews, and columns for trade and consumer magazines. He can be reached at pete@hdtvexpert.com.
www.phpcom.com

Ramin:

Contrast ratios describe the difference between white and black and are measured off patterns which usually contain only white and black. The two main kinds are on/off CR which is a sequential measurement which is made on a full black screen and a sull white screen sequentially. ANSI contrast or related checkerboard contrast measurements are simultaneous measurements made on a black and white checkerboard.

Sequential CRs quantifiy usually the largest spread between black and white a display can achieve over time, while simultaneous CR quantifies how well a display can maintain contrast within a single image by resisting light spill from the white portions of the image into the black portions of the checkerboard. They are both very important.

For more detail see this article on contrast ratios:
http://www.hometheaterhifi.com/volum...06-part-1.html

Note that these common measurements do not attempt to describe the bit-depth or other grayscale capabilities such as gamma which are also important in an image and also important to things like image-depth.

As mhafner mentioned, bitdepth is akin to the resolution of the grayscale itself in terms of how many shades of a gray a display can display. Sufficient bit-depth is important in preventing any introduction of false contouring/banding in the image. You can have high contrast performance but low bit-depth. An extreme example would be a light bulb, which when measured in an on/off CR method achieves a practically undefined/infinite on/off CR since the ratio is between some significant output of light when turned on divided by no light when turned off. But a light bulb is a simple 1-bit system can can't even produce one shade in between white of black! So there, you have no gray tones, but practically infinite contrast.

In displays similar situations may arise where you have significant contrast ratio performance but limited bit-depth capabilities. You ideally want great performance in both, but it is important to understand how they are unrelated attributes.

It is also important to realize that ANSI and on/off CR are distinct: a display may have high ANSI CR performance and low on/off CR, or high on/off CR and low ANSI. An ideal display would have excellent performance with both sequential and simultaneous type measurements. The strengths and weaknesses in a display's CR performance will determine loosely how a display achieves instantaneous CR in a scene based on APL. This is discussed in more detail in the article I linked.

I notice that user "tbrunet" has responded in this thread. I don't know what is response is as he is on my ignore list. Presumably it's some nonsense about CR which he has continuously attempted to confuse people about. My apologies for his presence.

Provocative info Chris!..I'll email/forward Peter Putman your 'scientific research' / conjecture.

BTW, nice physic response to my post

Chris I've been doing that very thing at the movies ever since I was a teenager (e.g. late seventies). I can't remember how I discovered it, but it's truly amazing. If you cover one eye at the movies (or even on a good HT display) and give your eye time to adjust the depth just starts getting deeper and deeper and more realistic in the image.

Our brain uses many different cues to establish depth and distance, including of course calculating the off-set provided by our stereoscopic vision. My explanation would be that, in taking away one eye you are taking away the stereoscopic vision cues that let your brain distinguish all that information coming from a flat plane (the screen).
Which, with two eyes open, keeps the image 2-D. Once that binocular cue is gone - when you cover one eye - I presume our brain starts having to rely on all the other depth cues. And since a good film/video image contain much of the same type of depth cues (shading, perspective etc) our brain uses to determine depth, our brain starts re-calculating on those cues...which results in the feeling the image is becoming more 3D. It's amazing because, once the "one-eye" effect kicks in, I can move my vision from my display (my plasma) to the "real world" and they look very similar, because my brain is essentially working off the same cues for each.

Neat stuff.

Thank Chris for the information and i will certainly check the article. Now i understand why the contrast ratio can be so dramaticly different from projector to projector and why it can indeed be a very large number. If one can achieve pure black levels and very bright whites, the contrast ratio can go large.

I am now wondering if there is any inofrmation source around on the bit depth of projectors? Which type of display has in general better bit depth? LCD or DLP or the others? Which models are relatively good or better in this regard? My impression is that this is a more technical concept and requires more advanced devices to be measured comparing to CR or brightness, for example! That is why, one often does not read any information on bit depth in projector or display reviews around.


Since i do not want to highjck the thread, if anyone knows sources that answer my questions, they may send me a private message. I would apprecuiate that.

Thanks, Ramin

Best is xenon iluminated 3 chip dlp with an excess of a billion highly linear colors, then there are plasmas like fujitsu who use a dedicated processor to control the scaling and the plasma mechanics of pulse and discharge in random access subfield domain, all the other CRAP can have the OFTEN EXPANDED color depth but it is highly unlinear therefore CRAP, or it is 16 million colors as stated on the box and therefore is CRAP.

I am not saying that everyone that buys this CRAP is CRAP nor that the companies that sell this CRAP are CRAP, but that we should put a STOP to all this CRAP.

Too bad that 3chip DLP is currently out of my budget.

Ramin

Then buy a Fujitsu plasma or change hobbies.

You are aware that any 8 bit source when mapped 1:1 on a pixel based projector will have only these 16 Mio colors including 3 chip DLP so how is it crap??

Well, I think that (depite the fact your answer doesn't really address his question and instead contains some technical voodoo talk), that there just MIGHT be some pleasing devices in the range in between exactly ONE plasma and a 5-6 digit price DLP.

As a matter of fact... this very section of the forum just happens to be dedicated to such devices! What are you doing here?

A) these devices I am speaking of will expand the color gamut (much like audio oversampling is done),B) like in Audio: Bandwith and Throughtput, you need a bigger pipe than the signal you are trying to pass through it otherwise things get lost or truncated.

The Titan 1080p 3 chipper emulates xenon extremely well for 40. This type of image cost 4 times that one year ago, id call that progress. If everyone spending 18k on a projector stopped buying the LCOS LOW MTF nonsense, I am sure the price would halve on the 3 dlp, in just a couple of years an average videophile will be able to buy a 1080p 3 dlp second hand for 10k or less. THAT IS PROGRESS.

Sticking your heads in the SAND is what most of you are doing by calling the mediocre image of the LCOS "pleasing", anyone that sees the pearl next to a 3 chip dlp and has the audacity to still term the pearl "pleasing" is doing this forum more harm than good.





Raising awareness of that very same issue. We need 3 chip dlp to come down in price BADLY. The ************ forum is populated by individuals that will spend 300,000 on the speakers and 10,000 on the projector. So it gets lonely for the videophile in there.

You would think that because the Human brain processes 70% of the information derived from the eyes, that the expenditure on the display versus the speakers would be more proportionate. Nope, the audio industry is so well entrenched, one would think that there is a "special handshake" characters conspiracy behind it.

When will we see home theater done right for the average consumer? That is my goal.

I suggest that many of you could put a second mortgage on your home and get a projector like the Titan 1080p. You have my "goodhousekeeping" seal of approval that it would be the very last projector you would ever need.

Does this mean that I should cancel my pre-order of the JVC RS1 and order the Titan?

I don't want to step on anyones deal, if the rs-1 is coming out first. But if available roughly at the same time, HELL YEAH (are you freakin kidding? YEAH).

I am neither going to buy a plasma TV nor change hobbies! TC even if it is plasam does not please me as much as a projector does. The size of the image and a real cinema experience, a projector generates is what that pleases me most. Besides a reasonably large plasma TV costs a lot more than a projector which can still produce an image two to four times larger!

Indeed right now, i am looking around for a good 1-chip DLP projector and once the 3-chip DLPs get cheaper, i would buy one of them. the options within my affordable price: Optoma H77, H79, HD7100 and Samsung 710AE. I am debating whether the extra money paying for a DC3 will gain me a noticeable improvement in image quality over HD2+ while stretching my budget even further. The two HD2+ ones have gained very positive reviews around and the Samsung is particularly famous for its good shadow details
(Yes, I am aware of its startup issues).

Now the question is that within this price range, which projector has better bit depth (i don't care if somebody wants to call them crap, crab or whatever else)?

It is funny that H77 would have costed me about $800 or so more than my LCD projector last December but now, i can buy one for the money i got from selling my LCD projector! H79 would have costed me 2.5 times more last December but now only 1.5 times more!

Ramin

Good thing you are not a financial advisor, this type of suggestion could easily earn you some jail time...

I wonder what a divorce lawyer could do armed with evidence of a $50K mortgage used for buying a projector with CINERAMAX seal of approval...

It's settled then.

After all, it's only money, right?

Right, there is more where it came from.

Unfortunately, the Treasury Dept doesn't hand it out for free and my boss isn't in a "giving" mood.

I'm all for 3 chip DLP, just can't afford it AND the wife AND the mortgage. Maybe someday...

And this adressed the person's question (to whom you responded to via quoted reply) about bit-dpeth capability of various PJ's, HOW exactly?

Oh, and :"Hey everybody Peter needs to keep his commissions up wants us to recognize that anything less than a $40K projector is crap, despite what our eyes may tell us, so go ahead and cancel those preorders and/or head on over to Videogon and get those ads up to unload the trash currently hanging on your celings!"

Thanks Petey!

I have to say I with CINERAMAX on this one. I would rather buy a projector that will completely satisfy me and spend months (years?) in poverty enjoying the best image I could get. Personally if I spent £2000 on a projector that I wasn't completely satisfied with, the year after I would spend another £2000 and the year after etc. Eventually I would get something really satisfying but I know I would always be thinking "Hmmm, if only I saved up a few more months, I would be some way to enjoying the ultimate home cinema experience..."
For me, if you can watch nature documentaries and be thrilled, then 3 chip has done its job. So now I am broke.
You only get one go, no regrets and all the rest of that stuff that makes people spend lots of money!

We are all with him like you!! It is our pockets that don't go well along with his suggestion!

Well, just to look how funny this is, i only mention prices of two 1080p projectos he compares: $40k versus $4k! Ha, ha.

Added note: Well, actually some people are so rich that to them, the ratio of $40k to $4k is the same as the ratio of $400 to $40 to me, or even like the ratio of $40 to $4 to me! So my point is not valid since this is a relative concept! ;p

Could you provide any links with regards to A? I am not aware of current displays doing this at all, nor do I feel it's really desireable for current content. There are standards for future imaging that includes enhanced gamut capabilities (ITU BT.1361), but I think you are confusing gamut and bitdepth here.

As for B: well of course when you're dealing with linear displays like DMDs you need to use much greater bit-depth because you have to de-gamma. But beyond providing sufficient bit-depth(greater than 10 bit) to avoid any banding issues, there's no improvement to be had, unless you are referring to something else I am not aware of in part A.

I'll look for a link. There are discussions in the scaler arena to do gamut expansion algorythms. Yes I am convoluting gamut and bit depth (you can't have one without the other).

I think depth is something to strive for but be very careful we don't devolve into the Speaker comparison language of something that can't be measured.

i.e. Frequency response can be measured in a speaker "Punch" or "Air" cannot.

Lumens, contrast ratio, etc can be measured, "Depth" cannot.

As soon as we let ourselves compare these products with the non-measureable metrics, we are doomed to buying molybdenum speaker cables because they are more "transparent".

-Allen

No, but we can measure things that will contribute to the perception of "depth", correct?

There is no reason to think we can not messaure the image depth even if it has not been done before.


Ramin

You can have one without the other, just not well.

You can have a B&W display with superb bit-depth but basically no gamut at all. Or you could have a display with ridiculous gamut and horrible bit depth. You want both, but they don't hinge on one another.

Sure, but what we lack is a good model for just what all contributes to image depth and how those contributions are weighted. If we had such a model, then we could come up with some way of quantifying depth to a somewhat objective degree. That understanding is not really complete, and quite frankly it would be an endeavor that would be so complicated(including the nature of the image displayed itself), that it seems at this point easier to discuss subjective impressions of a sense of 'depth.'

I don't think it's dissimilar to things like imaging and other things in speaker and audio systems. There are many objective things that can be measured which can contribute to things like good imaging, but it's so complex to do all those measurements and really understand how they work. It's the art side of the science. We're stuck for now with the subjective descriptions, inadequate though they may be.