JVC RS1 review!!! (of sorts) - Page 11  

About the new "ANSI" patterns with lower APL
They will still be a measure of light dispersion more than intra frame contrast. They will however have a for film more typical APL.

This new measure will say little of the contrast look in a dark chase scene without shining objects. That is because very different video levels are in such a frame compared to ANSI patterns.

I think we need.

A On/off as usual to measure contrast potential
B chose a new light dispersion measuring pattern with a content typical APL
C1 Chose one interesting real video frame and measure in that one.
or
C2 Construct a pattern with spatiallly dispersed varying video levels. If projectors handle the same video level differently, gamma, but that is a part of the projector.

And i agree, but still in the practical sense you are reporting numbers based on ONOFF and your modified ANSI board which gives the reader a hint for contrast performance.

It is the ANSI board as such that is being questioned by people since the numbers often doesn't give a fair hint of contrast performance, combine that with that people think that ANSI is the real measurement and you have a problem.

Often people shout out for ANSI contrast and often refer all the depth of DLP to the ANSI contrast, so people have a wrong picture of what the ANSI number really is.

People think that just because you have black and white boxes in a pattern means that the measurement is for "mixed scenes".

Often you here "The DLP had much greater depth in mixed scenes due to its high ANSI contrast"

Mixed? what is that, a mixed scene in a typical movie isn't near what the ANSI board measures in terms of APL.

And as have been said, ONOFF and ANSI are measuring the "end" points of the scale, however IMO ONOFF lies closer to movie material than ANSI.

If we where to measure computer applications which tends to be close to 40 in APL then the ANSI measurement would be a better indicator. And a LCD monitor is perceived as more "contrasy" compared to a CRT monitor when using Word for example, even though that depends on many factors the ANSI number in this situation gives a hint of what we see. Put on a movie on the same screens and the LCD is blown away.

All of this is numbers, all of it is a simplification of the "real world" none is completely accurate. But you are supposed to make the models as close to the thing your are trying to represent with the model.

If you are representing movie material in numbers than measuring on a standard 50APL ANSI board doesn't make any sense?

Even ONOFF may not make any sense but still i believe that it lies closer to the reality you are trying to model.

There has to be a balance between measurements and practical applicability, you could probably make the measurements and patterns very complicated but then you wouldn't have to many sources of numbers. Just as you can add a parameter to an equation in physics to get closer to the reality.

So we still need these "basic" patterns and measurements like the ANSI board, but that doesn't mean that we cant change them, bring them closer to the thing we are trying to measure.

Even though you and many here may know how to interpret ANSI numbers and other figures, MANY people dont, and therefore it is good to give them numbers that make sense on their own to have a parameter which you could compare.

Often people don't see the numbers in a context, you may see ANSI numbers and think MTF, sharpness and many other things etc. But people usually look for a measurement to use simply comparing products.

Lets say breaking distance for a car, usually measured from about 60mph, why do you do that, because often you face situations where you have to break from those kinds of speeds. Now the ANSI board is like measuring the breaking distance from 200mph.

Or even horse power of a car, usually horse power is good if you are racing and constantly lying at top rmp where you usually get max horse power, it still effects lower rmp though. However in normal driving situation you would like to know the power you get from about 2000-3000 rmp, for example disel engines have less horse power but much more torque earlier, and they usually feel more powerful in normal driving situations but they are useless if you are only going to be at high rpm.

So having the torque at 2000-3000rpm usually gives a better performance indicator for normal use, that is why turbos feel so powerful, they give you the torque early on but they may not have much more horsepower.

Car turbos is almost like an iris, which gives you a boost in ONOFF, and thus creates more depth in film material, the iris doesn't do any good for computer applications just as many turbos are not as useful if you are constantly in higher rpms. But since we are at lower APL and since we are driving at lower rpms they make a difference in performance.

Example from BMW

530i at 190BHP
300 @2500-4000rpm


0-60 = 6,5 sek, (here you rev out all gears until you reach 60mph)
50-75mph = 6,6 sek (in 4/5 gear which is a usual overtake)

530d at 170BHP
Tourqe 500 @ 1700-3000rpm

0-60 = 6,8
50-75 = 5,1

Now you tell me which performance parameter torque or BHP gives the best indicator of how these cars feel at normal use.

I would argue that BHP is like ANSI and Torque here is like ONOFF in this example.

BHP gives you potential power at the high rpm and these torque figures gives you potential power at the lower rpms, bot are important and both affect the rpm scale, however normal driving is closer to the torque numbers in this example. Thus that number is better for comparing the above cars for normal use.



I'm sorry for the incoherent babbling and these terrible examples

There is no pattern that provides a number that isn't simply a figure of merit. It appears to me that what you are trying to do is create a new contrast ratio figure of merit for a combination of the full-field field contrast ratio with the intra-image contrast ratio of an image with a specific collection of objects, of specific sizes, at specific intensities, at specific locations. I don't know if you intend to measure the contrast ratio at one specific point within said image, or as weighted sum at various points. But either way, when you are done you will have a figure of merit for that pattern, not a "typical" film image. You will not be able to say that because projector A measures better than projector B in that scene, that it will also measure better in another scene, because the measured scene may rely more heavily on full-field contrast ratio thus favoring projector A, while another scene relies more on intra-image contrast factors that favor projector B. Since your measurement combines both factors, by itself it tells you less than you would know with both the separate full-field and modified-ANSI measurements. But even if you have all three measurements, the new measurement only tells you what happens in your new test pattern, or a film frame that can be reasonably considered equivalent. It doesn't tell you about other film frames. You can only infer that by having the full-field and modified-ANSI measurements.

Hence, if you were to measure a series of projectors and only supply that new number it would certainly be misleading. If you also supplied the full-field and ANSI values it would be better, but people that don't understand the full-field and ANSI numbers would simply rely on your "typical" film pattern number and still be mislead.

A somewhat more useful approach would be a series of test patterns that transition from heavily dependent on full-field contrast ratio to heavily dependent on ANSI contrast ratio. Those patterns would show the transition of performance on test images from projectors that favor one parameter vs the other. But of course that can be simply inferred by the current full-field and ANSI numbers. You would still need to provide some actual film images that could reasonably considered equivalent to each test pattern to visualize how the measurements relate to actual images.

One might ask why not simply define a set of actual film reference images and measure them for each projector. Besides the obvious problem of ensuring calibrated images, the measurements can only be based on measuring reference white and reference black (which aren't likely to occur sufficiently isolated in real images) else other projector errors (gamma, colorimetry, etc) would destroy the measurement accuracy.

Yes this is what I am trying to say, you could make it very complicated and really try to measure a real scene, but that would not be applicable for many sources of numbers just as you have said.

What I any many in the other thread is trying to do, is take down the ANSI measurement to a level where the number can reflect movie content better. See my poor example above with car torque and horse power.

ONOFF and ANSI represents each end of the scale, i know that, i know and understand much of what you are trying to say and i think my point is being lost somewhere in my argumentation (sorry for that)

I am not trying to say that ANSI is not important and i also know that depth depends on many factors I am not arguing on that.

I am simply trying to improve the patters we have today, there is a need otherwise the other thread wouldn't have gotten any response, i think many people have been thinking about it.

Will it be an improvement in how we show contrast figures i dont know, but i like the discussion and trying to improve the measurements.

I am taking what we have today, the numbers usually shown in every review and try to see if there is a way to give people better number or numbers for comparing contrast more directly connected to what they will see in movies (again see my car example)

I will not discuss a subject based on analogies. Let's stick to video. There is plenty of video terminology to express the issues clearly.

English is not my native language and i don't posses all the knowledge you have in video, so sometimes i may use examples to perhaps illustrate my point better. They may be poor but may also serve a purpose.

I will not try to explain what I mean again, here you have a break down of what I am trying to say.

1. We may need to try and define an alternative to ANSI contrast measurement what I mean about that you can read in this thread, they explain it much better.

Link to thread

2. IMO of the two ONOFF gives a better hint of contrast performance for movie material, again i know that there are many factors in play, but i am discussing the parameters usually used for presenting contrast performance for people.

That is all, i am not arguing with you on the other stuff.

If you don't agree and you think that the ANSI measurements is fine as it is, than ok, but my point is better explained in the link.

I understand (I think) what you are trying to do. I also understand why you think the two numbers, full-field and ANSI (or my modified-ANSI) contrast ratios create confusion with novices. I see the same thing. I believe it is because people want to analytically combine the two numbers into one value that describes what they see. You can not do that. (There is even a calculator web page somewhere that computes a contrast ratio by combining the two numbers.) That does not work for real images. The ANSI number is nothing but a figure of merit. You might as well always divide it by 10, or always multiply it by 100. Bigger is better. 2x bigger is 2x better projector performance for that image parameter. Does that mean the image has 2x more contrast? Not unless it is a 4x4 checkerboard. Is this a problem? No. Image quality is not determined by any one factor. You can't combine the grayscale accuracy with the black level to determine image quality numerically either. Nor can you measure MTF and add that to CIE x,y colorimetry accuracy to get the final image quality "value".

greg just made it clear that HE "will not discuss a subject based on analogies".

You, however, are free to express yourself as you see fit, and I encourage you to do so. I am a "lurker" with a JVC on order and am learning a lot from this. Thank you!

Yes it is a figure of merit, but a figure used to present something for people. And it does represent something, even though it is measured on a 4x4 pattern it does say something other than just about that 4x4 pattern, just as any equation in physics where you try and model reality. Beacuse that is what we are trying to do, model reality to have performance paramters in numbers that we can compare and analyze.

When the Auto iris came into play you changed the way you calibrate greyscale, why did you do that? was it because the old way didn't represent how the iris works in real movie material. By lowering the APL and checking with windows did you not better reflect how the iris would work in movie material.

By using windows did you not come closer to "reality" with the iris, did you not more accurately model the real situation in which the iris functions and thus you improved your calibration.

ANSI is a figure of merit, but why not try and improve that figure to more accurately represent the thing we are trying to measure. Not just improving ANSI, we may come up with a whole new way of presenting contrast, which I think is needed.

Im I wrong, perhaps I am, but there seems to be other that have similar thoughts (see link to thread) and if im wrong i have at least learned

Ekkerhart
good to know you will review the PD 1080p (the EUR20K or 10K model ?)
the dual color wheel is probably the only "almost" solution to dithering, which to me remains THE big problem of mono DLP. (some even say there see something similar with 3DLP, from the movement of the mirrors).
Sharpness (perhaps being "excessive") of DLP of course if the optics are excellent is a big plus of the 1-DLP but frankly watching a good blu ray tonight on a Pearl and soon on an HD1 that will replace it, the sharpness is striking.
on the contrast: there is some "brilliance" sometimes on a DLP, there's something. most likely because of the mirrors principle.
I haven't seen yet the HD1. but delivering natively without a brigthness-punch crushing IRIS system, 15000-18000:1 is a giant step. I hope they can also deliver a model with 1500-2000ansi and a CR of 15.000:1 some day.

The "cleaness" (= no added dithering noise) of SXRD/D-ILA is very very pleasing and natural compared to DLP.

Your English is great! I would never have known it wasn't your language.

I didn't mean to sound harsh about using analogies. Since English isn't your native language I understand now why you might want to use some other examples. But I think I've understood your points fine. I understand why you want to find a better test pattern that more closely models film. I would too. I moved away from the standard ANSI contrast measurement process to find a process that better modeled how a projector affects image quality. At the moment, I don't see how another pattern, or even a couple of patterns, that could be standardized, will provide a better indication of how a broad range of projectors (with and without a dynamic iris, with high or low full-field contrast, with high or low ANSI contrast) will compare over the range of film images. I see more problems created by new contrast ratio measurements, than trying to educate people as to what the existing measurements mean. I'm not saying it wouldn't be great to have such pattern(s), but I tried to explain why I don't see a practical solution to that goal.

(The reason I won't discuss analogies is because I've seen thread after thread in this forum destroyed because analogies quickly break down, then the discussion revolves around augmenting the analogy, offering new analogies, arguing about the new analogy, etc.)

I see what you mean, i think we have the same view on the matter itself and I reckon your concern about the practical problems of new patterns on a broader scale. And as I said there may not be an improvement in the way we present contrast performance numbers but it still is very interesting, at least to discuss and try to improve. People following these discussions learn about how the numbers and measurements interplay and that is good.

We started a similar thread over at our Swedish forum, about importance of ANSI, ONOFF contrast etc, and it was fun to see this thread here at AVS discussing the same thing (the one i linked to). What i learned from our Swedish thread is that people find this very interesting and many don't really know what the numbers represent.

Not to worry about the analogies thing, i get what you mean and i have seen the same things before where the analogy itself becomes the discussion.

Yea, I agree. Arguing about analogies is like ...

I have an idea I want to float by you all WRT patterns for contrast on real world images...

Would it be possible to construct a pattern that has subtle details on a sliding scale (like not-so-subtle details that then move toward more subtle details) and have those details marked. Then we can display the pattern and have people say that from their seated position they can see the details up to point X and then it disappears. Then someone else says they can see the details up to point Y in which case we know their intra-scene contrast is then better.

Now I know this would be more of a reflection (no pun intended) of the room conditions and reflectivity, but even so that's what ultimately matters (the detail we see on the screen). Is there any merit to such a pattern?

As technology changes, there will normally be a need to change the testing methods. It is just much harder to implement new testing standards than new technology. The only point I have for sure is that ON/OFF and ANSI CR are two different measurements and only equate to those specific tests. When you start to measure anything other than ON/OFF the environment comes into play. Mainly any bounced light to affect readings. A CR test in a totally black light absorbing room, starting with say a ANSI layout 4x4, however start with small white squares say where the 8 white squares total 10% of the screen area, then 20% then 30% .. 50% (ANSI), then switch to the black squares at 40%, 30%..... 0% (ON). This could possibly give a CR performance curve that would go from OFF to ON in 10% steps, similar to doing a Gray-scale. These (maximum performance level) results may rarely be replicated in a home theater due to wall treatment, color, screen and room size, but it would give a measure of performance at different levels, something you cannot get with ON/OFF and ANSI alone.

Since some references were made to cars and performance, when did you last buy a car and get a full HP/Torque curves from a dyno, or acceleration curve from 0-150 MPH? Some magazines may do it, but generally you see 0-60, 0-100, ¼-mile, max lateral Gs, stopping distance, the typical eye-catching or standard tests. When did you last run the ¼ mile? The good one is fuel mileage, why don't we see a fuel mileage curve from 0-70 MPH for both 1 200-pound driver and loaded 4 passengers and baggage?

The ultimate factor is how well it actually performs to the owner.

I like the idea. Then use the area under the curve and you get a measurement for ANSI type contrast performance.

Now we talk all about the contrast from the HD1, Cine4home previewed the HD1 wich give's very good contrast numbers, but what about the Blacklevel how does the blacklevel compares with a very good DLP ( Optoma H79, Marantz 12S4, Sharp Z20000 ) . i Just wonder if the JVC also has a good blacklevel, This i for me also important.

greets, and a happy coming new year from the Netherlands....

The black level is dependent on the white level, the ONOFF number tells you the range meaning it tells you how far apart the black and white can be.

So for example

If you have 15000:1@500 lumens on a 80" screen you will have

Approx 25fL, take this number and divide it my 15000 and you have your corresponding blacklevel

25/15000 = 0,0017fL

Now imagine a DLP at 4000:1@400lumens, on the same screen you would have

20fL with a black level of 0,005, which is much higher then then JVC

Now imagine a DLP with 8000:1@250 lumens on a 80" screen that is

12,5fL with a blacklevel of 0,0016, so now you have the same blacklevel as the JVC but with much less lightoutput.

If you put a ND filter on the JVC to half the 25fL you get twice the better blacklevel as the DLP above, or you could increase the screensize for the JVC and still have a good light output.

So you see its all dependent on your "white level", the ONOFF tells you the possible range.

But the higher the ONOFF the deeper the possibility is for the black level, meaning that if you match the lightouput of two PJ, the one with the greater ONOFF will have the better blacklevel.

Here you have some additional help, i dont know your screensize, but here you have some very ruff numbers.

80" divide lumens with 20

100" divide lumens with 30

110" divide lumens by 37

Now you have the "white level", divide that by the ONOFF and you have the black level for your screen.

Example again
Here you have two PJ which you asses that the numbers are good, perhaps measured by WSR.

JVC = 15000@600lumens
Sharp Z21k = 7000@400lumens

You have a 100" screen

You take the lumens and divide by 30, this gives you.

JVC = 20fL
Sharp = 13fL

Now you divide that number by the corresponding ONOFF, this gives you the black level.

JVC = 0,0013
Sharp = 0,0018

So there would not be that much difference in black level, however if you would to match the JVC lightoutput by an ND filter or if it had an manual iris, and you could get the JVC to 13fL the black level would be.

JVC = 0,0009
Sharp = 0,0018

Now this doesn't say all about the image, as have been pointed out the sense of contrast and depth is dependent on many factors.

Hope it made any sense.

PS: all numbers above is using fL, you could divide the lumens directly to see which PJ has the better blacklevel, however by converting to fL also gives you a sense of the white level at the screen, and you are aiming for 12fL, which is the "recommended" "white level".

So for 80" you probably must use a ND filter for the JVC, at least i will, why don't they put in a manual iris to give the user control over the light output, it is very appricated, especially for PJ which such a high calibrated output.

All above is ruff numbers, just to give an idea.

Wow, thanks for the great explenation


greets

Very nice ZoomAir. The only thing I would add is that smoeone's fL number may be different than what you posted for sample screen sizes, depending on the gain of their screen. Looks like you used a gain of 1.0? So for example where ZoomAir lists 30 as the number to use for 100", you would instead use 30*1.3 = 39 if you had a 1.3 gain screen. Also ZoomAir I assume the units for the numbers you listed such as 0,0013 for the JVC are foot candles not lux.

Yes very ruff numbers based on 1,0 gain, it wont matter for determining which has the best black level, you only need some kind of light output unit for that.

I used ruff fL numbers to give an hint about the light output in relation to the recommended 12fL, i thought that that is a number people recognize.

All the numbers are fL e.g. 0,0013fL

Yes, very good. I only wanted to point out about the screen gain in case people were curious what their actual black level would be like with the RS1 (as opposed to just using the numbers of the RS1 vs. Sharp for comparison purposes, in which case it does not matter as you point out).

excellent discussion
could it be said the gain on the screen however makes the whites punchier but what is the effect on the blacks and the black level ?

again the HD1 having NATIVE very high on off CR for a digital projector is a huge advantage over any iris-based machines which either make the transition visible and/or lower the brigthness, the "punch" of the projector way too much.

so, it has a very good blacklevel. !!

i just sold my H78 and i wanted to buy a marantz 12S4 or a Sharp Z2000, but in that case i think i better would wait for the new coming JVC HD1

Yes, the screen gain makes the whites brighter (i.e. punchier). However it raises the black level the same amount. So for a 1.3 gain screen the white level is raised by 1.3 and the black level is raised by 1.3. The contrast ratio of course remains the same.

lovingdvd--> very true



What the screen gain does it that it "narrows" the viewing cone. 1.0 gain in theory spreads the light equally in all directions, the higher in gain you go the more light is focused in one direction (simplified answer).

You can never add more light then the projector is giving out, so a higher gain will only look brighter if you sit at the "right" position, if you move towards the sides it becomes dimmer.

Its all about what you do with the light you have, with a higher gain you choose to prioritize an area in which the light is "focused".

So having a higher gain and if you sit at the "right" position means that you will have a greater amount of light directed towards you compared to a lower gain screen, this in turn also means that the reflected black level will be higher.

This focusing of light also means that less light will be directed towards walls etc. which in turn means that reflections from the nearby walls etc will be less visible on the image. However if you go to high in gain you will have many other problems.

For me the Studiotek 130 is the optimal screen, it has a specified gain on 1,3, and just as with any manufacturers numbers the actual measured gain differs, but i think that the Studiotek is very close to 1,3.

To sum up, gain of the screen tells you how much light is focused in a direction, imagine this:

Stand up and hold your arms out (like you want to hug someone) imagine that you are the screen, now as the gain gets higher, you start to move your hands together (your palms move towards each other). That is what is happening with the light. As you move your arms together you focus the light towards a more narrow area. You don't add anything, it DOESN'T make your projector any brighter, you just choose to prioritize the resources you have in one direction.

And since "black level" is simply light just as "white level", you also focus the "black level" towards you and thus as the white level increases by this "focusing" so does the black level.

Hope this answered your question

I am getting very excited about this projector, and February can't come soon enough. I have recently found that I am very satisfied with the image my Sharp 20K is putting out with the iris clamped down. Since I am using a 120" Silverstar, the image is reasonably bright, and the black level is considerably improved. I thought I needed a projector as bright as the Sharp with the iris open, but I realize that I was not giving my eyes enough time to adjust to the image at the lower level.

Now, I am starting to think, that I may end up using the RS1 with an ND filter, until the bulb ages a bit, and am very excited about how dark the black level will be. I can't wait to dig out all of my dark movies to watch, and experience in a whole new way again. Hey, maybe by the time the RS1 is here, the Descent will play on the Pioneer Blu Ray player that I am getting.

Generally speaking gain is gain so if you push peak white, you also push minimum black. They're both innately tied to the gain of the screen. If you have a peak white of 25 ft lamberts, the minimum black level will be Peak White / CR. There is still light at the lowest value.

Further If you raise (or lower) peak white, you raise (or lower) minimum black by the same amount since the CR is a constant in the equation.




Well hopefully the production machines realize the promise. I have little doubt they will; but it has happened before.

Cheers,

I will need to use an ND filter as well. Are you planning on tilting it a bit to avoid light reflecting back into the lens? I believe Mark P suggested this on one of the RS1 threads - to avoid the problem of having light reflect back into the lens and in turn reducing contrast (hope I have this right). I'm still not sure how best to implement this when using an ISCO3 lens and where I can buy the best quality ND filters. Any advice would be appreciated. Thanks.

I am actually not happy that JVC didn't include a manual iris, it is a simple option for giving the user control over the light, especially when you have such a high light output as the JVC seems to have.

I am NOT happy about adding things in the the light path, anything that is added will potentially degrade the image. However 500-700 lumens is way to much for my 82" screen, 20fL+ is IMO to much for movie viewing in a dark room, i would like around 12-15fL.

The good thing is that even though you have to add a filter now, later when the lumens drop due to the lamp you can still have great light output by removing the filter.

Exactly how i will do with the filter i don't know right now, i don't have much experience with the "placement" of filters, and i haven't practically analyzed how it affects the image, all i know is that i don't like to add things in the light path.

All this could be solved with a manual iris, WHYYY

Sorry i couldn't help with the ISCO lens and the ND filters, i don't have the practical experience with them to offer any good advise

PS: I know that by adding a manual iris you also add an extra element in the light path, but i believe that JVC would have done it in a better way then me adding some ND filter.

I have tried a lot of ND filters, and have only been satisfied with the glass absorptive filters. I found that the Gel, or Plastic filters degrade the image, but the glass filters, which you can get with anti-reflective coating, work best.