Sony VPL-vw1000 - Page 34

Speaking of the 95, when does yours go up for sale?

Pip

I agree we should either merge the threads or start an actual "Owner's Thread" soon.

Home Theater Magazine had 860 lumens in their review.

Art

I think you missed the thrust of my point. I was emphasising that blu-ray is still not the limit of what 1080p material is capable of containing and thus a 1080p display is capable of representing. This was a couple of years ago but if I recall it was a 4:4:4 12-bit 24fps uncompressed data stream at around 350MB/s. It was being fed from an SSD. The content was demonstrated in an equivalent compression as found for HD Satellite, then the blu-ray encode, then the uncompressed, and then back to satellite. It was quite shocking to be honest.

This is zooming, center reading , Minolta LS 110 luminance meter,after calibration,three hours on the lamp.

Art

My motivation is to discuss the science of video. I have been professionally and personally been involved in it for > 20 years so I thought it is useful for people know the science of what is possible and what is impossible. And what did I tell you that Sony's own manual didn't?

Or are you saying that I mistook the word "science" in the name of the forum?


I am not here to tell you why you are impressed. That is a personal opinion and impression.

I tried to address a specific scientific point: that BD has more resolution than 1080p. We can discuss that objectively. And I did and happy to drill into the mathematics if you think that was not enough.

What you saw subjectively, I also tried to explain in simulations with the pictures that I had, and posted snapshots of the projector itself. I showed how perceptually you can think the projector is sharper but how that is just contrast improvement which comes as part of interpolation being turned up at the high-end.

If you tell me you rather not know the science and like to think something is there that can mathematically be shown to not be there, and confirmed by the manufacturer of the projector to not be there, be my guest. But please don't ask me to play along .

I should note: this is one of the oddest comments I have ever received on something like this .

That sounds like the feed from a Panavision Genesis digital cinema camera...and others.

Also from a 20ft throw per the reviewer. Seems like the common them is the lumens drop considerably from 20ft. At least for some us, including me.

For the beginners it's important to understand the distance from the screen is irrelevant without screen size. What affects the light output related to throw is the relative position of the zoom lens. If it has 10 clicks then which click it is on will define the lumens at that setting. That will be constant regardless of the distance to a screen. As screen sizes used at that zoom position change then screen brightness will change, lumens will not. It is the THROW RATIO or ZOOM POSITION ON THE PJ that correlates with the lumen output. It works just like a variable aperture zoom camera lens.

The HT review was also from a pre prod machine that Sony only claimed top end lumens of 1500 on. So it was supposed to be 500 lumens low from the production models right from the start, before any calibrations.

If you took a 480p image and displayed it with larger pixels on a 480p projector and then showed that on a 12 foot screen, what would bother you is the size of said pixels which would become very visible. Upscaling then to a 1080p projector reduces pixel sizes and with it, gives an image that looks less "digital."

If you have a 1080p projector today and are not seeing the pixel edges, then making the pixels yet again smaller gives you no benefit. Further, the smaller pixels reduce an artifact of display pixels which is to provide a sharpness boost. So all else being equal, an interpolated 4K image will look somewhat softer than on the 1080p. Sony knows this so I suspect by default it is sharpening the image as it is upscaling it. That is what I showed from the review snapshot. There will be artifacts as clearly seen in those images in the way of increased halos.


That's right. That is the pixel edge effect I explained above.


It doesn't depend on anything. Honest! The math doesn't provide for any dependency. All frequencies > 1080p were thrown away at the time the video was scaled to 1080p to store on BD disc. No amount of wishing that those pixels coming back makes them come back . Scalars as used in this projector has no intelligence with respect to image content to synthesize new pixels as their own manual admits.


Well, you wouldn't be the first guy who likes a manipulation of something instead of having it be transparent . Here are the two side-by-side images again:



I am sure if I conducted a poll, everyone would say the image on the right is higher resolution and better. Problem is, if I apply the same thing to a different image, then you would complain. Didn't I hear people say film grain was exaggerated? Well that is one of the problem areas.


There is no "examination" going on. That would require image knowledge and the Sony scalar does not have such knowledge. If I gave you two dots, how would you know that there should be a direct line between them or a curved one?

I appreciate your layman description but the science here is precise and tells you that you can increase your interpolator taps but that is not going to give you more detail. It simply changes the frequency response of the "filter" that it is, and with it, can cause the image to not lose some of the resolution it already has.

Remember, the *ideal* interpolator's (scalar's) job is to do no harm to the source. Everything is a step down, not up! In reality, interpolators can be dialed to produce extra sharpness and artifacts or softer images and less artifacts. The mathematics of the filter (again that is what a scalar is), tells you this.


One key thing: it is NOT the image that was viewed and approved by the talent when the video was produced! If your goal is to color the video, then by all means go for it. Turn up the reds, it will make roses redder. Turn up the green, it will make the grass, well, greener . I thought our goal in this forum is to discuss how not to do things this way.


It is not even taking a baby step to do that. The resolution of what you are seeing is 1/4 of 4K. But sure. Why not humor me. Show me a test pattern that pattern on BD discs that on this projector resolves 3K pixels. You have something like this?

Again, you are talking about something that Sony itself says it is not possible. Why would they say this if what you are saying is true?


Only if you captured it that way. Otherwise, it is wishful thinking and anti-science talk.


There is no fuzzy logic. There is no Hal 2000 in there either. There are techniques for creating super resolution from multiple frames of something. But they tend to misfire and create horrendous warping artifacts. And at any rate, can be very expensive to create.


And I don't want to be fooled . I want exactly what is on disc. I don't want my projector to boost contrast, create halos, exaggerate compression artifacts and film noise. But sure, it is good to have the tool for people who crave such things and that is why we have the adjustment dials on displays.


You mean other than science?

Guys, nothing wrong with loving this projector. Say that you love the picture and we are done. But please, please, let's not pretend there is some magic explanation or increased resolution here. These are basics of signal processing I am talking about.

Let me finish by saying that the demos I saw at CES showed this projector to be a fine specimen. And it may be the best there is in that price range. So please you all, don't be defensive . Just trying to make sure your back is protected as you sing the praises of this projector, lest we want to sound like the hallways of Bestbuy .

So this is short throw? Amazing......was this after full calibration, if so did you measure prior to calibration? Sorry for so many questions but this is very disturbing.....

Amir, thank you for injecting a bit of science in this thread so overflowing with emotions (nothing wrong with those) and subjectivity (nothing wrong with that either).

Heh. I spent half an hour writing a long reply to this but in the end realised I had totally forgotten the point I was trying to make. In short: yes, there's more to 1080p (and Blu-ray can do 1080p just fine as far as raw pixel resolution is concerned) than resolution alone and Blu-ray's image quality is hobbled by its 4:2:0 chroma system. But that's the best we as consumers have to feed to our consumer displays that are similarly hobbled by 8-bit processing and/or panels. I don't know how VW1000 would react to real 4:4:2 or 4:4:4 1080p content but I'm quite convinced there's no true new information being created when VW1000 is displaying 1080p 4:2:0.

Thanks for keeping the discussion honest Scott. I bit my tongue a little as I posted knowing full well that it's meaningless without screen size. The problem was, all I saw listed in his review was that it was a 120" screen. So I didn't know if that was 120" wide or diagonal. Nor did I know what aspect it might be. Not to mention I saw no mention of screen hight.

My observation with the available information and my point was that his recorded higher lumens measurement was associated with a short throw vis-a-vis Art's, HT Magazine, or adidino's lower lumens with a longer throw.

i know nothing but aren't larger pixels surrounded by a matrix of no information and if you have more pixels with less no information matrix doesn't that result in more information or detail to your eye?

So... you don't calibrate your displays, then?

I sense a great disturbance, as if a million DoPs and cinematographers suddenly cried out in frustration.

I stopped by a local dealer this morning because I heard that they had just received a VW1000. They very kindly demoed it for me for about 90 minutes on my own content, plus a 3D disc of theirs since I didn't have one with me.

Wow.

This projector was essentially straight out of the box without any kind of proper calibration, projecting (zoomed) onto a StudioTek 130. The room was totally light controlled. I didn't get the exact size, or throw.

I was very impressed with the image. Plenty of punch, plenty of brightness, excellent blacks. The motion handling was superb. I was sitting so close to the projector that my head actually bumped it at one point. I could clearly hear the fan (and the iris!), but I was thinking, "This thing is really quiet on low lamp. I wonder how it will be on high." Then I found out it was on high! We set it to low and it was just a whisper.

There were only two things I noticed, and I'm sure one of them is down to the lack of calibration:

1. The blacks were very, very slightly crushed. This is, I'm sure, fixable with a little adjustment.
2. On Super 8, with Reality Creation off, watching chapter 3 (the chapter with the train crash), film grain seemed to be emphasized. We turned Reality Creation on and set both sliders to zero, and this actually reduced the apparent grain, but it was still more noticeable than on a recently-calibrated C3X 1080. I wonder whether one of the many settings would affect this.

The 3D we watched was the animated movie with the owls, and it looked very good indeed. I noticed a strange phenomenon in some of the shots and I don't know how to describe it. Perhaps it was related to some false contouring in the image causing some weirdness with the 3D process. I didn't notice if the smoothing setting was on, so perhaps this could also be tweaked. Otherwise the 3D was among the best I've seen -- dramatically better than the JVC. Possibly comparable to the magnificent 3D Solo, although I've never seen that movie on that projector.

Anyway, the VW1000 is now at the top of my list for projectors, and I'm wondering whether I really need my Prius.

Any thoughts on the film grain observation?

Sit one of those DPs or cinematographers down with a DVD, a player, and two large displays, one with 480p resolution, and one with 1080p resolution, and ask him or her which display is truer to the original projected film image. He or she will always say the 1080p display has greater verisimilitude. It won't matter that the video is deinterlaced and upscaled from 480i to 1080p. And if the deinterlacing and scaling is adaptive of picture content, the more the end result looks like the film. Who is the purist in this example, the one who insists that the 480p display is true to the original video, or the one who insists that the 1080p display is truer to the real source, the original film?

Anyone who argues that the 480p display is preferable or closer to the mark is plain silly, and they know it. Why do they think they get to make the same argument when the scaling goes from 1080p to 2160? What matters is verisimilitude to the original picture, not some digital intermediate that is technologically limited. If tricks are employed to get there, such as sampling and compression on the one end, or scaling on the other end, what does it matter?

Of course, the above only applies if you are close enough to actually see the difference between 1080p and 2160p.

Properly tweaked, black crush is non existent. I demoed some scenes from the RR Hall of Fame Concert. Sting was wearing a black sport coat with a black tshirt. I could clearly distinguish both and could and even clearly see some black fuzz on his jacket and t. Black levels are amazing on this thing. Amazing how it handled the various shades of black.

Art is not at long throw with 14' wide screen at 19'-7" throw. This puts him at 1.4 to 1 throw ratio.

Just one? I thought everyone posting in this part of he forum is of that type. Pretty sure there are a lot of people here who would not want distorted pictures.

This is what happens when you try to sharpen an image. Top half is the original, the bottom is the sharper version:



See the distortion in the high-frequency patches on the right? Even in the low frequency patches you now see more issues. Look at the blocking errors and ringing next to the thick lines (which were probably were there mildly in the original image I found online but now heavily exaggerated).


It is news to me that we no longer care about fidelity and science in this forum. Is there a consensus view that we should no longer care here?

I agree that the destruction of fine detail on a 1080 display, which is displaying a 1080 image, is highly undesirable and so sharpness controls should be avoided in that instance. But what if you use the "extra" pixels instead to provide things like edge enhancement. You are not removing any of the original image detail, but complimenting it with extra available pixels. This is only something you can do when you have more display pixels available than the original source. From a purist point of view it is still blasphemy. But there is a middle ground that is better than what you are illustrating in your picture.

So who's using or planning to use an AT screen with this projector? Which one do you have or plan to own? I should make final decisions tonight.

I'm torn between the Stewart ST 1.3 MP, Enlightor 4K, or Center Stage XD

This projector is very addictive. I am hooked. I snuck in to see some of Star Wars 3D and the image was just not sharp enough. Brightness some scenes was lacking severely. I think just after a few days I am spoiled.

Only when it gets in the way of a good circlejerk.