Why Screenies don't matter.

**Harpmaker** · **05-09-08, 08:54 PM**

Actually, the title of this thread isn't 100% accurate. It should be more like "Why Most Screenies don't matter"; but I'm getting ahead of myself.

First, a few definitions:
Screenie:
A digital photo taken of a projection screen displaying a paused movie or television scene.

Reference image photo:
Similar to a screenie, but the still or paused image is not meant to entertain, but is designed to show an attribute of the screen or projector. Most such images are meant to aid in correctly calibrating television sets or projectors for contrast, brightness etc..

Reference photos are necessary to determine screen performance. For an example see this post. http://www.hometheatershack.com/foru...html#post97077

Unless you are total newcomer to DIY screens, and internet forums dealing with such, you have seen a number of examples of screenies. The things are all over the place as people try to show off their home theaters, home-made screens or newest projector. There is nothing wrong with this, it can be fun to see such images; but, what do they really tell us? Most of the time, not much.

I want to be clear that I'm not talking about photos that show how to build a screen, or images of entire home theaters. These usually are quite informative, and some are even down-right gorgeous. Many HT's I've seen on forums should be in Home and Garden magazine! They are works of art!

One of the primary reasons most screenies don't give us much usable data is the fact that taking them is not an easy thing to learn; it ain't Rocket Science, but the only place to learn how is in forums like this one.

Most people just use the auto setting on their camera. That's a bad way to take screenies or reference photos, and will lead to great frustration if the person is trying to show anything like the image they are seeing on the screen with their naked eyes.

The first step is to use whatever mode your camera has that will allow the user to manually set the white balance. Go into that mode and then project a solid white image onto your screen. There is one on the Video Essentials DVD, and probably other video calibration DVD's as well. After you have this image on your screen, go through whatever procedure your camera uses to set this screen as the white reference. This is the only way to be sure your camera will produce photos close to what you see.

Now we come to setting the exposure time for the photos. Many times the camera will automatically set the proper exposure for the shot, but many times it won't either. The only thing to do is to take multiple photos of the screen at different exposure times and later select the one that looks closest to the true screen image.

Even with white balance and exposure taken care of we still have a problem, one I haven't found a good solution to yet. I'm talking about color saturation. I have taken screenies of a movie that seem pretty good as far as proper colors and exposures are concerned, but dad-burn it, the things are over-saturated! I'll show what I mean by this in a moment. Sometimes a camera will let you compensate for this, but most won't. In my case, my camera will only let me choose to have a more saturated picture!

Now comes the two biggies; contrast and dynamic range. Very few camera's, at least in the "point and shoot" category (read that as the cameras most of us have

), will let you manually set contrast levels, and NO camera I know of can manually adjust dynamic range; you either have an adequate amount or you don't (most don't).

Contrast and dynamic range are related, but it will take someone smarter than me to explain it properly. My first set of photos will show you what I mean though. they're coming up shortly.

In the end, our digital cameras are designed to make the best "normal" photo of whatever the camera is seeing. This is at odds with what we are trying to do with screenies. We are trying to show what our screens really look like to our eyes, but the camera is trying to make each photo look like it was taken in a perfect textbook setting, such as outside on a clear day in the middle of the afternoon with the sun behind the camera. In short, the camera is trying to make each photo "perfect" and not real. Sorry, at the moment I can't explain it any better than that.

All the following photos are from the movie The Fifth Element full-screen version.

The first two will show my camera's lack of dynamic range, meaning I can't get proper shadow detail and highlight detail at the same time. The first photo is exposed the way the camera thought it should be. In general, it's a good photo for shadows and mid-tones, but the highlights (glare on foreheads and the doctor's shirt) are overexposed. The second photo had a shorter exposure time and the highlights look better, but the shadows have gone too dark. If my camera had a wider dynamic range this wouldn't happen.
Camera auto exposure

Manual exposure

This next photo also has dynamic range problems (I could see more detail in the mans eyes), it also has a more evident color saturation problem. The colors are too vivid and are not what I was seeing on the screen.
Auto exposure

Colors again too vivid.
Auto exposure

While this next photo looks like it was properly exposed, in reality, the image on the screen was brighter yet.
Auto exposure

One more example of lack of DR (dynamic range) and over-saturation by the camera, not the projector.
Auto exposure

Now another example of why most screenies don't mean much. The screen used in these photos was made up of 5 panels, each about 12 inches wide and 4 feet tall. Each panel was painted with a different paint. There isn't a lot of difference between them, but there are differences. Especially in the second panel from the left, it hot-spots. A bit of the bare wall of my shop can be seen as the far-right edge of the screen image.
Composite screen under white PJ light

The more discerning of you might have noticed the composite screen in the photos above. Here are some photos in which it is very hard to see the separate panels. If 5 different screens can all look the same, what does that tell you about using screenies to show how one screen is better than another? That is what reference photos are for.

Several shots where the composite screen is quite apparent.

One last photo that just plain looks good

.

The final kicker is that NONE of the above photos looks like what I was seeing on the projection screen. The real image didn't have the contrast or saturation the photos have. The camera was doing it's job and making a silk purse out of a sow's ear, or at least doing it's best. Most cameras do the same thing.

There are times when screenies can show a relative difference between two or more screens and give meaningful results, but these are few and far between on the forums. The scenes must be chosen with care and photographed properly. Such photos will almost never look "good", they will be lacking in contrast or color; but they should, many times, if they are truly showing what the screens looks like. My hat goes off to the folks that take the time, and have the expertise, to take such photos.

**wbassett** · **05-09-08, 09:50 PM**

Another problem is compression. The camera does it... then usually you have to resize the image for forum use, and some online photo hosting sites even add more compression and some even change the file format.

Screenies fall more in the eye candy category than anything, but certain ones can be used to demonstrate performance conditions, such as with ambient light levels. Some people rely solely on screen shot images as proof of 'empirical' data. That's fine, but most of the time it tends to be subjective.

Also it is really impossible to compare one picture to another. The only time any 'comparison' can be done is within the same picture.

I also have seen a lot of 'comparisons' between white screens and gray screens and those really don't prove anything. Of course blacks will look better on the gray and whites will look better on the white screen. The only way to do a comparison is like harp demonstrated, and that is by using test screen panels that are all similar in shade so the projector can be calibrated to all the panels.

**wbassett** · **05-09-08, 10:12 PM**

Another common thing I see a lot of are screenies of animated movies like Pixar. These always look great. That's why I like a wide variety of images to be shown. It gives a better 'feel' for empirical performance.

**Blaser** · **05-10-08, 02:09 AM**

Screenies are for fun and illustration only IMO. They are not meant to demonstrate a PQ, but a rough idea.
Screenies can be good to illustrate for ex. the viewing angle of a screen when comparing it at 0 deg and off-axis...

**Barney Fife** · **05-10-08, 07:29 PM**

For the most part, screen shots are the equivalent of teets on a bull.

**Prof.** · **05-10-08, 08:00 PM**

I think the end result of the accuracy of reproducing a projected image, is also determined by the quality of the camera being used..

If you took the same screen image with two different camera's..one being a $3000.00 Nikon SLR and a $150.00 Fuji or the like..you would have two totally different images..both in accuracy of colour and definition..
Even between the cheaper compacts, there can be big differences..

Harp..in your instance, I'm guessing that you have a very good quality camera..
The sharpness of your images are way beyond what my little Nikon compact can produce..

I totally agree about the selection of images, to show off what you're trying to reproduce..
I have had to spend a whole evening taking numerous shots to find maybe three suitable pics.that come as close as to what I'm seeing on the screen..
I see an image on the screen and I think, that would look nice, and the camera reproduces something that doesn't look anything like it!!

I think for the most part, we are all careful to find images that reproduce well, because after all, we want to show this projector or screen, in the best way possible..

Generally speaknig, If someone puts up a number of screenies, I can generally get a feel of how that projector or screen is performing..unless the poster is saying that it doesn't look anything like this..
Then it's meaningless...
But overall, it's just good eye candy, that I think most people like to see..

**cynical2** · **05-10-08, 09:20 PM**

Great thread, Harp!

I've taken enough screenshots now that I can attest to the general premise. Without very careful planning/execution (and lots of photographic knowledge), I've found it extremely difficult to accurately capture what I'm seeing on the screen. There are methods to get accurate (at least much more accurate) screenshots, but quite frankly I know of only one person that follows that rigorously follows that methodology (and that's mech).

I do think screenshots are useful for showing the relative difference in "range" (brightest white to blackest black...analogous to perceived contrast ratio) that screen A can display vs screen B. Some of Bill's photos in the PFG thread are a good example of that. While they may not be "accurate" in the absolute sense (meaning vs what he acutally sees), I think that it's still possible to gain insight into the screen performance. Even then, scene selection/variety, camera quality/settings, etc etc are still extremely important.

**mechman** · **05-11-08, 12:50 AM**

I have to say that I agree with just about everything being said here. Screenies serve a purpose in my mind. And that purpose is to augment the data. Data being the RGB values, xyY values, Lab values, and the spectrum image. I think the data itself tells you most everything you need to know. The rub comes when there needs to be a comparison. I think since the dawn of Black Widow we've done a ton of comparisons to paints that were supposed to be as good or better. And those paints gave us data readings that were in line with Black Widow. However, when it came time to see how they performed, they failed in every occurrence to live up to their billing. That's where photos come in handy.

The problem in photos though, is that they can be doctored. We've seen this in the past over at avs. Especially when one's business relies upon them - again avs. If one's livelihood creates a motive for this or one's ego, then that should be taken into account. Enough of that though! Time for some comments.

Quote:

Harpmaker wrote:

Unless you are total newcomer to DIY screens, and internet forums dealing with such, you have seen a number of examples of screenies. The things are all over the place as people try to show off their home theaters, home-made screens or newest projector. There is nothing wrong with this, it can be fun to see such images; but, what do they really tell us? Most of the time, not much.

I wouldn't say not much. But when compared to the data side, I'd agree. Photos are useful for comparisons and great ways of showing what you're trying to say, such as hot spotting. But photos are definitely not the be all end all!

Quote:

Harpmaker wrote:

Most people just use the auto setting on their camera. That's a bad way to take screenies or reference photos, and will lead to great frustration if the person is trying to show anything like the image they are seeing on the screen with their naked eyes.

Auto is what I use on my camera as it has shown me that it's capable of taking the shot depending upon the image being displayed. One f-stop may not be the right f-stop for the next shot. And same goes for shutter speeds. Constantly battling with the camera ends up being a battle with time. On my old Olympus 4000 I used to do manual adjustments, but not on my E-500. But you have to try and display what's being seen and that's a difficult task in extremely low lighting situations.

Quote:

Harpmaker wrote:

The first step is to use whatever mode your camera has that will allow the user to manually set the white balance. Go into that mode and then project a solid white image onto your screen. There is one on the Video Essentials DVD, and probably other video calibration DVD's as well. After you have this image on your screen, go through whatever procedure your camera uses to set this screen as the white reference. This is the only way to be sure your camera will produce photos close to what you see.

Another way to do it is to use a neutral gray reference card such as the WhiBal. You can peruse the ins and outs of using that here. The nice thing about it is that it's not exclusive to screen shots and can be used in all of your photographing exploits. I've found mine extremely handy and very accurate!

Quote:

Harpmaker wrote:

Now we come to setting the exposure time for the photos. Many times the camera will automatically set the proper exposure for the shot, but many times it won't either. The only thing to do is to take multiple photos of the screen at different exposure times and later select the one that looks closest to the true screen image.

Even with white balance and exposure taken care of we still have a problem, one I haven't found a good solution to yet. I'm talking about color saturation. I have taken screenies of a movie that seem pretty good as far as proper colors and exposures are concerned, but dad-burn it, the things are over-saturated! I'll show what I mean by this in a moment. Sometimes a camera will let you compensate for this, but most won't. In my case, my camera will only let me choose to have a more saturated picture!

Now comes the two biggies; contrast and dynamic range. Very few camera's, at least in the "point and shoot" category (read that as the cameras most of us have

), will let you manually set contrast levels, and NO camera I know of can manually adjust dynamic range; you either have an adequate amount or you don't (most don't).

Contrast and dynamic range are related, but it will take someone smarter than me to explain it properly. My first set of photos will show you what I mean though. they're coming up shortly.

In the end, our digital cameras are designed to make the best "normal" photo of whatever the camera is seeing. This is at odds with what we are trying to do with screenies. We are trying to show what our screens really look like to our eyes, but the camera is trying to make each photo look like it was taken in a perfect textbook setting, such as outside on a clear day in the middle of the afternoon with the sun behind the camera. In short, the camera is trying to make each photo "perfect" and not real. Sorry, at the moment I can't explain it any better than that.

With regards to exposure, contrast, dynamic range, etc, you get what you pay for. There are many things at play with each component including firmware, optics and the image sensor. The key here is tinkering until you get it right, all the while doing what you can to get them as accurate as possible - ie color balancing.

And this is also why I really disregard pretty much any photo - for judgment's sake - unless they're comparison photos. Don't get me wrong I love the screen shots of people's finished products and all. But they really are not usually the full story.

Quote:

wbassett wrote:

Another problem is compression. The camera does it... then usually you have to resize the image for forum use, and some online photo hosting sites even add more compression and some even change the file format.

Compression is a problem. Especially with the online hosting and what not. But most cameras will let you shoot in RAW mode, which is the uncompressed image. The caveat being that the files are extremely large. I shoot all of my photos in RAW, as it's much easier to color balance them post production in Adobe Bridge.

Quote:

wbassett wrote:

Screenies fall more in the eye candy category than anything, but certain ones can be used to demonstrate performance conditions, such as with ambient light levels. Some people rely solely on screen shot images as proof of 'empirical' data. That's fine, but most of the time it tends to be subjective.

This is why I think it's good to be discussing this! They are subjective and they have been abused... a lot (I'd say mostly elsewhere)!

But you have to cater to your 'market' - for lack of a better term.

Quote:

wbassett wrote:

Also it is really impossible to compare one picture to another. The only time any 'comparison' can be done is within the same picture.

I'm not so certain on this. If the lighting, the image, the settings on the camera, etc. don't change... I'll say maybe!

Quote:

wbassett wrote:

I also have seen a lot of 'comparisons' between white screens and gray screens and those really don't prove anything. Of course blacks will look better on the gray and whites will look better on the white screen. The only way to do a comparison is like harp demonstrated, and that is by using test screen panels that are all similar in shade so the projector can be calibrated to all the panels.

Comparisons of white to grays is something that should never be taken seriously. The whites on my FG screen and on all the Black Widow panels look just fine to me. Right up until I put up a piece of Designer White.

Apples to apples, oranges to oranges.

As far as what type of content to show, the movie shots are primarily just eye candy. Most everything that needs to be shown via photos is readily available on the DVE, AVIA, or some other calibration disk. I prefer the checkerboard pattern for blacks and whites. There are solid color patterns for colors or there are color bars. But most of this stuff is moot as if you follow the guidelines we're trying to set forth - calibrate pj to D65 and project onto a D65 neutral screen - you will get everything you're looking for with regards to accurate color reproduction, skin tones, etc. Then it just becomes a question of 'do you want a bit more pop'? But that's a whole other can of worms...

**mechman** · **05-11-08, 12:58 AM**

Quote:

cynical2 wrote:

There are methods to get accurate (at least much more accurate) screenshots, but quite frankly I know of only one person that follows that rigorously follows that methodology (and that's mech).

I'm not sure if it's rigorously or blindly!

That is one of the reasons why it takes me a long time to get pictures up though. It's not exactly an enjoyable experience for me!

There's nothing quite like spending a good 3 or 4 hours taking shots and downloading them to the computer. Then another hour or more white balancing the bunch and converting to jpegs. Then another hour uploading to photobucket. And then I find out I royally screwed something up! It happens a lot more than you folks think!

Anybody can take a bunch of pictures and throw them up as quickly as possible - we see that a lot! But to take the time and try to get it right.... takes me a few days just to try and psyche myself up for it lately!

**wbassett** · **05-11-08, 11:27 AM**

Another thing that hurts things when it comes to screenies is even the pros (commercial screen advertisements) doctor their images. Take a look at some of the online sites as well as ads in magazines. They usually show a stunningly perfect image in a totally lit room, screen images that often look like a plasma display. People see that and at 100" plus sizes and they think "WOW! I want that!" If you think about it, that's the goal of an advertisement- to get people to want to buy the product.

Screens are also something we see very little on in the magazines when it comes to reviews and extensive testing. They are often the secondary purchase and more often than not don't get as much thought as the rest of the setup does.

The classic screenie of a still image from a movie is fun and always catches people's eye, but there are a lot of variables that people don't realize. Also, even if a screenie is an accurate representation of what the screen looks like when an image is projected onto it in real life, that doesn't mean it will look the same for each person and their setup. Still, it would be hard to get people's interest in DIY screens without screenies.

Way too over used are Pixar and animated shots in my opinion. Those types of shots always look amazing, but most people watch live action more than animated, so how well a screen captures the actual look, colors, flesh tones, shadow detail... all that comes into play and makes a difference.

I think there are two distinct categories here... performance shots and eye candy shots. Like I said before, when everything is combined it is a powerful statement.

The biggest problem I see in DIY is education, and I like this thread because it is informative and educational for people to read. It is so easy for a person new to projectors and screens to get caught up in the excitement of the eye candy though.

**wbassett** · **05-11-08, 12:31 PM**

Here is a comparison shot. These have been contested at times but it does demonstrate how one screen can look and perform different from another.

In this example we can clearly see that blacks are deeper and darker than the main screen, but whites aren't taking the classic hit that happens with a lot of gray screens. In this example I would say that the screenie does show the difference between a good quality matte finish and a slightly more advanced application.

Here is one that is definitely eye candy, but I would say is a bit of an unusual thing to have taken a screenie of. The reason it was picked was the plaque. The image depth of the reflection is what I was trying to show. Keep in mind, this was taken using a Sharp presentation projector with a CR of just 250:1 on a simple OTS matte finish screen.

And then there is the argument of whites on a gray screen. Everything is relative. With a direct comparison to a white screen, yes whites will look less than white on the gray screen, but once that white reference is removed, our eyes and brain no longer have a reference point to compare things to so since we know something is supposed to be white... or even a specific color for that matter, our brain will tell us it is that color. Of if the screen itself is way off and pushing the image in a certain direction then we will start to see the color push.

Nobody can argue that this doesn't look white, but if a white screen reference was introduced then all those uniforms would really look white on the white reference and a duller dingy grayish on the gray screen. It's all relitive to our references that our eyes see.

Calibration shots and color bars though are really what demonstrates a screen, but they are boring to look at, and as Harp pointed out, they still can be very subjective and prone to a lot of variables.

Some of the variables can be the camera. Mech mentioned that a lot of camera's allow you to shoot in RAW mode, many don't though, mine included. Even if the camera does have that feature, the operator themself becomes a variable. Some people know how to use cameras, other's don't and just point and shoot... even if their camera has a ton of professional features.

I also see most screenies that are just the screen. I always like to take a bit of a wider shot that included part of the room itself so there is a reference in the shot. Not all people do this which does cause an inconsistancy between what people are looking at. Without references it is very hard to tell what we are looking at when it comes to lighting and the room conditions.

Comparing two screens that are vastly different basically can only show deficencies and sometimes that technique has been used to make one screen look better than another when in reality it's an invalid test. Neither screen is being shown in its optimal conditions, and when it comes to calibration one screen is going to suffer, or at best neither screen will be shown in its best 'light'. (Pun intended

)

So I am firmly in camp with both sides of the debate. Overall screenies do tend to be eye candy, but we also can get some information out of certain ones. I'll go back to the broken record again and that being that screenies alone are fun, but when combined with data such as color balance, spectral reflectance curves, and some of the other hard data available, then we can get a very good idea of how the screen performs overall.

**tonyvdb** · **05-11-08, 12:33 PM**

Screen shots do show one thing very clearly and that is if there is any hot spotting as well as if done right can illustrate how good or bad the contrast is in a particular frame of a movie.

If nothing else they are a good way to show others who dont understand what we are trying to do what is going on.

**<^..^>Smokey Joe** · **05-13-08, 06:05 AM**

I might add...

The main issue with photos not withstanding the obvious ones already pointed out is the array of varibles along the whole path.

Consider; for the viewer to see the result as the original observer all stages must be at the same reference. ie Calibrated.

All these stages have to be at the same reference(what ever chosen).

1.Source material/Player output
2.Projector/screen
3.Camera
4.Computer/screen(data crunching)viewing
5.Computer/screen at final viewers end.

Any experiment that is considered real is repeatable, a one off occurance is really an indicator read wrong can lead one astray.

1. Although many use calibration DVDs as a source, alot dont. Mistake one. Reference material must be used. Then the player, how accurate is yours? Mine truncates 10% off the bottom end, which I didnt know until I referenced it against a video generator. I know not everyone has a video generator at hand , but you get my point. Minimalise the in accuracies by using reference DVDs.

2. This is more complicated. Alot dont do a full calibration(greyscale). Alot do. But what to calibrate too?
The PJ direct, so iliminating the screen surfaces? Feasibly the best to show a camera shot, however I proved to myself that this also doesnt work for the reason that until each surface as a reflection is calibrated the surface is not at its full potential. The is the fundemental issue with screen shots!
I actually found that calibrating to each surface under experiment then photographing each of them with the same parameters(exposure, ISO, fstop and so forth(no white balance,ill get to that).Then cut up the photos in the PC so one can compare each surface to the same reference visually. This yielded the best image differences.

3. The white balance on a camera. How many know you can calibrate a camera? (a bit like calibrating a printer). The overall white balance can be the least of the cameras issues.
Anyway, I asked a semipro top photographer here(wildlife) about how he would shoot screenies.
He said the problem with doing a white balance on the fly is that one can be off distorting the captured raw material. Alot of cheap/mid priced cameras don't offer RGB white balance, only predifined white balances which are generic.
Here he said it is best to use your PC graphic programe to set the white balance. (of course you need a reference card or white balance material shot). Capture the shot as is, concentrating on getting the exposure time correct with the least noise(digital noise). Use those built in histograms to get the exposure right. I always use ISO80 setting with long exposure to get less noise for example.
I tend to agree with his theory. Others may not, but i found his method fruitful.

4. Calibrate the monitor/PC. The meters bought for calibrating your Projector is designed for PC's...use em.

5. If you the final viewer has a meter for calibrating and you havent calibrated all the pc's in your home, shame on you

I do more PC calibration than home theatres, most are blown away how monitors compare before and after.

As you can see(pun) the varibles can add up, so what are you really looking at? what is someone else seeing relative to you?

In some ways you can compare some relative differences, however one has to trust the original observer in their observations which one can easily be led astray by enthusiasim(not directed at anyone, a general comment).

Overall the above issues are the reason why numbers measured with traceable equipment to known references remove error and varibles. One has to visualise of course what this means.
Screenshots really document proceedures and show proof of work.

**mechman** · **05-13-08, 09:28 AM**

Quote:

<^..^>Smokey Joe wrote:

Overall the above issues are the reason why numbers measured with traceable equipment to known references remove error and variables. One has to visualise of course what this means.
Screenshots really document procedures and show proof of work.

I agree with this whole heartedly!! Screen shots are nothing more than an addendum to the data. If there's no data, then they're (screen shots) nothing more than useless fluff.

**kbgl** · **05-15-08, 09:21 AM**

I agree that screen shots are mostly for entertainment. I've taken a bunch in the last month while trying to learn how to get the camera to cooperate. I've had lots of pictures come out tinted green. After reseting the white balance, only a hint of green tint is there on occasion. Most shots are oversaturated. But every once in a while I get a good one. I think they can be handy to show certain aspects of picture quality, or to show to someone who has not seen front projection other than at a retail store with tons of ambient light. The shot below is my favorite, and just fluff as mentioned above.

Advertisement