| Dwiel |
What I am proposing is a thread where we can decide on some standards about how we can compare results. Right now, it is very hard for someone to tell from pictures how good the picture is, what the contrast is like, what the screen is like, etc.
I think that if we had some standards as to how we compare our pictures, we could compare much mroe easiely. This would allow us to more easily determine how much better screen A is to screen B that is on the other side of the world ;-).
For screen comparison, I suggest that we all put a piece of standard paper in front of the screen and take pictures from some predifined angle relative to the screen, and maybe even a certain distance relative to the size of the screen.
Moslty I think that if we all had a piece of paper in the picture we could tell how much better the screen is than the paper. We could then compare the differences between the paper and screen of each person instead of trying to compare screens directly.
Also, some other diyvideo place for germans or something had a picture that everyone used, a kind of standard test pattern, to proove the picture quality of their setup. This might be helpful again so we can compare the output a little better. This way you dont have to worry about if the picture is of a dark sceen or w/e.
I think that with more standards, we will be able to better compare our results and thus learn from eachother's work more easily.
Let me know what you think... any ideas/comments/questions/suggestions?
post any other standardization ideas you have also
Thanks! |
|
|
| dracul |
This is a great idea. Here are my recommendation.
I started making a list of shooting standards for photos and then realized there are to many variables to control by the time we see the image here.
But since I dont think we have anything better to do on we should go this route. Digital camera owners only I think.
1. Set your digital camera to 200 asa sensitivity. Shoot at 1024x768 res or whatever rez but everyone shoots at the same rez.
2. Set your monitor at the best contrast and black levels first.
Blacks are as black as possible while whites are still bright white.
3. Since you took various exposures with your digi camera at small exposure increments, pick the one that has the best image .
This means blackest blacks while stil brightest whites. You are not to mess with anything else.
4.Upload the picture in original tif format or uncompressed format for analysis.
5. Viewer first makes sure his monitor is optimized for optimum contrast and colors are set as realistic as possible. You do this with a test or color chart available online I think.
6. Analize the uncompressed digi camera shot.
7. Everyone should take shots in complete darkness on blackout cloth of the same size.
8. Photographer should also take note that there are no harsh hits of spill hitting the screen such as spill leaks from the wide diylabs triplet hitting the floor and than on the screen or from the ceiling. Basically the setup should be ideal before you take the shots.
Now there will still be variables since not everyone will use the same digi camera but I think this is the best we got. |
|
|
| Dwiel |
I do not know much about digi-cams because I dont have one, but what you have suggested sounds good to me :-)
One thing that I am unsure about is what to do with people with different sized screens. I screen that is projected 13' is goinf to be much different than the same setup onto a 6' screen. Also, if you have a really nice screen, your pictures might look better than someone else's who might even have a better projector but worse screen... What do you think about this difference? I think that if there is a piece of paper in there, or some other surface that everyone can compare their screen with, we will be able to tell if the screen is making the difference or the projector. This also kind of fixes the problem we have when someone with say an alluminum screen wants to share his projector's results. If he puts a piece of paper up there too for comparison, we can see what the difference bettween his projected image on paper and everyone else's. It also introduces a good way to see the scale without blocking the picture, as long as a standard sized piece of paper is used. This is better than putting a person in front because that blocks some of the view and kinda makes it more difficult to compare...
Just some thoughts. Again, I cant say much about the digicam ideas because I dont know enough about them, but all of you digi-cam owners out there? Does dracul's proposal sound good? |
|
|
| dracul |
The piece of paper wont cut it. We need to see the image big and full screen or larger on our monitor when we analyze.
We also want to present the analyzers with aproximately the same setup that we would be actually watching the movies.
As soon as we have different screens , sizes it throws everything out of wack way to much.
Unfortunately those wishing to contribute will have to make the effort to setup a blackout screen cloth or something else that is cheaper. I dont think it needs to be 100 inches. Maybe
50 inches will be ok.
Those using stronger or weaker bulbs will result in more light on the screen but the digi camera should compensate for since you choose the photo that is the same darkness/brighness , contrast level. So a brighter screen image will have a brighter image but the camera will expose it darker. ( You pick the most accurate exposure). You must pay close attention to everything when selecting the image here.
Take shots in small exposure increments both ways (brighter and darker).
The list should also include:
9. A standard source still image at dvd resolution. Preferably a jpg photo. Not dvd video movie paused or guessed at roughly the same frame.
10. Dvd desktop progressive player may be better. There may be less variables than various video cards if we go with the pc standard for output. Not sure on this one.
This may seem anal but we want to minimize as many variables as possible. I `m sure some of you might be saying screw this it aint worth it. If you guys things its too much speak up for something in between. |
|
|
| Takusama |
| quote: | Originally posted by dracul
This is a great idea. Here are my recommendation.
I started making a list of shooting standards for photos and then realized there are to many variables to control by the time we see the image here.
But since I dont think we have anything better to do on we should go this route. Digital camera owners only I think.
1. Set your digital camera to 200 asa sensitivity. Shoot at 1024x768 res or whatever rez but everyone shoots at the same rez.
2. Set your monitor at the best contrast and black levels first.
Blacks are as black as possible while whites are still bright white.
3. Since you took various exposures with your digi camera at small exposure increments, pick the one that has the best image .
This means blackest blacks while stil brightest whites. You are not to mess with anything else.
4.Upload the picture in original tif format or uncompressed format for analysis.
5. Viewer first makes sure his monitor is optimized for optimum contrast and colors are set as realistic as possible. You do this with a test or color chart available online I think.
6. Analize the uncompressed digi camera shot.
7. Everyone should take shots in complete darkness on blackout cloth of the same size.
8. Photographer should also take note that there are no harsh hits of spill hitting the screen such as spill leaks from the wide diylabs triplet hitting the floor and than on the screen or from the ceiling. Basically the setup should be ideal before you take the shots.
Now there will still be variables since not everyone will use the same digi camera but I think this is the best we got. |
Haha, now number 7 would be pretty hard to follow.
Anyway, I think that most of us have different screens so I purpose that with the picture, you just give general information on what screen/DIY screen you are using.
Also what camera you are using and what settings. Shutter speed and ISO is quite important. And 1024*768 is a bit too big, 640*480 or 800*600 is little more fitting, wouldn't you say?
Also any other kind of information, like cables for example.
Oh and that official test-pattern is a pretty good idea ;) |
|
|
| Dwiel |
Is there any way you guys know about so we not only make up some of these good ideas to use as standards, but for people to actually follow them more or less...
I think that it will be much harder to get people to use the standard than to actually come up with the standard...
Any ideas?
Thanks |
|
|
| Takusama |
You know what?
The only standards I really would like to have is information on all the related stuff in the picture. What LCD are you using, what bulb, what kind of screen is that, what cables are you using, what camera are you using to take these pictures and so on.
That's the kind of standard I want, because you can never have too much information on these things.
A official test-pattern would be pretty neat though. |
|
|
| Dwiel |
OK, well for the test pattern, what about one of these:
    
Maybe a combination of these
Some are used for convergence on CRTs, but might still be helpful |
|
|
|
