Sensitivity test - try to understand bad results

[vicmortelmans]

Hi,

This is an experiment I did:

- displayed an SMPTE test image on a PC monitor (LCD), such that (to the eye) the 5% white and 5% black squares are visible. (look here for an example: http://brighamrad.harvard.edu/research/topics/vispercep/tutorial.html; you may have to adjust monitor brightness/contrast to see the squares!)

- took a picture of the monitor (no ambient light) with average metering on my regular b&w film

- developed and scanned the film according to normal processing

Looking at the raw scanning result, the 5% white (max density on film) square is visible, but the 5% black (low density) is not. This area is quite grainy as well. I should have bracketed the shot to try out higher exposure, but it's actually confirming what I see in shadow area's in my regular shots.

Would this be normal? I thought that the brigthness range of a monitor is much lower than of a real-life subject. If my film/developing/scanning process can't reproduce detail that is visibly displayed on a monitor, how can it ever satisfactory reproduce a real life scene?

I'm having trouble finetuning my processing to improve quality, but looking at this result, it seems like I'm still far off...

Maybe someone can redo the experiment to check against his film/developing/scanning process, or check with digital?

And how can I improve shadow detail? Just rate the film at half or 1/4 ISO?

Groeten,

Vic

 

[ClaremontPhoto]

By photographing a computer monitor you're introducing a new varaiable and muddying the waters.

You can get a test card (on paper) to test your set up, and I suppose you might be able to download one from the internet also.

How are your regular photos? I don't know your work but your avatar seems rather dull. Is this typical of your photos?

 

[vicmortelmans]

I agree that my test setup is quite random, but downloading a test card and then printing it? That's even more 'muddying' I'd say.

Actually, I assumed that my negative would be capable of reproducing monitor detail without any problem, because to my knowledge, monitor brightness range is much smaller than real life scene brightness range.

And in general, most of my pictures are quite OK, but I have a hard time reproducing shadow detail.

Here's some recent material (only look at b&w for this discussion), but obviously this only shows the best results...:

http://users.pandora.be/vicmortelmans/fts/2007/2007_oefening/index.html

Groeten,

Vic

 

[ClaremontPhoto]

Your photos look fine for tonal range and exposure.

Carry on making photos like these and you'll be ok, or better.

 

[willie_901]

Vic,

I too have a lot to learn about capturing shadow detail.

The SMPTE procedure you referenced is to test a video display monitor, not how tour your camera, film and scanner can photograph a monitor.

Your results are confusing because display monitors are designed to work well with the human eye and brain. They are not designed to work well with a lens and film (or a digital-camera sensor).

If you are interested in testing how well your negatives are modeled by your scanner, you need to do a different test. You would have to purchase a SMPTE (or other test card); photograph it, and scan it.


I tend to agree with Jon Claremont though. Just go out and shoot. You can experiment with exposure technique, film selection and development methodology to understand about shadow detail.

My advice is to obtain and study "The Black & White Handbook" by Roger Hicks and Frances Schultz. I'm sure others can recommend books and on-line articles that would be equally useful.

willie

 

[Finder]

SEMPTE test is a video test. So naturally it won't work to test film sensitivity. There is are several good reasons - camera metering does have a problem with self-illuminating objects. Black is simply off and white is not always white - monitors cycle at a certain frequency and we only see a unflickering image because of persistance of vision. And you meter is averaging the target to 18% gray, which can cause this unusual scene to be underexposed.

Looking at the raw scanning result, the 5% white (max density on film) square is visible, but the 5% black (low density) is not.

How are you judging these are maximum and minimum densities? Where do you get these numbers? You just said the monitor contrast is well below and average scene so you are not approaching those densities ranges in the film. So naturally your image will have low contrast.

I would give up the monitor test. It won't work as you clearly found out. Get a gray card and gray scale and photograph them outside. Also bracket. A test of a single exposure is not going to tell you much as exposure controls shadows and development controls highlights. And there is a volume of litarature telling you how to determine your personal film speed based on your camera, lens, development combination, which is what you are doing.

 

[willie_901]

Vic,

I just stumbled upon this link:

http://www.butzi.net/articles/scannersoft.htm

I think you might find it very useful.

willie