Logical Bracketing Method Needed

[John Locke]

One thing I'm definitely guilty of is staying within a limited range of what I know works...and also trusting the auto settings too much on the XL-1. Like a lot of people, I read things here or elsewhere, test them out, and if they work just stick with them. But then when I start reading about variances in gain, AE Shift, shutter speed...and all the combinations of them, it gets a bit confusing...and frustrating.

Well, it's time to get around that. I've been wanting for a long time to do some testing...problem is, I've never found a logical bracketing or testing method to do this extensive testing, so without guidance it's hard to know how to go about it correctly. I've searched the internet and books trying to find something that gives "assignments" to help you learn the logistics of the camera. For instance, "Today you'll learn how to shoot in low light situations. First, set your mode as..."

So, without anything to guide me, I'll tell you what I plan to do, and if you think there's a better way to do it, please chime in. Here's the plan:

I'll divide the testing into 3 main areas...outside (daylight), indoors (indoor lighting, and then outside at night with a lot of lights (like a brightly lit street scene).

For each shoot, I'll set everything at "0" and start with fully automatic and take a shot. Then, for each shot thereafter bracket the AE Shift up or down until I've covered that full range.

Next, move the gain to -3 db and do the whole AE Shift test again at each gain setting, and consequently move up and down the entire gain range until all is done.

My thinking is that once I find a combination I prefer with gain and AE Shift, I should keep them as they are and then start bracketing different aperture and shutter speeds, one cycle for each different camera mode from fully auto all the way to spotlight mode. (I figure shooting in photo mode will be equivalent in exposure and resolution to the way the shots would appear in frame mode and will help with quick indexing...am I wrong in this?)

Once all is done, I'll chart the stills side-by-side in a graph-style setup to compare differences.

Then, after I've run through these tests, I'd like to chart the differences between the same shot with various filters and filter combinations.

This is no small task, I know...but I really want to know how to get the most out of my XL-1 in all situations...especially since I've never gotten any really good low light shots. With your help in keeping me on a logical path, maybe this testing set up can serve as a model for others who want some kind of guide to follow to learn the ins and outs of their XL-1s. After all is said and done, if Chris is interested, I'll zip the whole thing up and send it to him.

And then after that...more testing...with my true nemesis, audio!

[Bill Ravens]

Z....

BRAVO!!! I like your idea. AFAIK, the XL1s is a wonderful tool....but, that's all it is...a tool. The real talent and vision comes from within the artist. Like a painter's brush, the ability to get the inner vision onto a physical medium is a function of how well the artist can manipulate his/her tools. A good musician has spent hundreds, if not thousands, of hours in some venue, learning to be one with his guitar, piano, whatever. The same is true with photography. The more flexible the tool, and the XL1s is very flexible, the harder it is to use it in a professional way. Understanding how the tool responds to all its variables is a critical step to being professional. In my own mind, anyone can crank out content...kind of a production diahrea...that really means very little. Just watch a local news broadcast to see the garbage being produced on video. A well constructed piece really shows the skill, talent and vision of the creator.

ahhh....philosophy.....there's too much emphasis on quantity over quality....guess my age is showing, eh?

I would offer the suggestion to include playing with a light meter as a comparison with the internal metering of the XL1. An incident light meter is best, but, sometimes, long distance shots make an incident meter unreasonable.

[John Locke]

Thanks for the quick input, Bill...and the encouragement!

I'll add that to the menu. That raises another question, though. My light meter (used extensively for still shots but not once for video) is a Minolta IVF. It can be used for film, but the fps is set at 24 fps (I believe...I'll have to check my manual...whatever it is, it's different than the XL-1 fps rate).

So, I'm not sure if there's a way around this, or if I've just got the wrong light meter.

[Bill Ravens]

LOL.....one of my own "old" light meters is a Sekonic L206 spotmeter. It's a terrific light meter, calibrated for the Zone System. Unfortunately, it runs on 1.35v Mercury batteries that are no longer made. I haven't been able to find a replacement battery, so the meter is retired. I've looked at film light meters. Besides their very high cost, it would add to my learning curve. For now, my Weston Master 6 gives me a simple baseline to use as a comparison with the XL1 internal meter.

On another thread I've read questions about the "factory setting" on the XL1 light meter being off by as much as one f stop.
This is a disturbing thought, especially for random point and shoot.

[John Locke]

Bill,

Out of curiosity, I dug out my Minolta manual and meter and checked the "Cine" feature it has. Here are the problems related to video and the XL-1:

1. You first have to input an ISO (who knows?)
2. The available frame speeds jump from 24 to 32 to 64 with nothing in between.

So...looks like light meter testing won't be a possibility. I can, however, make sure the XL-1 meter and the Minolta show the same value range.

Oh well...

[Bill Ravens]

Yeah, the shutter speed vs frame speed analogy is a little bewildering....at least I find it so. In the days of film and mechanical shutters, frame speed, shutter speed and ISO were intimately tied together.

In a DVCam, shutter speed is a virtual term since "exposure" of the CCD is electronically controlled and there is no mechanical shutter. Ignoring aperture for this discussion, actual exposure is controlled by a buffer in the electronics. The ability to adjust the gain, on the fly, is like a variable ISO/ASA film rating. After tests on my own XL1s, I've found the ASA rating to be approx. 160....with zero gain and zero AE setting. Changes in the AE setting are worth 1 stop for every unit (1) value on the AE dial. As we know, one f stop is worth doubling or halving the ASA rating..so, for example, changing the AE setting to +1 effectively changes the CCD ASA rating to 320. I still haven't run tests to try to relate db's of gain to f-stops. For a rainy day, I guess.

In terms of actual use, I am of the current opinion that the frame speed, as referred to film motion cameras, is a little different than frame speed when referenced to a digital vidcam. I also think that a still camera light meter has relevance to DV for this reason and a film camera meter does not. One of these days, I'll have to delve deeper into the meaning of shutter speed/frame speed when referring to a DVCam. I know that the aperture(f stop) setting is still relevant to film, that's a comfort...<g>

[Chris Hurd]

John, of course I'm highly interested and would be most anxious to add this stuff to the Watchdog. Well done,

[Charles Papert]

Some things that I do know and others I believe that I know:

To test for the ASA/ISO reading of a given camera that has built in metering, you need to point the camera at a 18% reflective gray card (available at photo stores) and put the camera on an auto setting. Then meter the card with an incident meter facing the light source (or a spot meter pointing at the card, they should read the same). Adjust the ASA/ISO of the meter until the stop on the meter matches the stop indicated in the camera, that will deliver the ASA/ISO rating.

However, my experience with digital video is that it does not follow a standard sensitivity curve, and that as you move towards underexposure the ASA rating will effectively climb. Thus a camera that is measured at a fat exposure (say 2000 footcandles) as having an ASA of 160 may actually deliver more like an ASA of 500+ in low light, say 25 footcandles. This is why an HD camera was used to shoot the opening scenes of Michael Mann's recent film "Ali", when Will Smith was running at night. The HD camera captured the details of clouds against the moon in a way that current film emulsion cannot without pushing several stops and resulting in excessive grain.

As far as the effect of gain, I believe 3db of gain represents a single stop.

Cine cameras require a meter setting of 1/50th of a second, rounded off from 1/48th since the shutter effectively halves the exposure time of a film frame, 24 times a second. Video speed is, as I recall, effectively 1/30 of a second, regardless of progressive or interlaced (happy to be proved wrong on this).

Myself, after years of using a meter for video work, I now will just meter the key exposure and adjust the rest of the scene to eye if prelighting, and from the monitor while shooting. I rarely even bother with a waveform monitor anymore, since the digital signal processing is not as sensitive to issues such as clipping etc.

[Bill Ravens]

Charles.....

Thanx for some great input. I was not aware that the sensitivity curves were different. When using a light meter, I usually expose for the highlights(zone 8 or 9) and let the shadows fall where they may. Occassionally, if it's a full face shot, I'll expose for the skin at Zone 7. The non-linearity you describe is apparent at greater than 2 f-stops difference so it shouldn't have a great effect. Altho', you're right that it's a 4 stop difference from that 18% gray card to zone 9. Does the non-linearity extend to overexposure, as well? I tend to rely on the in camera meter for low light situations, anyway.