Experiments with older flash "Auto" modes

[reakeener1970]

So, out of curiosity yesterday I connected my Nikon SB-24 and SB-28 speed lights to my D7100 to see how well the flashes' auto modes work.

I figured I wanted my histograms to be on the high side (a little overexposed) (I am doing this to see how well they will work in auto mode on my film cameras -- so I wanted histogram peaks in the upper part of the histogram to account for B&W film vs digital sensor.) (Basically using my digital camera as a *real* (film) camera simulator (ha ha).)

Anyways, over about an hour and many shots with repeatable results (despite nonscientific method), I found that I needed at least two stops, and usually three or four, of "overexposure" to get the correct exposure in the camera.

I had my camera settings locked in at:
400 ISO (I usually use 400 speed B&W film) and
1/60 sec (common sync speed for older cameras)
(all auto stuff: Active D lighting, Auto ISO, etc. etc. disabled and camera set to monochrome mode -- not that that mattered since I was primarily looking at the histogram)

I had to tell the flash (pretty much same results on both the SB-24 and SB-28) that I was shooting either 25 or 50 ISO (remember my camera is set to 400 at this time still) in order for the flash to fire powerfully enough to give me the exposure I wanted. I got similar results regardless of whether I fired the flash directly at the subject (which was a standard gray card, btw), or bounced it horizontally or vertically.

As long as I kept it at 25 or 50 (3 or 4 stops over), the auto function did give me pretty consistently good results. Similarly, I also played around with leaving the flash set to 400 and telling it I was shooting at f/8 or f/11 (while I was really at f/2.8) and it predictably did the same at setting its ISO to 25 or 50 (figured it would, just experimenting here).

I just found this interesting because there are some situation for which I would like to use the Auto function. It might also account for why some of the film shots I have taken on other cameras with other auto flashes (ones I can't test on my D7100 for fear of frying its circuits) are chronically underexposed.

Anyone else played around with this at all -- I would be curious to hear your results.

 

[reakeener1970]

Also, fyi, this was outside in an overcast late afternoon, so 1/60 f/2.8 and 400 was not too far off ambient background light and I was using this to test how well the auto function would work for fill lighting as I was under an area of shade and my gray card was also under the shade area with me.

 

[MartinCrabtree]

Were the SB-24 and 28 consistently the same?

 

[BrianShaw]

I always applaud experimentation but why not use them with the intended technology, a film camera? They worked as designed back in the olden days and should still... If used as intended.

 

[reakeener1970]

...Aaaandd now I am trying the same thing again in a different setting -- in my house -- and it works best if I use the proper settings (no tricking it into overexposure). If I do my "trick" I get wildly overexposed images now.

So I don't know. The range is further today than yesterday (2.5 to 3 meters instead of 1.5 and in a larger space.)

 

[reakeener1970]

Were the SB-24 and 28 consistently the same?

yes

 

[reakeener1970]

I always applaud experimentation but why not use them with the intended technology, a film camera? They worked as designed back in the olden days and should still... If used as intended.

I do, but usually in TTL mode and not auto. Like I said, I was using my digital as sort of a simulator so I could see the results right away and play with settings.

 

[analoguey]

What about flash exposure settings in the DSLR itself? The metering it did, or what info it sent to the flash?
That could have skewed results?

sent from tapatalk

 

[MattKing]

In my film camera experience with negative films, sensors on flashes tend to assume indoor exposures, complete with at least some contribution from flash output bouncing off nearby floors, walls and ceilings.

In the days when I used to do a fair bit of outdoor fill flash work (weddings mainly), my Metz or Vivitar flashes used to give pleasing results if the sensor was set for one stop larger than the aperture on the camera (e.g. f/11 when the camera was set to f/8).

 

[reakeener1970]

What about flash exposure settings in the DSLR itself? The metering it did, or what info it sent to the flash?
That could have skewed results?

sent from tapatalk

I had it set to center-weighted metering. It doesn't seem to send anything to the flash -- I still have to set the ISO, Zoom, Aprture, etc. settings in the flash manually. I would think that if it was sending ANY info to the flash, ISO would be it.

Speaking of which... a little plug for the older SB-24 over the SB-28 is that on the SB-28 you have to set the ISO first thing after you turn it on and the SB-24 you can set or change whenever you need to. Normally not an issue, but it made my life easier in these tests with the SB-24. Plus I like the ease of the switches over the multiple mode, etc button presses of the 28. Of course, on the F5 the SB-28 gives me pre-flashes and the fancy 3D metering, and is considerably smaller than the SB-24.

 

[David A. Goldfarb]

Your ambient exposure is about 4 stops over to begin with, so it shouldn't be surprising that you need 3-4 stops strobe exposure above expectations to balance your ambient exposure.

At ISO 400, 1/60 sec., you should be around f:11-16 on an overcast day.

 

[BrianShaw]

In my film camera experience with negative films, sensors on flashes tend to assume indoor exposures, complete with at least some contribution from flash output bouncing off nearby floors, walls and ceilings.

In the days when I used to do a fair bit of outdoor fill flash work (weddings mainly), my Metz or Vivitar flashes used to give pleasing results if the sensor was set for one stop larger than the aperture on the camera (e.g. f/11 when the camera was set to f/8).

I had the same experience using auto flash, Vivitar 285s, in underground mines where there was no ambient.

 

[markbarendt]

...Aaaandd now I am trying the same thing again in a different setting -- in my house -- and it works best if I use the proper settings (no tricking it into overexposure). If I do my "trick" I get wildly overexposed images now.

So I don't know. The range is further today than yesterday (2.5 to 3 meters instead of 1.5 and in a larger space.)

Tricking the Nikon strobes can help but for fill I normally have to trick it the other way.

Ok so a couple foundational things about shooting with flash.

1 - The shutter time is important to sync the strobe with the shutter (mechanically) but the shutter duration itself is normally irrelevant to the flash exposure as the flash duration is typically 1/1000th of a second or even a lot faster than that. Flash duration is what controls flash power.

Also, fyi, this was outside in an overcast late afternoon, so 1/60 f/2.8 and 400 was not too far off ambient background light and I was using this to test how well the auto function would work for fill lighting as I was under an area of shade and my gray card was also under the shade area with me.

2 - The ambient exposure still works like normal.

So, when in auto mode the Nikon speed lights (not the camera) measure the light that hits a sensor in the flash unit and it cuts off the flash when it gets enough light. Here's the rub, if it sees enough ambient light it won't even fire the strobe.

No surprise to me that on an overcast day you got very little fill without tricking the system. On a clear bright day with your subject under that tree the flash will work harder.

...Aaaandd now I am trying the same thing again in a different setting -- in my house -- and it works best if I use the proper settings (no tricking it into overexposure). If I do my "trick" I get wildly overexposed images now.

So I don't know. The range is further today than yesterday (2.5 to 3 meters instead of 1.5 and in a larger space.)

Range can be a factor, but if you are in the range the speed light indicates it shouldn't matter.

The difference you are seeing and the speed light is seeing is because of the difference in the lighting. In overcast light the clouds provide most of the light, indoors the speed light typically starts providing most of the light. The speed light is working perfectly.

What the speed light isn't designed to do in auto mode is balance your exposure like TTL can. All it cares about is "did I see enough light?".

Inside is where flash in auto mode gets fun, just don't point the strobe head at your subject. Tilt and swivel the strobe's head so that the light bounces off a wall or ceiling or both. This fills the whole room and provides directional light and then you and the strobe get control of the whole room. Play with that a while and you'll see.

 

[Sirius Glass]

The flashes measure the ambient light and are working as fill in flash. Also it will work correctly of something called FILM!!!!

 

[markbarendt]

The flashes measure the ambient light and are working as fill in flash.

In auto mode speed lights measure the total volume of light during the exposure (ambient and artificial) and fires the strobe until the "correct" total is reached.

Also it will work correctly of something called FILM!!!!

Yep.

 

[Chan Tran]

Also, fyi, this was outside in an overcast late afternoon, so 1/60 f/2.8 and 400 was not too far off ambient background light and I was using this to test how well the auto function would work for fill lighting as I was under an area of shade and my gray card was also under the shade area with me.

May be this is the problem. Underexposure when the ambient light is quite high. I have found my Sunpak PZ5000AF which was supposed to be compatible with my Nikon F5 and it refuses to fire when the ambient is bright.

I don't have your newer SB-24 and 28 but I have older SB-15 and SB-16b and they work quite well. Not underexposure and sometimes a little overexposure but at a degree correctable if I shoot RAW.

 

[Sirius Glass]

I don't have your newer SB-24 and 28 but I have older SB-15 and SB-16b and they work quite well. Not underexposure and sometimes a little overexposure but at a degree correctable if I shoot RAW.

What is this RAW of which you speak? Is it how you have your meat cooked? I hope that you are not talking dirty again!

 

[pbromaghin]

Your ambient exposure is about 4 stops over to begin with, so it shouldn't be surprising that you need 3-4 stops strobe exposure above expectations to balance your ambient exposure.

At ISO 400, 1/60 sec., you should be around f:11-16 on an overcast day.

Ya. OP is 8 stops from sunny 16 which, according to Fred Parker is equal to:

"Las Vegas or Times Square at night. Store windows. Campfires, bonfires, burning buildings. Ice shows, football, baseball etc. at night. Interiors with bright florescent lights."

If you're getting this in late afternoon, there is definitely something weird going on here.

 

[Steve Smith]

I had the same experience using auto flash, Vivitar 285s, in underground mines where there was no ambient.

My father used Viviutar 283s and 285s when they were first available and stated that at the time, they were the only flashes they coud find which actually put out the amount of light which the manufacturer claimed.


Steve.

 

[BrianShaw]

My father used Viviutar 283s and 285s when they were first available and stated that at the time, they were the only flashes they coud find which actually put out the amount of light which the manufacturer claimed.


Steve.

Last year I bought several very old and very used 285s. Surprisingly when I verified GN with a flash meter they were almost dead on.

 

[wiltw]

So, out of curiosity yesterday I connected my Nikon SB-24 and SB-28 speed lights to my D7100 to see how well the flashes' auto modes work.

I figured I wanted my histograms to be on the high side (a little overexposed) (I am doing this to see how well they will work in auto mode on my film cameras -- so I wanted histogram peaks in the upper part of the histogram to account for B&W film vs digital sensor.) (Basically using my digital camera as a *real* (film) camera simulator (ha ha).)

Anyways, over about an hour and many shots with repeatable results (despite nonscientific method), I found that I needed at least two stops, and usually three or four, of "overexposure" to get the correct exposure in the camera.

I had my camera settings locked in at:
400 ISO (I usually use 400 speed B&W film) and
1/60 sec (common sync speed for older cameras)
(all auto stuff: Active D lighting, Auto ISO, etc. etc. disabled and camera set to monochrome mode -- not that that mattered since I was primarily looking at the histogram)

I had to tell the flash (pretty much same results on both the SB-24 and SB-28) that I was shooting either 25 or 50 ISO (remember my camera is set to 400 at this time still) in order for the flash to fire powerfully enough to give me the exposure I wanted. I got similar results regardless of whether I fired the flash directly at the subject (which was a standard gray card, btw), or bounced it horizontally or vertically.

As long as I kept it at 25 or 50 (3 or 4 stops over), the auto function did give me pretty consistently good results. Similarly, I also played around with leaving the flash set to 400 and telling it I was shooting at f/8 or f/11 (while I was really at f/2.8) and it predictably did the same at setting its ISO to 25 or 50 (figured it would, just experimenting here).

I just found this interesting because there are some situation for which I would like to use the Auto function. It might also account for why some of the film shots I have taken on other cameras with other auto flashes (ones I can't test on my D7100 for fear of frying its circuits) are chronically underexposed.

Anyone else played around with this at all -- I would be curious to hear your results.

I just did this test so you have comparative information...

• Flash equipment: Metz 54MZ-4
• Target: Macbeth Color Checker placed on a sofa which measures -0.6EV compared to 18% grey card and which dominates the frame.
• Lens FL set to 'normal' AOV (30mm), flash coverage angle set automatically to 28mm
• Ambient light: ISO 400, 1/15 f/2.8
• Camera set to ISO 400 1/100 f/2.8
• Minolta Autometer Vf Flashmeter measurement of Auto mode (flash photosensor) as a 'control' measurement: ISO 400, 1/100 f/2.0 +0.5EV

ETTL mode shot, as comparative 'control' shot.

The Auto mode photosensor flash shot

The two histograms for the above shots

Conclusion: Virtually same results between ETTL vs. flash Auto mode.
I have always found Metz photosensor mode to be accurate and reliable, given the inherent limitations of a fixed photosensor AOV and the variable of background and subject brightness. I have, in the past, relied upon 135 and medium format film cameras with TTL capabilities for increased accuracy flash exposure because of the flash AOV being same as the lens on the camera. This comparison shows that my Metz Auto mode flash is equivalent to what I would get with digital camera ETTL, and the flashmeter measured exposure was within 0.5EV of the lens aperture.

 

[BrianShaw]

Interesting... not too unexpected.