Who uses a sky quality meter for estimating night time exposures?

Rudeofus · Nov 25, 2009

A recent thread here described in detail, how people estimate and bracket night time shots. This seems necessary because most light meters don't cover such dark scenarios. From my brief forays into astro photography I came to know about the "Sky Quality Meter" which essentially tells one how dark the night sky really is. Could this help me estimate accurate exposure times for night shots as well? Technically it seems like a very sensitive light meter, all one needs to do is establish a table for converting from their magnitude readings into exposure values. I could even imaging creating different hoods for the light sensor so I could restrict its angular coverage (would obviously require creating separate conversion tables for each hood).

Has anyone tried this? Does it work for this purpose?

Steve Smith · Nov 25, 2009

I have never heard of such a thing. This did remind me though of the method my father taught me for metering sunsets: Point the meter straight up vertically. I can remember trying it once but can't recall how successful it was.

Steve.

Rudeofus · Nov 25, 2009

Steve Smith said:
This did remind me though of the method my father taught me for metering sunsets: Point the meter straight up vertically. I can remember trying it once but can't recall how successful it was.

I read that for metering sun sets one points the spot meter about 6 solar diameters off the sun, this has worked very well for me at least. The difference between this and night shots is that with night shots your exposure meter simply says "there is no light at all" when an 1 hour exposure confirms the opposite. It may even be useful for estimating maximum useful exposure time for star trail shots, before the back ground brightness registers on my film.

If you search for "sky quality meter" you will find countless hits in astro forums, so it seems to be used in this area for predicting faint star visibility. I'm about to plunge down the money and get one, but would love to hear beforehand from people already using this device for night shots.

Lee L · Nov 25, 2009

Rudeofus said:
I read that for metering sun sets one points the spot meter about 6 solar diameters off the sun, this has worked very well for me at least.

Interesting idea. Which direction do you use for the offset? Is that recommendation designed to be metered on clear sky?

6 solar diameters is only 3 degrees, and I've seen published measurements on significant differences in flare among 1 degree spot meters, specifically the Minolta vs Pentax (although this was years ago and I don't recall which specific Pentax). In any case, not having tried this myself, I would expect the difference in flare would mean a variation of at least a stop or two when metering this way. Which spot meter are you using?

I'm not saying this isn't a good approach, just that the recommendation may be more specific to a particular model of meter than might be apparent at first glance.

Lee

PVia · Nov 25, 2009

Google for The Ultimate Exposure Computer...very, very accurate for a variety of situations.

Galah · Nov 25, 2009

PVia said:
Google for The Ultimate Exposure Computer...very, very accurate for a variety of situations.

Or you can just set your Olympus OM2N to "Auto", and press the shutter release (using a tripod, of course)

wiltw · Nov 25, 2009

The issue is not a meter to read low light, the issue is that film suffers from Reciprocity Failure...the emulsion response is down at the toe of the curve, not where it is relatively linear in response. So even if the meter said 50 seconds for the ISO speed that was loaded, it might really take much more time than indicated.

AgX · Nov 26, 2009

The Recipocity Failure is not linked to any part of the specific curve. You can aim for a certain luminance of the opbject to correspond to a certain density on the film at any part of the specific curve. But under certain conditions (exposures yielding high intensity illuminance at shortes duration and low intensity illuminance at longest duration) the density will be lower than expected.

Rudeofus · Nov 26, 2009

@Lee: The 6 diameters rule is not a scientifically verified law but just a rule of thumb which I apply for sun set shots with silhuettes and with which I have gotten excellent results on slide film so far. I don't know about dedicated spot meters, as my EOS 3 does this job very well with its 1° spot measurement function.

@PVia: The ultimate exposure guide just lists a number of light situations together with an exposure estimate for them. It does, unfortunately not tell me how long I can expose star trails before the back ground turns grey or how long to expose a very dark forrest or landscape (this is also where the in camera meter of my SLR falls short). For good negative film it's not a big deal if you are one or two stops off, for slide film being 2 stops off kills the image.

@wiltw: sure, reciprocity failure needs to be taken into account, but that's something you measure only once for every film (or look up numbers online). As long as I did night time exposures with negative film the results looked great, although I just guestimated the exposure time. Unfortunately slide film isn't that forgiving, so I started looking for a measurement device for extremely low light.

Lee L · Nov 26, 2009

Rudeofus said:
@Lee: The 6 diameters rule is not a scientifically verified law but just a rule of thumb which I apply for sun set shots with silhuettes and with which I have gotten excellent results on slide film so far. I don't know about dedicated spot meters, as my EOS 3 does this job very well with its 1° spot measurement function.

Thanks. I believe you'll find that the EOS 3 has a 2.4% spot meter. The viewing angle of the metering area in degrees will change with the focal length with an in camera TTL spot meter. That's why I assumed from your description that you were using a handheld spot meter.

Lee

markbarendt · Nov 26, 2009

Rudeofus said:
@PVia: The ultimate exposure guide just lists a number of light situations together with an exposure estimate for them. It does, unfortunately not tell me how long I can expose star trails before the back ground turns grey or how long to expose a very dark forrest or landscape (this is also where the in camera meter of my SLR falls short). For good negative film it's not a big deal if you are one or two stops off, for slide film being 2 stops off kills the image.

Actually it does, it essentially provides you a zone 5 setting/reading, like all incident meters try to.

It's up to you where to place the zones in your shot from there.

Rudeofus said:
@wiltw: sure, reciprocity failure needs to be taken into account, but that's something you measure only once for every film (or look up numbers online). As long as I did night time exposures with negative film the results looked great, although I just guestimated the exposure time. Unfortunately slide film isn't that forgiving, so I started looking for a measurement device for extremely low light.

Like reciprocity, the camera setting you might like for a given effect in a given lighting situation is only necessary once.

Rudeofus · Nov 26, 2009

Lee L said:
Thanks. I believe you'll find that the EOS 3 has a 2.4% spot meter. The viewing angle of the metering area in degrees will change with the focal length with an in camera TTL spot meter. That's why I assumed from your description that you were using a handheld spot meter.

Sorry for my confusion. You are correct, the spot meter of my EOS 3 covers 2.4% of the view finder screen, so it covers about 20mm^2 on the film. This corresponds to a circle with a diameter of 5mm. With my 70-210 F/4 I get angular coverage across the diameter of the spot metering circle of about 1.4°.

So while I freely admit that I totally mixed up the spot meter properties, the result at the focal length I used closely resemble the coverage of a spot meter. So my exposure rule of thumb should work with these as well, provided they have lenses reasonably free of flare.

markbarendt said:
Actually it does, it essentially provides you a zone 5 setting/reading, like all incident meters try to.
It's up to you where to place the zones in your shot from there.

The Ultimate Exposure Guide is a list of typical exposure situations, not a meter, and shouldn't be mistaken for one. IMHO it's exactly these kinds of guides and the need for bracketing which drove so many night shooters into the digital camp. My aim is to employ a device specifically made for measuring extremely low light szenarios for my photographic purposes. Bracketing is no fun if your exposures take hours.

markbarendt said:
Like reciprocity, the camera setting you might like for a given effect in a given lighting situation is only necessary once.

My aim is a device which tells me (directly or through conversion tables) what exposure an ideal film sans reciprocity failure would require for a zone 5 exposure. Adding or subtracting stops for specific purposes is trivial, and facturing in reciprocity failure should require much fewer failed shots than guestimating szene illumination with my bare eyes.

I guess I'll just try it out and report ...

Lee L · Nov 26, 2009

Rudeofus said:
I guess I'll just try it out and report ...

I'd definitely be interested in your results. Atmospheric conditions can be difficult to judge by eye, and can dramatically affect the outcome with astrophotography. BTW, the FAQ for the SQM indicates that it's sensitive to temperature, and has an internal temp sensor built into the circuit that compensates. But it needs some settling time, and greater lengths time when the meter is moved between environments of very different temperature.

Lee

markbarendt · Nov 26, 2009

Rudeofus said:
The Ultimate Exposure Guide is a list of typical exposure situations, not a meter, and shouldn't be mistaken for one. IMHO it's exactly these kinds of guides and the need for bracketing which drove so many night shooters into the digital camp. My aim is to employ a device specifically made for measuring extremely low light szenarios for my photographic purposes. Bracketing is no fun if your exposures take hours.

IMHO there are no magic bullet devices and the digi converts probably moved for other reasons because they face real limits for night use; their sensors and batteries tend to die or simply won't do what is asked when a truly long exposure, hours, is needed.

All I'm suggesting here, and at least what I find in my shooting, is that the lighting in a given situation, at a given day and time, at a given location; rarely changes.

Many wedding shooters set up at the equivalent of f4, ei400, at 1/400 for any backlit daylight shot; nail it every time. The reason this works is that the subject's surface that is facing the camera is simply in open shade when the sun is in position for this type of shot. The subject is just placed in the right zone for the shot. That's a subjective decision based on experience in getting what they want.

The light quality outside my home on a moonless night is pretty well fixed, same thing in the forests 50 miles from home, or 500 miles out to sea. Each of these has it's own setting but once that value is known and you add a note the your EV chart on the ultimate exposure calculator, your going to be darn close every time.

What messes me up most, on getting the effect I want in a given situation, is that I misjudge placement or I let a meter lead me astray. I don't think I'm abnormal here either.

PVia · Nov 26, 2009

Here's an 8 minute exposure using info from the Ultimate Exposure Guide. This was exactly the effect I was after. I took one shot, no bracketing...

http://www.flickr.com/photos/viapiano/3346880205/

AgX · Nov 26, 2009

Back to the original question: I can't find a sensitivity (stellar magnitude or whatever) given for that device.

Rudeofus · Nov 27, 2009

markbarendt said:
IMHO there are no magic bullet devices and the digi converts probably moved for other reasons because they face real limits for night use; their sensors and batteries tend to die or simply won't do what is asked when a truly long exposure, hours, is needed.

Obviously some vocal converts may be driven more by ample supply of marketing dollars than by real photographic benefits. Some of the problems with quickly discharging batteries and excessive noise have been overcome years ago (battery grips, image stacking), while analog shooters still resort to bracketing once our light meters prove too insensitive for the task. I wasn't looking for a magic bullet which makes my shots better, just for a metering device for extremely low light situations. If I can see details in a scenery with my bare eyes, a sensitive light meter should be able to deliver accurate numbers.

AgX said:
Back to the original question: I can't find a sensitivity (stellar magnitude or whatever) given for that device.

It's reading corresponds do star magnitude per square arc second. Someone could calculate how this corresponds to exposure values, but I rather let the spot meter of my camera find this out for me (compare spot meter reading of a moderately dark spot to the reading of the SQM). One star magnitude is about a factor of 2.5 (it's a logarithmic measure), so I could put together a table very quickly.

@PVia: I did plenty of night shots with negative film where I just random guessed the exposure time. These films have ridiculous latitude compared to slide film. It's my slide film night shots that look like crap so I started investigating devices for measuring very low light levels with some accuracy.

markbarendt · Nov 27, 2009

Rudeofus said:
Could this help me estimate accurate exposure times for night shots as well?

Maybe this could give you some data, but EV may not be the whole, nor even the biggest issue with your slide film providing poor results.

What is the difference between the real color of the scene and the way you want to render it?

How are you going to judge your filtering to get all three color layers exposed properly for the result you want?

Is your film sensitive to a broader or narrower set light wavelengths than the meter?

markbarendt · Nov 27, 2009

One other question here is; how you plan to adjust the development to correct for the narrow scene contrast?

Rudeofus · Nov 27, 2009

markbarendt said:
Maybe this could give you some data, but EV may not be the whole, nor even the biggest issue with your slide film providing poor results.
What is the difference between the real color of the scene and the way you want to render it?
How are you going to judge your filtering to get all three color layers exposed properly for the result you want?

Once I got the exposure nailed I can deal with color casts (in the worst case in a hybrid work flow, my scanner does much much better with slides than negs). Right now I look at pics with bright white mountains in front of white sky and dim star trails on light grey sky. And very short star trails on black sky. I'd rather spend 10 minutes measuring light and thinking it through than spray&pray.

markbarendt said:
Is your film sensitive to a broader or narrower set light wavelengths than the meter?

Since it's used to estimate observability of stars with human eyes (bare or through a telescope) I assume it's mostly sensitive to the wave lengths we see. I don't have this thing yet but if I get it I can try shining some IR diodes at it and see whether it gets fooled by them. I asked my question not in an attempt to advertise some gadgetery but to find out whether other folks have already tried this.

markbarendt said:
One other question here is; how you plan to adjust the development to correct for the narrow scene contrast?

I was not aware that scene contrast would be narrow, in my experience I had to battle excessive contrast as soon as artificial lighting exists in small parts of the view. Moon light should yield similar contrast levels to sun light.

markbarendt · Nov 27, 2009

Rudeofus said:
Once I got the exposure nailed I can deal with color casts (in the worst case in a hybrid work flow

This is the rub, you are dealing with three exposures not one; one for the Red layer, one for Green, one for Blue. Any layer that isn't exposed well will also have little or no detail. This isn't just a color cast issue it's a detail issue. If one layer is overexposed, one is normal, and one is underexposed, you only really have one layer that has decent detail and the color is wacky. Hybrid or Analog makes no difference here.

Rudeofus said:
, my scanner does much much better with slides than negs). Right now I look at pics with bright white mountains in front of white sky and dim star trails on light grey sky.

I think what you are seeing here is the lack of contrast that existed in the original scene. The SBR is very low and expansion development (pushing) would probably address some of that issue.

Rudeofus said:
I was not aware that scene contrast would be narrow, in my experience I had to battle excessive contrast as soon as artificial lighting exists in small parts of the view. Moon light should yield similar contrast levels to sun light.

I'm sure you know this but I'm going to state the obvious here, the moon is moving.

If you are after detail in the moon your exposure will need to be quite short and yes close to daylight settings. In that situation the rest of the scene's exposure just falls wherever it falls, the moon is the subject that matters. The only other choices in a single exposure are a- using a tracking mount but then the land/cityscape would blur if included or b- picking a day and time where the landscape and the moon are at similar EV's.

When the moon is included but detail isn't important, then the moon, like the stars, can be treated as point sources and a trail can be expected.

Similar thought on artificial light sources, just no trail. They can be treated as point sources too.

If you take the point sources out of the exposure calculation for a given shot, a night scene's subject brightness range can be very, very narrow.

I was at a workshop a week ago and we were all shooting a landscape scene late afternoon and over cast, the total SBR was about 3 stops when we arrived even shooting toward the horizon and measured by a variety of spot meters. That dropped to as little as 1 stop SBR before we left and it wasn't even dark yet.

Excluding point sources a night landscape may have an SBR well under 0.5. This is where I think your white mountain gray sky issue is rooted.

Kirk Keyes · Nov 30, 2009

markbarendt said:
I was at a workshop a week ago and we were all shooting a landscape scene late afternoon and over cast, the total SBR was about 3 stops when we arrived even shooting toward the horizon and measured by a variety of spot meters. That dropped to as little as 1 stop SBR before we left and it wasn't even dark yet.

Excluding point sources a night landscape may have an SBR well under 0.5. This is where I think your white mountain gray sky issue is rooted.

Could you go into more detail as to how you made these measurements?

markbarendt · Nov 30, 2009

Kirk Keyes said:
Could you go into more detail as to how you made these measurements?

Kirk,

I actually use and N90s with a longish lens that I spot with. Darkest shadow with detail, brightest highlight with detail.

About half of the participants had some form of hand held 1-3 degree spot meter and there was a solid consensus.

It was a really tough afternoon for shooting.

benjiboy · Dec 2, 2009

The answer to the O.Ps original question is, nobody.

mr. mohaupt · Dec 19, 2009

I understand what you are trying to do because I am about to venture into the astrophotography realm as well. As for shooting slides I can give you no guidance other then what I have read. Shoot Provia only because it doesn't block out as much of the red wavelengths as other films. Also What are you going for? Star trails or whole swaths of the sky? I think this will make a difference. That device more or less tells you the "quality" of the sky light.

Meaning if there is less "Sky glow" you have a higher quality of sky light. Ultimately your best bet is to try to find a very very very dark area only if you are trying to shoot photos of large expanses of the sky. The whole idea for this is for the people with there dobsonian and newtonian telescopes. How well will you be able to pick out details from deep sky DARK objects.

I have some experience with shooting photos of the night sky but I must admit they are all with my DSLR. I have produced great photos of the moon, of constellations and of subjects with constellations in the background. The key is knowing what you are trying to take a photo of because the advice for exposure is going to be different for the different types of shots.

Ultimately you are going to have stack photo's together (regardless of photo exposure length) if you want to get all of the possible details and colors from film. http://www.skyandtelescope.com/howto/astrophotography

That link might help.
~mike

Who uses a sky quality meter for estimating night time exposures?

Privacy & Transparency

Privacy & Transparency