Gossen Pilot Service

Discussion in 'Accessories' started by henry_finley|1, Nov 8, 2021.

  1. The thing to do while metering is first decide what part of the subject you want to expose for, and then meter that bit or meter the light illuminating that bit. Nothing else.
    When using relected light metering, you have to adjust for the subject´s reflective properties. Using incident metering you do not. And when using reflected light metering, you must take into account what the meter sees besides that part of your subject you decided yo meter.
    Differences between reflected light readings may be and often will be due to that: what does the meter see exactly. Not to things like meter accuracy or calibration.
  2. The effective aperture changes with focussing distance, so the EXIF datum is probably more precise than the manual setting to the nearest 'preferred' stop.
    But if you've found that the Sekonic tends to under-exposure, then why are you using it as your comparison standard?

    FWIW, my Sekonic L398 Studio didn't match other meters, nor digital cameras either. It differed by about a stop on the optimistic side. It was bought 2nd hand, and I thought maybe a previous owner had tinkered with it. So I inserted a piece of 0.3D ND filter between the diffuser and selenium cell. It now matches other meters perfectly and gives correct exposures.

    Together with your Twinmate experience; maybe Sekonic are just bad at calibrating their meters?

    Whatever. Technically there's nothing to beat TTL metering, which takes lens and camera body flare into consideration. Pose-wise, a handheld meter looks a lot cooler. :cool:
  3. It's under-exposing compared to my Canon Digital Compact when both are pointed straight ahead parallel to the ground. I intend to use the Sekonic L208 for all the folders in the stable, and I want to be sure I'm using the L208 in the right way to get the slightly under-exposed shots which I prefer, rather than over-exposed shots, I'm tired of blown out highlights. So the way I'm going to use the L208 is to angle it towards the ground in varying degrees depending on the whether there is bright highlights in the composition. If there's any of those, I'll lift the meter up a bit which will give me a faster speed to cope with the highlights better. Post processing will sort out the rest, but detail will be retained in the highlights. I've learned all this from you guys, you've been a big help.

    I'll be doing the comparisons over again now that my tests have shown that the L208 needs to be angled down, whereas inbuilt camera meters seem to be, and probably are, calibrated for straight ahead ... that's the way one holds the camera when taking a shot for film, and for digital, squinting at that damn LCD screen while composing the shot.

    I've started testing the L208 for film, ten shots were remaining in the Retina 111c so I shot those off and will develop the film (FP4) in a night or two.

    That sounds good, I might do the same to my old L98 somehow. It's only reflective but I may be able to do something to fix the two stops over, that way.

    Probably not bad at calibrating the meters ... they're just bad at telling us how to use them according to the way they were calibrated and any inherent quirks they may have.
  4. Wide-angle reflected light metering was always, and still is, fairly unreliable. Because the sensor takes in light from all over the place. Whereas camera TTL metering only 'sees' the light coming from the subject.

    Why not use the incident ability of the Twinmate? You just point the meter back at the camera from the subject position. (For landscapes and distant subjects, just point the meter backwards from the camera.) Much more reliable than guessing where to point a reflective meter.

    If you're having trouble retaining highlights in B&W negatives, then you're possibly over-developing. It's a common error to give a bit more developing time 'for luck' and go for a good, thick negative. But it's a surefire way to lose highlight detail.
  5. I've developed and scanned the test film from the Retina 111c. I got mixed results using the new Twinmate L208 with the last 10 shots. Strangely the Retina's meter, as old as it is, did a better job with the preceding 26 shots. I'll post some pics tomorrow and explain how I used both meters. The film was FP4 Plus developed in ID11 1:1 at 11mins.
  6. The meter in the Retina seems ok for the age it is. Here's two shots that only required the tiniest amount of post processing just to increase the contrast a bit. These were taken before I bought the Sekonic L208.

    Meter in Retina.jpg

    Meter test in Retina 111c.jpg

    With this third shot, still on the same film, I metered my hand with the Sekonic L208, keeping it down away from the sky, and hoped the reading would be slow enough to give a good exposure of the friendly parrot that drops in occasionally for a feed of sunflower seeds. I was figuring that my hand in shadow would be much the same shade as the red body of the parrot. The L208 reading was still is too fast for my liking. Perhaps the red on the bird needed a filter to brighten it up ... I'd need advise on that.
    Sekonic L208 test.jpg

    Color shot of the parrot that's shows the bright red.
    Say "Hello"
    Red parrot.jpg
    m42dave likes this.
  7. Red and green are going to appear very similar on panchromatic film. A red filter will darken the green leaves and blue(?) sky, it won't brighten the red parrot. Filters remove light, they don't add it or brighten their own colour, although applying the filter factor gives that illusion.

    As for exposures being 'correct'; that can only be confirmed by sensitometry/densitometry of the processed film. Scanners and film latitude can cover up quite a few stops in error.
    Last edited: Dec 9, 2021
  8. Correct exposure cannot be determined by densitometry, because such procedures know nothing about what you want. Maybe you want to compress the dark parts. Maybe you want an even, straight curve distribution of tones. Maybe you want less ... etc.

    It's rather nonsensical to point out that filters decrease exposure (in whatever selective way) and that an increase of exposure requires an increase of exposure.
    What filters do is change the relative exposure, selectively. So yes, whether you apply a filter factor or not, a red filter will change the relative brightness, make red things appear brighter relative to blue things.
  9. Thanks for that information.

    I've brightened this image to what I reckon the correct exposure should have been. Here's a side-by-side comparison of the corrected one and the previous one.
    1675573_3ae9117d525d6d33a23cd4340f51c07d.jpg 1675573_3ae9117d525d6d33a23cd4340f51c07d-1.jpg

    The new Sekonic Twinmate may be going back. Some readings are just too far on the under-exposure side, and it can be a little erratic as well I found out yesterday, two readings of the same object, withing seconds of each other, were different, 800sec and 500sec @ f8 - I'll give it one last chance, I'll trial it for one more day and send it back if it doesn't come up to scratch.
  10. To test a handheld meter, you need to know at least that the way you handle it is not a confounder. And especially when using reflected light metering, a small variation in how you point the thing may have a great effect. So begin testing using incident light metering only. Compare that to incident metering using another, known and trusted meter. Or to reflected, ttl metering using a camera pointed at a good grey card.

    I have and use quite a few handheld meters, and they all work great. Though it is rather hard to find two that agree to within 1/3 of a stop. You will have to get to know a meter, learn how it behaves in your hands (!). Incident metering is, for several reasons, almost fool proof and to be preferred over reflected light metering.

    Remember too that the way you interpret the results and translate those to speed and aperture settings will have an effect too. Using built-in meters on auto settings, you have no say over, hence no influence on that. So an extra thing to keep in mind when testing and using handheld meters. My personal bias, possibly in common with many others, is to be too optimistic when transferring. We all, i suppose, rather use, say, 1/30 than 1/15, so when a reading falls in between...

    Built-in meters only offer ttl reflected light metering, which may install a general believe that that is the best way of doing things. And though it works well enough, if you know and understand what the meter is doing and what to point it at, incident light metering is less complicated and faster. But ttl metering offers the advantage over reflected light metering using a handheld thingy of seeing what you point the meter at.

    So if you want a handheld meter, and do own a camera with built-in mdtering, use the handheld meter for incident, the camera for ttl reflected light metering.
    Last edited: Dec 10, 2021
  11. Feel free to chime in with some useful information then Q. G. instead of just creating confusion and an argument.

    Do you even own a camera? Because you're quite critical for someone that's never posted a single picture on Photo.net.
  12. Sure... You don't like it when people point out that you post nonsense in you quest to impress your greatness upon us all.
    Why not address the matter of fact points made in replies, Rodeo Clown? Well, that's why.

    And to help you understand: pointing out that you post useless nonsense is indeed very useful information. A forum like this is not for posing purposes.
    Last edited: Dec 10, 2021
  13. I'll do some incident metering tomorrow.

    I was out testing again today with reflective readings and I've concluded that the Twinmate is one stop under-exposing and that it appears to be calibrated for highlights. Every bright object in the photos from the digital cameras (I used two digital cameras on manual mode this time) was well exposed, just how I like them, but unfortunately the low light areas were too dark. Photos with no bright objects, simply came out one stop under-exposed all over. Today was overcast with no shadows and the Twinmate's needle hardly moved at all when I swung it around from one spot to the next, so I guess that lighting condition was good for testing the meter, no wild fluctuations to confuse things.

    This particular Twinmate I have is calibrated wrong, I'm sure of it. My Canon 980 digital has a light meter scale that I can manually set to ± "0", and every time, up close to an object, the shutter speed @ f8 differed from the Twinmate, which was always one shutter speed faster @ f8. I want an explanation from Sekonic, that's how things are traveling at the moment.
  14. Your digital cameras will be using a weighted matrix metering, i.e. not 'straight' reflected light metering, but an appraissal and adjustment of what the meter does, based on many readings taken in many parts of the field of view.
    Using a simple reflected light meter, you get nothing of that. It's all up to you, to take note of what you are metering, ie what is visible to the meter. How that determines what the reading will be. And what you have to do to turn that into the camera setting that will produce the desired result.
    It is not easy. You have to keep your wits... etc. So don't blame the (simple) meter too quickly.
  15. Yes, good point. One of my digitals' was set to "Pattern", the other set to "Center-weighted". I'll try the other modes and see what difference they make.

    I haven't had this trouble in the past with hand-held meters, generally they've performed well without having to do anything drastic to correct the exposures in post processing, the negatives have always looked good, bar a few films that were expired, or the chemicals were a bit off. I usually know what the problems are as they arise.
  16. The scanner was set the same for all frames on the film regardless of the metering, the Retina's meter, and the Twinmate's. The scanner didn't correct the exposures such that they all looked the same, some were darker and some were lighter. The Twinmate's exposures were all darker, and from what I know now about that meter, they were all one stop darker.
  17. I did a simple test using the Sekonic L208 to show the difference between three ways of metering. From left to right: 1) Reflective reading pointing the meter straight at the tree. 2) Reflective reading pointing the meter at the very base of the tree. 3) Incident reading by turning around and pointing the meter at the sky, angling the meter up just enough to clear a building in shadow (This incident reading matched the digital camera's meter reading, I'm happy about that, I might be saved, but what about the reflective readings, what am I supposed to do with this meter, use it only for incident readings?)

    1 Test 1.jpg 2 Test 2.jpg 3 Test 3.jpg
  18. AJG


    It looks as though you have a useful incident meter, which is not a bad thing. All of the selenium cell meters that I have used have similar issues with a very wide acceptance angle which makes precise metering difficult. For this image with this meter I would have walked up to the tree and taken a reading then stepped back to take the picture. Another option would be a spot meter, but this will be many times larger and heavier than your Sekonic L208.
  19. I did actually walk up closer to the tree for the two reflective shots, 1 and 2, for the reason you mentioned, the wide angle of acceptance. So what you see there in those two pics is how the tree itself metered. The under-exposure is very consistent with reflective metering using this L208. If I was to continue using it, I'd just have to adjust the camera to two stops slower speed every time. There's two stops difference between pic 1 and pic 3, they went 400sec, 200sec, 100sec, @f8 100ASA.

    The incident metering was pretty good each time. I took quite a few shots and they were matching the Canon digital's readings to 1/3 of a stop. Pic no3 is how my eyes saw the tree, highlighted leaves on the top half but still light green in color. However, I need to learn more about incident metering because I was not sure where to point the meter. So far I've been pointing it just up at the sky behind me. It seemed to work, but what if there was a wall or a building right behind me? I'll need to find out.
  20. That's the nonsense right there.
    Film speed - and hence exposure - is determined purely by densitometry. As seen in this graphic. ISOspeed.png
    But maybe you can't understand that?

    And did you not notice that I put 'correct' exposure in parentheses?
    The true correct exposure is, of course, that which gives the desired result, but calibrating or checking a light meter requires working to a standard exposure and development. Such as is set out in the ISO 1993 standard, and which can be read on Wikipedia.

    Kmac. You can check the meter reasonably easily without a densitometer or complicated procedure. You just need to find an evenly-lit white or grey surface.

    1. Load your film camera with the B&W film of your choice, fill the frame with the blank surface and set the exposure to the reading that you get from metering that surface.
    2. Give the film your normal processing.

    This should result in a frame with a standard exposure equivalent to a grey card exposed according to an incident reading.

    3. Compare the resultant negative density with this chart.
    You can just hold the frame up against the white areas of the (sRGB calibrated) screen and compare the density of the film against the squares.

    Your mid-density (Zone V if you must!) should be 0.6D above base+fog - according to Ansel Adams. For myself, (together with most film manufacturers) I'd prefer to see a mid density of 0.7D to within +/- 0.1D.

    The above density chart should allow you to gauge the base density using steps 0.1 to 0.4D. (If it's much more than 0.4D then either your film is stale or fogged, or you're overdeveloping the film.)

    Your mid-density + base + fog should fall somewhere between the two limits of 0.9D and 1.1D. If not, then something's amiss with the meter, your developing or the camera shutter. But just going by the look of scans or guesswork won't get to the root of any problem.
    Last edited: Dec 11, 2021

Share This Page