"What you see is NOT what you get"

You need the histogram displayed with the picture : then exposure will be 'predictable' and you can adjust the exposure up or down according to what the camera is telling you 'will be' the result.
hmmm ... that sound random.
Anyway, it works nicely on my G1 will any and all lenses.
With the proper m4/3rds lenses the cemeras are more naturally exposing correctly. With old lenses sometimes they do it wrong : the histogram can give you advance warning of that so you can adjust accordingly.

 

Have I understood you correctly, that yo set the exposure manually depending on what you see on LCD? This is a rather imprecise way of setting the exposure. It is better to let the metering of the camera do the job and learn how to use that (use, for example, spot metering in critical situations). With old manual lenses you can still set the aperture and the camera chooses the shutter speed automatically.

 

Do you calibrate your computer screen?

 

Do you have live view boost on?

 

Hmmm. Deep thought.
Sometimes my camera does give me a dark shot, when it exposes for maybe some backlighting instead of the subject ... but I've not had your problem exactly.
I have used my G1 on ...tokyo,helios.prektica,hanimex,fed,jupiter,yashinon,cosinon,hexanon,pentax110 ... well quite a few lenses. Maybe I really need an expensive lens to find your exact problem, sorry I can't be more help. I always have the little histogram on in the middle of the screen : it helps me recognise if the camera is trying to wrongly expose.

 

Try using centerweighted metering if you are shooting in very high contrast setting. The old lenses don't seen to work the same for me in these sitituations. Otherwise I get great results. I only have old glass at this time.

 

I can only reiterate much of what has been said before. I shoot almost exclusively with adapted lenses and to all intents and purposes suffer no bad exposures unless I deliberately take a picture in extreme light conditions.
Camera set to "multiple" metering mode, live histogram display "on". I usually dial in 1/3 of a stop under exposure to avoid blowing out the highlights. Use Aperture priority and set the desired f stop manually on the lens in use. The live histogram will alert you to any under or over exposure conditions and you can dial in correction accordingly.

Barrie

 

Definitely check your metering mode. You may be doing area metering, and in a scene with a bright background that will underexpose the area of interest in the photo. Try either spot metering, or try using the AE lock button once the metering looks OK in your LCD. Often times you can focus off to the side then move the composition off-center, and if you don't have your exposure locked it'll change.

 

With all due respect I have to take issue with the idea of using spot metering. This is a specialised metering method. Any meter is designed to render the scene it's metering as mid grey, thus if you meter a white card and expose for that reading the white card will be rendered mid grey. If you meter a black card and expose for that reading the black card will be rendered mid grey. Those 2 meter readings could be 6 or more stops apart. This is the principle of the zone system, meter the whitest area and place it on a high zone, meter the darkest area and place it on a low zone. Spot meters will often store the lightest area reading and the darkest area reading and produce an average reading which is what the camera should be set to. This is obviously a brief simplification, but using spot metering can lead to very misleading results, it needs to be used intelligently.
I refer you to the writings of Ansel Adams where, in "Camera and Lens" he states that when using a spot meter, unless the photographer is versed on the Zone System and visualisation his results can be bizarre indeed.

Barrie

 

On Olympus bodies, you can use the pen Hi & Lo spot metering to check out the range of the scene and then spot for something mid-grey... or not ! Spot metering is perfectly OK as long as you know what you're doing. I almost always spot meter, it's the best solution for rendering your intent in black & white. It's much more difficult to achieve good results in colour.

Cheers,

Mauve

 

Mauve, precisely what I said

Barrie

 

Sorry, I certainly misunderstood you. Blame it on my bad english. I thought you were against using spot metering as a general metering mode. I just wanted to say that in b&w photography, it's absolutely possible and useful to spot meter most of the time. It helps drive the eye of the (future) onlooker where you want in the picture.

Cheers,

Mauve

 

Spot meters are no different.. certainly not a specialized method of metering. Its simply the area which is metered (smaller circle) that is the difference. Measuring bright and dark areas to determine exposure is a technique. Spot, center, and average are different tools. You can apply a tool to whatever technique works best for you.

* Spot to meter middle grey
* Spot to measure bright and dark, then average.
* Spot the bright point and add certain exposure compensation...
etc..

All can be done with average and center average as well.

Certainly, Spot and Center average require a bit more thought than multi-segment (or whatever Panasonic calls it). I am assuming just4fun knows how to use a spot meter as he/she arrived at an acceptable exposure using their handheld Sekonic.

I believe this is the key here....

1) Using a native lens, set the GF1 to a center average. Meter a wall or Something that is uniform that fills the frame. Note the exposure reading.
2) Using an adapted lens, set the GF1 to center average. Meter the same wall filling the frame. Note the exposure.
3) Using a native lens, set the GF1 to Multi-pattern. Do the same.
4) Using an adapted lens, set the GF1 to Multi-pattern. Do the same.

1A) through 4A) Do the same with a high contrast area.... Its important to use the same scene throughout the experiment.

Comparing the exposures should give you a clue to as what's going on (I don't have a GF1 to perform the test myself).

My guess...

1 & 2 should agree.
1 & 3 should be close.
4 will be off a little against a wall and much more against a high contrast area.

The multi-pattern mode (evaluative, whatever they call it) is attempting to be more intelligent by adjusting automatically according to the different segments being metered. This feature is not working properly with none native lenses which "tell" the camera a variety of information including what focal length is set.


On my E-PL1, I have two "reset settings". One is tailored for native lenses and the other is tailored for adapted lenses. The metering mode for native lenses is ESP (the equiv of multi-pattern) and the metering mode for adapted lenses is center average. I often set to spot for high contrast scenes. Center average best matches my other system thus feels more "familiar" with manual lenses.. I've never had a problem.


Note....
1) Ignore the "brightness" of the Life view. It doesn't reflect exposure accurately only composition.

2) Digital cameras "ISO settings" are a toss back from film days for easier understanding but are not a true reflection of the ISO. Some camera's ISO settings will be slightly under or over in terms of sensitivity and will be different for each camera. DXOmark reports this discrepancy in their test reports for this reason. As such, I have a sticker under my cameras that tell me how "off" the ISO settings are. I have to take this into consideration when transferring exposure readings from my handheld meter to the camera.

 

Now that's interesting... hmmm? Not sure what's causing the difference between the two Leica glass.

I just swapped lenses between my Olympus kit 14-42 and a 35mm Summarit against the wall nearby on my G1. I didn't notice any obvious differences in exposures in multi-segment/pattern.. If you don't find time for the experiment, simply try shooting only in either spot or center average when shooting with adapted lenses and see what happens. IIRC, you can set one of the custom presets to make it easier to remember.

 

...yet you agree that it's a good method when used properly. Which is it? It's a perfectly good method to use, which is why I suggested it.