M8 with Metz 54MZ-4i and SCA 3502M4 in TTL mode

[peterv]

Advertisement (gone after registration)

Nearly everyone is talking about underexposure and hoping that the IR filters will cure it.

Wilson

 

And the IR filters do cure it.

 

As for the exposure control, Riley said it earlier, we just have to use M or A mode...

[Riley]

My Olympus is TTL flash, if I need to shoot inside a room with a view, and keep that view well exposed. TTL wont do it, it will always select an inconvenient aperture on S mode, or speed on AE mode, P mode wont do it at all... always sets F2.8 or something silly like that.

 

What I need is a aperture that accommodates the DoF I need, at least 5.6 or 8

and a speed that corrects the exposure for the outdoor scene exposure, 125-250, 180 is sync. No automatic mode P, AE, S, comes close enough to allow a simple EV adjustment on the flash.

 

So I set the TTL flash to auto, set a complementary aperture, and shoot at a speed I need

I dont believe that situation is uncommon. Once I have an exposure that works, I can move freely thru a house and shoot all rooms.

 

When I thought about it, I realized I was better off with the power of the Metz 54 on either AUTO or manual, rather than the expensive TTL Olympus flash. So now I use that, in the method described.

[wparsonsgisnet]

Is a UVa filter the same as the ir filter you need for proper flash?

 

Ronnie, the filter being discussed is the infrared-blocking filter, aka IR-cut filter, B-W 486 filter, forthcoming Leica IR-cut filter, and others.

 

The sensor in the M8 is more sensitive to IR light because of the thinner-than-usual cover glass over the sensor. This glass needs to be thinner because the Leica lenses extend so far into the camera body, and the thinness of the cover glass means it blocks less IR light than is desirable.

 

To correct this, an external filter is required. This is the filter that is being referred to in this thread, when discussing flash.

 

These IR-cut filters also include a UV filter. The filter glass is multi-coated on one side for IR reduction and on the other for UV reduction.

 

If you were wondering if a UV filter would suffice, the answer is no; it wouldn't help the IR problem in the M8.

 

If you look over several pages in this forum, you'll see many threads discussing this problem, in which other have explained this very clearly.

[wlaidlaw]

I am sorry guys, maybe I am being dense here (and I am sure you WILL tell me if that is the case) but I don't see what IR sensitivity of the CCD sensor has to do with the underexposure problem on the 54 MZ-4. If I have understood it correctly,the sequence of events on a GNC exposure is as follows;

1) The shutter release is pressed

2) A signal is sent by the camera to the flash which fires a pre-flash.

3) Depending upon the reflectance of the subject, the distance of the subject from the camera and the lens focal length and aperture, a varying amount of the light from the pre-flash reaches the shutter through the lens.

4) This light is reflected from the closed shutter onto a CDS array which measures the intensity of the light

5) The camera already knew the ambient light level either from the blue dot or the usual TTL sensors and does a computation of the required flash level to achieve a satisfactory exposure from the combination of ambient and flash light.

6) The camera sends a signal to the flash telling it the required flash intensity and maybe duration to arrive at that ideal exposure.

7) The camera shutter opens and the flash is fired.

 

So where does the IR sensitivity come in?

 

Wilson

[peterv]

I am sorry guys, maybe I am being dense here (and I am sure you WILL tell me if that is the case) but I don't see what IR sensitivity of the CCD sensor has to do with the underexposure problem on the 54 MZ-4. If I have understood it correctly,the sequence of events on a GNC exposure is as follows;

1) The shutter release is pressed

2) A signal is sent by the camera to the flash which fires a pre-flash.

3) Depending upon the reflectance of the subject, the distance of the subject from the camera and the lens focal length and aperture, a varying amount of the light from the pre-flash reaches the shutter through the lens.

4) This light is reflected from the closed shutter onto a CDS array which measures the intensity of the light

5) The camera already knew the ambient light level either from the blue dot or the usual TTL sensors and does a computation of the required flash level to achieve a satisfactory exposure from the combination of ambient and flash light.

6) The camera sends a signal to the flash telling it the required flash intensity and maybe duration to arrive at that ideal exposure.

7) The camera shutter opens and the flash is fired.

 

So where does the IR sensitivity come in?

 

Wilson

 

Wilson,

 

please read this thread:

 

http://www.leica-camera-user.com/digital-forum/13580-metz-54mz4i-underexpose-m8.html

 

Hope this helps,

 

Peter

[wlaidlaw]

Wilson,

 

please read this thread:

 

http://www.leica-camera-user.com/digital-forum/13580-metz-54mz4i-underexpose-m8.html

 

Hope this helps,

 

Peter

 

Peter,

 

I had read all that but it still does not explain how the IR sensitivity of the sensor can possibly have anything to do with the exposure. At the time the exposure is calculated, the sensor is hidden behind the closed shutter. Now it is obvious from people's results that fitting an IR filter alleviates the problem but nevertheless it must be something to do with the way the camera calculates the exposure and nothing to do with the sensor. Perhaps the 54MZ-4 has a weird spectrum. You would need to fire it onto a colour positive through a spectroscope to determine that. Does the SF24-D underexpose without IR filters or does it overexpose with IR filters - if not, why not? I think there are factors at work here that we have not yet fully explained. If I have time tomorrow, I will try and shoot some identical shots on a tripod with the SF24-D and 54MZ-4i. Sadly I don't yet have my IR filters to get before and after results on both flashes.

 

Wilson

 

Wilson

[Riley]

Advertisement (gone after registration)

well I cant explain it but its clear that is the issue

what I do find interesting is the camera seems optimised for the fitment of filters

it was never a solution to a problem, and was therefore intended to be like that all along

[wlaidlaw]

Well my SF24-D is back and as far as I can judge, it underexposes in GNC mode, to pretty much the same degree as the 54MZ. At least, I suppose, it is consistent. Of course on the SF24-D, you can adjust the EV very easily by holding down the "p" button on the flash and then using the + or - keys, followed by a final push on the "p" button to confirm. About +1 1/3 stops seems pretty close. I am wondering if it could be a lens or lens coding issue, as I am using a non-coded Zeiss Biogon 35. Are the people who seem to be getting closer to ideal exposure without EV compensation, using coded lenses? Please tell me.

 

If I can find, at a reasonable price, a black 43mm IR filter, I might get it, as Leica can only supply me with a silver one and at some vague time in the future. The alternative is just to set up a batch processing profile, to pull all my flash photos up by one stop or maybe one and a half.

 

As I have said before, better underexposed than over exposed, especially with the latitude and depth of information on the Leica images. The amount of detail you can pull out of apparently pitch black shadow is amazing. I thought the 10.3 Mp CMOS sensor in my Sony R1 was good, until I got the Leica.

 

Wilson

[peterv]

Peter,

 

I had read all that but it still does not explain how the IR sensitivity of the sensor can possibly have anything to do with the exposure. At the time the exposure is calculated, the sensor is hidden behind the closed shutter. Now it is obvious from people's results that fitting an IR filter alleviates the problem but nevertheless it must be something to do with the way the camera calculates the exposure and nothing to do with the sensor. Perhaps the 54MZ-4 has a weird spectrum. You would need to fire it onto a colour positive through a spectroscope to determine that. Does the SF24-D underexpose without IR filters or does it overexpose with IR filters - if not, why not? I think there are factors at work here that we have not yet fully explained. If I have time tomorrow, I will try and shoot some identical shots on a tripod with the SF24-D and 54MZ-4i. Sadly I don't yet have my IR filters to get before and after results on both flashes.

 

Wilson

 

Wilson

 

Good point!

 

Perhaps the photo diodes that measure the pre flash are also sensitive to IR and/or UV just as CCD? Of course the measured IR, which is taken into calculation does not show up as brightness in the photo, hence the underexposed pictures...

 

It would be interesting to see the results of such a test. I read somewhere that the SF 20 underexposes on the M8, but the SF 24 D is reported to work fine.

 

Peter

[peterv]

well I cant explain it but its clear that is the issue

what I do find interesting is the camera seems optimised for the fitment of filters

it was never a solution to a problem, and was therefore intended to be like that all along

 

Could it be that this was 'fixed' in 1.09?

 

Cheers,

 

Peter

[Riley]

wonder why they dont regulate flash as OTF,

or OTS as it where in real time

[pascal_meheut]

Because the CCD reflects too much light. Read LFI.

[wlaidlaw]

Good point!

 

Perhaps the photo diodes that measure the pre flash are also sensitive to IR and/or UV just as CCD? Of course the measured IR, which is taken into calculation does not show up as brightness in the photo, hence the underexposed pictures...

 

It would be interesting to see the results of such a test. I read somewhere that the SF 20 underexposes on the M8, but the SF 24 D is reported to work fine.

 

Peter

 

If the subject has not bored everyone totally into an early grave yet, I will have a B+W 486 filter here tomorrow. I will then post on my Picasa site, a series of comparative tests with the two flashes on GNC and A mode, with and without filter. Any other tests we could run?

 

Wilson

[peterv]

Because the CCD reflects too much light. Read LFI.

 

That's true, but I've googled OTS (didn't know what is was ) and it seems possible. Although I wonder what advantage OTS would have above the LFI-explained way of measuring the flash. (apart from eliminating the pre-flash)

[pascal_meheut]

I'm not sure but I remember the Leica guy in the LFI interview saying OTS was not possible because the sensor can act as a mirror.

[Riley]

they can borrow my sunglasses

[wlaidlaw]

After playing around a lot with the M8 yesterday and both Metz 54 and Leica SF24-D flashes plus an IR 486 filter on the Biogon 35, I came to the conclusion that the 54 was really not going to do the job. Sure it is much better with the IR filter on but now if you have someone in a white dress in front of a dark background - it overexposes in GNC mode. It is also not very accurate in bounce mode if the ambient lighting is marginal for a flash being needed and the use of the secondary flash can still throw the exposure out quite a bit. Metz confirmed to me this morning that their engineers will NOT be working on EV control for this flash in GNC mode for future firmware updates. They now recommend using the small, simple 44MZ-2 or the huge professional 76MZ-5 units with the M8.

 

I have therefore reached a gentleman's agreement with Metz's UK agent, Intro 2020, and they are going to take the flash + SCA unit back and refund me, which I have to say, is as good a result as I could have hoped for in the circumstances. I will be a bit disappointed nevertheless, as apart from the exposure problem, the 54 was an ideal unit for my requirements.

 

I will buy a SFILL diffuser from Lutz or similar for my SF24-D as the best available alternative solution.

 

Wilson