Magenta "flare" with APO Digitar 47 and Aptus II 12

[KevinGSaunders]

I have been shooting with my Aptus II 12R, Rollei X-Act2 body, Schneider APO Digitar 47 with Schneider center filter, and discovered a magenta cast on one the left half of the image. I did some more testing today and was able to duplicate the effect three times although I uploaded two sets. I have not gone out without the center filter but that is the next project.

Has anyone experienced this before?

Thank you very much.

Kevin

[gazwas]

Sync problem (cable?) or incompatible CF card.

Probably card as the Aptus 12 is renowned to be very fussy about CF cards.

[EricWHiss]

I was going to ask the same thing - did you shoot tethered or to a card?  I've got a similar setup - AFi-ii 12 and sometimes shoot on a Rollei Xact2.  I was guessing CF card too seeing the images. But I'd check all the usual stuff - power to the S-Controller if you are using electronic shutter and also the sync cables.

[KevinGSaunders]

Thanks all. I am running a 32g Sandisk Extreme Pro CF. I did clean the synch cables with alcohol and was unable to duplicate the problem later in the day. I will return to the scene of the crime this morning at the same time and see what happens.

Best,

Kevin

[KevinGSaunders]

I should ask, what cards seem to work best with the Aptus II 12R?

Best,

K

[KevinGSaunders]

I went back to the scene of the crime and was able to get the half frame magenta flare, and this one was without the cf. So it appears that I am getting flare only with the sun high in the sky and at a particular angle. The weird thing is, the half sensor flare. I am hoping someone out there was able to duplicate this.

[KevinGSaunders]

I have to tip my hat to Yair, who kindly worked with me to send raw files to Leaf. He says that the Aptus II 12 needs calibrating and needs likely to go to Israel. What is calibration and how can a back go out of calibration?

[EricWHiss]

I was worried that would be the issue.  This can be a two week to 6 week affair.  Mine has gone in for that service.   I guess the data is read off the chip by different circuits - one for each corner but because they are different, they need to be calibrated to match.  I don't understand what could cause them to get out of calibration, but hope yours is returned quickly.

[torger]

As far as I understand most digital backs have two or more separate analog readout channels from the sensor. A 100% perfect calibration does not seem to exist, if you shoot an LCC frame and increase contrast to the maximum on that you will probably see some split (centerfold) in the frame even on a recently calibrated back. I have the same on my ancient Aptus 75, but the calibration is good enough to not cause any problems in practical photography at normal contrast levels.

How they can go out of calibration I don't know. As far as I know when you send in the back for standard service (€500 or so) one of the things they do is to recalibrate the back.

[KevinGSaunders]

Thank you folks for your kind replies. I have a call into the MAC group and am very thankful I have my Aptus 75s as a backup.

Best,

Kevin

[ondebanks]

As far as I understand most digital backs have two or more separate analog readout channels from the sensor.

"Most" is correct. In this post, I will point out specifically which do and which don't.

If you want an MFDB which does not (cannot) have this split-readout calibration problem, look for ones with Kodak 16.7MP or 22MP sensors...like the P20/+ and P25/+. All pixels are readout through the same, single corner amplifier.

The other Kodaks all have dual (18MP, 31MP, 39MP) or quad (40MP, 50MP) outputs, and as such are in principle susceptible.

That said, the ones I've most commonly seen such problems posted about are the DBs with Dalsa sensors...like Kevin's Aptus in this thread. Every Dalsa DB sensor (right back to the 35mm-format 6MP and 11MP ones) has dual registers and quad outputs...which is why the Dalsa DB frame-rates have often pipped their Kodak counterparts...although the DB makers did not necessarily use all 4 outputs; using only 1 or 2 is possible. This trades off readout speed for calibration consistency.

Ray

[LenR]

This happens when you use swings or tilts.
Use the white disc that came with your back to apply a lens cast correction.
You do it after you have set up your shot with camera movements.
You can then shoot through the calibration file you just created or you
apply the correction to the shots afterwards and the cast will go away.

[KevinGSaunders]

Thanks Ray, for an excellent explanation. I wondered why the problem was on only half the image. FYI I do always do a shot with the white plate to apply an LCC, which I did prior to processing the images shown in this thread. The Aptus II 12 has some significant color casting with the APO Digitar 47, and as such I am now looking at Rodenstock lenses for the wider view.

The back is now on its way to Israel and it will return when it does! I am glad to have the Aptus 75s as I almost sold it, not thinking that something like this would happen.

I appreciate all your inputs.

Best,

Kevin

[EricWHiss]

Kevin,
When your back returns you can test that they got the problem resolved satisfactorily by shooting a white wall or or LCC plexi plate several stops under exposed at ISO 400 and then lifting the exposure in C1 or whatever program you use.  You don't need to mount any particular lenses or anything.  By amplifying the signal several stops of exposure (a couple in the ISO setting already) you'll amplify the differences in the two sides if there are any. The homogenous white will make it easy to spot any differences.   I'm pretty sure you can make all of these  80mp dalsa chiped backs show the line when pushed to the limits this way - ISO 800 lifted 3 or 4 stops and shadows etc.  so its a judgement call on what's reasonable and good enough.