Major Major Aptus 75 Problem

[pixjohn]

I did not want to post this problem under the JTFOTO major problem post because of the negativity that has been created. This topic is about solving an issue not bashing. If you have read other post by me on LL you know I have had a handful of issues. I am now at a desperate point to solve a new  issue or cut and run.

I am on my 3rd or 4th Aptus 75 back? (I lost count) The new back I have has created a MAJOR MAJOR  problem with highlights turning magenta in the center of my images. I talked to Leaf America but they had no answer at this time. Anyone else ever come across this problem?

I can definitely understand the frustration of other users with problems with Leaf and the software. I also know at least 2 Leaf users that returned their Aptus 75 for other brands. I  have been trying to be very  patient, but I am also getting to the end of my rope. I wish I would have stayed with my first back and just dealt with the centerfold issue with the gain adjuster. I will also say Leaf America has tried hard to help, but the results have  not been fruitful.  

I have been thinking of going with a Phaseone 45+ now that it has live video.  The  advantage with the Leaf back for me is the custom gain utility in V8. The gain utility acts like a software center filter and also correct the famous leaf centerfold issue at the same time. The newest back I am shooting does not seem to have as clean a file as the first backs. I made my purchase based on a Leaf Aptus 75 demo unit supplied by Leaf. I have never been able to duplicate the same high quality as that back.

Don't get me wrong when everything works I get great files.

With a weeks shooting coming up I am very nervous to shoot with this back.

Work Flow: tethered

1) Use live video and gain adjuster in V8 then disconnect
2) Reconnect in V10 - shoot in V10

Signed HELP ME PLEASE!

or

pixjohn (real web address in profile)

[yaya]

John I'm assuming that these images were shot with the custom gain file?

If yes, was the gain file exposed at +2 or less/ more?

I would try to over expose the gain file, maybe by another 0.5 a stop. It looks as if the Magenta bits are being "over corrected" by the gain process and then one of the channels goes too high up.

If the Gain file is a bit brighter, then the corrected file will be a bit darker.

Please let us know if this helped or not

Yair

[pixjohn]

Hi Yair, I have no way of knowing with interior shots and strobes  if I am shooting the gain file at  +2. My work flow to shoot the gain file is to open up 2 stops and shoot a mini strobe directly at the white disc on the lens. Sometime the gain utility tells me  i am to bright or 2 dark and I adjust according to what is needed. I have used the same work flow on the other backs and never had a problem.

[yaya]

Hi Yair, I have no way of knowing with interior shots and strobes  if I am shooting the gain file at  +2. My work flow to shoot the gain file is to open up 2 stops and shoot a mini strobe directly at the white disc on the lens. Sometime the gain utility tells me  i am to bright or 2 dark and I adjust according to what is needed. I have used the same work flow on the other backs and never had a problem.
[a href=\"index.php?act=findpost&pid=113380\"][{POST_SNAPBACK}][/a]

Understood, we'll require further follow-up (I have a few ideas that I need to run by R&D) and I'll make sure you're being taken care of.
Yair

[Eric Zepeda]

John, just trying to help...

Could the coverage of the mini-strobe used to create the gain adjuster file be causing the magenta over correction in the center? Is it possible to open up enough using the existing strobe lighting to create a gain file and then compare the results? You've obviously done extensive testing, so you've probably covered this base. It's just that I have noticed a fairly major difference when using the Cambo WDS and Phase Lcc's. I always try to use the same lighting to create the Lcc file as in the shot. I know it's a different platform, but the situations are somewhat similar.

Best,

Eric Zepeda

[pixjohn]

Hi Eric, Its an interesting theory but I am not sure its the problem. I never use the same mini strobe  to create the gain file. I have used the mini strobes and tungsten bulbs in the past on another Aptus 75 without any issues. Unfortunately I don't  have enough light to use unless I point a strobe head at the calibration disc. I could try a longer ambient exposure of 2-3 seconds but the time to shoot 8 shots and wait for the back to create the file would double my time.

[mtomalty]

John

Does the pink highlighting in the central areas only occur with artificial
light sources or can you make in happen with natural light sources.

It's difficult to detect the nuances on a small web image but it seems to me that,especially
in your second example,that the lampshade appears overexposed and is obviously pink
while the  overexposed window areas immediately surrounding the lampshade seems,as
far as I can determine,very neutral

Mark

[pixjohn]

The sky and clouds  also have magenta

[Cfranson]

The sky and clouds  also have magenta

It might just be the web image, but the magenta areas I see are very definitely overexposed to the point that there is no detail in them.
It's not unreasonable to think that any digital camera will give unexpected results in those instances.
Also, that magenta tint is typical of what happens when the image is overexposed in one or more, but not all, color channels.

[pss]

It might just be the web image, but the magenta areas I see are very definitely overexposed to the point that there is no detail in them.
It's not unreasonable to think that any digital camera will give unexpected results in those instances.
Also, that magenta tint is typical of what happens when the image is overexposed in one or more, but not all, color channels.
[a href=\"index.php?act=findpost&pid=113444\"][{POST_SNAPBACK}][/a]

i often blow out highlights completely (especially windows or shooting into lightsources) with my P30 and i never get the magenta you are experiencing....haven't seen that since my kodak 14n....purple fringes (CA), yes, but magenta highlights, no.....

[pixjohn]

Yes the overexposed parts could have something to do with it but why do I not see the same thing in the clouds out the other windows in the image? I never had this problem before this shoot?

It might just be the web image, but the magenta areas I see are very definitely overexposed to the point that there is no detail in them.
It's not unreasonable to think that any digital camera will give unexpected results in those instances.
Also, that magenta tint is typical of what happens when the image is overexposed in one or more, but not all, color channels.
[a href=\"index.php?act=findpost&pid=113444\"][{POST_SNAPBACK}][/a]

[ericstaud]

John,

I can only imagine that your gain exposures are over-exposed.  It looks like something that could happen by clipping one of the channels in the gain process.  It also looks like it is happening more in the center of your images.  This would make sense with the light fall-off from that wide lens.  The exposure could be clipped in the center area but not towards the outside.

Maybe you could bracket some exposures with the small flash and get the exposure nailed down before entering the custom gain dialog in LC8?

[eronald]

In this particular case I believe you may be looking at a software issue, NOT a hardware problems - have you tried running these files through several converters ?

The advice above (underexpose, then lift exposure in software and check ) seems sound. If this gets rid of the magenta cast on teh highlights you could then try assembling multiple exposures with the HDR feature in PS.

Edmund

[brumbaer]

Hi John,

the following results from my experience, still your problem might be caused by something different.

"Pink blowouts" like yours happen, when you blow the hilights in the green channel only and than "darken" the picture, while still working on the raw data.

E.e.  if you use Camera Raw and reduce the exposure, you might get the pink cast. If you don't change the exposure in Camera Raw, but continue and open the image in PS and reduce the exposure than, you will not get the pink cast.

This has to do with internal processing and can be explained with enough time and space.

Blowing all channels will not help, because you usually will get pink fringes at the edges of the blowout.

I'm not familiar with Leaf products, so I do not know where gain adjuster fits into the processing path, it might have an effect on the cast, it might even create it or it might have nothing to do with it. Did you try processing/creating images without using gain adjuster ? I think it's worth it, and if it has an effect, it will be great help to the Leaf developers to know this.

If this doesn't help, all you can do is to wait for Leaf to come up with a solution or expose your images differently (I understand that this may not give you what you want in other areas).

You could also increase the brightness of the image (using curve contrast brightness, whatever) until the pink turns to white, save the file and process in PS. Again I understand that this may not give the final results that you want.

Why it happens with this back and not with others I can't say, probably different sensitivity in the green channel.
But it's also possible that  you expose differently than before, or you use a different profile or something else changed. I'm not trying to move the pressure from Leaf to you. No matter what, Leaf and any other manufacturer has to make sure this doesn't happen with their products, I just want you to consider whether you changed something in procedure, which might have an influence.

Regards
SH

[vgogolak]

the red sensors are the least sensitive (also least energetic) light rays. Most SW compensates for the fact within normal exposures, but if the sensor is WAY oversaturated, maybe the compensation is off and leaves a bit of excess red rather than defaulting 255,255,255 to white

could be just a profile issue

[brumbaer]

the red sensors are the least sensitive (also least energetic) light rays. Most SW compensates for the fact within normal exposures, but if the sensor is WAY oversaturated, maybe the compensation is off and leaves a bit of excess red rather than defaulting 255,255,255 to white

could be just a profile issue
[a href=\"index.php?act=findpost&pid=113720\"][{POST_SNAPBACK}][/a]

What happens is that you should have linear values like (using 255 because most people think in that scale) 250, 290, 250. But because of the 255 limit you get 250, 255, 250.

After processing you get something like 290, 255, 300. Which reads as 255, 255, 255 a nice and friendly white.

Now you darken the image, which will work on the linear data tranforming the values to 125, 128, 125 after processing it will be like 145, 128, 150, which is a Magenta.

Of course the factors are not correct and depend on the white point and other processing settings, but thats what's most likely happening.

Changeing the profile will only help in special cases and when yo remove the Magenta by using the profile, you will most likely get a green cast in other areas, because there are other values of 250, 255, 250, which are real and not the result of clipping.

Depending on the way gain adjuster works an where it fits into the processing path, it may have an influence. I.e. some values get multiplied by a value greater than 1 and will overflow because of that. Or parts of the image get darkend and so the magenta cast gets visible (it would be there anyway, but you wouldn't see it until you darken the image, which you probably wouldn't do, because the exposure is ok or too dark.

Or something else or nothing, without knowing anything about gain adjuster and it's internas I'm just speculating.

Regards
SH

[brumbaer]

By the way the same is true for the Dalsa sensor when all channels are overblown for most color temperatures, because the multiplier for R and B are greater than the multiplier for G for most color temperatures.

The effect is the same as described in my last post.
Regards SH

[nicolaasdb]

I think you got to calibrate your back before shooting...the magenta in the blow outs can easily be jsut white when you system is calibrated. The rest of the image looks perfect. Or prevent the blowouts......shoot from a tripot and shoot for highlight and shadows.

[pixjohn]

I am still trying to figure out why I have been  getting a different image file with the same work flow. I have been shooting the same way with the aptus back for almost a year?

This week I am on another shoot and started the day with another Major Major problem. Half the image was green tint the other had magenta tint. This is after I did the gain adjuster to solve the centerfold and color cast. While I was talking to Leaf in Los Angeles and NY I thought about the only thing different with my work flow  was trying the 10.1 beta. When I asked if this could be the problem I was currently having I reset the preference in the Leaf software and the tints went away.

I have to say I was very impressed with the Leaf rep when he offered to drive 2 hours outside LA and bring me a new back at a min notice. (Thanks Bill for the offer) After deleaing with the magenta issue with the highlights and then the gren/magenta tint I was at a frustrated point.

I still have the magenta cast in the highlight and need to solve this problem soon. You definitely see the cast worse in the thumbnails then the rendered image but its  real.  I am working with leaf to solve this issue and hope to get back to my normal run around workflow using V8 and V10. My dream is to have one working v11 with live video and the gain adjuster built in that only takes 1 shot to create.  I can dream!

[narikin]

I am still trying to figure out why I have been  getting a different image file with the same work flow. I have been shooting the same way with the aptus back for almost a year?

This week I am on another shoot and started the day with another Major Major problem. Half the image was green tint the other had magenta tint. This is after I did the gain adjuster to solve the centerfold and color cast. While I was talking to Leaf in Los Angeles and NY I thought about the only thing different with my work flow  was trying the 10.1 beta. When I asked if this could be the problem I was currently having I reset the preference in the Leaf software and the tints went away.

I have to say I was very impressed with the Leaf rep when he offered to drive 2 hours outside LA and bring me a new back at a min notice. (Thanks Bill for the offer) After deleaing with the magenta issue with the highlights and then the gren/magenta tint I was at a frustrated point.

I still have the magenta cast in the highlight and need to solve this problem soon. You definitely see the cast worse in the thumbnails then the rendered image but its  real.  I am working with leaf to solve this issue and hope to get back to my normal run around workflow using V8 and V10. My dream is to have one working v11 with live video and the gain adjuster built in that only takes 1 shot to create.  I can dream!
[a href=\"index.php?act=findpost&pid=113944\"][{POST_SNAPBACK}][/a]

Jeez - you put up with all this?!!
its terrible that you are having these problems. no matter that the rep is coming out to drive 2 hours to you - these things should NOT be happening. Leaf should not be releasing backs or software with such disastrous issues.
I'm amazed that you haven't gone over to Phase, with their rock solid software + backs, as this is utterly unacceptable in any serious professional workflow.