Log In
Register
The Digital Bolex Forum
Topic RSS
12:26 am
June 22, 2012
Offline
After looking at some Ikonoskop samples in another thread, I tried to find out what dynamic range the DII has (11 stops). In doing so, I stumbled over a thread on another forum that mentioned and showed that dark portions on the Ikonoskop frames may turn magenta. Normally, I would not have noticed this, but this gave a little click in my brain, since it reminded me of the same problem tha Leica M8 had. As far as I know, didn't the M8 also use a Kodak sensor? In Leica's case, it turned out to be too weak a UV filter in front of the sensor. Leica ended up in sending front filters to every M8 owner who asked for one.
So, it might well be that the D16 could have the same issue. In this case, Pomfort should, imho, provide an automatic "de-magenta" function for the footage.
8:10 pm
Camera Owner
February 28, 2012
Offline
We will definitely look into this.
We are including a UV filter over the sensor.
If it is of appropriate strength, will this take care of the magenta problem do you think?
I have heard of this problem with Ikonoskop. I was under the assumption it was from the calibration stage.
Thanks! Joe
8:45 pm
March 27, 2012
Offline
I have always thought the color matrix calibration of the Ikonoskop was not the best. Though Dan Hudgins did comment that the Kodaks can be a bit of a bear to calibrate. Reasonable IR and UV cutoff is a good idea I think. Two questions:
Are you using the engineering grade model of the sensor with the non-reflective cover glass?
Will the camera firmware have a good black shading routine?
I am hoping you can achieve something much closer to Leica's color quality.
To my eye the Kodak sensors are capable of a dense saturated color quality that has a unique look. Similar in character to some of Kodak's films. The grading response in the few Ikonoskop samples seems to be very fast or sensitive. Small adjustments in photoshop can make quite significant differences in the image.
11:13 pm
June 22, 2012
Offline
joerubinstein said
We will definitely look into this.
We are including a UV filter over the sensor.
If it is of appropriate strength, will this take care of the magenta problem do you think?
Thanks! Joe
From what I heard back then, the problem was indeed attributed to the UV filter filter simply being too thin. I am not sure why the particular glass thickness was selected, but I am pretty sure that the Leica engineers had a reason. Which might have to do with flange and incident angle. And since the flange is also rather short between the back lens and the sensor, in c-mount, you may have to solve a similar problem. But this is just speculation, of course.
11:34 am
Camera Owner
Kickstarter Backer
March 17, 2012
Offline
Speaking of Ikonoskop Dii issues, I've shot with it a number of times and there was a ghosting issue. This was a result of over exposure in one quadrant in tandem with under exposure in an opposite quadrant. The result is some green noise ghosting of silhouetted objects in the shadows. The only post solution is to crush the blacks. Not ideas if you're trying to maintain a healthy level of contrast... I believe Ikonoskop found a workaround that requires affected cameras to be shipped back to Sweden. Might want to keep an eye out for this/contact those guys.
Thanks listening.
3:05 pm
March 27, 2012
Offline
Ari Davidson said
Speaking of Ikonoskop Dii issues, I've shot with it a number of times and there was a ghosting issue. This was a result of over exposure in one quadrant in tandem with under exposure in an opposite quadrant. The result is some green noise ghosting of silhouetted objects in the shadows. The only post solution is to crush the blacks. Not ideas if you're trying to maintain a healthy level of contrast... I believe Ikonoskop found a workaround that requires affected cameras to be shipped back to Sweden. Might want to keep an eye out for this/contact those guys.
Thanks listening.
Green ghosting is generally produced by internal reflections between the sensor and one or more lens elements. Hence my question about which version of the sensor The D16 is using. The engineering grade has a non reflective coated glass over the sensor vs clear glass.
10:42 pm
June 22, 2012
Offline
As far as I could take from Truesense's website, there is a variant of the KAI-04050 that has a cover glass with AR (presumably for anti reflective) coating on both sides (KAI-04050-CBA-JD-BA, which is in "Standard grade"). There are in fact also variants without AR coating, in Engineering Grade and Grade 2 (more defects allowed).
Incidently, from the "Product Naming Convention" PDF on their website:
Designation E:
Engineering Grade. Electrically functional and meet most, but not necessarily all, product performance specifications, however there are no limitations on the number of or size of cosmetic defects (points, clusters, columns, glass defects, etc.) allowed. Intended for evaluation purposes only and have NO warranty. Quantities are strictly limited and sold only “as available”.
And designation A:
Standard grade
which is apparently what we should ask for 
4:32 am
March 27, 2012
Offline
Thyl said
As far as I could take from Truesense's website, there is a variant of the KAI-04050 that has a cover glass with AR (presumably for anti reflective) coating on both sides (KAI-04050-CBA-JD-BA, which is in "Standard grade"). There are in fact also variants without AR coating, in Engineering Grade and Grade 2 (more defects allowed).
Incidently, from the "Product Naming Convention" PDF on their website:
Designation E:
Engineering Grade. Electrically functional and meet most, but not necessarily all, product performance specifications, however there are no limitations on the number of or size of cosmetic defects (points, clusters, columns, glass defects, etc.) allowed. Intended for evaluation purposes only and have NO warranty. Quantities are strictly limited and sold only “as available”.And designation A:
Standard grade
which is apparently what we should ask for
Thanks for correcting that. The AR cover is important. Ghosting is not unique to this sensor. Red has had issues with it with some lenses, especially vintage glass. It can happen with just about any digital camera under bright high contrast scene conditions.
10:37 am
Camera Owner
February 28, 2012
Offline
Razz16mm said
Thyl said
And designation A:
Standard grade
which is apparently what we should ask forThanks for correcting that. The AR cover is important. Ghosting is not unique to this sensor. Red has had issues with it with some lenses, especially vintage glass. It can happen with just about any digital camera under bright high contrast scene conditions.
Hey Guys,
I just asked about this and got the same answers you did. Engineering grade is not as good; Standard grade with the AR (Anti-reflective) coating is what we have ordered!
Apparently the AR coatings are a relatively new feature, so maybe cameras that had ghosting problems were made before the AR coatings were available? Just a guess.
BTW I'm here in Canada right now, going over the design in minute detail 
Thanks everyone!
9:02 am
July 10, 2012
Offline
Look at this post on ikonoskop
http://www.ikonoskop.com/forum.....entID=2137
the blown out street lamps are pretty bad - looks very CCTV-ish!
What can be done to avoid that?
2:06 pm
March 27, 2012
Offline
RobertGL said
Look at this post on ikonoskop
http://www.ikonoskop.com/forum.....entID=2137the blown out street lamps are pretty bad - looks very CCTV-ish!
What can be done to avoid that?
All sensors have their limits. In extreme contrast situations like this you have to accept some blown highlights or expose for the higlhights and accept more noise in the shadows. Or buy an Alexa or Red. Most of it looks pretty clean to me considering the conditions. The fundraising trailer looked overexposed in several places. Be interesting to see how the D16 handles similar circumstances. My GL2 would not even produce a visible image of that Turkish street scene in those conditions. Just pure noise in the shadows.
TheTurkish one was shot with a vintage Cooke Speed Panchro, I have seen examples of similar color fringing around specular highlights with that lens on other digital cameras too, so it could at least partially be the lens. Haloing from lens flare or even atmospheric conditions is not that uncommon.
6:24 pm
March 27, 2012
Offline
Here is a low light frame sample from the BMCC.
I think the Ikonoskop handled that Turkish street scene better. Cleaner around the light sources even with the halos, less cross color pollution and and less noise in the shadows. No camera is perfect. I saved this as a jpeg from the original DNG frame with no corrections. It is as shot.