Jump to content

New Band is Firmware Bug


sean_reid

Recommended Posts

Advertisement (gone after registration)

OR their dog!

 

Who is Michael Jones?

 

Michael Jones is actually Terry Jones, not to be confused with Terry Gilliam, or alternatively he is Michael Palin. Hope this is clear! (Dont't mention Eric and Graham, I mentioned them but I think I got away with it.)

Link to post
Share on other sites

  • Replies 107
  • Created
  • Last Reply

It was just a joke that I am sure most got, just meant to lighten the mood a little. I am well aware of the problems and I am just preparing to send my M8 in for the repairs. I am trying to hold off a little in case any other problems do come out but my dealer wants to send all the 1st batch back together on the 15th.

Link to post
Share on other sites

Hi Jon,

 

Here is a picture posted by Yoshinori in the thread:

 

http://www.leica-camera-user.com/digital-forum/12233-black-horizontal-band.html

 

In this real world situation it is difficult to avoid the banding since the probability is reasonably high that there is going to be some light at the edge that will cause the banding.

 

Furrukh

 

Furrkh, noted and understood. Maybe they can fix this; I hope so. If not, for me, it's just one of the M8 quirks that we may have to live with. I just feel that its strengths far exceed its weaknesses. Whether professionals can live with this paradigm I cannot say, but amateurs can and will. I suspect many professionals will do so too. Let us hope so.

Link to post
Share on other sites

It was just a joke that I am sure most got, just meant to lighten the mood a little. I am well aware of the problems and I am just preparing to send my M8 in for the repairs. I am trying to hold off a little in case any other problems do come out but my dealer wants to send all the 1st batch back together on the 15th.

 

I think we got it. Good luck with your M8. I have an upgraded one now - it works great - no bands so far - except the one I shot at the Xmas fair.

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

... for me, it's just one of the M8 quirks that we may have to live with. I just feel that its strengths far exceed its weaknesses.

 

Jon, I totally agree with you. The camera is a pleasure to use and the results are very good. Personally, this particular banding is just a minor inconvenience which is insignificant compared with the strengths of the camera (small size, well built, excellent optics, excellent image quality).

 

Furrukh

Link to post
Share on other sites

Advertisement (gone after registration)

Jon, I totally agree with you. The camera is a pleasure to use and the results are very good. Personally, this particular banding is just a minor inconvenience which is insignificant compared with the strengths of the camera (small size, well built, excellent optics, excellent image quality).

 

Furrukh

 

Well put, my sentiments entirely and I believe that of many others. Jon

Link to post
Share on other sites

Hi Mark, I was wondering about that myself. Maybe it is two degrees off the normal to the sensor measured to the outside edge the frame? Somehow intense light gets under the protective baffle for the reference pixels from that angle? Light really has to be strong (more than 4 stops over exposed) to cause this. Just a guess I am not a hardware guy. Wondering how they will fix this in firmware.

 

I used my 28mm Summicron Asp as well as 50mm Summicron for those lamp tests.

 

Furrukh

Link to post
Share on other sites

Steve, I agree, it's one of a long list of Leica things I'm waiting for - information on how the filter/lens hood/lens cap is going to work if a filter is required all the time and on how the cyan correction is going to work when the camera doesn't know the focal length.

 

Afaik all Leica lens hoods are designed to accomodate a filter. No "slim" filters are required for any Leica lens. Cyan correction (as is vignetting correction) can only work with coded lenses. I imagine there will be a parameter to be set to tell the camera if a filter is mounted.

Link to post
Share on other sites

Light piping between surface dielectric layers is a known problem with CCD imagers. That could explain why the angle of incidence is so critical. Light has to enter the "pipe" at the correct angle for it to be propagated by total internal reflection to the reference sites.

 

Bob.

Link to post
Share on other sites

Maybe it is just a few edge pixels causing this. I'm no expert, but I imagine a sector could then be switched off in firmware.

 

I don’t think switching the peripheral pixels off is a solution since these pixels provide important reference black point, white balance etc. info i.e. they set the reference for the visible pixels in the offending row (or half row until the middle of the sensor). With the reference gone it would be difficult to handle those visible pixels without producing other serious artifacts.

 

Furrukh

Link to post
Share on other sites

Wondering how they will fix this in firmware.

 

I used my 28mm Summicron Asp as well as 50mm Summicron for those lamp tests.

 

Furrukh

 

The only firmware workaround that I can imagine is to take a quick scan around the reference pixels and if there is a sharp narrow peak somewhere, disregard it and interpolate from the two sides of it. I'm also puzzled by the two degrees comment. The fact that you got the effect with two lenses as different as 28mm and 50mm, where the light at the edge of the image comes in at two very different angles, doesn't fit this line of thought.

 

scott

Link to post
Share on other sites

Afaik all Leica lens hoods are designed to accomodate a filter. No "slim" filters are required for any Leica lens. Cyan correction (as is vignetting correction) can only work with coded lenses. I imagine there will be a parameter to be set to tell the camera if a filter is mounted.

 

Jaap, to avoid vignetting on film, the WATE uses an over-size filter mounted on the lens using a step up ring and the standard lens hood and cap cannot be used. If you're using the lens on an M8, if you put an IR filter on all the time, the supplied hood and cap are redundant and there's no hood/cap solution. The cyan issue is that the lens doesn't tell the camera what the selected focal length is so unless they re-design the lens, the user will have to select this manually or else Leica will shoot for one-size-suits-all in their optimisation which, given the WATE is going to encounter the most extreme cyan, doesn't appeal.

 

Personally, I'd like to see them pull the WATE for a mechanical redesign and offer a smaller IR filter for M8 use which retains the use of the hood and cap - even though it will vignette on film.

 

Bottom line is that if you are spending €3500 on a lens, these compromises should not be required.

Link to post
Share on other sites

Michael Jones is actually Terry Jones, not to be confused with Terry Gilliam, or alternatively he is Michael Palin. Hope this is clear! (Dont't mention Eric and Graham, I mentioned them but I think I got away with it.)

 

and don't mention the War either...

( I assume you must be John Cleese/Basil Fawlty..)

 

BTW I'm not Manuel, but I'm from Barcelona :D

 

Oh well, If you aren't a Fawlty Towers "connaisseur", just forget about it..

Link to post
Share on other sites

The only firmware workaround that I can imagine is to take a quick scan around the reference pixels and if there is a sharp narrow peak somewhere, disregard it and interpolate from the two sides of it.

scott

 

Hi Scott, the band can be quite broad depending on the light source. Here is a picture posted by Eric in the thread (Post #35):

 

http://www.leica-camera-user.com/dig...al-band-2.html

 

So I think that interpolating around the spike (if you can call it a spike here) will not do the trick.

 

Furrukh

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

Dark Reference Pixels are described on p 7 of Kodak's specs on the KAF-10500, ending with: "Under normal circumstances, these pixels do not respond to light and may be used as a dark reference."

 

So it sounds as if the dark reference is a part of the sensor, and not something determined by shielding or baffling added by the camera manufacturer--i.e., no hardware changes needed.

 

However, what does it mean to say that "under normal circumstances" the pixels are unresponsive to light?

 

Anybody with understanding of chip construction?

Link to post
Share on other sites

Presumably "under normal circumstances" = "except when used in the Leica M8"... LOL

 

I believe these pixels are masked off in the sensor so that they are real pixels but kept in the dark. Somehow, it looks like light is finding a way in...

Link to post
Share on other sites

and don't mention the War either...

( I assume you must be John Cleese/Basil Fawlty..)

 

BTW I'm not Manuel, but I'm from Barcelona :D

 

Oh well, If you aren't a Fawlty Towers "connaisseur", just forget about it..

 

Don't tell Sybil I bought an M8 - Que? And don't mention the bands.

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...