Home | Contact Us | FAQ | Search & Site Map | Link to Us
Sign In | Join | Other 45 Sites in Network
PhotoKB Home
Discussion Groups
Digital Photography
Digital PhotoDSLR CamerasZLR CamerasPoint & Shoot Cameras
Film Photography
35 mmLarge FormatMedium formatDarkroomFilm and LabsOther Equipment
Photo Technique
Nature PhotographyPeople PhotographyTechnique General
General Photo Topics
General TopicsAustralian PhotographyUK Photography
DirectoryPhoto Clubs

Photo Forum / Digital Photography / DSLR Cameras / May 2005

Tip: Looking for answers? Try searching our database.

20D banding and DNG Converter 3.1

Thread view: 
Enable EMail Alerts  Start New Thread
Thread rating: 
Ben Rosengart - 26 May 2005 21:38 GMT
JPS said that DNG Converter 3.1 takes the "bands" out of 20D images.
I tried it out and it seems to work.  Why/how?  And since Adobe
apparently knows how to fix 20D banding, why don't they do it in ACR?

Signature

Ben Rosengart                                            (212) 741-4400 x215
    Sometimes it only makes sense to focus our attention on those
    questions that are equal parts trivial and intriguing.
                                            --Josh Micah Marshall

JPS@no.komm - 27 May 2005 00:03 GMT
>JPS said that DNG Converter 3.1 takes the "bands" out of 20D images.
>I tried it out and it seems to work.  Why/how?  And since Adobe
>apparently knows how to fix 20D banding, why don't they do it in ACR?

They supposedly do, in ACR 3.1.  It's supposed to be the same logic;
line-by line blackpointing.  Of course, they keep the RAW data at
12-bit, so small offsets up to a half RAW level, are still present.

It doesn't seem that they do it vertically, only horizontally.  A step
in the right direction, anyway, and let's hope the other converters
follow suit.
Signature


<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
  John P Sheehy         <JPS@no.komm>

><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Ben Rosengart - 27 May 2005 00:25 GMT
>>JPS said that DNG Converter 3.1 takes the "bands" out of 20D images.
>>I tried it out and it seems to work.  Why/how?  And since Adobe
>>apparently knows how to fix 20D banding, why don't they do it in ACR?
>
> They supposedly do, in ACR 3.1.

I see.  Sigh.

> It's supposed to be the same logic;
> line-by line blackpointing.  Of course, they keep the RAW data at
> 12-bit, so small offsets up to a half RAW level, are still present.

How's that?

> It doesn't seem that they do it vertically, only horizontally.  A step
> in the right direction, anyway, and let's hope the other converters
> follow suit.

That's interesting.  By far the most visible banding on my camera is
vertical.  But DNG Converter 3.1 still seems to help.

Since the problem is addressible by software, I do not understand
why Canon don't fix it in a firmware update.

Signature

Ben Rosengart                                            (212) 741-4400 x215
    Sometimes it only makes sense to focus our attention on those
    questions that are equal parts trivial and intriguing.
                                            --Josh Micah Marshall

JPS@no.komm - 27 May 2005 00:58 GMT
>> It's supposed to be the same logic;
>> line-by line blackpointing.  Of course, they keep the RAW data at
>> 12-bit, so small offsets up to a half RAW level, are still present.

>How's that?

The lines are not offset by integer numbers.

>> It doesn't seem that they do it vertically, only horizontally.  A step
>> in the right direction, anyway, and let's hope the other converters
>> follow suit.

>That's interesting.  By far the most visible banding on my camera is
>vertical.  But DNG Converter 3.1 still seems to help.

>Since the problem is addressible by software, I do not understand
>why Canon don't fix it in a firmware update.

Canon denies a problem.
Signature


<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
  John P Sheehy         <JPS@no.komm>

><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Ben Rosengart - 27 May 2005 01:28 GMT
>>Since the problem is addressible by software, I do not understand
>>why Canon don't fix it in a firmware update.
>
> Canon denies a problem.

That's remarkable, given how easy it is to reproduce the problem.
Do you have a pointer to any statement from Canon on the topic?
Do they participate in any web bulletin boards or the like?

Signature

Ben Rosengart                                            (212) 741-4400 x215
    Sometimes it only makes sense to focus our attention on those
    questions that are equal parts trivial and intriguing.
                                            --Josh Micah Marshall

JPS@no.komm - 27 May 2005 01:34 GMT
>That's remarkable, given how easy it is to reproduce the problem.
>Do you have a pointer to any statement from Canon on the topic?
>Do they participate in any web bulletin boards or the like?

No, but everyone on DPReview who sent their camera to Canon because of
it got it back with no change.
Signature


<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
  John P Sheehy         <JPS@no.komm>

><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Ben Rosengart - 27 May 2005 01:47 GMT
>>That's remarkable, given how easy it is to reproduce the problem.
>>Do you have a pointer to any statement from Canon on the topic?
>>Do they participate in any web bulletin boards or the like?
>
> No, but everyone on DPReview who sent their camera to Canon because of
> it got it back with no change.

About what one might expect.  Perhaps the rise of ACR 3.1 will
eventually embarrass Canon into action.  I must say, this certainly
puts arrows into the quiver of anyone out to criticize Canon.

Signature

Ben Rosengart                                            (212) 741-4400 x215
    Sometimes it only makes sense to focus our attention on those
    questions that are equal parts trivial and intriguing.
                                            --Josh Micah Marshall

briansgooglegroupemail@yahoo.com - 27 May 2005 02:33 GMT
> About what one might expect.  Perhaps the rise of ACR 3.1 will
> eventually embarrass Canon into action.  I must say, this certainly
> puts arrows into the quiver of anyone out to criticize Canon.

What exactly is the black banding problem?  (I have an XT)

Thanks.
Ben Rosengart - 27 May 2005 19:29 GMT
>> About what one might expect.  Perhaps the rise of ACR 3.1 will
>> eventually embarrass Canon into action.  I must say, this certainly
>> puts arrows into the quiver of anyone out to criticize Canon.
>
> What exactly is the black banding problem?  (I have an XT)

In low-light or underexposed images, one can see bands of noise,
especially in the less-exposed regions of the image.  For example,
look closely at the sky in this picture:

 http://narcissus.net/BAM-raw.jpg

Signature

Ben Rosengart                                            (212) 741-4400 x215
    Sometimes it only makes sense to focus our attention on those
    questions that are equal parts trivial and intriguing.
                                            --Josh Micah Marshall

JPS@no.komm - 27 May 2005 03:46 GMT
>About what one might expect.  Perhaps the rise of ACR 3.1 will
>eventually embarrass Canon into action.  I must say, this certainly
>puts arrows into the quiver of anyone out to criticize Canon.

Canon is the Hare in the story of the Hare and the Tortoise.
Signature


<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
  John P Sheehy         <JPS@no.komm>

><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
RichA - 27 May 2005 05:12 GMT
>>About what one might expect.  Perhaps the rise of ACR 3.1 will
>>eventually embarrass Canon into action.  I must say, this certainly
>>puts arrows into the quiver of anyone out to criticize Canon.
>
>Canon is the Hare in the story of the Hare and the Tortoise.

Given sales figures, it looks like Kodak is the tortoise.
-Rich
Ryadia@Home - 27 May 2005 11:09 GMT
>>>That's remarkable, given how easy it is to reproduce the problem.
>>>Do you have a pointer to any statement from Canon on the topic?
[quoted text clipped - 6 lines]
> eventually embarrass Canon into action.  I must say, this certainly
> puts arrows into the quiver of anyone out to criticize Canon.

Certainly the errors in Canon's fabled "L" series 24~70 f2.8 which are
so easily corrected during conversion from RAW to an image file have
alarmed some people. DxO fix the errors transparently and they are
hugely noticeable between a corrected image on an uncorrected one.
Canon's response when I returned the lens? "It has been tested and is
within the manufacturing tolerance for that lens".

I think replacing the cameras with newer models every couple of years is
sweeping the problems away. When their lenses have faults too, some
maker only has to come up with a decent camera body to match Nikon's
lenses and we'll see a lot of former Nikonians heading back that way.

Signature

Douglas...
It's traditional, painter's use it, Rembrandt used it.
Now you can put your photos on it too!
http://www.canvasphotos.com.au

Bart van der Wolf - 28 May 2005 15:57 GMT
SNIP
> Certainly the errors in Canon's fabled "L" series 24~70 f2.8 which
> are so easily corrected during conversion from RAW to an image file
> have alarmed some people.

What errors are you referring to?
I'm not saying that the lens (any lens, especially any zoom lens) is
perfect, I'm just curious as to what you mean.

> DxO fix the errors transparently and they are hugely noticeable
> between a corrected image on an uncorrected one.

Converting with DxO is one way of correcting residual lens
aberrations, and Low-pass filter effects. Other methods are possible,
if beneficial for the intended image use.

Bart
Frank ess - 29 May 2005 05:48 GMT
Here are three conversions of the same CR2 file.
Single change from default: increase exposure to +1.95
Canon 20D  24-70 2.8L  70mm  ISO 1600
1/200 sec @ f/3.2

PS CS ACR

PS CS2 ACR

Adobe DNG converter

http://www.fototime.com/inv/C5A10FA74EEEE99

Signature

Frank ess

Ben Rosengart - 27 May 2005 00:25 GMT
>>JPS said that DNG Converter 3.1 takes the "bands" out of 20D images.
>>I tried it out and it seems to work.  Why/how?  And since Adobe
>>apparently knows how to fix 20D banding, why don't they do it in ACR?
>
> They supposedly do, in ACR 3.1.

I see.  Sigh.

> It's supposed to be the same logic;
> line-by line blackpointing.  Of course, they keep the RAW data at
> 12-bit, so small offsets up to a half RAW level, are still present.

How's that?

> It doesn't seem that they do it vertically, only horizontally.  A step
> in the right direction, anyway, and let's hope the other converters
> follow suit.

That's interesting.  By far the most visible banding on my camera is
vertical.  But DNG Converter 3.1 still seems to help.

Since the problem is addressible by software, I do not understand
why Canon don't fix it in a firmware update.

Signature

Ben Rosengart                                            (212) 741-4400 x215
    Sometimes it only makes sense to focus our attention on those
    questions that are equal parts trivial and intriguing.
                                            --Josh Micah Marshall

Stacey - 27 May 2005 03:05 GMT
> That's interesting.  By far the most visible banding on my camera is
> vertical.  But DNG Converter 3.1 still seems to help.
>
> Since the problem is addressible by software, I do not understand
> why Canon don't fix it in a firmware update.

Probably the same reason olympus hasn't fixed the "problem" with the jpeg NR
on the E300, they are ignoring it as being a problem. For most people it
probably isn't?

I've given up any hope of OM addressing this but I shoot RAW anyway, for me
it was better they added MLU and worked on the exposure issues which they
did fix!  :-)

Signature


 Stacey

 
Sign In
Join
My Latest Posts
My Monitored Threads
My Blog
My Photo Gallery
My Profile
My Homepage

Start New Thread
Enable EMail Alerts
Rate this Thread



©2009 Advenet LLC   Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.