• UPDATE



    The forum will be moving to a new domain in the near future (canonrumorsforum.com). I have turned off "read-only", but I will only leave the two forum nodes you see active for the time being.

    I don't know at this time how quickly the change will happen, but that will move at a good pace I am sure.

    ------------------------------------------------------------

Puzzling 1DX Mark II Sensor Issue?

Jun 25, 2012
829
222
10,473
Canada
I saw a thread on DPReview where a user was reporting dust issues with his new 1DX Mark II: http://www.dpreview.com/forums/thread/4022765

I noticed these strange "brush-like" marks in the DPReview user's image in certain places. When I checked my camera, it is also exhibiting this phenomenon too. I took a sensor loop and peered inside, but there appears to be nothing on the sensor other than dust and oil splatter -- no scratches whatsoever.

What do you all make of this? The picture is a screen grab of a section of the frame from the DPReview user's image. I turned up DeHaze to 100% in Lightroom to emphasize what I'm talking about.


I should state that these marks appear at narrow apertures on my sensor, and this picture below the OP said was taken at f/32.

EDIT: Please see images below.
 

Attachments

  • Screen Shot 2016-07-20 at 10.06.37 PM.jpg
    Screen Shot 2016-07-20 at 10.06.37 PM.jpg
    767.7 KB · Views: 324
Someone on DPReview suggested that these might be oil smears and recommended using Smear Away by Visible Dust.

I've looked very carefully at my sensor using a Visible Dust loop and can't see any signs of smears or, worse, scratches (my main concern). If these were smears, how would they appear from normal use? I have not touched the sensor.

The other fellow on DPReview also has similar marks/smears too.

I feel a little deflated with this issue now. :(

Here are two screen shots of a portion of an image I took at f/32 with DeHaze maxed and without any modification to show the smears/marks.
 

Attachments

  • Sensor Marks.jpg
    Sensor Marks.jpg
    387.4 KB · Views: 256
  • Without DeHaze.png
    Without DeHaze.png
    1.4 MB · Views: 286
Upvote 0
Sure you're not applying too much scrutiny? You'd possibly find as much to criticize apply the same scrutiny and dehaze to just about any sensor on the planet.

If it's not showing up in your work then it's a non-issue. If it is interfering, then it's definitely something to follow up with Canon.

-pw
 
Upvote 0
I'm assuming you actually got your sensor professionally cleaned before posting this, of course.

If it's affecting your work, I'd get canon to have a look at it. This is a 1D, not a t3. If not... who really cares. Any sensor will show some dirt if you look closely enough, like pwp says. I agree that this looks like it would show up in anything monotone like a sky or a white sheet or dress or something...
 
Upvote 0
Gentlemen, it's not the spots that concern me on the sensor. It's the "brush-like" marks on the sensor that I am worried about, as I have never seen anything like that before.

It's an $8000 camera here in Canada, and I have had it a month. I'm concerned there's damage that was not caused by me.

The lines also do appear in my photographs. I noticed this from macro shots at f/20. It's just at f/32 they are more prounced, and I used the Dehaze function to show what I'm talking about as it exaggerates every spot. Dehaze is only to draw attention to the area.
 
Upvote 0
Send it in.

I had a problem with dust spots. Sent it in for a cleaning. Unfortunately, even after cleaning when I shot more outdoor scenes, I could see at least one huge dust spot. After much apprehension, I ended up using a Giottos to blow the sensor off. Seems to have taken care of the problem, except for some very tiny spots in a corner that won't be nearly as difficult to correct.

I'm going to monitor it and see it it continues.

Those lines don't look good to me. I wouldn't hesitate to contact Canon.

I'm not a believer in using anything that actually touches the sensor, so if it is oil or something, I'd rather have Canon do the cleaning.

I'm reading that this is a common problem with the 1D series. I hope not. I'd be curious about the experience of others and what they do about it. I don't care so much about dust in blue sky scenes, but don't want it showing up in other pictures.
 
Upvote 0
unfocused said:
Send it in.

I had a problem with dust spots. Sent it in for a cleaning. Unfortunately, even after cleaning when I shot more outdoor scenes, I could see at least one huge dust spot. After much apprehension, I ended up using a Giottos to blow the sensor off. Seems to have taken care of the problem, except for some very tiny spots in a corner that won't be nearly as difficult to correct.

I'm going to monitor it and see it it continues.

Those lines don't look good to me. I wouldn't hesitate to contact Canon.

I'm not a believer in using anything that actually touches the sensor, so if it is oil or something, I'd rather have Canon do the cleaning.

I'm reading that this is a common problem with the 1D series. I hope not. I'd be curious about the experience of others and what they do about it. I don't care so much about dust in blue sky scenes, but don't want it showing up in other pictures.

I just had the sensor cleaned this morning. Most of the spots are gone but the lines remain. The technician at Vistek didn't know what to make of the lines as he could see nothing on the sensor itself.

I'm wondering if this is a production issue with the AA filter or something.

As Pwp said, maybe I'm being overly critical, but it's a strange phenomenon that is a bit disconcerting IMO. I might send it to Canon, but I have had a bad experience with their repair centre in the past. Plus, I can practically hear them already telling me I have done something to the sensor.

It would be nice if other 1DX Mark II owners could do an expiermental shot at f/22 or f/32 to see if theirs' exhibit the same thing.
 
Upvote 0
Like I said above, I had the sensor cleaned today. The oil spots were largely removed, although new ones are already appearing (a fact of life with this series of camera). The brush-like marks remain, which is what I am upset over.

Here's a series of screenshots at different apertures. Notice how the sharpness and definition of the lines/brush-marks/smears do not change, but the dust and oil spots become more and more defined as I stop down. This leads me to believe it's a sensor defect, or something going on with the AA filter.

I'm also happy to provide RAW files too.
 

Attachments

  • f:25.jpg
    f:25.jpg
    501.6 KB · Views: 231
  • f:22.jpg
    f:22.jpg
    556.6 KB · Views: 242
  • f:18.jpg
    f:18.jpg
    499 KB · Views: 240
  • f:16.jpg
    f:16.jpg
    505.6 KB · Views: 236
  • f:14.jpg
    f:14.jpg
    480.2 KB · Views: 230
  • f:13.jpg
    f:13.jpg
    478.1 KB · Views: 239
  • f:32.jpg
    f:32.jpg
    577.5 KB · Views: 229
Upvote 0
pwp said:
Sure you're not applying too much scrutiny? You'd possibly find as much to criticize apply the same scrutiny and dehaze to just about any sensor on the planet.

If it's not showing up in your work then it's a non-issue. If it is interfering, then it's definitely something to follow up with Canon.

-pw

This.

First time I had sensor dust, I freaked out. Spend time reading about it and cleaned my sensor for hours.

Now if I see too much sensor dust and I get tired of photoshopping them out, I'll take my swab and 2 minutes later ~50% of the dust is out. Keep shooting until it really affects your photos. Shooting blank wall at F32 and using 100% dehaze is sure to bring out all sort of things. Almost certain if you send it in, you'll get exactly same or worse back.
 
Upvote 0
tpatana said:
pwp said:
Sure you're not applying too much scrutiny? You'd possibly find as much to criticize apply the same scrutiny and dehaze to just about any sensor on the planet.

If it's not showing up in your work then it's a non-issue. If it is interfering, then it's definitely something to follow up with Canon.

-pw

This.

First time I had sensor dust, I freaked out. Spend time reading about it and cleaned my sensor for hours.

Now if I see too much sensor dust and I get tired of photoshopping them out, I'll take my swab and 2 minutes later ~50% of the dust is out. Keep shooting until it really affects your photos. Shooting blank wall at F32 and using 100% dehaze is sure to bring out all sort of things. Almost certain if you send it in, you'll get exactly same or worse back.

My friend, it's not the dust. It's the brush-like smears that are there that bother me.

I used Dehaze only to draw attention to them for those that are unfamiliar where to look. The lines are still visible without Dehaze applied too, but are faint; I can still make them out. It's not normal.

The sensor was cleaned, with now only 3 or 4 dust particles remaining at f/32, which I don't care about. The lines remain.
 
Upvote 0
R1-7D said:
tpatana said:
pwp said:
Sure you're not applying too much scrutiny? You'd possibly find as much to criticize apply the same scrutiny and dehaze to just about any sensor on the planet.

If it's not showing up in your work then it's a non-issue. If it is interfering, then it's definitely something to follow up with Canon.

-pw

This.

First time I had sensor dust, I freaked out. Spend time reading about it and cleaned my sensor for hours.

Now if I see too much sensor dust and I get tired of photoshopping them out, I'll take my swab and 2 minutes later ~50% of the dust is out. Keep shooting until it really affects your photos. Shooting blank wall at F32 and using 100% dehaze is sure to bring out all sort of things. Almost certain if you send it in, you'll get exactly same or worse back.

My friend, it's not the dust. It's the brush-like smears that are there that bother me.

I used Dehaze only to draw attention to them for those that are unfamiliar where to look. The lines are still visible without Dehaze applied too, but are faint; I can still make them out. It's not normal.

The sensor was cleaned, with now only 3 or 4 dust particles remaining at f/32, which I don't care about. The lines remain.

I didn't say those were dust, I was just giving example.

To be clear, you see those lines at F32 on a solid color wall? Any "normal" condition where you'd see those?
 
Upvote 0
tpatana said:
R1-7D said:
tpatana said:
pwp said:
Sure you're not applying too much scrutiny? You'd possibly find as much to criticize apply the same scrutiny and dehaze to just about any sensor on the planet.

If it's not showing up in your work then it's a non-issue. If it is interfering, then it's definitely something to follow up with Canon.

-pw

This.

First time I had sensor dust, I freaked out. Spend time reading about it and cleaned my sensor for hours.

Now if I see too much sensor dust and I get tired of photoshopping them out, I'll take my swab and 2 minutes later ~50% of the dust is out. Keep shooting until it really affects your photos. Shooting blank wall at F32 and using 100% dehaze is sure to bring out all sort of things. Almost certain if you send it in, you'll get exactly same or worse back.

My friend, it's not the dust. It's the brush-like smears that are there that bother me.

I used Dehaze only to draw attention to them for those that are unfamiliar where to look. The lines are still visible without Dehaze applied too, but are faint; I can still make them out. It's not normal.

The sensor was cleaned, with now only 3 or 4 dust particles remaining at f/32, which I don't care about. The lines remain.

I didn't say those were dust, I was just giving example.

To be clear, you see those lines at F32 on a solid color wall? Any "normal" condition where you'd see those?


Tpatana,

I posted a few above examples starting at f/13. The lines don't really increase in definition going through the aperture range; they are just 'there', which I find somewhat strange. If you look at the dust in the images they become more defined and sharp, but the lines do not.

To answer your question: I saw the lines first in a macro flower shot I took at f/18 with my Twin Lite MT-24EX attached and firing.

Would I notice the lines in most normal every-day shooting type scenarios? Probably not. This is still disconcerting, however, for an $6000 ($8000 CAD) camera that I've only been enjoying for a month.
 
Upvote 0
R1-7D said:
tpatana said:
If you see them in normal pictures, then for sure contact Canon and ask for repair. Could be feature though.


If it is a feature, I hope it's one that's removed later in the production line or in future models. I don't care for it.

:)

My 7D had the noise banding feature too. Some of the "features" are there for reason we don't know, it could be there to mask or fix something worse.
 
Upvote 0
tpatana said:
R1-7D said:
tpatana said:
If you see them in normal pictures, then for sure contact Canon and ask for repair. Could be feature though.


If it is a feature, I hope it's one that's removed later in the production line or in future models. I don't care for it.

:)

My 7D had the noise banding feature too. Some of the "features" are there for reason we don't know, it could be there to mask or fix something worse.

Unfortunately, I genuinely feel like these streaks or smears are not supposed to be there or hide anything. :(
 
Upvote 0
The streaks are actually probably a part of the sensor itself. Sensors are cut from a crystal, and those crystals often have concentric radial patterns that occur as the crystal is grown. Usually the patterns are invisible, however with very deep cooling (-35C or colder) and long exposures, the patterns can sometimes become visible due to a phenomena called Resibual Back Image or RBI. This is when electrons are freed into the substrate of the sensor, and bleed back into the pixels during longer exposures.

It is entirely possible that the technological design of the 1D X II is allowing such RBI and bleedback to occur without the deep cooling, and for much shorter exposures. It would be rather odd, however a lot of new per-pixel technologies are being used in sensors these days, and all the implications of those technologies are not necessarily known.

Outside of RBI, I do not know of any other known phenomena that would cause such streaking.
 
Upvote 0
jrista said:
The streaks are actually probably a part of the sensor itself. Sensors are cut from a crystal, and those crystals often have concentric radial patterns that occur as the crystal is grown. Usually the patterns are invisible, however with very deep cooling (-35C or colder) and long exposures, the patterns can sometimes become visible due to a phenomena called Resibual Back Image or RBI. This is when electrons are freed into the substrate of the sensor, and bleed back into the pixels during longer exposures.

It is entirely possible that the technological design of the 1D X II is allowing such RBI and bleedback to occur without the deep cooling, and for much shorter exposures. It would be rather odd, however a lot of new per-pixel technologies are being used in sensors these days, and all the implications of those technologies are not necessarily known.

Outside of RBI, I do not know of any other known phenomena that would cause such streaking.

Jrista,

Thank you for your post. That's extremely interesting. As an experiment, if I were to take a shorter exposure, say like 1/400, would that potentially have an effect on whether these patterns appear or not? Might be an interesting experiment for tomorrow.


Your explanation could also explain why I saw these similar patterns on MacNeutal's sensor-dust image he posted on DPReview. Theoretically, it could also mean that while there are similar patterns shown on other 1DX II sensors, the patterns could manifest themselves in different locations too, correct?

I wish other people who have this camera would take some small aperture photos and see if they can replicate this. I've been unable to find anyone else who have posted f/13+ full resolution pictures.

Again, thank you for taking the time to post.
 
Upvote 0
Just doing a bit of reading on RBI -- isn't it a bit like an image burn in, to some extent like what you'd get on an older plasma tv? I'm seeing examples of silhouettes of images for RBI if I google.

At the moment I'm not completely convinced what I am experiencing is RBI, but I do suspect it is indeed a property of the sensor.

I was just reconfirming tonight that my macro shots from the other day, before I even had the sensor looked at (let alone touched) did exhibit the same patterns. The cleanings should have at least smeared the lines if it were oil on the top glass. Therefore, it's either the top glass which has defects, or the sensor itself, or something akin to RBI like what Jrista suggests.

What's interesting is the Dehaze tool lets me over-hence the problem, and I can see these markings stretch across the Center of the sensor, although it's not nearly as pronounced in the Center as it is along the left edge of the image. It's the only left edge where I can make the markings out without any adjustments; the Center smears are non existent without Dehaze.

I can post a raw file in the morning.
 
Upvote 0