Green band (fixed-pattern noise) in photos with Canon 5D Mark IV: Canon EOS-1D / 5D / 6D Talk Forum: Digital Photography Review

Green band (fixed-pattern noise) in photos with Canon 5D Mark IV

Started Aug 2, 2017 | Discussions

yadavtarun03 • New Member • Posts: 14

Green band (fixed-pattern noise) in photos with Canon 5D Mark IV

Aug 2, 2017

TLDR;

Photos from Canon 5D Mark IV show a band (fixed-pattern noise).
Canon Factory Service Center said unit is ok.
Looking for confirmation that unit is defective and suggestions for what to do to convince them of the problem and get it repaired/replaced.

Problem

I recently noticed that long exposures taken at higher ISOs with my Canon 5D Mark IV (purchased in Oct 2016, still in the warranty period) demonstrate a green band close to one edge.

This is quite visible in under-exposed images that are edited in Lightroom to increase the exposure (starts showing up with +1-stop, often when coupled with other changes).

Sample Images

Here's an Imgur album with a few example images showing the problem (deliberately edited to make the problem more obvious). The band is typically faintly visible in the original image and in almost all cases, starts becoming more obvious as I edit the photo.

5DIV, 200s, ISO 800, f/16 Exposure +0.66, Dehaze +100 in Lightroom to show green band at bottom

5DIV, 420s, ISO 800, f/16 Dehaze +100 in Lightroom to show horizontal green band along bottom edge

5DIV, 120s, ISO 640, f/11 Exposure +3, Dehaze +100 in Lightroom to show vertical green band along right edge

The full set of images from the 5DIV and the 70D (RAW and JPEGs) can be found here (Dropbox).

So Far…

I contacted Canon Customer Support and after exchanging a few emails they asked me to ship the camera to a service center After about 1.5 weeks, the body was shipped back and the accompanying document detailing service details said:

Your product has been examined and it was found that the product performed according to specifications. The unit was cleaned and checked. Product functions were confirmed. **note: Please shoot the image on with proper exposure and avoid the extreme exposure adjustments with software.**.

I should note that the test images I'd sent them were deliberately under-exposed to make the problem obvious. And I personally wouldn’t have considered an post-processing exposure change of 1- or 2-stops to be “extreme" when editing RAW images from a top-of-the-line, full-frame, professional DSLR. In fact, my Canon 70D from 2014 appears to perform better under the same conditions with the same settings - definitely more noise but no banding.

When I first reached out to Canon, I didn't know what this band was. I've since found out about Fixed-Pattern Noiseand the Long Exposure Noise Reduction option in Canon (and Nikon) cameras. Searching online, I see several people seeing this with older Canon cameras but many reviews and forum posts about the 5D Mark IV are raving about the lack of banding To me that's further confirmation that I have a bad unit.

Questions

Do folks with Canon 5D Mark IVs see this? I'd like to establish if this is common to this model or if it affects my unit in particular.
Am I wrong in expecting not to have to deal with this on this camera?
Any suggestions on what to do next to convince Canon that this is a problem with my unit?

Notes

[1] Conditions for reproducing:

I noticed the problem with a few images from the field, taken at ISO 600.
I’ve been testing with ISO 800.
Exposure durations have varied from 150s to 510s.
I fixed White Balance at 3100K - primarily to be able to easily compare images.
Long Exposure Noise Reduction is disabled.

[2] The first response was what looked like standard boilerplate asking me to:

try a different memory card
clean the lens contacts
try different lensesI tried all of these these (even though this looks like a sensor problem to me) and continue to see the issue.

[3] Canon covered return shipping.

[4] Long Exposure Noise Reduction

This was set to Auto on my camera (and I didn't realize what it was for).
When taking test photos, I found found that it only kicked for exposures that were 450s, and not for longer and shorter ones.
For those that don't know, when this is enabled the camera will take a second exposure with the same settings as the original, first shot except with the shutter closed. This allows it to compensate for any temporally constant noise - fixed pattern noise. The downside is that it keeps the camera busy for the same duration as the original exposure and so each exposure effectively lasts twice as long which isn't ideal when trying to take, say, 10 minute exposures out in the field.

[5] Forum posts about similar issues (other models):

[6] Canon 5D Mark IV reviews and forum posts about lack of fixed-pattern noise with 5DIV (sorry, none of these had anchors that I could use so you'll have to search the page for 'fixed pattern noise' to find the relevant text):

Google search results

Note that I'd posted this on Reddit but didn't get too many responses - I'm hoping to get some more help here.

Canon EOS 5D Mark IV Canon EOS 5DS Canon EOS 70D Canon EOS 7D Mark II

If you believe there are incorrect tags, please send us this post using our feedback form.

Reply to thread Reply with quote Complain

MOD Mako2011 • Forum Pro • Posts: 24,978

Not.....banding noise

In reply to yadavtarun03 • Aug 2, 2017

yadavtarun03 wrote:

TLDR;

Photos from Canon 5D Mark IV show a band (fixed-pattern noise).

Problem

I recently noticed that long exposures taken at higher ISOs with my Canon 5D Mark IV (purchased in Oct 2016, still in the warranty period) demonstrate a green band close to one edge.

This is quite visible in under-exposed images that are edited in Lightroom to increase the exposure (starts showing up with +1-stop, often when coupled with other changes).

.

5DIV, 200s, ISO 800, f/16 Exposure +0.66, Dehaze +100 in Lightroom to show green band at bottom

5DIV, 420s, ISO 800, f/16 Dehaze +100 in Lightroom to show horizontal green band along bottom edge

5DIV, 120s, ISO 640, f/11 Exposure +3, Dehaze +100 in Lightroom to show vertical green band along right edge

When I first reached out to Canon, I didn't know what this band was. I've since found out about Fixed-Pattern Noise

What you are seeing there is not Banding or Fixed Pattern noise. It's always in the same place and only related to one row of pixels. Likely an issue with the electronics associated with that row. Will require a sensor change to fix.

-- hide signature --

My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)

Reply Reply with quote Complain

Steve Balcombe • Forum Pro • Posts: 13,047

Re: Not.....banding noise

In reply to Mako2011 • Aug 2, 2017

Mako2011 wrote:

What you are seeing there is not Banding or Fixed Pattern noise. It's always in the same place and only related to one row of pixels. Likely an issue with the electronics associated with that row. Will require a sensor change to fix.

Agreed - there is something wrong and it's not what we would normally refer to as "banding". Faulty sensor (or electronics associated with it) is a good guess.

-- hide signature --

https://www.flickr.com/photos/stevebalcombe/

Reply Reply with quote Complain

OP yadavtarun03 • New Member • Posts: 14

Re: Not.....banding noise

In reply to Steve Balcombe • Aug 2, 2017

Steve Balcombe wrote:

Mako2011 wrote:

What you are seeing there is not Banding or Fixed Pattern noise. It's always in the same place and only related to one row of pixels. Likely an issue with the electronics associated with that row. Will require a sensor change to fix.

Agreed - there is something wrong and it's not what we would normally refer to as "banding". Faulty sensor (or electronics associated with it) is a good guess.

On re-reading then Wikipedia article, it sounds like the distinction between what I have and FPN is that FPN presents as a more regular grid pattern?

Additionally, as I noted in the post, Canon has already looked at this and said it performs to spec. I’d welcome any suggestions to help prove more definitively that the unit is defective.

Thanks!

Reply Reply with quote Complain

MayaTlab0 • Senior Member • Posts: 2,612

Re: Not.....banding noise

In reply to yadavtarun03 • Aug 2, 2017

yadavtarun03 wrote:

Additionally, as I noted in the post, Canon has already looked at this and said it performs to spec. I’d welcome any suggestions to help prove more definitively that the unit is defective.

Thanks!

I can't help you with that specific case, but Canon's position isn't definitive obviously and I'm not even sure that they actually believe what they answered you. It's just one of their BS standard starting point for "negotiations". If I had never pushed passed the usual "it's within specifications" BS that nearly all customer services always serve me, I'd have had nothing done, ever.

When I know that something of this kind is hard to prove (for example if I don't have two copies of the same product), I usually try to elevate the issue. One of the most efficient way I've found to get things done quickly is to find the company's email format, re-construct the mails of all the top management (regardless of whether they're qualified to answer or not), and then bombard all of them with a carefully explained mail where I clearly state the problem, why it affects me, and what I want them to do.

Since 99% of the time I was right and my product wasn't actually "within spec", I then always make a point to prove to them afterwards that their first answer was complete BS by showing them the effect of their own repair.

That being said, can you have access to another 5DIV from a friend, a shop, or from another source ? If you have found a way to reproduce your 5DIV's problem consistently, then comparing two 5DIVs could help in starting to build up evidence.

Reply Reply with quote Complain

jsmiller • Contributing Member • Posts: 826

Re: Not.....banding noise

In reply to yadavtarun03 • Aug 2, 2017

yadavtarun03 wrote:

Steve Balcombe wrote:

Mako2011 wrote:

What you are seeing there is not Banding or Fixed Pattern noise. It's always in the same place and only related to one row of pixels. Likely an issue with the electronics associated with that row. Will require a sensor change to fix.

Agreed - there is something wrong and it's not what we would normally refer to as "banding". Faulty sensor (or electronics associated with it) is a good guess.

On re-reading then Wikipedia article, it sounds like the distinction between what I have and FPN is that FPN presents as a more regular grid pattern?

Additionally, as I noted in the post, Canon has already looked at this and said it performs to spec. I’d welcome any suggestions to help prove more definitively that the unit is defective.

Thanks!

Regardless of what you call it, it is either a sensor fault or associated electronics fault affecting the readout of a several rows as others have said. (It is probably not appropriate to call it FPN, as it is not really a pattern extending over a good-sized region. I wouldn't even call it noise, but that's just my opinion.) The camera is faulty. You have to try again with Canon. Call up and ask to speak to a supervisor. Insist on it, don't give up. I had a problem with a Rebel years ago that front focussed, and that camera had no ability to adjust focus. I sent it back three times. One technician advised me that the problem was easy to fix- always shoot at f/8 or slower. When I said in that case why should I even bother to buy expensive fast lenses. He had no answer. That can be the quality of understanding their techs have. Then I got a high-level supervisor on the phone. He said send it to him, and it was fixed. It shouldn't be that way, but it can be.

I think your images illustrate the problem quite well. If you didn't before, I would send a highly cropped image that just shows those rows and a bit of the surrounding ones as well as one showing the whole image. It's totally obvious it has nothing to do with exposure levels. But what's obvious to some of us may escape a Canon tech completely.

Joe

jsmiller's gear list:jsmiller's gear list

Canon PowerShot ELPH 360 HS

Reply Reply with quote Complain

OP yadavtarun03 • New Member • Posts: 14

Re: Not.....banding noise

In reply to jsmiller • Aug 3, 2017

Thanks for your thoughts and observations Joe. I'm also considering reaching out to their social media accounts on Twitter and Facebook - hopefully that'll help.

Reply Reply with quote Complain

OP yadavtarun03 • New Member • Posts: 14

Re: Not.....banding noise

In reply to MayaTlab0 • Aug 3, 2017

Thanks for your thoughts and observations. I'm also considering reaching out to their social media accounts on Twitter and Facebook - hopefully that'll help.

As for a second 5DIV — I'd thought about that and that was one of my reasons for posting online. I don't have easy access to one and I was hoping someone might be able to try and reproduce. I will try and escalate and get this solved again but if that fails, I might rent a body for a day or two to test with. I'd rather not spend more but if that's what it takes to get a fix, so be it

Reply Reply with quote Complain

J A C S • Forum Pro • Posts: 12,323

Re: Not.....banding noise

In reply to yadavtarun03 • Aug 7, 2017

yadavtarun03 wrote:

Steve Balcombe wrote:

Mako2011 wrote:

What you are seeing there is not Banding or Fixed Pattern noise. It's always in the same place and only related to one row of pixels. Likely an issue with the electronics associated with that row. Will require a sensor change to fix.

Agreed - there is something wrong and it's not what we would normally refer to as "banding". Faulty sensor (or electronics associated with it) is a good guess.

On re-reading then Wikipedia article, it sounds like the distinction between what I have and FPN is that FPN presents as a more regular grid pattern?

Additionally, as I noted in the post, Canon has already looked at this and said it performs to spec. I’d welcome any suggestions to help prove more definitively that the unit is defective.

Call again. Ask to speak to a supervisor. Call another service center if needed.

I had to send my 135L three times to Canon. The first two times I could not convince them that there was a problem.

Reply Reply with quote Complain

MarshallG • Veteran Member • Posts: 6,300

Re: Not.....banding noise

In reply to yadavtarun03 • Aug 7, 2017

yadavtarun03 wrote:

Additionally, as I noted in the post, Canon has already looked at this and said it performs to spec. I’d welcome any suggestions to help prove more definitively that the unit is defective.

Try the exact same thing on another 5D Mark iv.

You're taking some very, very long exposures. It may be that ALL the Mark iv's do this; your 150 second "corner case" could have escaped Canon QA. Or maybe it's a glitch in the software; who knows? The first thing you should do is use another 5D iv's side-by-side with yours to see if the problem is unique to your body or not. There's no point in exchanging your camera if the problem will follow it.

MarshallG's gear list:MarshallG's gear list

Canon EOS 5D Mark IV Canon EF 50mm f/1.4 USM Canon EF 85mm f/1.8 USM Canon EF 16-35mm F2.8L II USM Canon EF 70-200mm F2.8L IS II USM +2 more

Reply Reply with quote Complain

Keeef • New Member • Posts: 8

Re: Green band (fixed-pattern noise) in photos with Canon 5D Mark IV

In reply to yadavtarun03 • Aug 7, 2017

Ask in the Astrophotography talk forum if anyone has seen anything similar.

Keeef's gear list:Keeef's gear list

Canon PowerShot SD940 IS Sigma DP1 Merrill Sigma DP3 Merrill Canon EOS-1Ds Canon EOS 500D +8 more

Reply Reply with quote Complain

OP yadavtarun03 • New Member • Posts: 14

Re: Not.....banding noise

In reply to J A C S • Aug 7, 2017

J A C S wrote:

Call again. Ask to speak to a supervisor. Call another service center if needed.

I had to send my 135L three times to Canon. The first two times I could not convince them that there was a problem.

Thanks, I'll keep at it!

Reply Reply with quote Complain

OP yadavtarun03 • New Member • Posts: 14

Re: Not.....banding noise

In reply to MarshallG • Aug 7, 2017

MarshallG wrote:

yadavtarun03 wrote:

Additionally, as I noted in the post, Canon has already looked at this and said it performs to spec. I’d welcome any suggestions to help prove more definitively that the unit is defective.

Try the exact same thing on another 5D Mark iv.

You're taking some very, very long exposures. It may be that ALL the Mark iv's do this; your 150 second "corner case" could have escaped Canon QA. Or maybe it's a glitch in the software; who knows? The first thing you should do is use another 5D iv's side-by-side with yours to see if the problem is unique to your body or not. There's no point in exchanging your camera if the problem will follow it.

Thanks! I'd definitely love to do that but I don't have easy access to another unit. I could rent from something like BorrowLenses.com but it's USD 123 (3-day minimum) before tax + delivery and I'd very much like to avoid that if possible. It's certainly a last resort option though!

By posting, I was hoping that 5DIV owners might be able to try and report if they see this problem with their cameras.

@Keeef has replied suggesting posting to the Astrophotography forum and I think I will try that to see if there are any users who can confirm or refute that this problem is endemic to the model.

Reply Reply with quote Complain

OP yadavtarun03 • New Member • Posts: 14

Re: Green band (fixed-pattern noise) in photos with Canon 5D Mark IV

In reply to Keeef • Aug 7, 2017

Keeef wrote:

Ask in the Astrophotography talk forum if anyone has seen anything similar.

Thanks for the suggestion - that sounds like a good idea!

Reply Reply with quote Complain

Blake Cook • Senior Member • Posts: 2,857

Try DPP to open the photos

In reply to yadavtarun03 • Aug 7, 2017

See what happens if you open the photos with DPP.

-- hide signature --

Blake in Vancouver
Canon and Zeiss Stuff. Mac Stuff & annoying PC & Windows stuff.

Blake Cook's gear list:Blake Cook's gear list

Canon EOS 5D Canon EOS 7D Mark II Canon EOS-1D X Mark II Canon EOS 5D Mark IV Canon EF 300mm f/2.8L IS USM +8 more

Reply Reply with quote Complain

OP yadavtarun03 • New Member • Posts: 14

Re: Try DPP to open the photos

In reply to Blake Cook • Aug 8, 2017

Blake Cook wrote:

See what happens if you open the photos with DPP.

Thanks for the suggestion - I gave this a quick try:

I haven't used this editor before and I couldn't tell how to edit the photo to see the same final result as my edits in Lightroom
As with other viewers, I do see the green band without any edits (albeit faint). I can make it a little more obvious by fiddling around with the contrast and color temperature.
Why do you think opening with this program might make a difference? I expect the problem to be inherent to the image itself.

Reply Reply with quote Complain

AmbleYonder • Senior Member • Posts: 1,792

Re: you are not the only one with the issue

In reply to yadavtarun03 • Aug 8, 2017

http://www.canonwatch.com/sensor-banding-issue-canon-eos-5d-mark-iv/.

check the long expose examples in the comment section .

Reply Reply with quote Complain

MOD Mako2011 • Forum Pro • Posts: 24,978

Not the...

In reply to AmbleYonder • Aug 8, 2017

AmbleYonder wrote:

http://www.canonwatch.com/sensor-banding-issue-canon-eos-5d-mark-iv/.

check the long expose examples in the comment section .

That is not the same issue the OP is having. Note the difference in color/size/edges

-- hide signature --

My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)

Reply Reply with quote Complain

Blake Cook • Senior Member • Posts: 2,857

Re: Try DPP to open the photos

In reply to yadavtarun03 • Aug 8, 2017

yadavtarun03 wrote:

Blake Cook wrote:

See what happens if you open the photos with DPP.

Thanks for the suggestion - I gave this a quick try:

I haven't used this editor before and I couldn't tell how to edit the photo to see the same final result as my edits in Lightroom

As with other viewers, I do see the green band without any edits (albeit faint). I can make it a little more obvious by fiddling around with the contrast and color temperature.

Why do you think opening with this program might make a difference? I expect the problem to be inherent to the image itself.

DPP is a Canon product that is designed for its sensors. Adobe Raw seem to take at least year to start to get colours right for Canon bodies in recent years. 1Dx processing isn't right yet.

-- hide signature --

Blake in Vancouver
Canon and Zeiss Stuff. Mac Stuff & annoying PC & Windows stuff.

Blake Cook's gear list:Blake Cook's gear list

Canon EOS 5D Canon EOS 7D Mark II Canon EOS-1D X Mark II Canon EOS 5D Mark IV Canon EF 300mm f/2.8L IS USM +8 more

Reply Reply with quote Complain

MOD Mako2011 • Forum Pro • Posts: 24,978

DPP

In reply to Blake Cook • Aug 8, 2017

Blake Cook wrote:

yadavtarun03 wrote:

Blake Cook wrote:

See what happens if you open the photos with DPP.

Thanks for the suggestion - I gave this a quick try:

I haven't used this editor before and I couldn't tell how to edit the photo to see the same final result as my edits in Lightroom

As with other viewers, I do see the green band without any edits (albeit faint). I can make it a little more obvious by fiddling around with the contrast and color temperature.

Why do you think opening with this program might make a difference? I expect the problem to be inherent to the image itself.

DPP is a Canon product that is designed for its sensors. Adobe Raw seem to take at least year to start to get colours right for Canon bodies in recent years. 1Dx processing isn't right yet.

DPP can't fix the OP's issue though. It's not a processing artifact. The line can be seen on the back of the camera when one looks

-- hide signature --
Blake in Vancouver
Canon and Zeiss Stuff. Mac Stuff & annoying PC & Windows stuff.

-- hide signature --

My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)

Reply Reply with quote Complain

Keyboard shortcuts:

FForum MMy threads

Latest sample galleries

Tamron SP 15-30mm F2.8 Di VC USD G2

Nikon Z6 sample gallery

Hasselblad XCD 80mm F1.9 Sample Gallery

Leica M10-D sample photos

See more galleries »

Latest in-depth reviews

No Nikon camera we've tested to date balances stills and video capture as well as the Nikon Z7. Though autofocus is less reliable than the D850, Nikon's first full-frame mirrorless gets enough right to earn our recommendation.

Nikon's Coolpix P1000 has moved the zoom needle from 'absurd' to 'ludicrous,' with an equivalent focal length of 24-3000mm. While it's great for lunar and still wildlife photography, we found that it's not suited for much else.

As a stills camera the Fujifilm X-T3 is a pleasant update to one of our favorite APS-C cameras, significantly improving the autofocus. If you're interested in stills and video, though, it's knockout.

Nikon Z7 AF tested: Tracking lags behind the D850, mirrorless competition

The Nikon Z7 is slated as a mirrorless equivalent to the D850, but it can't subject track with the same reliability as its DSLR counterpart. AF performance is otherwise good, except in low light where hunting can lead to missed shots.

The Canon EOS R is the first full frame mirrorless camera to use the new RF mount. We're well underway putting it through our range of standard tests – take a look at how it compares to the competition and our thoughts on using it so far.

Latest buying guides

Buying Guide: The best cameras under $500

What's the best camera for under $500? These entry level cameras should be easy to use, offer good image quality and easily connect with a smartphone for sharing. In this buying guide we've rounded up all the current interchangeable lens cameras costing less than $500 and recommended the best.

Buying Guide: The best lenses for Sony mirrorless cameras

Whether you've grown tired of what came with your DSLR, or want to start photographing different subjects, a new lens is probably in order. We've selected our favorite lenses for Sony mirrorlses cameras in several categories to make your decisions easier.

Buying Guide: The best lenses for Canon DSLRs

Whether you've grown tired of what came with your DSLR, or want to start photographing different subjects, a new lens is probably in order. We've selected our favorite lenses for Canon DSLRs in several categories to make your decisions easier.

Buying Guide: The best lenses for Nikon DSLRs

Whether you've grown tired of what came with your DSLR, or want to start photographing different subjects, a new lens is probably in order. We've selected our favorite lenses for Nikon DSLRs in several categories to make your decisions easier.

Buying Guide: The best cameras under $1000

What’s the best camera for less than $1000? The best cameras for under $1000 should have good ergonomics and controls, great image quality and be capture high-quality video. In this buying guide we’ve rounded up all the current interchangeable lens cameras costing under $1000 and recommended the best.

Check out more buying guides »


Tokina Issues Service Advisory for the Opera 50mm F1.4 FF on 1DX Mark II cameras Nov 7, 2018	Canon EOS 6D Mark II firmware 1.0.4 update fixes 'rare' operational bugs Jul 11, 2018	Canon interview: 'increased competition allows us to level-up' Mar 18, 2018	Full-frame showdown: Nikon D850 vs Canon 5D IV vs Sony a7R III Jan 30, 2018

12	Dull images on canon 5d mkiv
11	The 1DXMKII Rocks!
3	6D paired with a sigma 35 ART
3	6D ISO
2	Fall, End Of
1	Gray Import 5d MKIV

21	1Dx4me
20	bmike
18	MikeJ9116
14	Marek M
13	(unknown member)
11	RedFox88
11	Neoptix
11	Light Pilgrim
10	John Sheehy
9	mmarian
8	Makingtrax
8	SteveAnderson
7	Jura S
7	fishy wishy
7	George Zip
7	Andy01
6	Mark B.
6	hotdog321
6	Silvex
6	Dr_Jon


EveningOut by catastrophe from Your City - A Night Out	The Queen by TheGlennPalacio from red

Green band (fixed-pattern noise) in photos with Canon 5D Mark IV

You may also like

Latest sample galleries

Latest in-depth reviews

Latest buying guides