D3/D700 banding is real, but only after big user mistake...(with samples)

Started Jul 31, 2008 | Discussions
photoforfun Veteran Member • Posts: 6,084
D3/D700 banding is real, but only after big user mistake...(with samples)

D3/D700 banding is real, but only if you underexpose 2 stops while shooting over iso 3200 with strong highlights in the image and while you want to recover that image in postprocessing..., which is not a realistic scenario and a big "user mistake".... Upto that - and with any camera - correct exposure is one of the basics to get good pics...

I was reading about this as well when I had my D3(as Toa.design mentioned) and I did test it today with my D700. The issue is still there but only if you underexpose strong.

As most cameras have already have serious issues when you underexpose @ iso 200-400, a little imperfection after you made a mistake @ + iso 3200 is really a non-issue...

Underneath a sample

The image @ iso 6400 @ 0. EV, straight out of the camera, really outstanding iso 6400 IQ TMHO...

The image underexposed by 2 stops

The underexposed iso 6400 image recovered in NX2. Here you can clearly see banding.

-- hide signature --

Kindest regards,
Stany
I prefer one really good picture in a day over 10 bad ones in a second...

http://www.fotografie.fr/

Jono Slack
Jono Slack Forum Pro • Posts: 20,713
What an excellent post

Hi There
it's really nice to see a concept put so succinctly with good examples.

Thank You!
--
Jono Slack
http://www.slack.co.uk

 Jono Slack's gear list:Jono Slack's gear list
Leica M10 Leica SL2 Leica M10 Monochrom Leica M11 Leica Super-Vario-Elmar-SL 16-35mm F3.5-4.5 ASPH +2 more
J0seph Regular Member • Posts: 166
Re: I have to agree

I have to agree with this. I could replicate the "banding" issue, but only when pushing the exposure by 2+ stops for a file that was originally shot at 6400.

These are conditions that I don't think I would face in real life anyway.
--
Joseph Allen
http://www.JallenImages.com

Gabriele Sartori Veteran Member • Posts: 4,483
I think you have a point

With almost any camera is possible to have some type of banding when underexposed. On the other side people like to get certain situation with certain type of light.

I guess that in this case, taking the RAW version of your correctly exposed picture and then reducing the 2 stops in post would have given the desired result without the banding.

Thanks for the test.
Gabtiele

photoforfun wrote:

D3/D700 banding is real, but only if you underexpose 2 stops while
shooting over iso 3200 with strong highlights in the image and while
you want to recover that image in postprocessing..., which is not a
realistic scenario and a big "user mistake".... Upto that - and with
any camera - correct exposure is one of the basics to get good pics...

-- hide signature --

Regards
Gabriele
California, CA

 Gabriele Sartori's gear list:Gabriele Sartori's gear list
Leica M Typ 240 Nikon D7200 Panasonic Lumix DMC-GX8 Nikon Z7 Nikon 500mm F5.6E PF +2 more
ovrebekk
ovrebekk Veteran Member • Posts: 4,872
Thank you!

For clearing this up.

If you expect great results when underexposing an ISO 6400 shot two stops you are expecting to much!

I couldn't spot it at all in your first ISO 6400 shot, which looked great despite the high ISO.
--
Looking forward to the day a D700 will grace my sweaty hands.

AdamJRed Contributing Member • Posts: 545
Re: D3/D700 banding is real, but only after big user mistake...(with samples)

I'm sorry but you couldn't be MORE WRONG.

This issue has nothing to do with pushing the exposure in post processing.

I can duplicate the problem with out-of-camera JPGs.

Your post only proves that pushing an photo beyond its limits will result in problems, we all know that.

The problem does not have to do with post processing.

The problem is not about an incorrect exposure.

The problem is not about doing anything 'wrong'.

The problem is walking up to a DARK scene with BRIGHT 'point' highlights, and photographing the scene as you see it.

The effects will show up on out of the camera JPGs - so you can't blame PP as the cause.

I can take the exact same scene and photograph it with at least 4 other cameras that DO NOT do this, these cameras are not designed for high ISO use, yet the D700/D3 is.

 AdamJRed's gear list:AdamJRed's gear list
Fujifilm X-T1 Fujifilm XF 18-55mm F2.8-4 R LM OIS Lytro Light Field 8GB Fujifilm 16-55mm F2.8R LM WR
Dan Nikon Senior Member • Posts: 1,503
Re: D3/D700 banding is real, but only after big user mistake...(with samples)

I took this issue to task a bit more last night.

It did end up happening in two out of over 100 tries. But the kind of shots it happened in were really cr@ppy lighting situations, really cr@p photos, not even middle of the road ones.

You are making this more than it is, or using really cr@p technique, sorry to break it to you man...

AdamJRed wrote:

I'm sorry but you couldn't be MORE WRONG.

This issue has nothing to do with pushing the exposure in post
processing.

I can duplicate the problem with out-of-camera JPGs.

Your post only proves that pushing an photo beyond its limits will
result in problems, we all know that.

The problem does not have to do with post processing.

The problem is not about an incorrect exposure.

The problem is not about doing anything 'wrong'.

The problem is walking up to a DARK scene with BRIGHT 'point'
highlights, and photographing the scene as you see it.

The effects will show up on out of the camera JPGs - so you can't
blame PP as the cause.

I can take the exact same scene and photograph it with at least 4
other cameras that DO NOT do this, these cameras are not designed for
high ISO use, yet the D700/D3 is.

rbmphoto Senior Member • Posts: 1,837
AdamJ...

Do you even have a D700? Looking at your previous posts, it appears that you do not have one. Everyone is entitled to an opinion, but your FF Nikon posts seem based on nothing but bile...

Oh, how's the Canon/Sony forum these days?

cheers

vFunct Regular Member • Posts: 445
I've always expected banding with my D3

Had my D3 since last year. But, I'm an ASIC designer, and have actually designed consumer CMOS image sensor chips in the past.

Basically banding happens on the D3 because there are 12 read channels, to increase speed. Each of these channels are going to be slightly different, and so you're going to have variations in the lines that each channel reads. This is going to show up through small differences between each channel.

It's basically another source of noise.

The best way to remove this is to only have 1 channel of read out. Another way to cover it is to have only 2 channels of read out, since the bayer pattern on the sensor pairs each line together anyways to effectively cover up the individual line noise. But, once you start to have higher number of parallel channels, you'll see the noise differences between each a/d sensor, hence the banding.

This is always going to show up among highly parrallel A/D sensors that have fast frame rates. Underexposed images are where you're going to see it the most, since the noise differences are going to be exxagerated there, which will make it clearly visible when you recover the exposure in post-process.

The best way thing to remove banding is to MAKE SURE YOU EXPOSE YOUR IMAGE PROPERLY IN THE FIRST PLACE.

OP photoforfun Veteran Member • Posts: 6,084
extreme lightning example @ iso5600...

Underneath picture is an extreme bad/difficult lightning sample. A nearly dark billiard hall with 3 x 100 watt above the players head...D3 with 80-400VR @ F5.6 handhold 1/100sec. The image is unprocessed, only resized for the web.High iso NR was set to "low", and the in-camera sharpening was set to 5, which is normally too high for such a high iso.

a big crop of the area of the highlights. No banding whatsoever...

In this 100% crop there is noise in the dark areas because high iso NR was set to "low", and the in-camera sharpening was set to 5, which is normally too high for such a high iso.

While the center of the image was quiet good for iso5600(66% crop)...

-- hide signature --

Kindest regards,
Stany
I prefer one really good picture in a day over 10 bad ones in a second...

http://www.fotografie.fr/

OP photoforfun Veteran Member • Posts: 6,084
Re: I think you have a point

I guess that in this case, taking the RAW version of your correctly exposed picture and then reducing the 2 stops in post would have given the desired result without the banding.Thanks for the test. Gabtiele Regards Gabriele California, CA

Hi Gabriele. Nice to see you on this forum. Thanks for your reaction.

You are 100% correct, if I reduce the exposure on the correct exposed sample I get a darker image without any banding or noise, despite iso6400....

-- hide signature --

Kindest regards,
Stany
I prefer one really good picture in a day over 10 bad ones in a second...

http://www.fotografie.fr/

AdamJRed Contributing Member • Posts: 545
Re: AdamJ...

Yes, I own a D700.

I also have owned a Sony a700, Sony a200, Pentax K20D, Pentax K10D, Nikon D300, Canon 40D, Canon 5D, Canon 30D, Rebel XTi. I've had an assortment of lenses to go with them all.

The only post with 'bile' seems to be yours. My D700, less then a week old, and I've noticed a problem. A problem that doesn't occur on other cameras. I simply came here to share that I also have the problem and I get a bit frustrated with others brush it off.

I don't want anyone to think I am in the least bit unhappy with the D700. This is by far the most amazing camera I have ever used, period.

I'm not a bitter person, and I'm not bitter over this issue. I may be a tad frustrated when others insist there is no problem, while they might not see it or be able to reproduce it does not prove the problem is not there. I am glad others don't see the problem, or they see it but it doesn't bother them, and I'm not exactly sure how much it bothers me either, BUT there is still a problem there that does not occur with other cameras.

I don't post here much because of the cynicism and attacks I see so often, but being this is probably one of the most widely viewed forums I thought it was important to participate.

rbmphoto wrote:

Do you even have a D700? Looking at your previous posts, it appears
that you do not have one. Everyone is entitled to an opinion, but
your FF Nikon posts seem based on nothing but bile...

Oh, how's the Canon/Sony forum these days?

cheers

 AdamJRed's gear list:AdamJRed's gear list
Fujifilm X-T1 Fujifilm XF 18-55mm F2.8-4 R LM OIS Lytro Light Field 8GB Fujifilm 16-55mm F2.8R LM WR
ImTheDude Junior Member • Posts: 28
Re: AdamJ...

so what your trying to say is that if you're a really bad photographer, the D700/D3 takes bad pictures?

xrdbear Veteran Member • Posts: 3,961
Re: AdamJ...

AdamJRed wrote:

The only post with 'bile' seems to be yours. My D700, less then a
week old, and I've noticed a problem. A problem that doesn't occur
on other cameras. I simply came here to share that I also have the
problem and I get a bit frustrated with others brush it off.

I'm sorry if this also seems to be a brush off but I investigated this effect on my D3 last December when the first person reported it. In order to make it visible at all at iso6400 I had to take a shot in a room lit by a single 5W bare Luxeon LED shooting directly at it from close range. These things damage your eyes if you stare at them. Even then the effect was only just visible. It has never turned up in any real shot I have taken with bright light sources at high iso. IMO the problem, if one at all, is hugely overstated and doesn't seem to have affected the D3 becoming a best selling high iso workhorse.
--
Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/
Pbase gallery Pictures from Isle of Skye
http://www.pbase.com/xrdbear

brentj Contributing Member • Posts: 713
What I find interesting...

Is that the D3 has the exact some problem, but it was not a major issue until now. Why?

Probably because most poeple who purchased a D3 know how to use a camera, more often then not correctly expose their photos, and don't try to kill their photos in post.

The D700 comes out and you get a bunch of newbies who can't figure out how to get the exposure right in camera. As a result, they find "problems" that a better shooter would never see.

I see all this pictures of banding in night cityscapes at ISO6400 and my first thought is... "Who shoots a night cityscape at ISO6400?. Haven't these people ever heard of tripods?"

I'm not being a gear snob. I'm not a D3 owner. I have a D700. But I see some questions on this board relating to the D700 and I think to myself... This person really would probably be getting better photos with a D60.

xrdbear Veteran Member • Posts: 3,961
Re: I've always expected banding with my D3

Just to put the record straight. The sort of banding they are talking about is not what you are talking about. This banding is along a line parallel to the longest axis of the sensor emanating from a point in the image where it is hugely overexposed by a point light source. As I say in a post above I only ever managed to recreate it in deliberately extreme circumstances.
--
Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/
Pbase gallery Pictures from Isle of Skye
http://www.pbase.com/xrdbear

xrdbear Veteran Member • Posts: 3,961
Re: What I find interesting...

brentj wrote:

I see all this pictures of banding in night cityscapes at ISO6400 and
my first thought is... "Who shoots a night cityscape at ISO6400?.
Haven't these people ever heard of tripods?"

That's not really fair because it potentially turns up when doing candid street photography at night, good iso6400 territory. Nevertheless I find the effect minimal in the extreme.

-- hide signature --

Brian
Fine Art Print sales of the Isle of Skye at:
http://www.eyeofskye.co.uk/
Pbase gallery Pictures from Isle of Skye
http://www.pbase.com/xrdbear

J0seph Regular Member • Posts: 166
Excellent Obversations

Both very good and accurate points.

xrdbear wrote:

That's not really fair because it potentially turns up when doing
candid street photography at night, good iso6400 territory.
Nevertheless I find the effect minimal in the extreme.

-- hide signature --
CamasJC
CamasJC Senior Member • Posts: 1,382
Thanks Stany

Most persuasive argument = photos + settings
--
John
http://www.JChristopherGalleries.com

CamasJC
CamasJC Senior Member • Posts: 1,382
Thanks Vfunct

I have an EE background, too, so I appreciate your explanation... btw... should all this give us some clues of what we may expect from a 24MP sensor? (Everything becomes more citical, especially exposure at high gains?)

(?)
--
John
http://www.JChristopherGalleries.com

Keyboard shortcuts:
FForum MMy threads