Nikon D3 quality issue

DeanPhoto

DeanPhoto

Messages
332
Name
Dean Smith
Edit My Images
No
Hi all,

I'm after a bit of advice on where to start looking at image quality issues on a D3. I've just spoken to one of the editors that I work for and he has said that in the latest issue just printed there is a noticeable difference between the quality of images from my D3 to another photographer (who I believe also uses a D3), apparently this isn't the first time. This has been backed by the printers who have compared the images of mine to others.

This is obviously extremely worrying for me as I don't want to be supplying naff quality to anyone!

Now, before anyone replies I just want to make it clear it's an issue with the quality of the image and not the photography! I haven't seen the magazine as of yet so I can't actually see what he is talking about. From what I understand it's just overall detail, or lack of it, everything seems a little softer.

This isn't a brilliant example but is relevant as it's one of the images he used as an example. The following image is the cover which he has said is lacking detail in the black car at the rear and that the rear of the white car is out of focus. This was shot at 1/13, f/10.0, 24 mm @ ISO200. Considering I focussed on the red audi, and that it was only 6ft away from the camera max, I would expect the Porsche to be slightly softer as I'm nowehere near the hyperfocal distance, even though I'm shooting at f/10.0, right? I would also expect a lack of detail in the black car as the sun hasn't risen yet and therefore there's a lack of ambient and also, the light is rising from the rear.

5011859931_6c9cdec08a_b.jpg


Anyway, when I get home the Magazine should be with me and so I'll have a better grasp of what is being said.

There are two major things worth noting. He has said that a previous issue printed fine, the photos were taken with the same camera however, was photographed before my D3 was rain damaged and repaired by Nikon. Also, the photography in that issue was using a prime lens and not my new 24-70, which the most recent was shot with. In a studio the 24-70 was producing hazy results, I've heard there have been some troublesome lenses?

So, here is a list of things I presume I need to check.

Camera:
Image quality - I shoot in RAW 14bit
Colour space?


Processing:
RAW conversion
JPEG conversion

Lens:
Test lenses by using same setting sat on a tripod?


Is there anything else I can do, other than bang Nikon's door (NPS scheme member so not really an issue). I'm more than happy to supply RAW images if anyone would like to take a look?
 
Wow. Good luck getting to the bottom of this. I think you'll have an incredibly hard time finding out unless you took your current D3 along with another and a variety of lenses and used them both for exactly the same shots on the same shoot in controlled conditions. And then anally analysed the results to death like a digital Columbo.
 
Last edited:
Tell the picture ed to get real and stop pixel peeping, there's bugger all wrong with that image.
 
A full res file would confirm any issues, it's impossible to see anything wrong with the above image at this resolution.

To be fair though, if both the ed and the printing lab are under suspicion of a quality issue then something could be a miss.

However, what your explaining in the OP seems to be a misunderstanding of the focal plane.

:shrug:
 
Dangleman said:
Wow. Good luck getting to the bottom of this. I think you'll have an incredibly hard time finding out unless you took your current D3 along with another and a variety of lenses and used them both for exactly the same shots on the same shoot in controlled conditions. And then anally analysed the results to death like a digital Columbo.
Click to expand...


This is the problem!

I think I need to look at the RAW of the other photographers images to see whats up (although photos will be completely different).

Am I right in my thinking in regards to the DOF with the white car and the light issue on the black car, I am aren't I?
 
Tyke Tiler said:
A full res file would confirm any issues, it's impossible to see anything wrong with the above image at this resolution.

To be fair though, if both the ed and the printing lab are under suspicion of a quality issue then something could be a miss.

However, what your explaining in the OP seems to be a misunderstanding of the focal plane.

:shrug:
Click to expand...

Indeed, I'm more than happy to supply RAW's if anyone is willing to take a look? My problem is that I'm now used to the images so I can't really tell if anything is a miss, I don't go comparing them to my old images, I also replaced my old monitor at roughly the same time as the camera went to Nikon.

As for the focal plane comment, do you mean I have a misunderstanding or he does?
 
DeanPhoto said:
I think I need to look at the RAW of the other photographers images to see whats up (although photos will be completely different).
Click to expand...

Therefore a rather unhelpful comparison unless you both take the same pictures with the same lenses, under the same conditions and with the same settings.

DeanPhoto said:
Am I right in my thinking in regards to the DOF with the white car and the light issue on the black car, I am aren't I?
Click to expand...

Most likely, you've mentioned that your focus was on the red car in the foreground, so all cars on the background won't be as sharp, even at f/10 there will be some depth apparent.

The comment regarding the lack or loss of detail on the black car I'm struggling with though. Considering it's a black car, there is plenty of definition, even at the resolution above, the highlight spans from front to back, what more detail should there be with a gloss black object?
 
Last edited:
DeanPhoto said:
Indeed, I'm more than happy to supply RAW's if anyone is willing to take a look? My problem is that I'm now used to the images so I can't really tell if anything is a miss, I don't go comparing them to my old images, I also replaced my old monitor at roughly the same time as the camera went to Nikon.
Click to expand...

You can send me a raw file if you like, I shoot a D3, have been for over 2 years, if there's anything irregular I'm sure I'll spot it.

info@tomaswhitehouse.com

DeanPhoto said:
As for the focal plane comment, do you mean I have a misunderstanding or he does?
Click to expand...

The ed ;)
 
1/13 is pretty slow, could there have been a tiny little movement?
 
Tyke Tiler said:
Therefore a rather unhelpful comparison unless you both take the same pictures with the same lenses, under the same conditions and with the same settings.
Click to expand...

Yep. I guess the only thing I can really do is test my lenses out in a controlled manner.

Tyke Tiler said:
Most likely You've mentioned that your focus was on the red car in the foreground, so all cars on the background won't be as sharp, not even at f/10.
Click to expand...

I know it sounds ridiculous as DOF is probably one of the easiest things to understand in photography but just having that assurance has calmed me down slightly (was starting to think I was going mad).

Tyke Tiler said:
The comment regarding the lack or loss of detail on the black car I'm struggling with though. Considering it's a black car, there is plenty of definition, even at the resolution above, the highlight spans from front to back, what more detail should there be with a gloss black object?
Click to expand...

Well exactly, this was my point to him over the phone and I think he understands that. I think he was just expecting to see more details in the wheels etc but that side is effectively in a shadow.

I guess I need to take a good look at that mag tonight and see exactly what this problem is. You can imagine my disapointment and embarressment though at having to take a call like that!
 
woof woof said:
1/13 is pretty slow, could there have been a tiny little movement?
Click to expand...

It is possible of course, I took a fair few frames waiting for that cloud to come over so I'll take a look (tripod), but he's saying each photograph is lacking quality and is softer than he would expect.
 
You're not seeing his wife are you?

Sorry :bang: I hope you get it sorted or manage to shut them up.
 
What point on the Audi did you focus on?
Wouldn't this be a situation where using the hyperfocal distance would yield better results?
i.e. focus 1/3 into the scene, like the AM's plate and use F/16.
Send me the raw if you like:
d3Karen@hotmail.co.uk

Edit: sorry I skimmed your post and missed the bit about hfd. :geek:
 
Last edited:
I wonder if it's the lens as you've said that was the only thing that had changed from the previous issue to the current one.

The 24-70 does rear it's ugly head from time to time when it comes to sharpness issues.
 
woof woof said:
You're not seeing his wife are you?

Sorry :bang: I hope you get it sorted or manage to shut them up.
Click to expand...

I hope not, she's preggers!

They don't need shutting up, if it's a genuine issue then I'm thankful to them!

d3Karen said:
What point on the Audi did you focus on?
Wouldn't this be a situation where using the hyperfocal distance would yield better results?
i.e. focus 1/3 into the scene, like the AM's plate and use F/16.
Send me the raw if you like:
d3Karen@hotmail.co.uk
Click to expand...

Hi Karen,

I'll be honest; it was a little while ago and I can't remember off the top of my head. In hindsight it would have been better to shoot at the hyperfocal distance, as long as the Audi was perfectly in focus (not entirely sure it would work?), it was 5am in the morning when we were setting up.

It's not just this image though, as said above it appears to be all my images :(

I'll send you the RAW when I get back tonight.

Thanks for your help!
 
ryanyboy said:
That was my thinking too :)
Click to expand...

If the publisher only mentioned the rear of the white car and not the rear of the AM, then it could be a lens issue though. Maybe the lens elements are off-centre?

Edit: Google decentered lens elements
 
Last edited:
For what it's worth, the manual for the 24-70 says DOF at 24mm and f11 is:

1.03m to infinity (if lens is focussed at 2m) - see page 116
2ft 11in to 29ft 10in (if lens is focussed at 5ft) - see page 119

Is the black car further away than 29ft 10in?
 
Ozei said:
For what it's worth, the manual for the 24-70 says DOF at 24mm and f11 is:

1.03m to infinity (if lens is focussed at 2m) - see page 116
2ft 11in to 29ft 10in (if lens is focussed at 5ft) - see page 119

Is the black car further away than 29ft 10in?
Click to expand...

So I must be fairly close at f10 and 6ft I guess.

Black car is easily over 29ft away, struggled to see it!
 
http://www.dofmaster.com/dofjs.html

For 24mm at ƒ/10 the hyperfocal distance is 6.33 feet.

If there was an error so that the focus point was 4 feet away, then the far limit of acceptable focus would only be 10.7 feet. Similarly, focusing at 5 feet gives a far limit of 23.5 feet.
 
Right, after looking at the photos @100% in RAW it appears obvious to me that it's a lens issue. Not only are the edges really soft but the overall quality is shocking when using the 24-70 compared to my other lenses. Here's 2 examples:

The following are 100% crops from the left side of the frame, I've tried to show something that runs horizontal in each frame so you can see the fall off.

1. A photo from the same composition as above, I have brigthened her up a tad.
5018224676_c486d575e3_b.jpg


2. Photo using 80-200. Quality compared to the above is noticeable.
5017617745_a404a8fa17_o.jpg


So, am I right to think I might have a duff lens? Or do I need to conduct a more controlled test between lenses?
 
Ooh, forgot to mention the soft edges are indeed on each edge, not just the left side!
 
I'm no expert but the number plate and front bumper on the RCZ looks to be sharper then the back of the Porsche. Should the RCZ not be less focused than the Porsche as its further back for the DoF. Makes me think lens, but as I said no expert and it might just be my eyes.
 
PPalmer said:
I'm no expert but the number plate and front bumper on the RCZ looks to be sharper then the back of the Porsche. Should the RCZ not be less focused than the Porsche as its further back for the DoF. Makes me think lens, but as I said no expert and it might just be my eyes.
Click to expand...


Exactly what I thought and the first thing I noticed.
 
I normally try to avoid such things, but maybe try downloading a printing a test chart?
Then you'll be able to compare sharpness in different areas at the same distance from the sensor plane.
 
DeanPhoto said:
Right, after looking at the photos @100% in RAW it appears obvious to me that it's a lens issue. Not only are the edges really soft but the overall quality is shocking when using the 24-70 compared to my other lenses. Here's 2 examples:

The following are 100% crops from the left side of the frame, I've tried to show something that runs horizontal in each frame so you can see the fall off.

1. A photo from the same composition as above, I have brigthened her up a tad.
5018224676_c486d575e3_b.jpg


1. CA is not corrected for (easy to do in Lightroom). 2. Well it is out of focus. - Use f/16 and focus roughly to the middle / end of Audi, not the front. I guess the Audi is nice and and sharp. 3. Have you used remote and mirror lock-up?

2. Photo using 80-200. Quality compared to the above is noticeable.
5017617745_a404a8fa17_o.jpg


Well, yes - you have focused on right on it. xx-200mm are typically the best zooms available.

So, am I right to think I might have a duff lens? Or do I need to conduct a more controlled test between lenses?

Your test is inconclusive and doesn't really prove anything. You could try focusing on distant object and recomposing to the different parts of the frame at 2.8-5.6 to test for any centering defects.
Click to expand...
 
The CA does seem excessive on the Porsche, although if you read this -http://www.photozone.de/nikon_ff/456-nikkor_afs_2470_28_ff?start=1 - perhaps it's normal for the 24-70 at 24mm.

Also see last pic on this page http://mansurovs.com/nikon-24-70mm-f2-8-review

As for softness, I'd say the rear ends of the Porsche & Aston are on the limit (or even beyond) the DOF shown in Nikon's charts - i.e. 29ft 10in @ f11, 24mm when focussed at 5ft.

Here's a rough estimate of the distance:

  • Lens to front of Audi - 5ft
  • Length of Audi - 15ft
  • Rear of Audi to front of Porsche/Aston - 8ft
  • Effective Length of Porsche/Aston (when angled) - 8ft
  • Total (lens to rear of Porsche/Aston) - 37ft

You were on f10 so the DOF limit would have been slightly less than Nikon's 29ft 10in figure for f11 - let's say it was 29ft. This would put the rear of the Porsche 8ft outside the DOF limit.

PS. Re CA, Nikon's Capture NX2 might sort it out better than Photoshop.
 
Last edited:
DeanPhoto said:
Right, after looking at the photos @100% in RAW it appears obvious to me that it's a lens issue. Not only are the edges really soft but the overall quality is shocking when using the 24-70 compared to my other lenses. Here's 2 examples:

The following are 100% crops from the left side of the frame, I've tried to show something that runs horizontal in each frame so you can see the fall off.

1. A photo from the same composition as above, I have brigthened her up a tad.
5018224676_c486d575e3_b.jpg


2. Photo using 80-200. Quality compared to the above is noticeable.
5017617745_a404a8fa17_o.jpg


So, am I right to think I might have a duff lens? Or do I need to conduct a more controlled test between lenses?
Click to expand...

Pretty much what I said in post #16 :bang:
 
You must log in or register to reply here.
Back
Top