Photo Resolution

emily_brozyna · September 27, 2016

Hello, I am trying to figure out what's going wrong with my headshots. I am shooting on a Nikon3200 with a Nikon AF-S FX NIKKOR 50mm f/1.8G lens. I am saving as .nef and exporting as jpg in full dimensions 4000x6000 and 300dpi. Is it my camera? Is it my lens? Is it something I'm failing at technically? Is it how I am handling the file? Thank you for any wisdom you can share! -E

<a href="https://birchtreefoto.smugmug.com/Headshot/Dahianna-2/n-d4G6Q4/i-SgVmDXb/A">Link to Photo</a>

DavidTriplett · September 27, 2016

Emily, I can think of a couple of possibilities:

1. If you are shooting at f/1.8, your depth of field will be razor thin. This can be OK for portraits, so long as you choose the correct focus point. In your linked image, that focus point should be the model's eyes. However, the actual focus point appears to be elsewhere. (Focus appears to be better at the left end of the model's jacket's slash pocket.)

2. On the D3200, the 50mm FX lens behaves like a 75mm telephoto. That, combined with the very large aperture, will further contribute to a shallow DOF.

3. You may need to do an AF lens tuning procedure. You can find a procedure for this on your D3200 at this <a href="http://www.brianparkes.com/nikon-d3200-lens-calibration/">link</a>.

Your D3200 is capable of taking outstanding photos, so long as all the settings are correct, and dependent upon the photographer's own ability. This is likely a small technical issue. Use point focus, and place it on the model's nearest eye, and then experiment with varying apertures to obtain the DOF you want. I like aperture priority for this ("A" mode). Good luck!

emily_brozyna · September 27, 2016

This is great! Thank you so much for your time and helpful wisdoms! I am amazed at how you knew something so specific about my lens and how it behaves on my specific camera. What would you recommend I study to understand them so well ?

How could I get a high res, super sharp photo with a very large aperture? Better equipment?

William Michael · September 27, 2016

The issues causing the soft image are most likely the Photographer’s technique and certainly can be corrected.

The sample image appears to be pulled at: F/1.8 @ 1/125s @ ISO100; EC = +0.7; Aperture Priority; Pattern Metering; No Flash. Tech: D3200 using a 50mm F/1.8 lens.

The Subject’s Face is soft. Most likely this is a result of not one, but rather a combination of any of the following causes:

> Inappropriate selection of APERTURE resulting in an Inadequate Depth of Field. Rationale: if that image is NOT cropped in the Vertical, then the DoF for that shot is about 80mm (3 inches). If we note that the depth of a Nose is about 25mm (1 inch), then it is easy to reckon that even if the front of the nose is in sharp focus then the eyes likely will not be in sharp focus, even if the face is full front-on. Obviously this is more critical if the face moves to any degree of Profile.

> Incorrect Focus Acquisition. Rationale – the Plane of Sharp Focus is at or near the line of sewing on the Blouse at the mid-line of the Subject’s Bust . It is likely that you selected or the camera selected AF acquistion using the Middle AF point. The Subject’s Face is situated behind this Plane of Sharp Focus.

> Inappropriate Shutter Speed. Rationale Part (a) - 1/125s is not a safe Shutter Speed to hand hold as there can be Camera Movement which results in blur. Noted that there is NO Camera Movement evident in the sample image but there is evidence that the camera is being hand held (i.e. a Tripod was not used) Rationale Part (b) – 1/125s is not a safe Shutter Speed for AVAILABLE LIGHT Portraiture as there can be Subject Movement which will result in blur. There is evidence of Subject Movement Blur at the line of the Upper Chest, above the Breast Bone and toward the R. Clavicle. This is evidenced by the Motion Blur on the necklace which appears to be moving from Camera Right to Camera Left. Usually this motion blur is associated with the Subject breathing, but considering the Shutter Speed was at 1/125s and that there is no blur evident at the line of the Subject’s Bust, then I suspect that Subject ‘struck the pose’ and and stiffened her upper body and the camera caught that very slight movement as she was so doing.

> Not monitoring of the Camera’s automatic function. Rationale – When selecting Aperture Priority Mode one must be aware that the camera will CHANGE the Shutter Speed depending upon the TTL meter reading. Noted that Exposure compensation was used to address the TTL meter reading for EXPOSURE, but when using Aperture Priority Mode the Photographer must also be aware of the Shutter Speed, such that is appropriate for the shot.

> Lens quality Rationale – the optical design of most of the 50/1.8 lenses is such that whilst the centre is reasonably sharp when the lens is used wide open, there is a fall off in optical quality away from the centre. If that image is not cropped in the vertical, then, even though the lens is being used on an APS-C sensor camera, the Subject’s Head is quite a distance from the centre of the lens, note that these lenses usually become quite sharp across most of the Image Circle at around F/2.2~2.5

WW

William Michael · September 27, 2016

How could I get a high res, super sharp photo with a very large aperture? Better equipment?

</blockquote>

The gear that you have is quite suitable. Your emphasis should be on your general technique; your understanding of Depth of Field; understanding the necessity of appropriate Shutter Speeds.

> Arguably a better set of Exposure Parameters for that sample image would have been: F/3.5~F4 @ 1/250s @ ISO800~ISO1000. That would have provided a DoF of about 160mm (6 inches) i.e. appropriate for the depth of the Face and also a safer Shutter Speed to arrest any Subject and/or Camera Movement.

> AF Acquisition using an outer AF Point or use the centre AF point and Focus and Recompose with a slight camera tilt, I would have used the centre AF point and employed F&R.

> Shooting around F/2.5 would have been good too, when the Face is in a “front on” position. Certainly at around F/2.5 or a smaller Aperture, that particular lens will render better Image Quality than when used at F/1.8.

> Use a tripod and remote release when it is suitable so to do.

> Practice your hand holding and breathing techniques when using slower shutter speeds.

> Ensure that the Subject is perfectly still and has indeed 'struck the pose' before you release the Shutter.

> Consider employing Flash as Fill

> Consider developing better Post Production skills, especially Post Production Sharpening

WW

DavidTriplett · September 27, 2016

+5 to William W's comments. As usual, he covered the issues in far greater detail than I could ever do.

William Michael · September 27, 2016

"As usual, he covered the issues in far greater detail than I could ever do."

</blockquote>

Thanks for the mention but . . .

Your point #3 regarding AF Tuning and the link supplied is of critical value to the OP and she should look at that.

WW

Matthew Currie · September 28, 2016

I think it's about covered here, but I will just add that, alas, the D3200 does not allow for AF tuning.

edit to add: I did not see the link referred to earlier. I should say, instead, that it does not allow for simple AF tuning without actual camera modification. Needless to say, if the problem is only with one specific lens, this will not be the best way to go. I'll leave the rest of my original post, even though it duplicates, in some ways, what is in the link....

If you think the AF itself might be off, one way to check is to lay a yardstick or other well marked measuring object on the floor. Using widest aperture, and flash if possible (or at least a high shutter speed and tripod if available), aim at a specific mark from a fairly low angle. Use single point, single servo AF. Needless to say, the focus should be sharpest at that spot, and if not, error is pretty easy to see. This will also give you a good demonstration of just how much depth of field you get, and how it is disproportionate, front to back. Use the specific lens you are wondering about, and try different distances. Closer will give you shallower depth of field, and be a more severe test, but also introduces more possibility of user error if you're hand holding.

You can also compare the AF between viewfinder and Live View. The AF systems are different, and LV will usually give you a very accurate AF. Do this in good light, though. In low light it may be more forgiving, and (undocumented feature) it reverts to release priority if AF cannot be found in about a second, so you can, in fact, shoot with the lens cap on in LV. I found (with considerable flak from a few people here who contended that it is not possible) that my D3200, which fortunately does not have a tuning issue, is more accurate in viewfinder view than LV in poor light.

By the way, another thing the D3200 and its ilk lack is a lock on the focus point. It is very easy to accidentally move this, so make sure you know where the chosen focus point is. Use the [OK] button to recenter it. If you're not paying attention, it's very easy to forget this, and to focus on the wrong thing.

DavidTriplett · September 28, 2016

Matthew, you might check the link included in the response. There IS an option for universal mechanical tuning of focus on the D3200. It is internal to the camera body, and is not lens-specific. However, if the body really is off for some reason, the procedure can make up at least some of the difference. Your other points are well taken and should be helpful to the OP.

Matthew Currie · September 28, 2016

Yes, sorry I missed the link on the first go-round, and edited a bit late for the back and forth here. I do realize it's possible to tune AF in the camera, but would hope that's not necessary.

bebu_lamar · September 28, 2016

One more thing! I think you should set the EC back to 0 rather than +0.7 as I think the image is too bright and there are quite a bit of washed out area. Also I think the color balance is off.

JDMvW · September 28, 2016

How could I get a high res, super sharp photo with a very large aperture?

</blockquote>

I think the only way to do this might be focus stacking.BUT

On the other hand, why use large aperture if you don't want the shallow DoF/"bokeh"?

Is large aperture some kind of "goal" in its own right? Why?

If the problem is not enough light,

<ol>

<li>add more light (flash, whatever), or</li>

<li>Up the ISO and get the lens down to something more in its "best range" (normally somewhere between wide-open and f/16).</li>

</ol>

Matthew Currie · September 28, 2016

Looking at the attached JPG file, it looks pretty clear that above comments on incorrect focus point and inadequate depth of field apply here. Whether it's the fault of camera AF or operator, you can see that some points on that image are good and sharp, but the DOF is very shallow.

jcuknz · October 1, 2016

I simply sharpened the image with Paint dot Net and the soft areas came up nicely.

Being able to do elementary editing is a useful 'technique' IMO. If she had focused on the eyes the jacket would have been soft whereas the soft face is quite pleasing.

emily_brozyna · October 10, 2016

I am overwhelmed by all of the time you all took to write me the thoughtful and helpful responses. Thank you!

Tim_Lookingbill · October 10, 2016

Why is that image over exposed using Aperture Priority (f/1.8) and Pattern metering which should not have chosen 1/125th shutter speed at ISO 100 with what appears to be overcast outdoor lighting. What would make that camera's meter choose those settings? I know my 6 year old 6MP DSLR doesn't do that. The subject's face is all washed out.

I adjusted the Exposure slider on the downloaded jpeg by -2.00 in ACR and the image really pops with clarity and presence. Yeah the face is slightly soft but the edges on the nose and other features are quite sharp enough to fix in post. The hair strands trailing off near the front side of the face are tack sharp but the clothes fabric detail around the woman's breasts are much sharper so that's quite a bit of smooth and gradual DOF for a lens aperture wide open at f/1.8.

Figure out what's making your camera meter the way it is with that type of scene.

William Michael · October 10, 2016

"Why is that image over exposed using Aperture Priority (f/1.8) and Pattern metering which should not have chosen 1/125th shutter speed at ISO 100 with what appears to be overcast outdoor lighting. What would make that camera's meter choose those settings?"

"Figure out what's making your camera meter the way it is with that type of scene."

</blockquote>

A simple comparison of the sample image to the EXIF details should not be used to draw an accurate and definitive indication of the functionality of Camera's TTL Metering System.

As it shown in-line here and the in the link that is provided, the Image has been Post Processed using Adobe Software and as such cannot be used to accurately define whether or not the image "as shot" was overexposed.

It is just as plausible that the OP’s camera’s TTL meter is functioning perfectly and the OP could have manufactured/manipulated those skin tones in the Post Production Process.

*

To interpret how the camera’s TTL system metered this particular scene, one must take into account the EXIF details including the 2/3 Stop Exposure Compensation which was set in camera and then work backwards from that beginning point.

See below here, for that procedure:

1. EXIF details of the image: Nikon D3200; F/1.8 @ 1/125s @ ISO100; + 0.7 EC; Aperture Priority Mode; Matrix Metering Mode

2. Equates to the meter's input reading: F/1.8 @ 1/200s @ ISO100 (adjust Shutter Speed for EC)

3. Equates to: F2.5 @ 1/100s @ ISO100 (adjust Shutter Speed & Aperture)

4. Approximately: F/2.8 @ 1/100s @ ISO100 (round to nearest Full Stop)

5. Equates to a scene EV = 10

So what we now can conclude is that the D3200 TTL Meter when in "Matrix Mode" reckoned that the 'correct exposure' for that scene was: F/2.5 @ 1/100s @ ISO100 and those exposure parameters equate (approximately) to a scene at EV = 10

Additionally, what we do know is, a scene at EV = 10 could have the appearance of soft diffused light with no distinct shadow area: exactly as the scene which is displayed in the sample image. Additionally, as we are viewing a JPEG file which has undergone Adobe Post Production, if the skin tones appear unnatural or odd, then it is equally possible that was a result of that post production process and not a result of any in camera metering malfunction.

Therefore, from the data at hand we certainly cannot conclude that the Camera's TTL meter is functioning inaccurately.

*

If the OP has concerns that her camera’s TTL meter is providing false readings, one way to make an evaluation of that functionality, is to make several A/B comparisons with another camera or Hand Held Light Meter, in scenes of different Light Levels.

WW

Tim_Lookingbill · October 10, 2016

As it shown in-line here and the in the link that is provided, the Image has been Post Processed using Adobe Software and as such cannot be used to accurately define whether or not the image "as shot" was overexposed.

</blockquote>

I was aware the image may or may not have been post processed. I've read the EXIF data. It's yours and my assumptions at this point if the OP intentionally made the image look over exposed as some type of effect or style. That lightening whether post processed or rendered straight from the camera is adding to the soft blurry look and a perception of a lack of resolution.

Quote from the OP...

I am saving as .nef and exporting as jpg in full dimensions 4000x6000 and 300dpi. Is it my camera? Is it my lens? Is it something I'm failing at technically? Is it how I am handling the file?

</blockquote>

No mention of post processing. What would handling the file do to sharpness and resolution? Not much info from the OP as to what "handling" the file means unless she means she edited it to look over exposed.

The OP needs to post the unedited straight from .NEF to jpeg rendered version.

Tim_Lookingbill · October 10, 2016

William, since you can calculate roughly how much light in EV's was actually in an overcast cloudy daylight scene as the OP's head shot, would you say the Auto exposure TTL metering exposure settings my camera chose to produce the unedited image below shot in both shaded AND cloudy (meaning quite dark) are the correct settings for that amount of light? I've embedded the EXIF data in the jpeg.<div>

00eAym-565807684.jpg.4b87e5edd3ed8fd8caa180ba1f8dd723.jpg

</div>

William Michael · October 11, 2016

Tim, I don't know if the TTL Meter on the OP's D3600 is working correctly or if it is not working correctly.

My response to your post was simply meaning that there is not enough information provided to concluded either way: I understood your post to be stating definitively that the OP's TTL meter was erroneous - I was simply stating that was not a proper conclusion.

And yes I agree with these two points:

1. "That lightening whether post processed or rendered straight from the camera is adding to the soft blurry look and a perception of a lack of resolution."

2. [for a more detailed response] "The OP needs to post the unedited straight from .NEF to jpeg rendered version."

WW

William Michael · October 11, 2016

Tim

> I did not "calculate roughly how much light in EV's was actually in an overcast cloudy daylight scene as the OP's head shot" - what I did was work backwards and calculate what the TTL Light Meter read as the EV of that scene and my result was the light meter calculated the scene as at (approximately) EV = 10.

Then I stated that:

"a scene at EV = 10 could have the appearance of soft diffused light with no distinct shadow area: exactly as the scene which is displayed in the sample image."

The point of the statement (and the point of the whole post) was, that the D3600's light meter could indeed be working properly.

WW

William Michael · October 11, 2016

We do not want this thread deviate off topic, but answering the quetion:

"would you say the Auto exposure TTL metering exposure settings my camera chose to produce the unedited image below shot in both shaded AND cloudy (meaning quite dark) are the correct settings for that amount of light? I've embedded the EXIF data in the jpeg. . .55mm, 1/30's, f/5.6, ISO 800. Is my camera's metering over or underexposing for cloudy daylight going by the numbers."

</blockquote>

I am not sure how this question about the snake photo is on topic and relevant to the Opening Post - however, as a quick and also a final response. . . Tim asked for an evaluation as to whether or not the TTL meter on his Pentax camera is under or over exposing and he provided one sample un-edited photo and the EXIF data.

*

For such an evaluation, I would take into account more than just the numbers, but I would start with the numbers: For the Snake Photo the TTL meter reckoned - F/5.6 @ 1/30 @ ISO 800 The going by the numbers - 1. F/5.6 @ 1/30 @ ISO 800 ≡ F/1.1 @ 1/800 @ ISO800 2. F/1.0 @ 1/800 @ ISO800 (rounding to closet full stop) 3. F/1.0 @ 1/800 @ ISO800 equates to a scene at EV = 7

*

Evaluation: 1. EV = 7 is quite dark, we’d expect that to not usually to be an outdoor scene, even in a shaded area on a cloudy day. 2. (Now I take into account other information) But considering that the Object is a snake then maybe that reptile liked a dark place to remain unobserved – so a scene at EV = 7 is plausible. 3. (Now I take into account more other information) The area in the RH half of the Photo seems to be concrete. If it is concrete, then it appears to be too dark in colour (a hint that the image is underexposed). The EXIF reveals that the camera was set to “pattern” metering. I am not intimately familiar with the Pentax metering systems and modes, but it is most likely that this is an evaluative mode which reads the whole scene and then uses an algorithm to compute the exposure: further it is possible that the algorithm read that concrete slab (if it is concrete) as “Photographic Grey” – whereas a slab of concrete is probably closer to “Off White” 4. Taking into account these ALL the three points above, I think that the image is probably under-exposed – at an educated guess by up to two stops. If the image is underexposed, then the Scene is possibly at about EV =9 which is more in concert with what we could expect in a dark shaded area, outside on a cloudy day.

*

Answers: Tim asked two questions:

a. Answering the question: "would you say the Auto exposure TTL metering exposure settings my camera . . . are the correct settings for that amount of light? I think that the Auto Exposure settings are incorrect for that scene and as a result the image is underexposed.

b. answering the question: "Is my camera's metering over or underexposing for cloudy daylight going by the numbers." Going by the numbers only and your description of the daylight scene, I think that the TTL Meter was underexposing for THAT PARTICULAR SCENE. * The question not asked:

c. answering the question: "Is the TTL Light Meter on Tim’s camera is working correctly?" I think that the TTL Meter is very likely working correctly, especially if that is a concrete slab in the shot – exactly as we would expect a TTL Meter to work, for that scene when the Metering Mode is set to Pattern or Evaluative Mode.

WW

Tim_Lookingbill · October 11, 2016

Thanks for the thorough explanation, William.

Actually to stay on topic this was what I was wanting you to confirm or agree on since it was not mentioned at the start of this thread...

And yes I agree with these two points: 1. "That lightening whether post processed or rendered straight from the camera is adding to the soft blurry look and a perception of a lack of resolution." 2. [for a more detailed response] "The OP needs to post the unedited straight from .NEF to jpeg rendered version."

</blockquote>

It's clear as others have stated that the camera's AF focus point is on the clothes rather than the face that is adding to the overall soft appearance.

But at least the OP will know what brightening an image does to an image that has nothing to do with a perceived lack of resolution.

BTW the snake image I'ld forgotten I had cropped out most of the top lit sidewalk which most likely affected the matrix metering to expose that scene more darker. I still don't know how to connect EV numbers to visual brightness levels that define standard scene lighting types such as Cloudy Day.

William Michael · October 11, 2016

"the OP will know what brightening an image does to an image that has nothing to do with a perceived lack of resolution."

</blockquote>

Yes, I think that is an important point to make.

Previous commentaries made the points about the MAKING of the image, such as DoF; AF function; Shutter Speed; etc. But is is important to note that mostly any post production technique such as: lightening; darkening; contrast, levels and colour manipulation etc, can have some affect on the perception of "softness" or "sharpness" in the final image. For example, especially for Portraiture, often by increasing the mid tone contrast we can make a "soft image" appear a little sharper.

So in answer to Emily's Opening Post Question - "Is it how I am handling the file?": That question might be referring to "Is it what I am doing in Post Processing?" and any Post Production or "handling of the file" certainly could be affecting the perception of or the actual "softness", especially on the face where it appears that area has been lightened considerably.

***

We are probably close to getting off topic again, and I encourage Emily to ask further questions about her image and her procedures if she needs to, but this might have relevance and be of assistance to her, also:

"I still don't know how to connect EV numbers to visual brightness levels that define standard scene lighting types such as Cloudy Day."

</blockquote>

Firstly note that relevant to the OP's Sample Image and my comment where I linked that to possibly being a Scene at about EV=10 it was not my assumption that the Scene is outdoors, nor that it was a Cloudy Day - I merely described the light as soft and diffused and with little shadows cast and which is could be typical of a scene at EV = 10.

There are many compilations which list “Typical EV of Scene”. An abridged version was (probably still is) contained in the box of most Film Stock. Kodak Professional Handbooks and The Ilford Manual(s) of Photography also have extensive lists. Some Photographers (including me) still sometimes use the “Typical EV of a Scene” as their exposure guideline to cross check with their meter reading(s) or sometimes I do not use a Light Meter at all <a href="/photodb/folder?folder_id=972502">(example here.) </a>

Part of my Diploma of Photography course work was mental computations and writing in the formulae sentences such as: “F/4 @ 1/100s @ ISO400 ≡ F/8 @ 1/400s @ ISO400 equates to Scene at EV = 13”

The piece of information that you might be missing is, in the above sentence, it is assumed or understood that the EV is at ISO/ASA100 and pedantically should be written as:

“F/4 @ 1/100s @ ISO400 ≡ F/8 @ 1/400s @ ISO400 equates to Scene at EV(100) = 13”

When EV numbers are (were) used to describe the Level of Light on a Scene or when EV numbers are linked to a set of Exposure Parameters it is (was) assumed that the ISO/ASA is set at 100.

(This same assumption of the ISO being set at ISO100, is often made for Guide Numbers of a Flash Unit. For one example here is a quote from Canon literature: “the Speedlite 580EX II is high powered with a Guide Number of 58 suitable for bounce flash and telephoto lens shooting”)

Here is a simple list of EV's (at ISO100) being related to Subjects situated in various Outdoors Scenes: EV 16 Subjects in front lit bright sun in sand or snow scene, very hard and distinct shadows EV 15 Subjects in front lit bright sun in typical outdoors, hard distinct shadows EV 14 Subjects in front lit sun with light cloud cover soft but noticeable shadows with edges EV 13 Subjects in sun with cloud cover, some shadow but not very distinct edges EV 12 Subjects in overcast scene, very little shadows are cast EV 11 Subjects in shaded areas, very soft and diffused light, very overcast EV 10 Subjects in deep shaded areas, very soft and diffused light for example in forest on cloudy day EV 09 Subjects in very deep shaded areas at times nearer to dawn or dusk EV 08 Subjects in very deep shaded areas at the bottom of dark forests EV 07 Rarely encountered in daylight scenes Subjects in deep shaded areas of forest EV 06 Rarely encountered in daylight scenes Subjects in extremely dark areas

Now that you know when I was linking EV numbers to visual brightness levels that define standard scene lighting types such as "Cloudy Day", I was assuming that the EV was assumed to be computed at ISO100.

Hence this this sentence might make much sense: "F/22 @ 1/400s @ ISO800 ≡ F/16 @ 1/200s @ ISO200 equates to Scene at EV = 15" Note that the above sentence articulates what is also known as: "The F/16 Rule"

WW

BTW - Re the snake image - yes a cropped image is not the most useful for an efficient forensic analysis. An extra area of concrete in the original frame most likely would have affected the camera’s TTL metering, especially when it is in “Pattern” Mode.

Sign In

Photo Resolution

Recommended Posts

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Create an account or sign in to comment

Create an account

Sign in