Sharpness

newb

First off, Im not sure that this is the right section for this. If its not, I apologize.

I recently got my new to me Nikkor 50mm f/1.8. I started shootin with it wide open, just playin around and checkin it out. I noticed that the images were pretty soft, which I understand is just part of shooting at wide apertures. I stopped down and took some more shots, and then noticed that the wide open shots seemed overly soft. I double checked myself by puttin the camera on a tripod, shooting with a wireless trigger and mirror lock up.

This image is at f/1.8


This one is at f/5.6


My question is, is the image at f/1.8 overly soft, or is it acceptable? Thanks!

Icebear

Could you have chosen a more difficult image to try to evaluate sharpness?? There are just too many variables for my ability.

pathfinder

Your f1.8 image looks pretty bad to me, but I am curious how you focused for your first image as the leafs behind the car seem in better focus than the car itself.

As expected, your lens is better at f8, but still not what I would consider acceptably sharp, although you did not state what portion of an image ( % crop ) these pictures represent. And the leafs still look better focused than the vertical red branches or the car.

I do agree with John, the image is not an ideal target to evaluate also. But might lead to your trying to shoot a better target.

Maybe Icebear has a link to an image of a target he prefers to use for lens testing....... I have used a sheet of newspaper once upon a time, or even a flat brick wall, but there are better Air Force lens resolution targets somewhere on the web. Marc recommends and uses this chart from Edmunds http://www.edmundoptics.com/onlinecatalog/displayproduct.cfm?productID=1665

Ziggy has made this suggestion for testing lenses and I recommend it highly.

newb

Fair enough. I can get somethin else today.

This lens is not AF, this an AIS copy. I meant to put it in my original post, but it slipped my mind. I dont have a focus chart to shoot, would the brick wall or stop sign still be acceptable targets?

These two shots were focused the exact same. All that changed was the aperture and shutter speed.

EDIT: As far as a Google search is concerned these are 100% crops. But, the info I followed may not have been the most accurite.

newb

Ok, since Ive got a stop sign a few houses down, I went ahead and took a shot of it. Same as before, nothing changed beside aperture and shutter speed. These are crops of the images straight off the camera.

@ f/1.8


@ f/5.6

pathfinder

You are using a D7000 or a D5000 with a non-AF lens?

Neither of these exposures look correct, and the first image shows more chromatic aberration than the second.

Are you shooting in camera jpgs or RAW files? If RAW, much of the chromatic aberration can be corrected in Raw processing.

The first image does not look stellar, but most 50 f1.8s are not stellar at f1.8 either.

I am moving this thread to Cameras as it is not really about shooting technique. Maybe you will get more responses there.

newb

Im using a D7000.

I know these exposures are a little dark, but you can still see that the image isnt very sharp. Just like the first ones I posted.

These are just JPG's, but I shoot in RAW/JPG all the time. I know it can be corrected, but I wasnt sure if that was the only problem.

I know that images get softer at wider apertures, but are my examples softer than most? If I need to shoot another subject I will. Im just tryin to get an idea of rather or not theres a problem with the lens.

Thanks for movin it, I wasnt sure where to put it to begin with.

Qarik

The lesn may not be the sharpest but I am also seeing A LOT of chromatic aberation that might be getting confused with being non sharp.

Stuart-M

You do realise that the depth of field will be different at different apertures?

angevin1

Icebear wrote: »

Could you have chosen a more difficult image to try to evaluate sharpness?? There are just too many variables for my ability.

John.....dood, don't ask!:D

pathfinder

Qarik wrote: »

The lesn may not be the sharpest but I am also seeing A LOT of chromatic aberation that might be getting confused with being non sharp.

I agree and the CA is usually easy to reduce substantially in ACR - I use the lens profiles for CA removal routinely.

I had some reservations about focus accuracy as well.

newb

Qarik wrote: »

The lesn may not be the sharpest but I am also seeing A LOT of chromatic aberation that might be getting confused with being non sharp.

Alrighty. Like I said, I wasnt sure that was my only problem. But, if thats what it is then my question will be, is that normal for the AIS 50mm f/1.8? Ive never heard it was bad, and my other AIS lenses sure arent that bad wide open.

pathfinder wrote: »

I agree and the CA is usually easy to reduce substantially in ACR - I use the lens profiles for CA removal routinely.

I had some reservations about focus accuracy as well.

Is ACR a Photo Shop plugin, or whole seperate program?

Focus is a lot of the problem in the first set. There is about a 20ft difference from the red twigs to the ivy behind the car. But, I didnt see any problems with the focus in the stop sign image. Am I missing somethin there?

angevin1

newb wrote: »

Focus is a lot of the problem in the first set. There is about a 20ft difference from the red twigs to the ivy behind the car. But, I didnt see any problems with the focus in the stop sign image. Am I missing somethin there?

It may seem a bit picky, but I think you chose to shoot in a way that does not aid anyone in really digging into the image for the focus as a pure thought. Did you check out the links provided earlier? If I think I have focus issues, I try and shoot something organic if I can and not with a bunch of lines and strange angles.

newb

Ya, I followed the links. Thats where I got the stop sign idea.

Thats not picky at all. Ive never had to test for focus accuracy or sharpness before. As this is the only lens Ive got thats ever given me trouble. The other 6 (that includes other MF lenses) function perfect. I have no idea what Im doin here. If anyone wants to tell me something to shoot, I will do it and post a full size image. I have the full sizes from the stop sign if that would help.

I dont completely understand the last sentence. A stop sign isnt organic, but its got straight lines and angles lol. What kind of organic things are you talkin about? Trees?

angevin1

newb wrote: »

I dont completely understand the last sentence. A stop sign isnt organic, but its got straight lines and angles lol. What kind of organic things are you talkin about? Trees?

Agreed! I ought to have been more clear. I prefer to shoot a face when I am testing simple focus issues, clarity and the like. For front/back focusing issues, I like to shoot a chart or even a measuring tape or ruler. But for focus, I prefer to shoot a persons face. I think that Chart in one of the links seems like a good way to go about it though~

davev

newb wrote: »

Ya, I followed the links. Thats where I got the stop sign idea.

Thats not picky at all. Ive never had to test for focus accuracy or sharpness before. As this is the only lens Ive got thats ever given me trouble. The other 6 (that includes other MF lenses) function perfect. I have no idea what Im doin here. If anyone wants to tell me something to shoot, I will do it and post a full size image. I have the full sizes from the stop sign if that would help.

I dont completely understand the last sentence. A stop sign isnt organic, but its got straight lines and angles lol. What kind of organic things are you talkin about? Trees?

How about this.
Go to the stop sign again, stay in the same place cause you're going to take lots of shots.
Make sure you have the ISO high enough to get a good clean shot, meaning the shutter speed stays up over 1/200th, no matter what aperture you're on.
Make sure you put the focusing shot on the area for each shot.
You already know that your lens is soft at f1.8, so there's really no need to revisit that mark.
Same with it being sharp at f5.6.
So what you need to is figure out where your lens starts to be sharp.
Use aperture priority, and move up the line from f1.8 and snap a picture, move up again, take another.

Most of the less expensive lenses are a bit soft wide open. The trick is to find out where the photos start to
get sharp, or at least to where you are comfortable with the sharpness you see in the shot.

I'll guess at f2.8 or f3.2 your shots will start looking pretty good.

Give it a try and see if the results are ones you can live with.

newb

Ahh, now I understand! Thank you. Well, the only person around now is my 2 year old son. He doesnt hold still very well lol. I can do the tape measure tho. Theres a video on youtube for focus testing that uses an Excel spreadsheet for testing. This one actually...

http://www.youtube.com/watch?v=pDN9muV-vcc

would that work in my case?

EDIT: Shot a focus chart that was brought up on my screen. Even with the screen tilted and my aperture wide open, the whole thing was in focus. Guess Ill hafta print it off later.

newb

davev wrote: »

How about this.
Go to the stop sign again, stay in the same place cause you're going to take lots of shots.
Make sure you have the ISO high enough to get a good clean shot, meaning the shutter speed stays up over 1/200th, no matter what aperture you're on.
Make sure you put the focusing shot on the area for each shot.
You already know that your lens is soft at f1.8, so there's really no need to revisit that mark.
Same with it being sharp at f5.6.
So what you need to is figure out where your lens starts to be sharp.
Use aperture priority, and move up the line from f1.8 and snap a picture, move up again, take another.

Most of the less expensive lenses are a bit soft wide open. The trick is to find out where the photos start to
get sharp, or at least to where you are comfortable with the sharpness you see in the shot.

I'll guess at f2.8 or f3.2 your shots will start looking pretty good.

Give it a try and see if the results are ones you can live with.

I can deffinitly live with the results from f/2.8. Typicaly worlds better.

But to get back to my first post. I already know that lenses get soft wide open. Thats not the issue. I was askin rather or not the softness of the this lens was typical.

It was said by others that it appears to be CA more than anything. So I asked then if this amount of CA is typical for this lens.

newb

In case you werent tired of hearin from me....

Shot a ruler, and this is what I got. Note that focus was directly on the number 7.

@ f/1.8


Crop


@ f/5.6


Crop (same size as above)


It seems to me that this proves that my problem in the other shots was CA. Am I wrong? Also, it appears that I have a slight bit of back focus. Is that correct?

ziggy53

newb wrote: »

... It seems to me that this proves that my problem in the other shots was CA. Am I wrong? Also, it appears that I have a slight bit of back focus. Is that correct?

I recommend that you use Live View and magnify to focus that lens.

If your AF lenses are functioning properly and if the AF lenses appear to be focused when they are used in manual mode then there is no way that a manual focus lens can be said to be back focusing.

Using Live View and magnify eliminates all user error resulting from trying to use the viewfinder and should increase your focus accuracy since you are using the sensor itself to check for focus.

To me, it looks as though the lens may have some internal haze and you may be seeing "dispersion". I suggest shining a single LED flashlight into the lens when it is off the camera. View through the lens both frontways and backways. If you see any haze then the lens may have a problem internally and may need professional cleaning or replacement.

newb

ziggy53 wrote: »

If your AF lenses are functioning properly and if the AF lenses appear to be focused when they are used in manual mode then there is no way that a manual focus lens can be said to be back focusing.

That makes sence. And yes, my AF lenses are usually spot on. Even AF lenses that friends let me try always seem to focus fine.

When I got the lens there was no haze, but Ill double check it. Along with tryin live view for focusing.

Thank you very much for the input!

NeilL

I see a funny kind of purple cast in the rooftiles behind the 1st "STOP" image. But no sign in any of the images of the typical CA along edges of high contrast where light dispersion is happening. CA also usually occurs more often at the edges of the frame rather than towards the centre. The rooftiles are not at the edge.

Neil

pathfinder

In the first car image shot at f1.8 there are purple magenta hints along the horizontal fence line in the forground, and in the first stop sign image there are green borders along the roof. It can be difficult to see CA at less than 200 - 300% views. When I process images for CA, I do it at 300-400% view in ACR or LR. I agree that the CA noted is not as bad as we see sometimes, but CA can vary a great deal and will vary within the same lens = whether the line in the image is horizontal, vertical or oblique. I wrote about CA correction in ACR here

I agree with the OP that when he checked the focus with the rulers, it seems pretty good to my eye. Indeed the images seem acceptable to me even at f1.8. With a little sharpening in Photoshop those images will be quite acceptable if the chromatic aberrations are corrected as well. One can see the faint green and magenta rimming of the tape near the borders of the images even on the web sized images.

Ziggy's comment about dispersion seems pertinent, but dispersion does not always improve with stopping down does it Ziggy? Sometimes it will even get much worse with a smaller aperture, not better.

ziggy53

pathfinder wrote: »

... Ziggy's comment about dispersion seems pertinent, but dispersion does not always improve with stopping down does it Ziggy? Sometimes it will even get much worse with a smaller aperture, not better.

Correct. A lens with a bad fungus infection will often get worse stopping down. Then again, an internal defect to the side can cause dispersion that can subside with stopping down (depending on the internal location of the problem).

newb

Good info! Thanks again!

I checked the lens out again, as you said. With a light shining through and looking through the lens both ways. There a some small, faint dust specs in the lens. But, theres no haze at all and unless fungus can be mistaken for dust particles (I dunno what lens fungus looks like), there isnt any.

As I play with the lens, I get shots that are crisp and CA free. Then Ill get one that looks like the others I posted.

Matthew Saville

If anything, the ruler test proves that your AIS lens is capable of resolving great detail, even wide open, on the D7000.

Keep in mind that you are at 16 megapixels on a CROP sensor. If you were to convert that to full-frame for example, you would be out-resolving even the mighty D3X!!!

Older lenses simply weren't made to put up with this kind of scrutiny.

There is also a difference between detail and contrast. This lens looks quite capable of resolving detail wide open, it is just that low-contrast haze that is poor, and CA can't be auto-corrected like on newer AF lenses. That's what you get for using a "kit lens" from the 80's on one of the finest resolution DSLR sensors available today. (Hey, I'm a serious camera geek and I LOVE AIS lenses! I have three of them including the 50 1.8 AIS and they serve me well. But I realize (and stay within) their limitations...

I can highly recommend simply stepping up to a better lens, if wide-open sharpness and overall performance is that important to you on the D7000.

I actually have used my 50mm 1.8 AIS in professional situations at f/2, and on the full-frame D700 it delivers GORGEOUS images. But on a full-frame 12 megapixel sensor, the pixels are both larger AND less tightly packed, and are therefore MUCH more forgiving compared to a 16 megapixel DX sensor. It's just physics...

Lastly, you may indeed have just gotten a poorer copy of the lens, just luck of the draw. So if you want maybe you could put this one up on Ebay and look for another. But my estimation is that if superior wide-open performance is what you're after, you should consider something newer like the Nikon 50 1.4 AFS G or the Sigma 50 1.4 EX...

=Matt=

NeilL

pathfinder wrote: »

In the first car image shot at f1.8 there are purple magenta hints along the horizontal fence line in the forground, and in the first stop sign image there are green borders along the roof. It can be difficult to see CA at less than 200 - 300% views. When I process images for CA, I do it at 300-400% view in ACR or LR. I agree that the CA noted is not as bad as we see sometimes, but CA can vary a great deal and will vary within the same lens = whether the line in the image is horizontal, vertical or oblique. I wrote about CA correction in ACR here

I agree with the OP that when he checked the focus with the rulers, it seems pretty good to my eye. Indeed the images seem acceptable to me even at f1.8. With a little sharpening in Photoshop those images will be quite acceptable if the chromatic aberrations are corrected as well. One can see the faint green and magenta rimming of the tape near the borders of the images even on the web sized images.

Ziggy's comment about dispersion seems pertinent, but dispersion does not always improve with stopping down does it Ziggy? Sometimes it will even get much worse with a smaller aperture, not better.

I had a second look and yes there is CA involved.

Neil

newb

My main concern was rather or not the results I was getting were typical for this lens. The poorest results I mean. If the worst of the worst from my lens follows suite with most other copies, then Im perfectly fine with that. There wouldnt be anything I could do, short of buyin a new lens, anyways lol. But, if mine seems worse than all others Id seriously consider finding another. I love my other 2 Nikkor AIS lenses, and Im hopin this one will fit right in with em.

After playing with it more today, it seems that problems start near or at infinity focus. Ive never heard anything about subject distance changing a lenses performance. Could someone enlighten me?

Thanks again for all the input here. Im learnin a lot.

Matthew Saville

newb wrote: »

...After playing with it more today, it seems that problems start near or at infinity focus. Ive never heard anything about subject distance changing a lenses performance. Could someone enlighten me?

Thanks again for all the input here. Im learnin a lot.

Absolutely! Mostly it's just macro lenses that you hear about being soft at infinity, but normal lenses can also be soft at macro focal lengths, and fast primes can ALSO be soft at infinity.

Here is a test of fast 50's in various scenarios; it is very eye opening to see the performance of various lenses in various conditions:

http://www.kenrockwell.com/nikon/comparisons/50mm-f12/#zmax

http://www.kenrockwell.com/nikon/58.htm

I would simply not use a 50 1.8 AIS at f/1.8 and focus at infinity, unless maybe I was photographing stars at night in which case I'd still strongly consider getting one of the newer, better lenses. Bottom line- don't shoot in harsh light wide open; stick to f/2.8 minimum to kill CA and general "haze"...

The 50 1.8 will shine at fast apertures in low light at distances up to 10-20 feet, or stopped down just a little bit it will give incredibly sharp images of any landscape etc...

=Matt=

newb

Cool, thanks for the links!