Lens Test With Four 50 mm Lenses

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

georgedingwall
Posts: 207
Joined: Tue Aug 01, 2006 11:15 am
Location: Invergordon, Scotland
Contact:

Lens Test With Four 50 mm Lenses

Post by georgedingwall »

Hi all,

I've recently picked up 3 enlarger lenses on ebay, and I thought I would do a comparrison test of them side by side with the same test subject. I also put one of my camera lens into the mix.

The details of the lenses used in the test are as follows,

Schneider Componon-S 50mm F2.8 @ F5.6
EL-Nikkor 50mm F2.8 @ F5.6
Rodenstock Rodagon 50mm F4 @ F5.6
Nikon AF-D 50mm F1.4 @ F5.6

I first tried to establish which was the best aperture to use for each lens. I couldn't really see much difference between F4, F5.6 and F8, so I decided to go with F5.6 for all four lenses.

I tried to keep all of the exposure and other physical settings the same, but the magnification of the AF-D lens was greater than the three enlarger lenses, and I had to remove about 25mm of the extension behind this lens to keep the magnification approximately the same as the other three. It is still a bit larger , but seems OK for the purpose at hand.

I used a Nikon D200 with a 1.4X Tele-Converter and a Pentax Bellows. All fours lenses were mounted in reverse. The camera was mounted on my home made focussing rail.

I lit the screw with four tungsten lamps which gave an exposure of 1.5 seconds @ F5.6. The subject was placed inside a diffuser.

The diameter of the screw was 4mm.

All four images were made from a stack of 30 frames with a 0.1mm adjustment between frames. Helicon Focus 4.03 was used for the stacking.

The first image shows the full frame results for all four lenses. The second image shows a crop at 100% of the same area of the image from each of the four lenses.

From this group of lenses I would say that the best results came from the Schneider Componon-S. The EL-Nikkor comes next followed by the Nikon AF-D camera lens. The Rodenstock brings up the rear.

I seem to have allowed a bit more reflection on the subject than I had intended, and I'm not sure if this makes it harder to get clean images for comparrison. However, the Schneider Componon-S seems the better quality result form this group test. I will try to get a less reflective subject and run this test again to see if it does make any difference.

Anyway, here are the two images. Comments are invited.

Image 1. Full Frame.



I look forward to hearing your comments.

Bye for now.
Last edited by georgedingwall on Wed Mar 05, 2008 1:32 am, edited 1 time in total.
George Dingwall

Invergordon, Scotland

http://www.georgedingwall.co.uk/

rjlittlefield
Site Admin
Posts: 24061
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

George,

I see it the same way you list, except that I'd give the nod to the Rodenstock over the AF-D. That's based mostly on the crispness of the fine detail right above the "AF-D" label, which seems clearly sharper for the Rodenstock. The Rodenstock image has an obvious reflection on the side of the thread that the other three images do not have. That makes me a bit nervous about effects of lighting on perceived sharpness, but the area I'm talking about seems to be lit pretty much the same in the two images.

In any case, it seems to me that all these lenses are quite similar. Even side-by-side, some study is needed to see the differences. If you presented them one at a time, I think it would be a rare person who could reliably distinguish even the best from the worst.

The situation with these four lenses is very different from what I found in my collection, where for example the Olympus 38mm f/2.8 and the EL Nikkor 50mm f/4 at f/5.6 had obviously different resolutions.

--Rik

Charles Krebs
Posts: 5865
Joined: Tue Aug 01, 2006 8:02 pm
Location: Issaquah, WA USA
Contact:

Post by Charles Krebs »

George... curious why you would add a 1.4X teleconverter to the mix. Generally the 1.4X's seem to be far better than the 2x, but if you are after the best possible quality you might be better without it. It would good to do a comparison without the 1.4X but extending a bit more to get the same magnification.

Epidic
Posts: 137
Joined: Fri Aug 04, 2006 10:06 pm
Location: Maine

Post by Epidic »

Where to start. This is not a lens test as a systemic test which includes the software and lighting. You changed the lighting among shots so it is hard to judge; certainly the Rodenstock image has different light. In the full-frame, the El Nikkor image has the most contrast, yet the crop does not. I would wonder what is going on. Be aware the lighting can change subject contrast which impacts the resolving power of the system.

What is the software doing? And is it doing it equally? That could be a large black box.

Also which enlarging lenses are you using? All these companies have been using these name designations forever, but the products have changed mostly to do with the lens coatings - a silver barrel El Nikkor is not the same as a black barrel El Nikkor.

Perhaps a flat subject would be a better target as you would not need to use the stacking software. Also be consistant with the lighting.

Anyway, an interesting test. All these lenses are close and so a slight variation in the process or conditions will bias the results. It is hard to make definative conclusions.
Will

georgedingwall
Posts: 207
Joined: Tue Aug 01, 2006 11:15 am
Location: Invergordon, Scotland
Contact:

Post by georgedingwall »

Hi Rik,
rjlittlefield wrote:George,

I see it the same way you list, except that I'd give the nod to the Rodenstock over the AF-D. That's based mostly on the crispness of the fine detail right above the "AF-D" label, which seems clearly sharper for the Rodenstock. The Rodenstock image has an obvious reflection on the side of the thread that the other three images do not have. That makes me a bit nervous about effects of lighting on perceived sharpness, but the area I'm talking about seems to be lit pretty much the same in the two images.

--Rik
The reason I put the Rodenstock last was the poor definition in the white material on the thread. Even allowing for any effect of the spurious highlight nearbye, it seemd to me that there was not as much detail in it.

I don't know where that reflection comes fron on the thread of the Rodenstock image, as I made no adjustment to the lighting setup between each frame sequence. I'm pretty sure I did not disturb the screw when changing lenses, and the exposure was the same for all images.

As I said in my original post, I think I need to get a less reflective test subject.

Bye for now.
George Dingwall

Invergordon, Scotland

http://www.georgedingwall.co.uk/

georgedingwall
Posts: 207
Joined: Tue Aug 01, 2006 11:15 am
Location: Invergordon, Scotland
Contact:

Post by georgedingwall »

Hi Charlie,
Charles Krebs wrote:George... curious why you would add a 1.4X teleconverter to the mix. Generally the 1.4X's seem to be far better than the 2x, but if you are after the best possible quality you might be better without it. It would good to do a comparison without the 1.4X but extending a bit more to get the same magnification.
I use the 1.4 X extender for two reasons.

First it acts as a dust barrier and prevents dust and other particles from getting inside the camera and onto to the sensor. I found that since I started using my old Pentax Bellows that I am get a lot more dust marks on my images which need to be cleaned during processing.

Secondly, I find that with the reveresed enlarger lenses fitted, I am very close to the subject and using the extender gives me a little more room between the lens and the subject.

I'm aware that the extra glass might have an effect on quality, and will take your advice and try a similar sequence without it to see if there is any noticable difference.

Bye for now.
George Dingwall

Invergordon, Scotland

http://www.georgedingwall.co.uk/

georgedingwall
Posts: 207
Joined: Tue Aug 01, 2006 11:15 am
Location: Invergordon, Scotland
Contact:

Post by georgedingwall »

Hi Will,
Where to start. This is not a lens test as a systemic test which includes the software and lighting. You changed the lighting among shots so it is hard to judge; certainly the Rodenstock image has different light. In the full-frame, the El Nikkor image has the most contrast, yet the crop does not. I would wonder what is going on. Be aware the lighting can change subject contrast which impacts the resolving power of the system.
Well I did try to keep the variables the same for each sequence of shots. I don't know where the reflection on the Rodestock image comes from. I did not change the lighting between shots. All exposures were the same. The actual sequence that the lenses were used was Schneider, EL-Nikkor, Rodenstock and then the Nikon AF-D. I'm just not sure how I could have created the highlight for the Rodenstock and then have it gone when I fitted the AF-D.

The lighting I used for these exposures was 4 x 100 watt tungsten photofloods that used to be fitted to an old copy stand. They were set at about 75% output and were locked in a fixed position.

What is the software doing? And is it doing it equally? That could be a large black box.
All the images in these sequences were shot in RAW format. In the Raw converter they all had the same brightness and contrast settings. The only individual tweeking that was done, was to balance the exposure of a few frames so that all frames were the same. This adjustment involved changes of less than a quarter of a stop. The images were saved as maximum quality jpeg.

Other than the stacking, I made no changes in Helicon Focus. The stacks were saved as tiff files.

When I opened the tiff images in Photoshop, all that was done was to assemble the composite images. No processing of any kind was done at this stage. The composite images were converted to sRGB colour space and saved using the Save for Web option in Photoshop. I used a compression level necessary to make the images less than 150 KB to keep within the image posting guidlines of the forum.

The full frame had a higher qulaity setting than the crop. I assume this was due to the unifrom grey background. It is possible that this has caused the difference you find between the full frame and crop images.
Also which enlarging lenses are you using? All these companies have been using these name designations forever, but the products have changed mostly to do with the lens coatings - a silver barrel El Nikkor is not the same as a black barrel El Nikkor.
I'm not sure I know enough about enlarging lenses to be that specific about exactly which model from the history of enlarging lenses these three might be.

Is there some specific information you would recommend I find that would be more helpful. I could post images of them if you think that might help.
Perhaps a flat subject would be a better target as you would not need to use the stacking software. Also be consistant with the lighting.
The reason I used the lenses in this way, is to see the results from the way I will use them in real life. It is unlikely that I will use the enlarger lenses to take single exposure images. I prefer to use my dedicated macro lenses for that purpose.

Thanks for your comments. Bye for now.
George Dingwall

Invergordon, Scotland

http://www.georgedingwall.co.uk/

DaveW
Posts: 1702
Joined: Fri Aug 04, 2006 4:29 am
Location: Nottingham, UK

Post by DaveW »

Yes it is rather curious, we seem to have two distinct areas here to assess sharpness. The one Rik uses and the white feature both George and I initially used.

Could Helicon Focus be reacting to different lenses differently? I ask as I have never used stacking software. I too would have placed the lenses in the order George did. However, I would not expect a lens with an f1.4 maximum aperture to usually perform as well as one that is only f2.8 given the same build quality, on the basis it is always easier to make a good little-un than a good big-un!

Could the flare shown on the Rodagon be due to poorer lens coating rather than a change in George's lighting? The main difference with digital lenses, as opposed to conventional 35mm lenses apart from their covering circle, is usually the coating on the back element of the lens to reduce reflections from the now highly reflective digital sensors compared to film. These reflect back and forth between the sensor and the rear element of the lens. George of course had the lenses reversed so it would be the front element in his case.

Pity DSLR camera's don't make provision to fit some type of clear filter between the lens and mirror box to stop dust getting into the works anyway. A friend of mine had a Fuji with a fixed zoom lens and still had sensor dust problems with it pumping in dust to the sensor. Prevention through a sealed mirror box behind the lens rear element would be better than cure?

I too would suggest not using stacking for the initial test but a sheet of newspaper with fine print to test the flatness of field of all these lenses. Obviously if the field of these lenses is not flat it will produce different results though as enlarger lenses take their image from a flat surface, the film, and project it onto another flat surface, the paper I would expect all to have a flat field.

DaveW

georgedingwall
Posts: 207
Joined: Tue Aug 01, 2006 11:15 am
Location: Invergordon, Scotland
Contact:

Post by georgedingwall »

Hi all,

I've did as Charlie suggested and reshot the Schneider sequence without the 1.4 X extender. I had to add 65mm of extension tubes to achieve approx the same magnification.

I couldn't find exactly the same area on the screw that I had previously used, so I picked an area with plenty of suface details.

The camera, exposure and post processing were all the same as in the previous shots. The only difference was that I placed the screw a bit further back from the front of the diffuser I used, and this seems to have reduced some of the reflections. I used exactly the same jpeg compression for both images.

It looks to me as if the image without the 1.4 X extender has a little more definition in the finer surface details.

Schneider Componon-S 50mm F2.8 @ F5.6
Last edited by georgedingwall on Mon Jul 26, 2010 6:32 am, edited 1 time in total.
George Dingwall

Invergordon, Scotland

http://www.georgedingwall.co.uk/

georgedingwall
Posts: 207
Joined: Tue Aug 01, 2006 11:15 am
Location: Invergordon, Scotland
Contact:

Post by georgedingwall »

Hi all,

As Will and Dave suggested, I have taken single frame images of a flat surface. In this case very small text from an advertisement in the back of a photography magazine. I can barely read this text, even with my glasses on.

I've kept the text as flat as possible by sticking it onto a microscope slide, mounting the slide on a precision hex head machine bolt standing on its head and squaring the slide as best I could so that it was perpendicular to the sensor plane.

The layout in the images below are the same as the original images.

top left - Schneider
top right - EL-Nikkor
bot left - Nikon AF-D
bot right - Rodenstock

I'm not sure how accurate the focus is on each of these images. Each lens seems to have its focal plane in a different place and each had to be refocussed when changing lenses.

I'm not sure I see much difference between these images. Maybe the Nikon AF-D is a little softer than the others, and the Rodenstock seems quite good in this series. When I view the full size images, I can't really separate the three enlarger lenses.

What do you think.

Here are the images.

Bye for now.
Last edited by georgedingwall on Mon Jul 26, 2010 6:32 am, edited 1 time in total.
George Dingwall

Invergordon, Scotland

http://www.georgedingwall.co.uk/

DaveW
Posts: 1702
Joined: Fri Aug 04, 2006 4:29 am
Location: Nottingham, UK

Post by DaveW »

On my computer screen in the 100% crop the Rodenstock is slightly the sharpest, but there is so little in it. In any case, though a standard camera lens like the Nikon f1.4 is usually one of the makers best corrected lenses, like all conventional camera lenses it will be computed for infinity focus, unlike "macro" lenses like the micro nikkors or enlarger lenses which are computed for close focus, so it has performed very well. It would be interesting to see how your micro nikkor performs in this test compared to the others because that will be computed for close focus rather than infinity?

When I first changed to digital I tested my 60mm micro nikkor with the newspaper test and the camera mounted on a copy stand. I ran through a couple of apertures and got a flat field in both at 1:1 using manual focus. What did impress me was when I backed off from 1:1 slightly to give the camera's autofocus room to work and tried it on autofocus the image was still perfectly focused close up. Of course newsprint is a contrasty target, something that autofocus needs.

DaveW

rjlittlefield
Site Admin
Posts: 24061
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

George,

About the bright reflection in the Rodenstock image, first series... It could be due to different positions of the entrance pupils from one lens to another. ("Entrance pupil" = apparent position of the aperture, as seen from the subject.) Even lenses with the same FL can have significantly different entrance pupil positions. I'd guess that the Rodenstock's was enough in front or in back of the others', to catch a direct reflection that the others did not.

Still, that same effect can easily change the appearance of a translucent subject, which I believe that chunk of white material is. If the Rodenstock's entrance pupil happens to be positioned to catch relatively more light coming through from the back, then contrast within the white stuff will be reduced, which will reduce the effective resolution.

About software effects...

I've tested single frame vs HF stacked frame using the same subject, same lens, same lighting. (See here for details.) In those tests, stacking slightly reduced the effective resolution at same f/stop. That's consistent with my understanding of how HF's algorithm works, because at each pixel position the stacked result will include small but non-zero contributions from slightly OOF images.

However, I would be surprised to see the software do anything noticeably different from one lens to another. I've never directly tested that, and on brief thought it seems like a hard test to do accurately. (Detailed thoughts omitted here. :D )

As you've noticed, it's often tricky to get critical focus at large aperture. With some lenses, it's even trickier to get best focus a stop or two below wide open, because the focus point shifts. The easy solution, these days, is to shoot a stack and pick the best single image. I'd be reluctant to interpret the single-frame series unless we're sure that all the pics are at best focus.

--Rik

georgedingwall
Posts: 207
Joined: Tue Aug 01, 2006 11:15 am
Location: Invergordon, Scotland
Contact:

Post by georgedingwall »

Hi Dave,
DaveW wrote:I would be interesting to see how your micro nikkor performs in this test compared to the others because that will be computed for close focus rather than infinity?

DaveW
I wondered that too, so I did a stack of the screw using that lens. I mounted it the right way round, but used the same amount of extension behind the lens. The magnification is a little less. I think this lens may be as good, if not a little better, than the other lenses used in this test.



Bye for now
Last edited by georgedingwall on Mon Jul 26, 2010 6:33 am, edited 1 time in total.
George Dingwall

Invergordon, Scotland

http://www.georgedingwall.co.uk/

georgedingwall
Posts: 207
Joined: Tue Aug 01, 2006 11:15 am
Location: Invergordon, Scotland
Contact:

Post by georgedingwall »

Hi Rik,

The more I get into this subject the more complicated things seem to get.

I'm get a certain amount of satisfaction from just trying a few things to see what results I get. These three enlarger lenses were comaratively cheap, (less than 100 UKP for all three), so I wasn't expecting sparkling results with them.

Saying that, the results aren't that bad for any of them considering the price.

I know that the software plays some part in reducing the quality present in the unstacked images, I always assumed it was my poor technique that caused it.

If you were put on the spot, what would you say is the best lens I could get for my sort of set up. I get about 8 to 1 magnification on the D200's 23.6mm sensor as a maximum if I use all of my accessories. Is there an ideal lens that would work best for this magnification, assuming money was not a limiting factor.

Thanks for the comments.
George Dingwall

Invergordon, Scotland

http://www.georgedingwall.co.uk/

DaveW
Posts: 1702
Joined: Fri Aug 04, 2006 4:29 am
Location: Nottingham, UK

Post by DaveW »

Would say the micro is the best image so far!


There is a list of true photomacrography lenses here and their magnification ranges:-

http://www.a1.nl/phomepag/markerink/microlen.htm

DaveW

Post Reply Previous topicNext topic