Nikon 10.5 & 12-24 vs Sigma 10-20 Photos EDITED

Discussion in 'General Discussion' started by Scott Sherman, Nov 19, 2005.

  1. I just received a Sigma 10-20mm lens and was wondering what the difference was between that lens and the Nikon 10.5 fisheye lens when defished. So I went into the back yard to take some comparison shots using a D2x set on basic jpg. Logically, the Sigma 10mm image should have been just a pinch wider than the Nikon 10.5 when defished in NC 4xx. The below images are basically out of the camera. These are not comparisons for color or image quality since I made some minor adjustments to shadows and and applied a bit of curve each a tad different. The only comparison here is for Angle of View. Each was saved for the web at the same size for comparison. The camera was locked in place on a tripod so, when I took the images I just very carefully replaced the lenses. The camera remained stationary and did not move between shots.

    I found it very interesting...

    One thing I did notice is that the nikon 10.5mm defished image while wider than the 10mm image with the Sigma lens did still show more distortion at the sides, (notice the leaves on the left and right sides and the bricks seem stretched on the right). It seems to me that the 10.5 defished image is either wider than 10.5 or the Sigma 10mm image is greater than 10mm in the final image. Or is the Sigma accurate and the 10.5 defished not accurate? This is what I think. I believe that the 10mm is supposed to be the equivilent of a 15mm in a 35mm camera. Can anyone do that comparison?

    EDIT: *** Please be sure to check out another post below in this thread in which I added a number of test shots with a link to PBase for a better comparison.

    This is a Nikon 10.5 fisheye
    original.

    This is the Nikon 10.5 de-fished in Nikon Capture
    original.

    This is the Sigma at 10mm
    original.

    And for fun, the Nikon 12mm
    original.
     

    Attached Files:

    Last edited by a moderator: Nov 20, 2005
  2. I conducted the same test using the same lenses and shots using NEF as an original file for each this time. These are in my living room with no flash, camera tripod mounted as before.

    One observation regarding Sigma compared to Nikon lenses. I had to adjust the temperature lower in the Sigma lens to acheive the same white balance as in the Nikon lens. Then for fun I used the temperature in the 12mm shot that I used in the Sigma and it came out warmer than the rest of the nikon shots. Also, it took a bit more sharpening to get the Sigma lens closer to the Nikon lenses. Nikon lenses are definitely sharper and more contrasty out of the camera all things being equal

    Nikon fisheye lens 3400 K
    original.

    Nikon fisheye defished in panotools in PSCS2 3400K
    original.

    Sigma 10-20mm @ 10mm3200K
    original.

    Nikon 12-24 @12mm 3150K
    original.
     

    Attached Files:

  3. I find this interesting.

    I like to see these comparisons. I am example kind of guy.

    Oh yeah - Nice backyard - And that TV - Very Cool!
     
  4. Me too Mark, glad you liked the pictures and the TV...

    Gadgets... MUST HAVE GADGETS...
     
  5. Is that REALLY a television? I though it was a large piece of modern art that one would see in the TATE MODERN.

    I have just been to the pictures to see Harry Potter and the Goblet of Fire and I am sure that our local cinema screen was not that big. Do you wear wide angle goggles to watch it.

    BW. Bob F.
     
  6. I have always subscribed to the theory that bigger is better and the one with the most toys at the end wins.

    I live in L.A. so I needed something to watch all those police pursuits on and with this tv it's like being there in the back seat of the cop car. Quite fun, also you haven't lived until you have played doom on a 72" high def tv in a dark room.
     
  7. The 10.5 de-fished has more coverage and the quality of the pictures is also very good specially in the picture made outside.
     
  8. Thanks for the examples Scott. That's a all I need, another case Of LLD, this thread has made the 2nd and 3rd cases today for buying the 10.5 fisheye plus throwing in the desire (uh, need) for either the Sigma 10-20 or the Nikon 12-24. I plan to buy the D200 in a few months but thought I had my lenses. Ohhhhhhh! I'm in deep.
     
  9. Thanks for this work, Scott. I'm shopping for a WA, thinking the Nikon 12-24, or maybe the Tokina (have read some good reviews of it compared to the Nikkor). The 10-20 wd be even better since I've got the 17-55 so there wouldn't be so much overlap. I see in photo #2 some CA on the left side picket but none from the Sigma. Any chance of posting a 100% crop from both pics for a further determination for me?? Thanks muchly.

    BTW, that's one honkin' HUGE television!!!! Be great to watch a slideshow of your fav shots on that! Eyecandy!
     
  10. Many Images With Exif Here

    Hi Sandi,
    This is for you (and anyone else interested). There was too much information to post here so I have provided a link at PBase below for images taken with my D2x and the following lenses and settings. Exif data is available for each image below the image for comparison.

    Sigma 10-20mm @ 10mm & 20mm @ f/4 through f/32 @iso 100 400 800

    Nikon 10.5mm fisheye @ most f stops and iso 100 400 800

    Nikon 12-24 @ 12mm & 24mm @ most f stops at iso 100 400 800

    you can view these images at full size for 100%

    All images were shot on a tripod @ approximately 18" from the subjects. The camera was stationary for all lenses and shots. No flash, images were taken indoors with the only source of light, sunshine coming through nearby windows. No incandescant light were on. WB was set to A in camera and all images were shot in NEF raw. The color space is RGB.

    All images opened in NC 4xx then batch filed into jpg images and then downloaded to PBase. No other adjustments of any kind were made to any of these images and all images were processed exactly the same. No sharpening etc. In other words they are pretty much exactly as they would appear out of the camera opened in NC.

    If you have any questions please let me know. I hope this helps answer some questions for those looking at wide angle alternatives.

    Here is the link to my PBase gallery for these images

    http://www.pbase.com/interactivephotos/comarison_wide_angles
     
    Last edited by a moderator: Nov 21, 2005
  11. Initially I intended to replace my Nikon 12-24mm and Nikon 10.5 fisheye with a Sigman 10-20mm to get an all in one wide angle lens. I tend not to like to change lenses in the field if possible or at least try to reduce changes, but after seeing this comparison, I will definately keep the 10.5 for those ultra wide shots, but I am going to sell the Nikon 12-24 since it makes no sense for me to have both. The 12-24 is rated as a superior lens over the 10-24, but I like the wider angle of the 10-20.

    Here is another interesting comparison of the lenses;

    http://www.nikonians.org/html/resources/nikon_articles/nikkor/af/wide_angles_shootout/
     
  12. Scott...I have a question. I have this lens too and I wanted to know why when the picture is defished, why doesn't it include everything that is in the fisheye picture? I noticed that when you took the the first shot of the living room, that the lens got the blinds in the picture but when you defished, it did not show up in the picture. Is there a reason for this?

    Thank you,
     
  13. Hi Mellisa,

    I'll give you my lay understanding of this. If you check the box in NC to "include areas where there is no image data" you will get a defished version of the image which includes all the information in the original image. When NC or Panotools "defish" an image they are basically distorting it from a circular or fisheye image to spherical or a normal image. This creates what is referred to as a "bowtie" effect because the top and bottom corners are higher and lower than the middle part of the image. This bowtie looking image is not usable for the most common formats like framing or website and most find it a bit distracting because the outside part of the image appears stretched. When NC and Panotools for example, distorts it to this effect, some of the information at the extreme edges are blurred and not usable for most purposes so it is cropped out. You can use some panoramic software to manipulate it further like Real Viz but it is still not as effective as a real panoramic image for extreme wide angle images. One way to give the fisheye image a panoramic look is to defish it and then crop it to an extreme ratio so the length of the image is much longer than the height of the image to make it long and skinny. This will effectively give you a panoramic image if the subject was shot at a large distance from the camera like in far away scenic shots.

    You can duplicate the defishing process somewhat in PS if you open your 10.5 fisheye lens picture and make a duplicate layer of it. Then open the edit>transform>distort tool. You will see some handles on the top layer which you can then pull sideways to bring the pincushion image into a more normal image but to make it rectangular, you will have to crop it and in that process you will loose some of the image at the sides of the image in a similar way as NC does in the automated process.

    Hope that helps.
     
  14. FishSauce

    FishSauce

    Aug 10, 2005
    Chicago, IL
    Scott,
    Very interresting comparison. I have Nikon 10.5mm and Nikon 12-24, usually I use Nikon 10.5 for a quick panoramic 360 degre by shooting 4 pictures ( handhold ) at 90 degree each then stich them together using Photo Vista 3.0

    Just wonder if Sigma 10-24 could do the same as nikon 10.5 ? Have you tried shooting 4 pictures at 90 degree and check out any overlapsing

    By the way the yard and the TV are very impressive :biggrin:
     
  15. Scott...Thanks a lot for the information.

    I prefer to use NC because it is so much quicker and I think it does a really nice job too.
     
  16. Mellisa,

    There are other tools in Photoshop CS2 that have an interesting effect on a fisheye image if you just want to play around. edit>transorm>warp and filter>distort>lens correction although I have not had much luck in that one for the 10.5. I think it is for less extreme pincushion or bowtie images. Also Panotools for small fee sells a plugin which works. I only mention this because I have begun to use PS to process my raw images and it is easier to do it all in PS if you are using the 10.5.

    NC is a quick fix and works quite well. I just like to try different things for fun.
     
  17. Thanks a lot Scott I will give those ideas a try too.
     
  18. Hey Ben,
    Haven't tried it with a 10.5. I mostley have done panoramics in single rows and for those you want the closest to aspherical as possible. I think you are talking about immersive interactive 360 like they use for real estate and cars ads and so forth. I played with that for awhile using Ipix but their license fees were to restrictive. Since then there have been many new advances in software that make what your talking about easier and better. Check out the Kaidan website, they have a lot on this type of photography and there are many other web sites dedicated to immersive 360 imaging. I think quicktime is also involved in that. I used to do something similar where you put an object like a coffee cup on a lazy suzanne and take pictures of it turning the platform in 10 degree increments and then put it on the web so people could virtually turn the cup around and turn it upside down and so forth. Excellent for ebay adds and so forth. So many gadgets... so little time.
     
Loading...
Similar Threads Forum Date
Fuji X-T1 vs Nikon D4 General Discussion Oct 3, 2014
Nikon D4 vs. iPhone 5s General Discussion Feb 4, 2014
Amount of activity on the forums - Canon POTN vs. Nikon Café: Why so different? General Discussion May 31, 2013
Nikon p510 vs. Olympus pen General Discussion Dec 30, 2012
The race is on... Nikon vs Sigma Service General Discussion Jan 11, 2011