Page 2 of 6 FirstFirst 1234 ... LastLast
Results 11 to 20 of 52

Thread: DSLR Scanning - Signal / Noise

  1. #11

    Join Date
    Sep 2003
    Location
    South Carolina
    Posts
    5,505

    Re: DSLR Scanning - Signal / Noise

    Quote Originally Posted by sperdynamite View Post
    I like and use the Panasonic S1R even though the Sony has a better sensor. The reason being that Panasonic has pretty great tethering software, a variety of shooting ratios (1:1, 4:3, 3:2), and the pixel shifted scans do not require an extra piece of software to open them in LR or ACR. IIRC the Sony files need proprietary Sony software. Not a huge deal but an extra step. Nothing can beat the Sony sensor though. The Panasonic comes close but it's not as high res.
    There appear to be support in Light Room for opening Sony pixel shift shots.
    https://www.dpreview.com/news/572728...s-to-lightroom

    Wish it were available in Bridge as I don't use LR. But the Sony Image Edge software is pretty easy to learn, and a free download.

    Sandy
    For discussion and information about carbon transfer please visit the carbon group at groups.io
    [url]https://groups.io/g/carbon

  2. #12

    Re: DSLR Scanning - Signal / Noise

    Peter, Sandy, Sperdy--Thanks for the replies! I do have a Pentax K-5 (ASP-C, 16MP, with AA filter) so was looking at the used Pentax market. At around $300, you can get a K-5ii without AA filter; with $200-$300 increments getting you later models with more MP and then pixel shift until you reach the FF K-1 at about a grand.

    At any rate, as Peter implies, that big negative absolves a multitude of sins--hopefully I'll finally decide on an output strategy soon.

  3. #13

    Re: DSLR Scanning - Signal / Noise

    Quote Originally Posted by CreationBear View Post
    Peter, Sandy, Sperdy--Thanks for the replies! I do have a Pentax K-5 (ASP-C, 16MP, with AA filter) so was looking at the used Pentax market. At around $300, you can get a K-5ii without AA filter; with $200-$300 increments getting you later models with more MP and then pixel shift until you reach the FF K-1 at about a grand.

    At any rate, as Peter implies, that big negative absolves a multitude of sins--hopefully I'll finally decide on an output strategy soon.
    I started my DSLR scanning journey with the K-1. A very fine camera in some ways. AFAIK the lower KP has pixel shift as well. I don't think APS-C vs FF matters much when it comes to scanning. In fact a smaller sensor has some advantages. Though when it comes to pure resolution the larger ones have the advantage.

  4. #14

    Join Date
    May 2019
    Posts
    20

    Re: DSLR Scanning - Signal / Noise

    Quote Originally Posted by Peter De Smidt View Post
    I never found noise to be a problem, even with quite dense negatives. As sperdynamite points out, this should be even less of a problem with pixel-shift/averaging systems. Using HDR exposure blending increases the density range that can be captured, but my old D200 could already do about a stop more than an Epson scanner when scanning a Stouffer step wedge, and so it might only be called for in the most extreme cases.
    Actually, was interested in exploring pixel shift of BNW 4x5.... Unsure if can/should convert the DSLR to BNW via Bayer removal... On quick read has both advantages and disadvantages..... Though would be curious to see if can get A7RIV PIxel shift with 25 exposures and - and debayered....

  5. #15
    Peter De Smidt's Avatar
    Join Date
    Jan 2001
    Location
    Fond du Lac, WI, USA
    Posts
    8,954

    Re: DSLR Scanning - Signal / Noise

    Having the anti-alias filter removed doesn't seem worth it. It's expensive. It might affect your camera for other uses. And it's unclear how big of a difference it would make. Even my really old D200 could do better than an Epson with a cmos, Bayer, anti-alias and all that. Personally, I'd rather spend the money towards a better camera. It's easy to shift from what's good enough for an intended purpose to 'what's the best I could do?' Diminishing returns set in rather quickly.
    “You often feel tired, not because you've done too much, but because you've done too little of what sparks a light in you.”
    ― Alexander Den Heijer, Nothing You Don't Already Know

  6. #16

    Re: DSLR Scanning - Signal / Noise

    Quote Originally Posted by Peter De Smidt View Post
    And it's unclear how big of a difference it would make
    Ah, definitely--I'm quite fond of the K-5 as-is...great ergonomics and a bomber build. For my purposes, it might make sense to just optimize the hardware (copy stand, light source, etc.) and see what a good ink set could do with a print I could afford to frame. Very glad you gents are pushing the envelope, though!

  7. #17

    Join Date
    Jul 2016
    Posts
    4,566

    Re: DSLR Scanning - Signal / Noise

    Quote Originally Posted by Peter De Smidt View Post
    Even my really old D200 could do better than an Epson with a cmos, Bayer, anti-alias and all that.
    LOL , this is totally false and missleading.

    It's about understanding very basic math: An EPSON takes 140 Effective megapixels of a 4x5 sheet and a D200 will take 8MPix effective from every shot. You may need 30 shots stitched of the D200 to equal the EPSON 4x5 scan, after discounting the overlaping loss.

    Don't tell that you wre not aware. I you don't recall it, Pali K made a proficient benchmark of a V700 vs two Creos and a Scanmate 11000 drum, being the Epson result totally equal to the expensive scanners, while resolving less the EPSON usually takes all what LF film is able to resolve in practice.


    https://www.largeformatphotography.i...=1#post1479178


    The V700 performs extraordinarily well with LF sheets, top notch, you will have a lot of problems to equal the EPSON for LF with a DSLR, a lot of stitching. Even with the Cezane you have to stitch crops to equal the Epson for 4x5", as the Cezanne sensor has only 8000pix.


    Yes, a modern DSLR can take more than the EPSON from a 35mm frame in a single shot, for LF situation is totally different, a lot of stitching is required to equal the Epson, and anyway the Epson takes all what the sheet has, so all those complications to scan with the DSLR are futile.

  8. #18

    Join Date
    Jun 2015
    Location
    Glasgow
    Posts
    1,009

    Re: DSLR Scanning - Signal / Noise

    A low MTF optical system (Epson) with a lot of pixels and higher noise will always perform worse than a higher MTF optical system (Nikon and lens) with lower pixel counts. Theoretical 'resolution' has long been known to count for very little compared to latitude, noise, MTF performance. I'm not surprised that Peter was seeing better performance off the Nikon sensor, as above about 600ppi the Epson MTF performance drops like a stone.

  9. #19

    Join Date
    Jul 2016
    Posts
    4,566

    Re: DSLR Scanning - Signal / Noise

    Quote Originally Posted by interneg View Post
    A low MTF optical system (Epson) with a lot of pixels and higher noise will always perform worse than a higher MTF optical system (Nikon and lens) with lower pixel counts. Theoretical 'resolution' has long been known to count for very little compared to latitude, noise, MTF performance. I'm not surprised that Peter was seeing better performance off the Nikon sensor, as above about 600ppi the Epson MTF performance drops like a stone.
    More LOL.

    That test is faithful, well made, self explanatory and the most important: totally honest.

    https://www.largeformatphotography.i...=1#post1479178

    It's the indian, not the arrow, if you are not capable don't blame the V700.

    In fact Pali measured the real optic perfomance of all those scanners, but the practical test shows that (specially for LF) scanner resolving power is not useful beyond a certain point, as limiting factor usually is not the scanner.

    You are lost in MTF elucubrations without understanding the real impact in real photography... see the test, it's self explanatory: for MF and up there is little benefit (or none) from better scanners than the Epson.



    By the way just consider that the V700 resolves 2900x5.9 = 17,110 effective pixels in a single row, so those are exceptionally good lenses:top notch... The Linos inside your X1 is not better. Didn't you know that? Surprised?

  10. #20

    Join Date
    Jun 2015
    Location
    Glasgow
    Posts
    1,009

    Re: DSLR Scanning - Signal / Noise

    Anyone who spends their time quoting 1951 resolution charts while dismissing MTF and information capacity has clearly had no useful engagement with the peer reviewed literature of imaging science since about 1952. The 1951 chart is popular because it provides simplistic answers for camera fondling amateurs suffering from sub-Freudian pathologies with no apparent understanding of where real world imagemaking places demands on imaging systems.

    The D200 sensor has better MTF than the Epson scan array, lower noise, & is designed to be able to take sharpening well - to the point that the Nikon delivers sharpness performance more in line with its sensor resolution, while the Epson gets rapidly bogged down beyond about 7-8mp for a 4x5.

Similar Threads

  1. DSLR scanning vs Dedicated flatbed scanning.
    By robertraymer in forum Digital Processing
    Replies: 89
    Last Post: 22-Jun-2019, 03:20
  2. Scanning Issues - White Noise
    By Sean Chilibeck in forum Digital Processing
    Replies: 2
    Last Post: 9-Apr-2015, 04:34
  3. Scanning Negative Film- Noise/ Grain in blue skys
    By mark.s in forum Digital Processing
    Replies: 11
    Last Post: 6-Jan-2013, 21:14
  4. More scanning: Best practices for DSLR scanning
    By sully75 in forum Digital Hardware
    Replies: 6
    Last Post: 3-Jan-2012, 15:37
  5. scanning with dslr?
    By boris in forum Digital Hardware
    Replies: 46
    Last Post: 18-Mar-2009, 20:38

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •