Film Negative Conversion Software Reccomdentaions
Quote from Lorenzo Rossi on May 17, 2024, 1:49 pmVery interesting comparison, I wonder if you manage to remove any color cast from the three samples (purplish the K-1, greenish the SDQ H, a tad greenish the SD-1) the SDQ H may well turn out to be the "better" colors.
In the first K-1/SDQ H comparison I agree with your findings that after a careful processing the two images look pretty similar, with only a very slight edge for the Foveon Quattro one to my eyes.
The SD-1/SDQ H is actually very close in terms of details, I'm curious about your experience of the SD-1 having more details in macro shots, as usually my experience with the two was the opposite (after processing), i.e. the SDQ H at the end was showing more fine details because of the increased resolution, which more than compensate the reduced acutance due to the Quattro vs Merrill sensor structure.
I still have my SD-1 and would love to do a comparison vs my current scanning workhorse, the mighty Nikon Z8, but I don't have a Sigma macro lens unfortunately..
My observation of it having more detail was more of a feeling than a methodical conclusion. It just felt like the files had more detail, it may not in actuality. But my thought was that it made sense given how much lower resolution the red layer 4mp vs 15mp made a bigger difference that the larger spacial resolution.
Here is my attempt to color match the two Foveons. It's not perfect and I'm not sure it ever will be. I probably spent close to 45 minutes trying to get it this close. Interestingly simply copying and pasting the same adjustments to from one to the other leads to radically different images.
SDQH:
SDIM2863-2 (SDQH Colormatch) by Cory Maben, on Flickr
SD1:
SDIM3895-2 (SD1 Colormatch) by Cory Maben, on Flickr
The SD1 was far easier to edit, no matter what I did to the SDQH, I felt like I could not remove the color cast entirely, just move it around. The SD1 is slightly blue and the SDQH is slightly yellow, but accounting for that I can no longer see any real differences in color tonality. If you like I could send you the either the tif or x3f files to play around with yourself if you want.
I would highly recommend the 70mm macro for the Foveons, it's a very nice lens, although it did seem hard to find in SA mount when I was looking for it.
At the end of this though I'm still not sure I answered my own question. I was really hoping for a clear winner here and I didn't get that. My personal preference is actually the K-1. But I feel like the Foveons were the most accurate to what the film "should" look like. Between the Foveons, my preference was was for the SD1, it required less work and looked great.
I have been thinking about getting rid of my Foveon stuff. Lately, I feel inclined to move more into film and away from digital for a variety of reasons, so it would have been really convenient if the K-1 was the clear winner. It would have allowed me to greatly consolidate my gear and just use K Mount. Which makes me think I should compare the SD1 to a pixel shifted K-1 and see how that shakes out. But I stayed up way too late last night doing these comparisons, so it'll have to wait a couple days.
Thank you for the efforts, I actually slightly prefer the SDQ H colors, I've always found Quattro's colors "richer" than the Merrills and usually much more pleasant (for my tastes at least) SOOC as a starting point.
Your impression that Foveon renditions are more "film-like" than the K-1's ones don't surprise me much, at the end if you think of it the Foveon sensor three-layers structure is much more similar to film than the single layer of Bayer sensors.
If I were you, if you prefer the SD-1 and would like to trim down your gear, I'd probably sell the SDQ H, which is still retaining its value quite nicely (they usually sell for 1000+ bucks). I did the same pretty recently to fund my first venture in the medium format world with the Mamiya ZD and so far not regretting it 🙂 (maybe I should try some scanning also with it and an extension tube, as I don't have any macro lens for it)
Quote from Lorenzo Rossi on May 18, 2024, 7:33 amThank you for the efforts, I actually slightly prefer the SDQ H colors, I've always found Quattro's colors "richer" than the Merrills and usually much more pleasant (for my tastes at least) SOOC as a starting point.
Your impression that Foveon renditions are more "film-like" than the K-1's ones don't surprise me much, at the end if you think of it the Foveon sensor three-layers structure is much more similar to film than the single layer of Bayer sensors.
If I were you, if you prefer the SD-1 and would like to trim down your gear, I'd probably sell the SDQ H, which is still retaining its value quite nicely (they usually sell for 1000+ bucks). I did the same pretty recently to fund my first venture in the medium format world with the Mamiya ZD and so far not regretting it
(maybe I should try some scanning also with it and an extension tube, as I don't have any macro lens for it)
That would be very cool and I'd be interested to see the results from that setup. I have been considering selling off the SDQH and I may even make money on it since I got it for a good price. I also feel like I could let a few lenses go too and put that money towards something else.
It does make sense that film looks more correct because foveons are more like film in that way.
Found another standalone one
Filmomat Smart Convertor
https://www.filmomat.eu/smartconvert
I would also (if you are not already) recommend shooting RAW when camera scanning, most negative conversion software works on the assumption that it will be dealing with a RAW file and also offers more flexibility for colour correction, noise control, etc.
Cool one!
Just bought the cellphone version to play a bit with it 🙂
Quote from Gideon Liddiard Photography on May 18, 2024, 9:11 amFound another standalone one
Filmomat Smart Convertor
https://www.filmomat.eu/smartconvert
I would also (if you are not already) recommend shooting RAW when camera scanning, most negative conversion software works on the assumption that it will be dealing with a RAW file and also offers more flexibility for colour correction, noise control, etc.
I am shooting raw on both cameras the problem is with the Sigma cameras is that most raw software will not open them or will but displays a very distorted image. I assume the programs aren't really sure how to interpret the foveon files. So I must turn them into tifs before I can convert the negatives. But thanks for another recommendation!
Okay. Got the comparisons shot for the K-1 with pixel shift vs the SD1. These are edited to taste/match. Here are the results:
SD1:
SDIM3897 by Cory Maben, on Flickr
K-1:
MFP_0086 by Cory Maben, on Flickr
I did manage to figure out a solution for the Foveon's green cast. By sacrificing a bit of resolution and leaving the perforations visable, I can use the white light as a white balance point in Sigma Photo Pro, before exporting the tif files, which made working with the color significantly easier when converting the negative.
I will not address the two individual photos characteristics again as I think nothing has changed. But I will say that the K-1's colors better match my memory of how the day looked, but I acknowledge that the SD1 is probably more accurate to what the colors should be.
The point of this comparison was to compare the detail and sharpness between the two cameras when allowing the K-1 to use it's pixel shift function and I am pretty shocked by the results:
K1PS-SD1(2) by Cory Maben, on Flickr
The small text isn't legible on either photo, but the text on the SD1 is more clearly defined. The larger text in the window is actually readable on the SD1, while it is not in the K-1. Also look at the brick work beneath the window, showing much more clarity and definition on the SD1
K1PS-SD1(3) by Cory Maben, on Flickr
Here I was focusing on the fine detail of the leaves of the tree. They are individually clear and well defined again for the SD1
K1PS-SD1(4) by Cory Maben, on Flickr
This was closer to the edge of the photo and this was really impressive for the SD1. You can make out some texture on the trees on the hill, which has been almost flattened entirely on the K-1.
I was so shocked by these results, I went back and retook the shot. I was sure that I must have missed focus with the K-1 and the result was exactly the same.
However, this doesn't really give me any clarity with regard to what camera to use for film scanning. I think I may default to the K-1 as I find the images more pleasant and easier to work with and keep the SD1 around for when maximum fidelity is important.
