About PhotonsToPhotos and DXOmark charts using Fujifilm FinePix S5 Pro as an example

About graphics

About graphics

Many photographers, amateur photographers and even photo equipment reviewers refer to photonstophotos while mentioning the dynamic range of the camera Fujifilm FinePix S5 Pro.

The site photonstophotos The Photographic Dynamic Range measurement technique does not take into account the R-pixels of the Super CCD SR II sensor, which is located in the heart of the camera Fujifilm FinePix S5 Pro... Their software (specifically 'NefUtil') does not' understand 'the sensor structure and only considers S-pixels. Most likely when testing Photographic Dynamic Range sensor Fujifilm FinePix S5 Pro used D-Range 100%, which turns off half of the sensor subpixels (turns off R-pixels).

It is interesting: graphs for Fujifilm FinePix S5 Pro and classic 6 MP cameras based on conventional CCD sensors of that time on the site photonstophotos very similar. The only difference is that the Fujifilm FinePix S5 has a base ISO of 100. Excluding R-pixels, sensor capabilities photonstophotos become almost equivalent in dynamic range to 6 MP CCD sensors at Nikon D100, D40, D50, D70, D70s, Pentax * ist DL2, K100D and the like. This is indirect evidence that the sensor Fujifilm FinePix S5 Pro tested with R pixels disabled.


Link. increase... It is clearly seen how the graphs for Fujifilm FinePix S5 PRO и Nikon D100, D40, D50, D70, D70s, Pentax * ist DL2, K100D (based on 6 MP CCD sensors of that time) are similar to each other. The graph does not show a drop in the curve after ISO 800-1000, but it must be. And also all users from personal experience with Fujifilm FinePix S5 PRO know that its dynamic range at low ISO values ​​significantly exceeds all old 6 MP cameras.

If photonstophotos took into account R-pixels, then after ISO 1000 there should have been a sharp drop in the curve of the graph down due to their forced shutdown. R-pixels can only work in the ISO 100-1000 range, at higher ISOs they are automatically disabled (this is an indisputable fact). This is clear evidence of the fallacy of the hotographic Dynamic Range plot for Fujifilm FinePix S5 Pro on the website photonstophotos... In this regard, data from DxOMark much more objective (although there are also problems with the SNR 18% graph).

Dynamic range

Dynamic Range Fujifilm FinePix S5 PRO according to DXOmark... There is a drop in the curve after ISO 800-1000, so it should be (above these ISO values, a part of the camera sensor is forcibly turned off).

By the way, I unsubscribed (screenshot) owner photonstophotos, and most likely everything is exactly as indicated above.

DXOmark data are also not true in everything. Roughly speaking, Fujifilm FinePix S5 Pro и Fujifilm FinePix S3 use the same sensor, which is clearly seen in the following graph:

DXOmark data

increase... The data DXOmark about the dynamic range of cameras Fujifilm FinePix S5 Pro и Fujifilm FinePix S3 Pro are similar to the truth.

But here SNR chart 18% for Fujifilm FinePix S5 Pro built incorrectly. It does not take into account the fact that after ISO 1000 there should be a sharp drop of the graph curve downward due to the forced disabling of R-pixels. SNR 18% is directly related to Dynamic Range. R-pixels can only work in the ISO 100-1000 range, at higher ISOs they are automatically disabled. This same feature is well displayed for the SNR 18% plot for the camera. Fujifilm FinePix S3 Pro.


increase... Schedule for Fujifilm FinePix S3 Pro goes down after ISO 800/1000.


  • photonstophotos does not give an objective picture of what is happening and reference their dynamic range data Fujifilm FinePix S5 Pro you need to be very careful. And if to generalize more roughly - photonstophotos gives false sensor data Fujifilm FinePix S5 Prowhich are not worth referring to at all.
  • DXOmark gives an erroneous graph for an SNR of 18%, although the Dynamic Range data is more likely to be true.
  • Fujifilm FinePix S5 Pro Is a very specific camera, and it is not always easy to understand its operation. Actually, this is why it is interesting, and therefore continues to attract the attention of many users.
Comments on this post do not require registration. Anyone can leave a comment. A wide variety of photographic equipment can be found on AliExpress, The Amazon и B&H Photo.

Material prepared Arkady Shapoval... Look for me on Youtube | Facebook | Instagram | Twitter | Telegram.

Add a comment:



Comments: 15, on the topic: About PhotonsToPhotos and DXOmark charts using Fujifilm FinePix S5 Pro as an example

  • UstasFritZZZ

    Yes, leave the old man alone, for God's sake!)))

    • in

      Ha, I knew for a long time that these charts are garbage in vegetable oil, apparently the old man is only one of the victims of these market-site charts.
      And the post is apparently directed to recent reviewers who reviewed the charts from the first site

    • Arkady Shapoval

      After the release of interview with Oleg (and, in principle, before) many wrote to me in a personal, referring to the photon-to-photos as proof of the mediocrity of the Fujifilm FinePix S5 Pro (including Vadim thoughtlessly refers to the photon-to-photos in his second video about Fujifilm FinePix S5 Pro). I released this note, where I clearly showed on the facts that photon-to-photos are wrong.

      The point is not the Fujifilm FinePix S5 Pro camera, but the fact that such sites can and do make serious mistakes. Then these errors are picked up by users. Fujifilm FinePix S5 Pro is just another very striking example of their ignorance of the intricacies or special sabotage / data fitting. Although these sites give a good overall picture, I also often refer to their data.

      • Alexander

        You are of course right, it was undoubtedly worth it to figure it out and measure the photon-to-photos.
        But if you dig deeper, then the question is not a trivial mistake. The sensors themselves obviously have a DD that was quite usual for those times, but they work in different lighting conditions and, roughly speaking, the addition of their signals gives such a noticeable result.
        In the absence of a standard technique, you can intend anything.
        But any information is needed for making decisions, in itself it is useless.
        And what decision should we make?
        I voiced my opinion above, today it is a camera “for the soul” (a toy for an adult boy).
        Actually for me all my cameras are such “toys”, a hobby is an irrational business. :)

        • Arkady Shapoval

          The sensors themselves obviously have a DD that was quite usual for those times, but they work in different lighting conditions and, roughly speaking, the addition of their signals gives such a noticeable result.

          No, in this case you are also mistaken. The whole essence of the sensor was built on getting an extended DD and solving one of the problems of the old STS. The second part of the sensor does not form an image (it does not participate in the formation of the resolution of the final image), it is designed only to expand the dd by 2 steps. And to measure something in isolation from this is a laughing matter to the chickens. Usually, users associate this misunderstanding with the erroneous opinion that both R and S affect the formation of details (that is, they work like two ordinary CCD matrices, which you, in fact, mentioned), but the R part of the sensor does not take part in the formation of the resolution 3 / 6 or 12 MP, it serves only as an additional adjustment to the S part of the sensor. I'll try another example: 12 MP can be obtained without the R part of the sensor, but the DD will be low.

          The method used by photons-to-photos is rather primitive and vulgar; it boils down to taking a picture of the world on a computer monitor and loading RAW files into a console utility. Photographing the monitor alone causes wild skepticism. In this case, it is advised to use 1/30 of the shutter speed, but at the same time it is advised not to bother if the shutter speed is shorter or longer. The console utility eats NEFs well (it actually has such a name), but RAF does not eat at all. It is stated that it should support TIFF / JPEG / DNG, but no matter how hard I tried, I did not manage to feed her test images from Fuja in this format.

          At the same time, the data for other Nikon cameras in NEF format, which I managed to run through the program, practically matched the data on their website (my result was usually half a step worse).

          • Alexander

            No, in this case you are also mistaken ...
            Maybe I won’t argue.
            One remark.
            Arkady, miracles do not happen. And if the level of technology at a certain time allows you to create a photodiode with a certain minimum noise level (and this level determines the DD in most cases), then it is so. Another question is, using different exposures, you can expand the DD. Fiuggi added photodiodes, HDR technology uses the ones that are, but with different exposures.
            You look at the results of photons-to-photos as a photographer (a user who cares about how important the result is), photons-to-photos as a developer who, perhaps, cares about this very level of technology (photodiode DD).

            • Arkady Shapoval

              I suspect that there is no point in talking further, since it seems to me that you did not understand the essence of what is happening. thanks

            • in

              HDR - it's ridiculous to include it here. Fuji has a completely different working principle.

    • Andrei

      And don't tell me, Arkady himself knocked this interest, and over the past couple of months I have heard more about this camera than in all the years before, from looking at my first camera D80. Why this is necessary, it is not clear enough ... To raise the price even higher on it, or what? The camera is already 15 years old, what to take from it now, xs, the era of the large-scale complex has come.

      • Arkady Shapoval

        The essence is simple - using c5 as an example, one can show the possibility of another evolution of sensors. If FAVEON is simple and straightforward, one hundred SUPER CCD SR is a very interesting and unusual solution.

        • Andrei

          But, unfortunately, this is already a “dead” technology for custom photographic equipment. (not industrial, not scientific, although maybe for them too, here I am not aware of all the events) There was just such a sensor, it was cool, everything is no longer relevant. It's like once upon a time, before the release of Android phones, there was a motorola that could be flashed and flashed, and it was a very, very popular phone, I wanted one for myself. But now only old people and zero sense will remember about him. Romanticizing technology is a bad idea. It was and passed, the end. 🤷‍♂️

          • Arkady Shapoval

            Your opinion, of course. also important, thanks.

        • Andrei

          Of course, I don't mind someone doing this - romanticizing and all that and discussing these sisters, but now it looks exactly like hyping on something, while blowing the news out of nothing. 🤳 Excuse me, Arkady, this is my opinion ✌😊

          • Arkady Shapoval

            No problem.

  • Rodion


Add a comment

Copyright © Radojuva.com. Blog author - Photographer in Kiev Arkady Shapoval. 2009-2022

English-version of this article https://radojuva.com/en/2021/09/zagadochnaya-fujifilm-finepix-s5-pro-2/

Versión en español de este artículo https://radojuva.com/es/2021/09/zagadochnaya-fujifilm-finepix-s5-pro-2/