XT-1 white balance: do i have a bum camera?

Discussion in 'Fuji X-Mount Cameras' started by rbelyell, Dec 8, 2015.

  1. rbelyell

    rbelyell FujiXspot Regular

    118
    May 14, 2013
    i just picked up a used xt1 and took it out to a local xmas walk for some night time street shooting. the lighting was just from typical street lights. i shot at 6400 raw plus jpeg in classic chrome and processed in latest lightroom. the jpegs came out fine, but the raw all had a greenish/yellowish color cast. very disconcerting. tbh, and not to start a debate because as its an opinion its not really debateable, but i have really disliked xtrans raw rendering in its first gen cameras. i was hoping the 'new' xt1 engine would be better than the first gen x100s and x20. on the whole it kinda is, but i'm wondering if this color cast in these conditions is an xtrans thing or maybe i have a bum camera.

    if its an xtrans thing, does anyone have a lightroom workflow setting to deal with it?

    thanks!
     
  2. KillRamsey

    KillRamsey Super Moderator

    Feb 15, 2013
    Hood River, OR
    Kyle
    I only shoot jpg, so I can't really help you. But I'm curious if you used Auto WB or set it manually? And when you say "typical street lights," do you mean metal halide (whiteish) or high-pressure sodium (orange)?
     
  3. rbelyell

    rbelyell FujiXspot Regular

    118
    May 14, 2013
    hi kyle
    i used auto wb. i do not know how to distinguish the types of streetlights you mentioned, sorry. i didnt know there were different types!
     
  4. KillRamsey

    KillRamsey Super Moderator

    Feb 15, 2013
    Hood River, OR
    Kyle
    Once upon a time I sold electrical supplies, including these high-output bulbs. It's not something most people have any reason to know about, I guess. :)

    HPS are those really, really orangey yellow overhead street lights that have been around since the 70's or so. The color temp is actually pretty far to one end of what the camera can balance to, I think. Mercury vapor is that blueish cheap overhead lighting that a lot of homeowners in the 80's would put off the roof of their garages to light the driveway at night. Metal Halide is much closer to "white," and is used a lot more often now.
     
  5. flysurfer

    flysurfer X-Pert

    Feb 1, 2013
    Nuremberg
    Rico Pfirstinger
    You did set Lightroom to Classic Chrome, right?
     
    • Like Like x 1
  6. rbelyell

    rbelyell FujiXspot Regular

    118
    May 14, 2013
    well rico, i did not realize that you could change 'camera profiles' in lightroom from adobe standard to another 'simulation' without losing the extra data in the raw file. i have since learned that many choose other 'profiles' offered for other cameras besides fuji when downloading raw files to lightroom. i mistakenly thought raw was raw, that any program would download raw the same way, and that the 'default' download would be somehow less than optimal. i never even considered looking at the 'camera profile' box. the errors of my thinking have been overwhelmingly pointed out on anther forum, so now i get it and will not be using lightrooms default adobe standard when i download raw xtrans files.
     
    • Like Like x 4
  7. Odille

    Odille FujiXspot Rookie

    20
    Jun 16, 2015
    Glenorchy, Tas, Australia
    Odille
    Having a small chuckle at your latest response :) Good luck!
     
    • Like Like x 1
    • Appreciate Appreciate x 1
  8. Amin Sabet

    Amin Sabet Administrator

    795
    Jan 31, 2013
    Boston, MA (USA)
    You misunderstood @flysurfer@flysurfer and @rbelyell@rbelyell . Regardless of your camera profile (for JPEGs), LR gives you the option to change your camera calibration profile from "Adobe Standard" to a different one such as "Classic Chrome". This changes the colors and contrast in your raw processing in LR, similarly to how the in-camera profiles affect the raw processing in camera.

    See screenshot below:

    profile.
     
    • Agree Agree x 1
  9. flysurfer

    flysurfer X-Pert

    Feb 1, 2013
    Nuremberg
    Rico Pfirstinger
    There are plenty of differences between the film simulations in the camera and the corresponding Lightroom profiles. It's far from a perfect match, but it's more or less workable. The differences are rather prominent, though, when it comes to images taken with the DR function (DR200%, DR400%), as Lightroom's highlight recovery works adaptive (similar to Apical Iridix), while Fuji's does not. You can use Iridient Developer as an alternative, though. Iridient offers all Fujifilm simulations, too (I actually played a minor part in creating them), but Iridient's highlight recovery is very similar to Fuji's DR function processing, so your Iridient results will look much more like the SOOC JPEG from Fuji's own JPEG engine.

    This is actually a quite interesting topic that's often ignored, even though it's practically more important than those endless discussions about sensor size or megapixels. Thats because you don't need pixel peeping to spot these differences, you can already see them on a thumbnail image. This is also why I startet to offer Fujifilm RAW workshops here in Germany, where we spend two full days to point out the characteristics and challenges of Fuji RAWs and RAW conversion options. For example, while everybody talks about Lightroom's issues with fine detail, color issues that are much more visible are widely ignored.
     
    • Like Like x 1
  10. rbelyell

    rbelyell FujiXspot Regular

    118
    May 14, 2013
    rico, you seem to know about this from the 'inside', as opposed to the rest of our 'outsider' speculation. so can you please help me understand what advantage xtrans offers over traditional bayer that compensates for these RAW download translation woes? i confess that i overwhelmingly enjoy RAW from the original x100/10 to the 'upgraded' xtrans sensor. i also confess imo as a consumer, fuji messed up badly and continues to do so by insisting on xtrans. i have used digital from leica, sony, olympus, panasonic, epson, nikon and canon and NEVER had RAW interpretation issues like that with xtrans. correspondingly, and importantly, i see absolutely no IQ advantage to xtrans over the sensors in any of those other manufacturers, nor over fujis own bayer-sensored cameras. i routinely contribute to several camera forums, four of which are devoted to fuji, and on none of them have i ever heard anyone choosing fuji to get their hands on xtrans files, while i have heard of many who have left fuji or will not try fuji in the first instance because of xtrans. so why this seemingly insane insistence on xtrans?
     
  11. Amin Sabet

    Amin Sabet Administrator

    795
    Jan 31, 2013
    Boston, MA (USA)
    Seems I misunderstood your earlier (now deleted) post in this thread. Sorry about that.

    I think @rbelyell@rbelyell doesn't mean that the RAW files themselves are bad. He means that they are not leading to good results in his raw processor of choice (LR) when using his usual LR workflow. Correct me if I'm wrong, @rbelyell@rbelyell

    I don't think there is any controversy that Fuji's use of X-Trans presents some challenges for third party raw processing support. And it certainly is possible to have a hard time getting accurate colors from a particular RAW processor color profile for a particular camera, irrespective of one's knowledge about WB, exposure, etc.

    @rbelyell@rbelyell - if the LR color profiles mentioned don't give you the results you want, you could also try the Huelight profiles for $15: Huelight FujiFilm Camera Profiles

    I'm happy with the results I'm getting both from built-in color profiles in both LR and from Capture One, so I can't say I share your frustration about raw processor results from X-Trans-derived files, but those are some things you can try.
     
    • Appreciate Appreciate x 1
  12. flysurfer

    flysurfer X-Pert

    Feb 1, 2013
    Nuremberg
    Rico Pfirstinger
    I have no inside information, but Fuji has communicated the advantages of X-Trans in its marketing. After all, it's an USP they think will sell more cameras, but if you do not like the "look" or the overall results, simply do not buy or use it. There are plenty of conventional Bayer options available, and there's also Foveon, which Sigma also markets as an USP (and yet, it only occupies a very small niche).

    As for X-Trans, the advantages are more resolution with less moiré, there's less high-ISO color noise, and all RGB colors are used in each single sensor line. The sensor is also a bit more light efficient, as it features more green pixels than Bayer, and as you know, it's the green pixels that basically make up luminance and detail.

    It's not difficult to extract plenty of detail with software like Iridient Developer. The only software with some issues I know of is Adobe Lightroom, but they fixed the color problems in version 6.1 (or 6.2?) and also promised to look into improving default sharpness/detail rendering in a future release.
     
    • Like Like x 2
  13. rbelyell

    rbelyell FujiXspot Regular

    118
    May 14, 2013
    thanks for your reply. i understand, like those who shoot parallel lines and charts, there is a segment of the community that puts its faith in hypertechnical explanations of why one thing is 'better' than another. i dont and therefore i was asking more of a real life observational question. sorry you took it the wrong way. obviously i dont have to use xtrans. that wasnt really my point.

    so let me be clear: my point was observational and anecdotal. i observe xtrans raw interpretation is inaccurate and finding the necessary workflow to correct it is a PITA. i observe no IQ benefit whatsoever when pitting xtrans raw over any other sensor kind, even when viewing the work of others who 'understand' and properly utilize the necessay workflow. anecdotally, as opposed to something like foveon, i havent seen any 'craving' in the photographic community for xtrans; indeed ive seen quite the opposite. so my question was directed at what we can see with our eyes as xtrans benefits vs the competition, not what the fuji marketing machine technobabbles about.

    and yes, my conclusion seems to be the same as yours as to fujis motivation here: it is to distinguish itself from every other sensor out there. we differ in that i believe they are distinguishing themselves in the wrong direction. i take the repetition all over the web of similar discussions about the negatives of this singular sensor as evidence in my corner.

    finally, i own three fuji digi cams and have owned several fuji film cams. i root for the company to succeed, and believe they did with their hardware concepts of manual control and the quality of their glass. their intitial digital foray via the bayer-sensored x100/10 were awesome. i fully lament their decision to abandon bayer in favor of xtrans, and see that decision as singularly holding back their sales and popularity. others views may rightly vary, i was simply seeking a factual observational basis for that variance that i might be missing.
     
    Last edited: Dec 13, 2015
  14. Amin Sabet

    Amin Sabet Administrator

    795
    Jan 31, 2013
    Boston, MA (USA)
    I think that Fuji's X-Trans cameras have some of the nicest looking, if not the nicest looking, in-camera JPEGs around, and I definitely see less aliasing with X-Trans sensor images than I do with Bayer sensor images or even Foveon sensor images (which have no color aliasing but lots of luminance aliasing).
     
  15. rbelyell

    rbelyell FujiXspot Regular

    118
    May 14, 2013
    thanks amin, and thank you also for clarifying my comments in a couple of your earlier posts. and i absolutely agree with you on both counts. fuji jpegs are sublime, really. and while imo the first gen xtrans jpegs were nowhere near as good as their bayer counterparts, i do think that whatever gen is in my xt1 produces better jpegs at high iso. as an aside and fully OT, i do believe that the natural progression of quality would mean that, had they chosen to stick with bayer, this x-new generation of bayer 16mp jpegs wouldve naturally been better than the original 12mp x100 bayer sensor. so i dont necessarily credit xtrans here, i credit how tech goes.

    my quarrel with xtrans is solely with raw, which i tried to convey. all of my observations, conclusions, criticisms and queries relate only to xtrans raw.
     
    Last edited: Dec 13, 2015
  16. Amin Sabet

    Amin Sabet Administrator

    795
    Jan 31, 2013
    Boston, MA (USA)
    But only with xtrans raw processed by Lightroom, right? Maybe part of that quarrel belongs with Lightroom.
     
    • Like Like x 1
  17. rbelyell

    rbelyell FujiXspot Regular

    118
    May 14, 2013
    well im not sure. i believe Capture One also has issues, and its adobe products across the board, not just LR. i feel like i have read about others as well, but i cant put my finger on 'em right now. ive no love for adobe, but i dont think its just them. and i'm not sure i'd blame them at all. fuji needs be aware of how their consumers work. its seems foolish for them to ignore how their consumers utilize their products in real life and decide to make life more difficult for people who buy their products. especially so when the 'benefits' vs their competition is not readily apparent. it would be like a car manufacturer putting out a vehicle they claim is in a performance class by itself, but it runs on petroleum jelly. well great, but their end users fueling options are limited to only gas stations!

    and at least for me, i see these raw interpretation issues, and wonder if in addition to all the other obvious negatives, the sensor is needlessly limiting what we can get out of their great glass. i dont know, but it makes me wonder. i just dont see what benefit either fuji or the consumer is getting from this sensor choice.
     
  18. flysurfer

    flysurfer X-Pert

    Feb 1, 2013
    Nuremberg
    Rico Pfirstinger
    Well, you certainly do, but here's the thing: I don't. I find it perfectly okay (and actually very smooth looking, but liking this is a matter of taste). By the way, I am comparing 8 different external RAW converters in my RAW workshop, where we see that each of them delivers a quite different default result when fed the very same sample RAW (an Imaging Resource studio scene with plenty of colored Siemens stars). Even different versions of the same converter (like Lightroom 5.7 vs. Lightroom 6.2) will show vastly different results.

    That's why I have long lost interest in blaming or praising X-Trans (or Bayer) for anything, as pretty much all demosaicing issues are problems of specific RAW converters and how they handle different things. They all have strengths and weaknesses.

    Of course, similar differences occur when you feed a Bayer test image to those 8 converters. Heck, even Foveon looks completely very different in Iridient Developer compared to Sigma Photo Pro (and again different when the JPEGs come straight out of a Sigma camera).
     
    • Like Like x 1
  19. Amin Sabet

    Amin Sabet Administrator

    795
    Jan 31, 2013
    Boston, MA (USA)
    I haven't seen any issues with C1. Adobe has just two products which open Fuji RAW files (LR and ACR), and they do it the same way. I see this as an Adobe issue. Iridient Developer has just one person doing development IIRC and has had great Fuji X-Trans file support for a long time.

    But back to the issue, have you tried the built-in LR profiles yet? It may be that the problem you are seeing has been solved with the introduction of those profiles. Maybe your favorite gas stations already carry petroleum jelly ;).
     
    • Like Like x 1
  20. flysurfer

    flysurfer X-Pert

    Feb 1, 2013
    Nuremberg
    Rico Pfirstinger
    All converters have different issues, but seeing or not seeing them depends on the subject and on the observer, so your mileage may very. Critical test charts (not the ones from dpreview, though) tend to reveal such issues, but it's debatable how strongly such problems affect real-world pictures.