Do different raw programs develop your camera's raw files the same? Do those programs treat various camera brands equally? Testing them side by side, the answer to both those questions is a definite no. Here's how Lightroom Classic stands up.
Sometimes, I start an article thinking this will be easy to write about, but then it grows, and I find too much information to include in one piece. That was the case for this article, which has grown into a series comparing Lightroom Classic with other raw development programs.
For a previous article, I downloaded several of those different programs, and some I already used. I decided to use this opportunity to put them through their paces and see what gave the best results. Moreover, would loading different cameras' raw files through these programs, without making adjustments, deliver different outcomes depending on the camera brand?
In this series of articles, I'll use the same images from different brands of cameras: Nikon, Canon, Sony, Fujifilm, and my OM System camera and how they look when opened under various raw development programs. The anecdotal evidence suggested that some programs performed better with some brands than others, and I wanted to see if this were true.
There are real-world tests intended as a guideline, giving you an idea of what to look for when investing in software. Minor differences can probably be ignored because they will be correctable in the software or plugins. However, there were some appreciable differences I came across, and it's these I concentrated on.
I'm not trying to develop the images for these tests to get the best possible final results. That, after all, is subjective. Instead, I started by looking at the details in the shadows and highlights and whether they were recovered equally using the adjustment sliders, especially how much noise was produced in the case of the shadows.
Of course, the results shown here won't be the only deciding factor in choosing a program. Price may be important to you – these were shown in the previous article – and you may find some programs more intuitive than someone else. Furthermore, each program has its unique features, and they might be necessary for some and not for others. In other words, it's perfectly okay for one person to love Lightroom (or insert another program name here) and another to hate it. The intention is not to tell you that any program is great or dreadful.
I'm using Lightroom Classic as a benchmark for these tests. I use it almost daily, so I am familiar with it. I find it intuitive, although I've met some who struggle using it; that's true of any program. It's also probably the most popular program. The layout is sensible too. Although not perfect, I consider it a good program. The cataloging was second to none, but others are catching up, if not yet overtaking it.
Lightroom Classic Test Results
Lightroom Classic and Nikon
Of all the cameras I tested, with Lightroom Classic, the straight-out-of-camera results were most impressive from the Nikon Z7 II. Close-up at 100%, the test image was clean of noise. There was no sign of over-sharpening, and the image result seemed to benefit from an increase in sharpening from the default. Highlights could be recovered, and shadows increased with little evidence of noise. If I were a Nikon owner, I would be pleased. However, in future articles, I will demonstrate that even better results are obtainable when it comes to fine detail.
Lightroom Classic and Canon
The Canon EOS-R results at default were slightly over-sharpened and benefitted from reducing the sharpening. Highlights were recoverable, but slight luminance noise was visible in the dark tones. When the shadows were increased, this was more clearly seen, and the noise reduction softened the image, thus requiring additional sharpening. With this in mind, when shooting with a Canon, I would be inclined to use a plugin for noise reduction, such as Topaz DeNoise or ON1's NoNoise. More concerning, the colors were inaccurate; I discuss this in a later article, and skin tones were dull at default values.
The following shot compares the straight-out-of-camera versions of a Canon photo with Lightroom (left) and DxO PhotoLab 6 (right). The skin tones are far more appealing with the default DxO result.
Lightroom Classic and OM System
Lightroom handles the OM-1 raw files poorly compared to other programs I tried. By default, there were many ugly artifacts in the image that were not in other programs. Even reducing the sharpening slider down to zero, there is a graininess that cannot be seen when loaded into ON1 Photo Raw or DxO Photolab 6. Removing this with Lightroom's noise reduction slider left the image looking muddy. That could be balanced with the sharpening slider, but that also reintroduced the noise.
Furthermore, increasing the shadows worked poorly, leaving a purple hue of chroma noise and not showing the detail.
Of course, one might blame the noise on the camera and the smaller sensor size. However, as you will see in future articles, this wasn't the case with some other programs I tried, plus the following two cameras with larger sensors suffered from detail issues too, when run through Lightroom Classic. My results seem to support the claim by OM System professionals that Lightroom doesn't do the OM System raw files justice.
Lightroom Classic and Sony
Similarly, the Sony a7 IV images appeared noisy at default values, which I could not see in other programs. Increasing the shadows accentuated that noise. In fact, increasing the shadows introduced more noise than with the OM-1, which was surprising because of the difference in sensor sizes. However, the OM-1 has a newer stacked sensor, which might account for that.
There was a stark difference between the flatness of raw file previews that were apparent in Lightroom and other programs. One can see a massive difference by placing the above image alongside the same image at default settings in DxO PhotoLab 6. Notice how dull the climber's face looks on the lefthand Lightroom conversion than the righthand PhotoLab version.
Lightroom Classic and Fujifilm
Images from the Fujifilm X-T5 didn't have any noise at default values. Still, the sharpening applied by Lightroom gave an odd painterly effect at default values, especially with the fine details of green foliage. The sharpening required reducing, but that seemed to soften more quickly than other brands.
Highlights were easily recovered, and shadows increased with little noise. However, it still seemed to struggle with the fine detail of the yellow/greens. I can only speculate whether this has to do with Lightroom having problems processing the greens from an X-Trans sensor.
What I Like and What can be Improved with Lightroom Classic's Raw results
Lightroom is a good program. I like its workflow, and it has a lot to be said in favor of it. The layout of the Develop module is excellent, and the raw results at default are not bad. However, neither are they the best.
I can already hear some screaming that these are just the default values with no adjustment settings applied. That is true, but here's the crunch: with some other programs, even those new to me, I could achieve better results than I could with Lightroom Classic, which I am reasonably familiar with. I found color, tone, sharpness, and noise control better elsewhere. Nevertheless, you will see in future articles, when I reveal those that did better or worse, that each other program has its advantages and disadvantages too.
I do like Lightroom Classic. Its library module is impressive, and for applying rapid developments, something I do for writing articles, I would happily continue to use it. The results are not bad. Furthermore, many people will be happy to address its shortcomings by opening their images into Photoshop and the various external plugins, especially the AI noise reduction and sharpening plugins. However, for my high-quality limited edition prints, I will now choose a different program to develop my raw files, and my conclusion will be in one of the follow-up articles.
The other limitation of these tests is that I have tried one high-end camera from each brand: what applies to the Canon EOS R might not to a Rebel T7, and less sharp kit lenses might give different results too. However, I hope these tests show that it is worth considering raw developing options other than the obvious ones.
Just as I always advise people buying cameras, don't take another's word that something they own will be the best option for you.
I should also stress that all the cameras performed well and can produce clear images with recoverable highlight and shadow details. They all stand up to professional standards; I would happily own and use any of them.
Perhaps the most significant learning point for me was that when we read camera reviews, especially when they look at detail and noise levels, we should consider the program the reviewer used to make their judgment.
I want to thank my fellow writers for generously sharing their images for me to play with: Used with the kind permission of Peter Morgan, Canon; Gary McIntyre, Fujifilm X-T5 and Nikon Z7-2; and Andy Day, Sony a7 III.
Sorry this has nothing to do with the article (which I will read in a moment—looks super interesting!) but this image comparison tool/widget/slide-over thing just does not work in Safari. DPReview also uses it too and it annoys me to no end that seemingly no one cares that it doesn't work on the browser every iPhone, iPad and Mac uses.
For what it's worth I am on MacOS 12.6.2, Safari v. 16.2. I have tried viewing the page with no plugins to be sure it wasn't something I was running in Safari. The widget does the same thing on my iPhone running iOS 16.2. What it does on each is when I click and drag (or tap and slide) on the arrows, they just snap to the left of the image only revealing the 'after' or right image. Hope that helps.
It's a browser issue, Colin. It works for some people with Mac's, but not everyone. It seems to be a common iPhone Safari issue. You might want to escalate it to Apple and ask them to fix it. Thanks for highlighting it.
Yes, it is very frustrating. Many articles all about the before/after work, and you can’t see the comparisons. Ggrrr
I had real issues with my EOS R images just looking "meh" in Lightroom Classic. When I tried Capture One, I was blown away by how much better the colors were and how much cleaner the results were in flat areas like a cloudless sky. I now use the R5 and it looks better in LR, but the colors are still not as good as my custom profile created with X-rite. Noise was still superior in C1, but my entire workflow is pretty tied to LR and PS so switching would be a real challenge. For most of my work, LR/PS is good enough, but I'll still pre-process in Capture One when I really care about getting the best out of my Canon files.
Oddly, when I had a Fuji X-E3, I found the colors in the various camera profiles Adobe offers were pretty close to what the camera spit out, whereas they weren't as good in C1. That said, the wormy look present in LR was not there in C1.
But yeah, it would be nice if Adobe cared more about having better base profiles for some wildly popular cameras.
--- "Similarly, the Sony a7 IV images appeared noisy at default values"
Just fyi…those images are actually from an a7iii.
I tried some A7 IV images too, but didn't have permission from the photographer to use them. The same issue existed.
Don't you think as an article, it's a little disingenuous, though? This is how fake news starts. Baby steps of false information.
Bingo! We will just take his word from it. Nice catch! Opinions are like ……..
Looks like a DXO fan article. No issue with that. I prefer the Lightroom results on your exemples, much better on the climbers. At least you could have adjusted the exposure settings, much (too) brighter in DXO.
I am using Lightroom as a benchmark and will be looking at several other programs, some of which perform better with image quality, including DxO and C1 (spoilers!) And some that were worse. DxO was the next on my list, and I had been testing it, which is the only reason I mentioned it. Also, as I say in the article, it may not be the only deciding factor.
While it’s temping to think of Lightroom’s “default settings” as a singular variable with which to compare images from multiple cameras, the reality is that it’s a little misleading because of the camera profile setting. It’s obvious that different cameras will encode their images differently from one another so it should be fairly obvious that one camera profile cannot give all of those variables the same treatment. People often ignore the camera profiles but they really are the true starting point for developing RAW files. They determine the baseline for the editor; the other adjustments (shadows/highlights, color, etc) are therefore not working from the same starting point for each profile. People really should take the time to find a good profile to work with first, then move on to other settings. Just because the “default” profile works better for some cameras (Nikon, Canon) doesn’t mean that’s the profile that should be used for all of them.
I do personally find that working with Sony files in Lightroom has been a struggle compared to Nikon/Canon. A lot of Adobe’s profiles for Sony tend to push either the contrast way too far (and desaturate the color) or push the color way too far on profiles where the contrast is more balanced. I think it’s just a steeper initial learning curve. In my experience the Camera Calibration tab goes a long way in helping with the color of Sony files.
IMHO FWIW Capture One and Sony work well together. Sony and Adobe Camera Raw...not as nice as C1
Yes and no. Capture One works great for tethering. I really wanted it to replace Lightroom in my workflow altogether, but there are a few features missing that just keep it from doing so. Camera Calibration being the biggest one... if you're working with color-saturated images I just find that there really is no substitute for that in C1. I wish there was. Also, I hate that there's no visual representation for sharpening (i.e.threshold) in C1 the way that LR has the masking slider. Local adjustments are also easier in LR.
Cataloging aside, LR still has a few advantages over C1. I just wish that their profiles for Sony were better.
Did Adobe address their "worming" issue with Fujifilm raw files? That's the number one reason why I switched to Capture One after making Fujifilm my main system.
I am very surprised that this is still an issue with Adobe and Fuji. I have not followed it a while. I thought they would have fixed it by now.
The "worming" issue was the reason I and my working partner switched cameras because it was annoying with the XT-2/3 and XT-20. We tried to alleviate it a while with C1P but this was not always possible in our workflow with certain clients. With our Nikons we never experienced any problems either with C1P or Adobe Lightroom/ACR.
I really liked the handling of my XT-20 because it was a fine small camera. But the postproduction was always problematic with Adobe. The XT-2 even produced weird lensflares where you could see the "sensor" pattern which I I often had to retouch in Photoshop since we used lensflares a lot as a stylistic choice. It was weird. The lensflare issue was even apparent in C1P.
From what I know this does not happen in the higher end cameras like the GFX 50 and 100 because they don't use the x-trans sensor.
It would be interesting to see the results if the author were to take all the cameras - with as close to equivalent lenses as possible - into the field and take the same shot with each. Of course the cameras would need to be set up to be as close to equivalent as possible in field of view, depth of field, exposure, etc. If you have different photos taken by different people at different times with different lenses and camera settings, it is hard to know just what the effect of each raw converter is. Of course, in the field the light might be changing from one shot to another, too. Perhaps a studio scene with controlled light might help to isolate the effect of the software.
That's why I encouraged everyone to compare the different programs with their own cameras. This was a real-world test and not a clinical, laboratory examination. It would have been impracticable to do what you suggest, The idea was to point out that there are differences between how different software performs between different camera brands.
Ivor, I think the huge issue here is that a disclaimer of ‘this was a real world test’ doesn’t cut it when you need to actually be more scientific and repeatable when trying to comment on levels of noise when recovered in an image. The fact remains, to make this tangible and anything of substance you would need to have the exact same image from all manufacturers bought into the programmes to do the exact same edit to each.
One area that I would like to see a detailed focus on, is the quality of the chroma channels on fine details using different raw interpreters, as well as the "enhance detail" function in lightroom and ACR (Which doesn't bring out any more fine detail, but does fix some demosiacing errors in some cases, which they have updated since its initial release). It seems like everyone has their own secret sauce to demosiacing which impacts how certain fine details are rendered.
I don’t get editing. Just learn your camera and try to get the correct settings for the shot you’re trying to create. To me, editing is like thinking I can go in and “fix” a painting after the fact. More importantly, 99.9% of people are looking at photos on a mobile device of some sort and can’t see the difference. Just my two cents…. Most of the “greats” never edited their photos
Most edits are not for the purpose of fixing a shot, instead they are designed to get the look you want. for example, if you look at many popular landscape photos, many have been post processed,ad there is often simply no way to get the resulting look in-camera. Furthermore, many of the design goals for a modern camera is not to get such a result in-camera. instead it is a process of providing you with image capture capabilities that well exceed your output medium's capabilities. For example, taking a 14 bit per channel image with 15-16 stops of dynamic range from your camera, and then processing it to display what you want in the resulting PC monitor or TV that may only display 8-10 stops of dynamic range at one time, and while managing that, also adjusting colors and other values to match your artistic vision, or if documenting something, then achieving as much realism as possible.
Even in the past with just film, many famous photographers heavily edited their photos, especially since they not only wanted to express their artistic vision in their photography, but also deal with the limitations of the film having a wider dynamic range than any output medium of the time, though it was much harder on them because the changes had to be done in the darkroom in order to dodge and burn various parts of an image and selectively process parts of get a desired lightness without blowing out or crushing desired parts of the image.
But again, those edited photos are obviously edited and should just be shot with an iPhone/Pixel. I don’t look at heavily edited landscape/modeling etc photos like they are art. I look at them like someone spent an inordinate amount of time to “create” something that wasn’t/isn’t real. My two cents… I think that is fortunately the way most people are viewing photos these days…tired of the hdr over exposed “pictures”
Marc Perino please explain how Ansel Adams or Peter Lindbergh edited their film photos…
With many dedicated cameras, they are designed around the user editing the image, as there is no one size fits all process.
When you take a photo with a smartphone camera, often they are taking a 12 bit raw image and running it through an image signal processor that applies a preset gamma and color curve, along with other things such as noise reduction and other adjustments. It then finishes with some compression before spitting out a jpeg or other compressed format.
Some may use scene recognition to select between different color profiles, for example, some will boost saturation if it detects food, only issue is for some foods, it will cause color clipping.
Take a decent raw files, and you are in full control of the luminance and color channels.
"Most of the “greats” never edited their photos"
The opposite is true. From Ansel Adams to Peter Lindbergh to Helmut Newton to Annie Leibovitz – just to name a few "greats" – all edited their photos heavily or had them edited by someone else. Sometimes analogue sometimes digital. Depends on the era they live(d) in.
I differentiate between developing a photo and editing it. When you shoot a JPEG in your camera, it is developed to look the way some clever technical team in (mostly) Japan decided how the image should look. Some photographers, including some greats (e.g. Cartier-Bresson), did just that. As Marc points out, others do. Most digital cameras give you the choice of more than one development style.
Developing it yourself allows you to produce a photograph how you want it to look. This includes changing the different colors' saturation levels, and adjusting tones to bring out details in the highlights and shadows, as well as cropping and leveling a photo, and removing unwanted noise, and so on.
For me, editing is when you move pixels around or change a photo so it appears differently, like smoothing skin, cloning unwanted objects from the picture, and so on.
What you want to do is entirely up to you, and you should not be pressured into doing anything, whether developing, or editing, or nothing at all. One of the best photographers I've ever met only uses jpegs from the camera. Another does substantial editing. I mostly just develop in raw and occasionally edit. Whatever works for you is all that matters.
LR is good, but good is not best, :) C1 is the best.
I'm in the middle of running the images though Capture One and others. Watch this space!
I can understand what looked like a reasonable sized article becoming a whole series.
With the third party Raw converters I'm reminded of the graphic and cpu optimisers used in computer bench marking.
At the danger of making what has become a series of articles even larger, how do the manufacturers own Raw converters compare to the third party software products you've started with?
Given your already varying findings, I would be interested in hearing your future conclusions as to whether camera reviewers over the past decade or more have been delivering good advice to prospective purchasers, or whether the third party work flows they undoubtedly use might have cast doubt on decades of their work?
I am very aware that some camera reviews elsewhere have based results on using, especially, Lightroom. As an OM System user (Olympus as was) I know that the results from using some other software is far superior to Lightroom, and in some other programs not as good.
Still, Lightroom doesn't give bad results; I still use it for quick edits and I don't think that anyone that uses it or any advanced raw processor will be throwing their edits out of the window.
Saying that, I've revisited some of my older raw files to see if I can get better results with other programs, and I can. I've been trying other cameras files, and I can with those too, to a greater or lesser extent. However, as I say in the article, these results are partly down to personal taste.
I should add that I rarely open a photo into Photoshop, and many photographers use extensive editing as part of their workflow, and that might negated my findings.
Thanks for the great comment, and Happy New Year.
Great article Ivor. Would you recommend giving On1 a try for the Fuji files?
Hi Gary, I haven't got round to testing it with the Fuji images yet, but will do in the near future. It's a free download and trial. I find with OM System files it works well.
EOS R user here...
You can customize your import procedure.
I as well do not like all that sharpening (mainly cos I apply sharpening after global editing) and the default Adobe Color profile is really bad: upon importing I set sharpness to zero and apply Adobe Neutral profile, so as to start from a very flat and dull picture, this having more editing latitude.