6
\$\begingroup\$

If I don't want to do post processing of the photographs, is it better to continue shooting in JPEGs instead of RAW? Or is there a push-button workflow which will produce better JPEG images by doing RAW to JPEG conversion outside of the camera? I.e., I am willing to give more computer time for the algorithm to perform the same outside of my camera.

I am specifically looking for potential off-camera RAW to JPEG conversion which will do a better job than my 5D Mark II. I don't want to move any sliders during the process nor am I planning to store RAW after it is done.

\$\endgroup\$
7
  • 1
    \$\begingroup\$ How do you define "better JPEG images"? \$\endgroup\$
    – mattdm
    Commented Sep 9, 2018 at 14:04
  • 3
    \$\begingroup\$ Post pro is an integral part of the photography workflow. It was in the darkroom and it is now on the computer. I think you should address the challenges you're facing that're causing you to throw in the towel on post instead of trying to find ways to skip it. \$\endgroup\$
    – OnBreak.
    Commented Sep 9, 2018 at 16:00
  • \$\begingroup\$ Eh, there's no shame in treating a digital camera more like a polaroid. \$\endgroup\$
    – mattdm
    Commented Sep 10, 2018 at 13:51
  • 1
    \$\begingroup\$ On your rollback: titles should be descriptive, not clickbait. They should summarize the specific question. The general idea of "workflow for lazy photographers" is much broader than the specific actual question you are asking. \$\endgroup\$
    – mattdm
    Commented Sep 10, 2018 at 20:52
  • 2
    \$\begingroup\$ If you don't like the specific title I'm suggesting, that's fine, but please do pick a more specific title that captures the actual question. \$\endgroup\$
    – mattdm
    Commented Sep 10, 2018 at 21:02

4 Answers 4

15
\$\begingroup\$

"Better job" is subjective. We certainly get lots of questions like

which by their very existence show that many people like the results they're getting out-of-camera. RAW conversion certainly does not automatically result in "better" images, at least not for some universal definition of "better".

However, every RAW program is different. It very may be that the default settings of some such program match your preferences better. (Or, the settings with a few tweaks which you never again touch.) If that's the case, then, sure, that will be the "best workflow" for you. But there's no universal definition of better.

I am willing to give more time for the algorithm to perform the same outside of my camera.

It sounds like you're thinking that the computer may allow more sophisticated or longer-running conversion algorithms, due to the bigger CPU and additional memory, thus getting more detail or better color or something. This isn't actually the case — computers allow more flexibility, but the hardware and software in the camera are finely tuned and optimized to be really, really good at this specific task. Fujifilm, in fact, recently released a feature which lets you use a USB-connected camera as a sort of co-processor in RAW conversion, sending the files to the camera to use its conversion pipeline. That is, the tuned processing in the camera's dedicated hardware is so much better than a general-purpose CPU for this one task that it's computationally worthwhile to send the files off-computer and back again.

In general, the real advantage of RAW is this flexibility. There is no one set of settings that will be "best" for every image for most reasonable definitions of "best". Having RAW files lets you decide that after the fact, including comparing different software and algorithms for the particular image.

However, with most cameras, you can get a good chunk of that flexibility by saving in RAW+JPEG. You can use the JPEG files in the common case, and use in-camera after-the-fact RAW-to-JPEG for fixing white balance errors or similar — and then you always have the RAW files if you have an image you really decide you want to give different treatment to.

(And, I'd encourage you to store the RAW files after the fact. Storage is relatively cheap. If you have Amazon Prime, you have unlimited online storage for images including RAW files; might as well stash them there in case you decide you want 'em later.)

\$\endgroup\$
6
  • \$\begingroup\$ Actually this is wrong. I know of JPG encoders that take a LONG time (minutes) to make a JPG that is about 10% of the size of what a cmaera outputs without visible difference. More CPU power and time definitely allows more sophisticatred algorithms and multiple approach systems to come up with a lot better encoding. \$\endgroup\$
    – TomTom
    Commented Sep 9, 2018 at 15:43
  • 2
    \$\begingroup\$ @TomTom As far as I know, that's entirely to do with optimized compression. I am guessing that Jean means something other than that by "better", but it's true that that is just a guess. Even if that is what's meant, I'm actually not aware of any RAW conversation software which implements Guetzli or similar, although it would certainly be possible. \$\endgroup\$
    – mattdm
    Commented Sep 9, 2018 at 17:18
  • \$\begingroup\$ @mattdm Although only those at Fuji who made such decisions know for sure, I've always suspected the USB connected camera co-processing thing was as much about insuring Fuji's actual algorithms, particularly for film simulation presets, were being applied to the raw data (instead of a 3rd party preset with the same name) as it was about anything else. As far as it may be about processing speed, it's probably aimed at users with lower horsepower devices, such as tablets and phones, rather than higher end desktops upon which most power users do raw processing. \$\endgroup\$
    – Michael C
    Commented Sep 9, 2018 at 22:05
  • \$\begingroup\$ @Michael In their press release for the feature Fujifilm claimed "high-speed conversion approx. twenty times faster* than before", and the * goes to "Mac Book Pro Model A1502 of [sic] Mac OS X 10.11.6". Unfortunately that's an ambiguous model number; in any case, it's not a powerhouse but not really a low-power tablet or phone either. \$\endgroup\$
    – mattdm
    Commented Sep 9, 2018 at 22:10
  • \$\begingroup\$ @mattdm So now the engineering department that actually decided to add the feature is writing the copy for the marketing department? Hahaha. As far as I'm concerned, a mid-level Mac notebook is relatively low powered due to the lack of cooling capacity compared to a true high power desktop. \$\endgroup\$
    – Michael C
    Commented Sep 9, 2018 at 22:15
8
\$\begingroup\$

In addition to what the other answers have already stated:

There may be specific shooting situations that you encounter on a regular basis that would benefit from batch converting raw files after the fact using a "recipe" for that shooting location that is far more detailed than what in-camera settings can apply. You would have to do the work of developing the recipe one time. But after you have an optimized conversion profile for images shot under the same conditions, you only need to do a batch conversion to apply that formula to all images shot in the same situation.

There are things I shoot regularly for which I use this strategy. One is high school sports. In fully enclosed gyms or outdoor stadiums at night where the artificial light is always the same, I create profiles for that specific facility for each of my camera bodies. Each time I shoot in that facility I use the same set of exposure parameters (or at least the same exposure value - if I increase Tv I reduce Av by the same amount). I can then apply the same exact raw processing recipe/preset to all of the images. These profiles that I've created using my raw processing application allow much finer adjustments of many things than the in-camera settings do, as well as allow things such as HSL adjustments that the camera doesn't even offer.

For example:

  • My Canon cameras only allow manual color temperature adjustments in 100K increments. I can select 4200K or 4300K, but I can't choose 4370K in-camera. With Canon's Digital Photo Professional 4, I can select color temperature in 10K increments, and thus can choose 4370K.
  • The same is true of white balance correction. In camera adjustments are limited to whole units that are roughly equivalent to a 5 mired color correction filter. DPP 4 allows adjustments as fine as one-tenth of the same units along the Green←→Magenta and Blue←→Amber axes, which translates into adjustment steps of only about one-half mired.
  • DPP 4 also allows brightness adjustments in increments as small as 0.01 stops, highly detailed contrast curves adjustments and separate adjustments for shadows and highlights as well as overall contrast. In contrast (heh heh), the camera only offers a general contrast setting with no separate highlight or shadow adjustment, nor any kind of customized light curves.
  • Noise reduction in camera is a choice between: 'Off', 'Low', 'Medium', or 'High'. DPP 4 has very fine control of both Luminance NR and Chrominance NR of 20 units in 0.1 step increments for each.
  • A single Sharpness setting ('Low', 'Medium', 'High') in-camera is similarly much coarser than the 0-10 in 0.1 steps 'Sharpness' control using DPP 4, not to mention the even more detailed 'Unsharp Mask' controls with independent 'Strength', 'Fineness', and 'Threshold' adjustments available in DPP 4.
  • This is also the case with lens correction. In camera is pretty much 'On' or 'Off' for correcting things such as chromatic aberration, geometric distortion, and peripheral light falloff (for camera models that even offer all three in-camera). In DPP 4 there is a highly configurable CA correction tool with 0% to 200% adjustments in 0.1% increments for overall CA correction as well as separate blue and red sliders with -1.0 to +1.0 in 0.1 steps. Geometric distortion correction and peripheral light falloff can also be adjusted in 0.1% steps from 0-120% and 0-100%, respectively.
  • The entire Hue-Saturation-Luminance module available in DPP 4 (and many other raw processing applications) has no in-camera counterpart, other than the general 'Standard', 'Portrait', 'Landscape', etc. 'Picture Style' development presets that seem to apply some HSL adjustments. Being able to independently control the hue, saturation, and luminance for eight separate slices of the color wheel in very precise increments is invaluable when working with images shot under less than full spectrum light sources such as found in high school gyms, stadiums, under sodium vapor streetlights, etc.

Is it possible to save Canon camera JPEG conversion preset in DPP?

Yes. They are called "recipes" by Canon and are saved as .dr4 files for DPP 4. DPP version 3 and earlier use a different file extension and those can not be used in DPP 4, or vice versa. You can also create a Custom 'Picture Style' using Canon's Picture Style Editor and load it into the camera, but there is not a direct one-to-one correspondence between all of the options available in DPP and the options in PSE.

To save a camera preset using DPP 4, set the default behavior of DPP 4 when opening raw files to use the in-camera settings at the time the image was taken. This is the default behavior of DPP, so if you haven't changed it, DPP is already set that way. You then just need to save the recipe from an image shot in your camera that used the settings you wish to save. It is as simple as right-clicking on the image in the 'Edit' window and one of the choices from the context menu is "Save recipe in file."

To apply a previously saved 'recipe', simply use the thumbnail preview window, select all of the images to which you wish to apply the recipe, then right click on one of the images and select "Read and paste recipe from file." You can then use the "Save"button at the top of the window to batch process all of the selected images.

\$\endgroup\$
2
  • \$\begingroup\$ Is it possible to save Canon camera JPEG conversion preset in DPP? \$\endgroup\$
    – Jean
    Commented Sep 10, 2018 at 20:21
  • \$\begingroup\$ @Jean Yes. Please see the additions to the answer. \$\endgroup\$
    – Michael C
    Commented Sep 10, 2018 at 23:33
1
\$\begingroup\$

If I don't want to do post processing of the photographs, is it better to continue shooting in JPEGs instead of RAW?

The whole and only point of raw formats is that they allow more in post-processing. If you're not post-processing then you're not exploiting that advantage. But by not using in-camera JPEGs you're incurring the cost of possibly losing the excellent built-in demosaicing that had been custom-tailored to your camera (that might be using proprietary data that is not included in raw). For this reason (if you like the in-camera results) auto-processing raws to JPEGs outside of camera is pretty much pointless: all costs, no benefit.

Post-processing is a very powerful tool and by rejecting it you're rejecting big part of what photography is today. Nevertheless, given your constraints the answer is:

Yes, continue shooting in JPEG.

\$\endgroup\$
5
  • \$\begingroup\$ @MichaelClark Yes. Often. 1) manufacturer might want to hide their algorithms. It basically cost them nothing, as professional photographers will use Photoshop or Lightroom anyway. 2) manufacturer might omit or encrypt some data in order to deliberately handicap 3rd party software (like Nikon did with RGB matrix data). Sure, the two would have to happen at same time, but we have all the components ready. My point is that better results without human interaction are impossible to guarantee and it appears this is what OP is asking for. \$\endgroup\$
    – Agent_L
    Commented Sep 10, 2018 at 14:57
  • \$\begingroup\$ @MichaelClark I'm not arguing with you. I understand that vendor apps do find use. I was only trying to rationalize manufacturer's way of thinking when some big guy starts worrying about "competition stealing our multimillion-dollar algorithms". My only point is that "same quality of demosaicing cannot be guaranteed in every circumstance". I've edited the answer \$\endgroup\$
    – Agent_L
    Commented Sep 10, 2018 at 15:06
  • \$\begingroup\$ While I agree with the gist of your answer with regard to the usefulness of saving raw image data if someone has no desire to do any post processing whatsoever, I fundamentally disagree with your assertion that external raw processing has to be inferior to in camera JPEGs, in that at least one manufacturer (Canon) does not do any proprietary in-camera processing that is not also included in their external raw editing application. I highly suspect there are others as well. \$\endgroup\$
    – Michael C
    Commented Sep 10, 2018 at 15:18
  • \$\begingroup\$ @MichaelClark That was not my intention. What I wanted to say that if the in-camera demosaicing uses dirty tricks, you can only lose them. \$\endgroup\$
    – Agent_L
    Commented Sep 10, 2018 at 15:22
  • \$\begingroup\$ But you don't have to lose them, which is what the answer strongly infers. Even if the camera uses proprietary processes (Canon's cameras do), those same processes can be included in an external application (they are included in DPP). It's no more difficult to crack the firmware of a camera than to crack a raw processing app. \$\endgroup\$
    – Michael C
    Commented Sep 10, 2018 at 15:43
-1
\$\begingroup\$

You can use Adobe Lightroom software for this.

  1. Upload RAW file to Adobe Lightroom software.
  2. Then apply auto adjustments or else you can create some templates with different adjustments by changing the saturation, exposure and etc...
  3. Then save it as JPEG. It will do the magic.
\$\endgroup\$
1
  • 7
    \$\begingroup\$ Will this "magic" be a "better job"? In what way? \$\endgroup\$
    – mattdm
    Commented Sep 9, 2018 at 14:14

Not the answer you're looking for? Browse other questions tagged or ask your own question.