When I first started taking digital photos I had no idea what the difference between JPEG and raw was, but now I know I'm never going back to shoot JPEG in-camera for these specific seven reasons.
On the whole, most photographers attempt to shoot raw format wherever possible. That is, the lossless image file that your camera automatically generates stores extra metadata related to the color, brightness, and more of said photograph. However, there are still some that shoot JPEG in-camera through either ignorance or necessity. I won't be getting too technical or scientific with my approach to this subject, rather I'll be going on real-life experience and what I've found in my time as a full-time photographer.
JPEG is a reliable image file format that's widely viewable across a range of devices around the world, so it's a useful file format in terms of sharing images, but shooting straight to JPEG in-camera, in my opinion, is a bit of a no-no. That's because we no longer require the smaller files that the compressed JPEG format offers because memory capacity has rather caught up with the larger file sizes that cameras are capable of nowadays (though we're starting to see that turn again with the advent of hyper resolution monsters that shoot 50MP+). Throughout this piece, I'll demonstrate with actual examples of JPEG compression using minimum and maximum compression standards. Unfortunately, because I'm sharing this with you via the web all the photos will be JPEG format, but I hope you get the gist.
I recognize that there are some photographers that find shooting straight to JPEG completely essential, such as photojournalists, sports photographers, and other media/news outlets that require the ability to publish images almost the instant they happen. I also realize there's the ability to capture both raw and JPEG simultaneously, or on different cards in dual-card slot cameras, but unless you're working in the aforementioned fields, rarely if ever will you need to turn something around so quickly that you have to bypass the editing stage entirely. For the rest of us, it's no longer a necessity to shoot like this, so here are 7 reasons why I'll never shoot JPEG, in-camera, ever again.
Changing Color Profiles is a Pain
My approach to color profiles and white balance is this: keep it the same, all the time, every time. On my Nikon I like to shoot in the "neutral" picture control (read: color profile) and with my white balance set to flash. I even keep the screen brightness at +1 at all times. This consistency allows me to manage my opinion over the exposure and featured color in a shot. Exceptions I make are when I switch to tungsten or fluorescent white balance and dim my screen to -5 when shooting astro at night, the former to reduce the impact of orange light pollution and the latter to stop the strain of my eyes as I fumble around in the dark.
But when it comes to editing my shots I tweak both aspects significantly if I feel the original shot isn't right. That's where JPEG normally lets me down. There have been times I've needed to shoot in JPEG in-camera (rarely) and I don't bother making a raw backup, only to head out a few days later forgetting to set it back. It's only when I've come to edit those shots and been flummoxed as to why the color isn't matching up like it always does that I realize what I've done. It's hard to explain but the color palette doesn't shift like I'd expect it to in raw. The gamut hits brick walls and I can never get that subtle, finely-tuned look I'm going for.
When working with shots at 100% zoom I've noticed on more than one occasion with JPEG files that there's a strange quantizing of pixels. Things start getting more blocky the more you zoom in and the reason for that is that in order to compress the file, the JPEG format actually discards information inside the image.
Depending on the amount of compression applied these artifacts are either more or less noticeable. Also, when editing images and adjusting things such as tone and color, due to the lack of information in the photograph, you're also prone to discovering other issues such as increased noise (or grain) across the frame, and also color noise where random sections of light or dark have speckled color highlights instead of the natural tone that's already there.
Ability to Save Shadows
It's incredibly hard to boost the shadows of a JPEG image without incurring some kind of obvious degradation in image quality. Quite often, the data just isn't there to support the boost and as such it looks muddy and even gritty in places. Take a look at a zoomed-in example below. You can see that detail is missing completely from the tree foliage and trunk. The needles are simply a mass of green blobs with no definition. Use the slider to compare minimal and maximum JPEG compression examples.
Ability to Save Highlights
For the same reason as above, it's much more difficult to save or edit highlights in shots taken in JPEG format. Let's take a typical example of an outdoors shot that includes a bright sky and darker foreground. Exposing for the foreground will overexpose the sky and in some cases lead to clipping. In a real-world scenario once clipped, a JPEG file is unrecoverable. You try and drag down the highlights or reduce the whites sections and all you'll get is a gray, blank mass.
However, do the same thing with a raw file and you'll find that those once-clipped highlights are now recoverable. In fact, it's probably a good idea to overexpose scenes in some cases when shooting raw because a small level of clipping is acceptable because you'll be able to recover it later in editing software and it can help reduce noise in shadowy areas.
Flexibility in Exposure
I use Adobe Lightroom to process 90% of my photographs and the current iteration gives users the ability to boost and reduce the exposure value of a shot by 5 stops. That's 10 stops from minimum to maximum, overall, which is huge. This works magnificently well with raw files as I can then tone highlights and shadows to suit the mood of the photo. But in JPEG, though it has the same function, the results are sub-par (to my eyes at least) and don't give the same dynamic range I would normally expect from my editing workflow.
The lack of data in JPEG files also means that correcting noise, whether from high ISO values or long exposures, is also quite inflexible. Shots seem to turn to smooth glass whenever I add a little noise reduction. You've probably seen the same thing when the image is suddenly devoid of noise only to be replaced with small facets of smooth, glass-like fragments that tesselate throughout the frame.
When I'm finished with an edit, I'll save it as a high-quality JPEG. I don't mess around with TIFF because they're much larger in file size and I've had some clients who don't know how to work with them (yes, hard to believe I know.) JPEG is widely compatible and probably the world standard in terms of image sharing online as I mentioned before. But I like to keep original raw files so that I can go back to them either a few days, months, or years later if I want to re-edit them.
Editing software is constantly becoming more sophisticated and new tools are being developed all the time that give me the ability to edit better and more accurately. I mean, what I wouldn't have given 10 years ago to have an object selection tool like Photoshop offers. Also, my taste changes over the years and the edit I loved five years ago may be too on-the-nose for me now. JPEG stunts that editing process significantly and that's one of the main reasons I prefer not to shoot it in-camera.