RAW vs. JPEG
We’re going to look at some of the issues revolving around whether to shoot in RAW or to shoot JPEG. This isn’t as black-and-white a decision as you may think.
Let’s first look at what we’re talking about here. Unless you have had your head in the sand, you know that pretty well all modern DSLR’s (and even some point-and-shoots) have the capability of shooting in RAW. Actually ALL digital cameras can shoot RAW, only some of them can deliver images in it. Some cameras automatically convert everything to JPEG before outputting them. All of your photo club buddies, the forums online, the sales people in the stores and the training sites like NAPP harp at you to shoot in RAW. “Anyone who shoots in JPEG can’t be serious”. We’re here to think about that.
So the RAW file is the basic full image that the camera has captured. Nothing has been done to it – every pixel, every level, everything that the sensor saw, is in the RAW file. A JPEG is the same picture after the computer (your camera IS a computer) plays with it. It does a bunch of stuff to the image that we’ll look at below, compresses it so that the file is smaller, and it does a lot of that by only affecting things you’ll never see.
Most of the time.
You can’t use your RAW file for anything. You have to convert it – to a JPEG, to a TIFF, to a GIF, to something else before you can output it anywhere. That means to print it, to look at it onscreen, to submit it to your club competition, even to process it in Photoshop. But it contains EVERYTHING so it’s your best starting point.
Sometimes.
A RAW file (by the way, there’s no real standard. Nikon has their NEF format. Canon has CR2. Other manufacturers have their own proprietary formats. Adobe is trying to get the world to go DNG which stands for “Digital Negative” but that hasn’t happened yet. And Sony wanted the world to go Betamax. Remember?) needs to be manipulated. You need to have a program that will do that – the most common are Photoshop and Lightroom, both of which use exactly the same engine: called “ACR” or “Adobe Camera RAW”. Yes there are other programs out there, but I haven’t used them for many years. And here’s where the problem arises.
When you import your RAW images into, say, Lightroom, you choose one of many presets – maybe your own custom one – to do some initial adjustments. Remember, you can’t use a RAW file without adjusting it: the image will be flat, dull, soft focus, and really big.
What prompted the writing of this article was a problem I’ve been having. All my images were oversharpened, but were actually out of focus because of the noise reduction I had to do to compensate, and were fraught with artifacts. I only discovered this when I tried to submit images to Shutterstock where they are REALLY fussy – another article will cover this topic – and they all got rejected. I was at the point where I figured there was something wrong with my camera and was preparing to bite the bullet and ship it to Nikon (along with my credit card info, making sure I had lots of available credit limit!). Then someone suggested that I shoot some comparisons between RAW and JPEG so I did: and almost fell out of my socks until I figured out what was going on.
Look how unsharp the RAW file is!
Here’s what’s happening: your camera is a lot smarter than you think. It’s a computer. Before it gives you back your picture in JPEG form, it applies sharpening and de-noise algorithms, it sets your white balance, even the colour space, changes the Gamma and modifies the relative contrast, and much more.
Now that big, bad, computing machine that’s sitting on your desktop can do that too. Much more powerfully, and with greater flexibility. So which computer should you use? The little one in the camera or the big one on your desk? Depends how much time you have and how much attention you are prepared to give to your pictures. Depends on your skill level too: are you as good as that team of software engineers over at Nikon? Maybe not, but of course not all images are the same, so how can they know what’s best for a particular picture? Well, they can’t… but they can give it a ‘go’!
My problem is, I’m not that smart. I seldom “RTFM” (Google it!). I just wing it. And sometimes I fly like a brick. Coming back to those initial adjustments in Lightroom, I was overdoing it. I was setting my sharpening to “100” where “25” would be a better starting point. I even pushed the clarity and vibrance up on all images as I brought them in because, well, I like them better that way. But it shouldn’t apply to all images.
Here’s an example of a picture I worked on for this article.
- Look how much the camera already did to create the JPEG
The one on the right is the JPEG as it came out of the camera. On the left is the RAW file as I imported it to Lightroom and where I had set everything to “General – Zeroed”, so no adjustments. About 30 minutes later, I got it looking like this.
- Two minutes of fiddling with the JPEG got me this. The workflow for the RAW file is below.
Gee. Almost as good as the JPEG! Well if you’re just going to look at it on a computer screen, that is. Or send it in to a newspaper for printing (their process is really low quality). But if you’re going to print it or project it on the side of a bus, it’s much better. It has a wider dynamic range, there are more colour nuances (JPEG is limited to 256 colours for each channel – R-G-B. RAW has 4096 or even 16384 levels per channel), more detail.
Does it make sense to process hundreds of images individually? A typical day at the ice races yielded about 400 images, of which some 250 or so were ‘keepers’. Imagine 30 minutes each and for what? Yes, you can batch process – you work on one image, save the settings and sync all the other pictures to those settings. You’re doing what those Nikon software engineers are doing, but you’re starting with your own custom settings. That can be good, or that can be bad, depending on how smart you are.
You can do a LOT to a JPEG image in Lightroom, or in ACR. You can change almost anything that you can do to a RAW file (although the camera profile is fixed), except you’re working on an image that has already been reduced in quality, compressed and had stuff stripped out of it. The RAW to JPEG algorithms in your camera has already done most of the work for you, but, as I said, at a cost.
So. Bottom Line. If there’s the remotest possibility that I might want to do high quality printing on an image, I’ll shoot it in RAW. That means most of the time, when I’m out shooting casually, or I’m doing landscapes or portraits, I’ll shoot RAW. But if I’m going to shoot the ice races, or a hockey game or an air show or any other event where I’m going to have hundreds of images to work through and tweak, and where I’m not looking to print anything or do high quality output, I’ll shoot JPEG. If I’m not sure: my camera has a setting where I can actually do both: RAW + JPEG FINE.
And those photographer friends who look down their noses and sneer at anyone not shooting in RAW? Ask them if they’re so serious, why aren’t they shooting with a Hasselblad H4D? Then go have a nap while they stay up all night processing their images.
-- 30 –
Have your say! Comments are super welcome!