The myths of digital imaging
The web is full of nonsense about various photographic matters. Most of it is evangelical rubbish, some of it is old-fashioned misinformation.
Let’s clear up a few things …
MYTH 1: Adobe RGB is for printing
Oh boy, hasn’t this myth caused a lot of people a lot of grief.
Adobe RGB is only useful for printing in specific circumstances, and only if you have a firm grasp of colour management.
sRGB is usually better, and always safer.
MYTH 2: Non-destructive editing in Photoshop
It’s widely understood that using Adjustment Layers in Photoshop is a non-destructive method of editing.
No. Every form of editing in Photoshop is destructive to the original pixels in some way.
Adjustment Layers simply allow reversibly-destructive editing – that is, you can undo the destruction if you want to.
At the end of your workflow, when your image leaves Photoshop for print or web, the layers are flattened and the destruction is irrevocable.
Even in 16-bit, there is still destruction – but it’s invisible to the human eye.
MYTH 3: Editing Jpegs in a Raw program
Raw processing programs such as ACR and Lightroom allow Jpeg editing as well as Raw.
So I hear people cry "Hallelujah! I can fix my Jpeg White Balance and Exposure problems!"
I drive a humble Holden hatchback. I could get it serviced at a BMW dealership – would that make it a luxury car? Of course not.
Editing Jpegs in a Raw program doesn’t give you any extra flexibility – it just gives you a different interface. If you’re comfortable with the interface, that’s terrific. Keep doing what you’re doing.
But don’t expect Raw results from low-dynamic-range 8-bit Jpeg data. It ain’t gonna happen.
MYTH 4: 72ppi is web resolution
This is both incorrect and irrelevant.
It’s been many years since monitors were even close to 72ppi. Most monitors are now 100ppi or much higher.
It makes no difference. Web browsers simply show the image at the resolution of the viewer’s screen, with no regard for the embedded resolution. All that matters are the pixel dimensions.
An 800px wide image will display at 800px wide, regardless of whether its resolution is 10ppi or 1000ppi.
MYTH 5: sRGB gives accurate colour on the web
Oh, if only this were true.
Yes, sRGB is definitely the safest option for web images. It’s the colour space that represents an "average" display. But of course, no two screens are the same, and some are wildly different.
If you have (a) a calibrated monitor, and (b) a colour-managed browser, then your web images should look correct … on your monitor. If another photographer looks at your website, and they also have (a) and (b), then your images will look accurate to them, too.
But Joe Public, with his straight-out-of-the-box cheap screen? Forget it. He sees whatever he sees, and there’s nothing you can do about it.
Don’t lose any sleep over it. Just be grateful that Joe is visiting your site, and be comforted by the fact that he is not particularly discerning – your photos look great to him.
MYTH 6: Red cast in skintones
I’ve encountered this one quite a lot recently, and it surprised me. Many people are trying to correct colour problems in their skintones by adjusting the red channel in Curves or Levels.
Skin is red. So it stands to reason that skin can’t have a red cast. Have you ever heard of a blue cast in the sky, or a green cast in grass? Of course not.
Skin can be too magenta (controlled by the Green channel) or too yellow (controlled by the Blue channel).
When editing skintones, the Red channel controls the saturation of the skin. Yes, it has a partial effect on colour, but only indirectly.
MYTH 7: Camera resolution
This is a common point of confusion. Many people see that their photos have a certain PPI value, and think it matters. Often I see the lament "My lab requires 300ppi, but my camera only shoots 240ppi".
As a kid, I remember falling for that old schoolyard riddle "Which is heavier – a ton of feathers, or a ton of lead?". I confidently replied "A ton of lead" before being laughed at (and scarred for life!)
Of course, they both weigh exactly a ton.
Let’s try the same riddle, but substitute camera specs … "Which is bigger – a ten megapixel file at 72ppi, or a ten megapixel file at 300ppi?".
Answer: They’re both ten megapixels. The resolution (PPI value) is arbitrary and completely irrelevant.
MYTH 8: Calling Adobe RGB “RGB”
Aaaaaaaaaargh!!!!!!
I don’t know why this one irritates me so much, but it drives me up the wall every time I see it.
- "Which is better – RGB or sRGB?"
- "How do I convert from RGB to sRGB?"
- "Should I send my images to the printer in sRGB or RGB?"
- … etc …
You probably mean Adobe RGB, and you should say so.
RGB is an all-encompassing colour mode. All photos are in RGB mode when they come out of the camera. But RGB comes in different spaces – sRGB, Adobe RGB and ProPhoto RGB are the most common.
Do these questions make sense?
- "Which is better – fruit or bananas?"
- "Should I drive to the shops in a Toyota or a car?"
Of course they don’t. You can’t compare bananas with fruit, because bananas are fruit. You can’t choose between a car and a Toyota, because a Toyota is a make of car.
It’s bad enough that Adobe have embedded the fallacy that Adobe RGB is better than sRGB – let’s not validate the fallacy by discussing Adobe RGB as if it’s the only colour space in the world.
MYTH 9: Mac is better/PC is better
They’re just different, that’s all. They do the same job, equally well. Get over it, people.
MYTH 10: Keeping highlights below 245
This one is surprisingly common. Often I see people advised to keep the highlight values in their images at 245 or lower, because "Anything above 245 will blow out in print".
This is complete nonsense. Your lab should print pure black at 0, pure white (ie nothing) at 255, and subtle shades of grey from 1 to 254.
If your lab can’t do this, the solution is NOT to reduce the tonal range of your images by 4%. The solution is to find a new lab.
Any lab worth their salt should have their equipment carefully calibrated and profiled. When they print, the 0-255 range in your image is precisely mapped to the 0-255 range of the printer.
If you’re not getting the beautiful white detail in your prints that you expect, go elsewhere.
MYTH 11: "I shot Raw – here is the SOOC"
Straight-out-of-camera, Raw data looks something like this:
11010001100110101110
00010001101110110001
01001101010100010111
01111100101110010011
etc …
In order for us to view the raw data as an image, it needs to be interpreted by your raw program.
Therefore, there is no such thing as a "SOOC Raw image".
So when you’re posting your image in a group for CC, you should state "Here is my image rendered using ACR defaults", or "Here is my image rendered using Lightroom’s Auto settings", etc.