1. Images for the web should be sized at 72 dpi. First of all, “dpi” refers to dots per inch, and computer screens have pixels, not “dots.” But “72 ppi” is also false. Pixels dimensions are the only criteria for computer images. 1200 pixels are 1200 pixels, whether they be 1/inch or 1200/inch. You still have 1200 pixels. Say you have an image that’s 1000 x 500 pixels at 72 ppi. How many total pixels is that? 1000 x 500 = 500,000. Resize to 1000 x 500 pixels at 300 ppi. How many total pixels is that? 1000 x 500 = 500,000. Exactly the same. If you’re resizing web images to 72 ppi, you’re simple adding a useless step to your workflow.
2. Mac monitors use 72 ppi while Windows ones are 96 ppi. Sorry, this is a myth. If it were true, a monitor would have to change resolution depending on whether it was connected to a Mac or a PC. Want to know roughly what the resolution of your monitor is in ppi? Measure the horizontal width of the screen and divide this into the horizontal pixel dimension at the monitor’s native resolution. My laptop’s screen is 1920 pixels wide, and measures about 13.5 inches horizontally. 1920/13.5 = about 142 ppi.
3. You should set Adobe RGB in your camera if you’re shooting RAW files. While this might affect the image displayed on the camera’s LCD, it does not directly affect a RAW file. After all, if Adobe RGB were actually applied, the file would no longer be RAW data.
4. For the best RAW file results, set a specific white balance in the camera. Don’t use Auto White Balance. RAW files have no white balance. Just as I said above, if a white balance were applied the file would no longer be RAW. A specific white balance is only set when the file is processed into a standard graphic file format such as .psd or .tiff or .jpeg; that is, when it is no longer a RAW file.
5. This RAW image is how it appears right out of the camera. Not true. A RAW image has to be rendered in some way before you can even see it. Exactly how it is rendered depends on the default settings of the RAW conversion software you use.
6. You can evaluate exposure by looking at the image on the camera’s LCD. You can adjust the LCD’s brightness on almost all DSLRs, so exactly which level of brightness would be “correct?” Sorry, not true at all. For that matter, the camera LCD most certainly is not a color corrected and calibrated monitor. You can evaluate composition; you cannot evaluate color or exposure. You definitely should use the histograms for exposure information.
7. Always use a UV filter to protect your lens. From what? Dirt and fingerprints? Then you must take the filter off for every shot, otherwise you’re shooting through a dirty, fingerprinted filter. Use one for “protection” only if you can state from what it is you’re protecting the lens. Salt spray? Yes, this might be an answer, but I live about 100 miles from the ocean and on the other side of a mountain range, and if there is salt spray here, protecting my lens will be the least of my worries.
8. 12 frames/second is better than 10/frames per second. In what way? Neither one guarantees you’ve caught the peak moment. Consider this: let’s assume a shutter speed of 1/1000 second. 12 frames/second captures 12/1000 of the action, and misses 988/1000 of it. Holding down the shutter button at the highest frame rate yields lots of images, but not necessarily the one you wanted.
9. Always underexpose a half-stop to richen the colors. Intentional underexposure with digital cameras is one of the worst things you can do. It simply adds noise.
10. Professional photographers get all their equipment free from the camera manufacturers. Boy, do I ever wish this were true, but it isn’t. For that matter, I wish it were true for cars and houses also.
11. All information on photography forums is true. You might remember back when there was an actual discussion about using Scotch® tape to clean camera sensors. I’m fairly sure someone fell for this, and actually tried it. We all know that it’s not Scotch® tape you should use, but duct tape, right? (Well, I read about using duct tape on the Internet, so it must be true.)
To think, for years I under exposed my digital images, it was a carry over habit from my slide shooting days. Live and learn, that’s why I read blogs like this.
You’re gonna end up getting sued by some fool over that duct tape joke!
Yeah, you’re probably right. So…IT WAS A JOKE, IT WAS A JOKE. DON’T DO IT.
It’s so refreshing to see someone with years of creditable experience set some things right.
Well done John.
But I’m sure there will be those that disagree and continue to do things as they have always done.
Great information. However, from a web designer’s point of view, I have to disagree with #1. It’s best to resize the resolution of your web images to 72 ppi because it creates a smaller file size, and therefore allows it to load faster online. It’s been shown that slow load times lead to higher abandon rates…meaning many people won’t wait around for large files to load. Web images don’t have to be as high-res as print images, so it’s best for your visitors if you optimize those pictures…especially if you’re posting many on one page or website.
Sorry, but that’s just not true. Pixels are pixels. Yes, you need to resize images for the web. But a smaller file size is the result of fewer pixels, not how many pixels there are per inch.
I have to admit, my first reaction was exactly the same as Kari’s. So I tried it in Photoshop. I opened a RAW file and sized it to 1,024 x 683 pixels at 600ppi and saved it as a JPEG. I did the same again but this time at 10ppi and was amazed that both files were the same size, 459KB. The amount of times I have read and heard that screen resolution should be 72 or 96ppi. I pays to think things through!
Great post. Many of these myths I seem to run across almost daily on the web! I began learning photography primarily from taking classes starting in 2001 at a community college, reading a couple of your books (most notably your Nature Photography Field Guide), and using common sense (which is also a cornerstone of your teaching in my opinion). I didn’t get involved in the online photo ‘community’ until 2010, and when I did it was a bit of shell shock, a bizarro universe where everything is upside down and ‘knowledge’ is often more akin to witchcraft and sorcery! 🙂 So much misinformation on the web…
Interesting post. As always I learned something new.
Thank you, John!
And, by the way, 1920/13.5 = about 142. The resolution of your laptop’s screen should be 142 ppi.
Thanks John, this was helpful
Great stuff John. I work a lot around salt water and never have any type of filter especially on any intermediate telephoto lens as they often hamper AF.
Thanks for busting so many myths. For me, first and foremost, I am removing that 72 dpi step from my workflow.
Thanks Arthur for posting this link in your blog.
Ah, but here is one reason why in certain circumstances I set Lightroom to export my RAW to jpg at 72 ppi… because I like to post my pictures on Facebook. I know some people hate Facebook, but if you like to use Facebook you have to be aware that their uploader will shrink your photo and change the dpi without asking. And it does a rotten job of it. Your photos will get weird aberrations, pixelization, and color casts. If you export a jpg solely for upload to Facebook you want to set it to sRGB, 2048 pixels on the long edge, and 72 ppi. Lightroom does a great job of shrinking your file and changing the dpi without screwing it up so when you post the picture it still looks good. The only thing Facebook does to the picture at that point is run a lossless compression algorithm to try to squeeze the file size down, but that usually doesn’t freak the picture out too much.
If you suspect that any program you are using to upload your file to the web is doing some kind of conversion in the background, you should do that conversion yourself with Photoshop or Lightroom since they probably have better math.
Since I don’t do Facebook, I cannot directly answer you. But, the only way to end up with a smaller file is to reduce the total pixel count. PPI has nothing to do with pixel count. To use your numbers, 2048 pixels at 72 ppi is exactly the same number of pixels as 2048 pixels at 2048 ppi, or 2048 pixels at 1 ppi. They all add up to 2048 pixels, and file size has not changed at all.
Thanks again to both you and Art Morris. There have been a few times when I’ve submitted an image for publication when the requirement was “72 dpi (not ppi).” I’m I right to assume that (a) there’s no difference, and (b) no matter how much sense it makes, or not, I should follow the specifications?
John’s books are always at hand, jic.
False statement no 7 must have been written by someone who has never caught the end of a lens and damaged the filter thread or scratched the glass.
No 11 is not even 50% correct, most is uninformed opinion!
“False Statement No4 is misleading. The white balance of an image opened in Adobe RAW is based on the white balance set in camera (certainly for Nikon and Fuji cameras) and if auto was set on the camera the colour temperature will vary. Set your camera to record in RAW, white balance to auto , shoot in doors by lamp light , shoot outdoor by sun light and look at images in ACR and you will see.
It is true that colour temp can be changed later with a RAW image and can’t with a jpeg (although open a jpeg in ACR and use the slider, can be useful.
Yes, if Lightroom/ACR has default WB set to “as shot” and WB in camera is set to “auto,” the appearance in the raw converter will vary. But this is not a “fixed in place” WB. The image has to be rendered some way so that you can see it. You can start at “auto” or any other setting in camera; it simply doesn’t make any difference. WB is only set when the RAW file is actually converted into an editable file format.
I just tested what changing the white balance does to a RAW image. The WB setting does change the white balance of the recorded RAW image when opened in Camera RAW. Even though this can be easily adjusted, leaving the WB set to Auto seems to work best for me. One less thing I have to think about when shooting.
I agree, especially for outdoor nature photography, where the light is changing all the time. You set the WB when you process the image into a true file format.
John – Another great post. I’ve learned a lot about film from your books and now learning about digital from your blog. Keep up the great work. BTW – Math on #8 is off a tad – 12 fps captures 200 of 1000 (not that it matters)
Please ignore my comment on the math. Forgot the shutter speed 🙁
On the otherhand, shutter speed is driven by aperture and ISO not the speed of the moving object (unless I want to capture a speeding bullet).
Thanks – you’re getting people to think, I’m sending more over…
Hello John, thanks for the great post! Good stuff here.
I would like to point out one thing about #4. You state that setting white balance in the camera has no impact on the RAW data. In one respect, this is true…RAW data is a digital signal representing what was exposed by the sensor.
HOWEVER…changing white balance in-camera affects the METERED EXPOSURE, and changes the distribution of exposure for each color channel in the camera. If you use manual exposure, this should be less of a problem, because you will probably be intimately tuned into the histogram and the EC scale. If you are a manual fiend, you will probably also be tuned into your white balance, and should be keeping it “correct” for the given illumination anyway.
There can be problems when using AWB. For one, white balance can shift from photo to photo, especially when shooting on days with inconsistent lighting (say landscapes with patchy clouds overhead). Without using something like a gray card or a color checker preliminary to each shot, the discrepancies between photos shot with AWB can be extremely difficult to correct and normalize in post, even with a powerful tool like Lightroom.
For any given time of day outside of the transitions between sunrise and daytime, and daytime and sunset, using a fixed AWB setting appropriate to the illumination you are working with can eliminate the quirky shifting of WB, eliminating it as something you need to correct in post. Normalizing the color balance and quality of your photos tends to be easier with more consistent WB strait out of camera.
Around sunrise or sunset, AWB can be a very useful tool, in that it will automatically compensate for the radically and rapidly changing light…however the color balance discrepancies will still generally be present in your photos. If you want consistent results, it might be better to simply be attuned to the changing light and manually set your WB in-camera as appropriate as the light changes. You will then have sequences of more consistent shots to work with in post.
Now, I shoot landscapes less often, and birds and wildlife more often. I am frequently shooting sequences anywhere from three frames to as many as 10-15 frames. With AWB, each of those frames can have slight to moderate inconsistencies with color balance. This may not be as much of an issue with landscape photography, but is still something to consider. As such, I would not necessarily call #4 a pure myth…there are reasons to use a fixed WB setting, even with RAW.
I guess I’d also like to point out that in #8, while 12 fps does not “guarantee” that you get exactly that right moment more than 10 fps…the higher frame rate does increase your chances of getting a better frame.
It is not about guarantees…frame rate never is nor has been. It is really about probabilities. A higher frame rate increases the probability of getting the best shot possible. Does that mean you will always get the best shot possible with 12 fps? Certainly not. Does it mean you will get more ideal shots on a more frequent basis than if you had 10 fps, or 8 fps? Sure!
If the myth you are addressing is purely about the notion that 12 fps will “guarantee” a better shot, then it is indeed a myth. But I don’t think anyone who knows their photographic trade truly believes any frame rate will actually guarantee anything.
The myth is that a higher frame rate makes a “better” camera. Frame rate, in an of itself, is just that: the frame rate. Would a high frame rate, on a camera with limited dynamic range, be better than one with a lower frame rate but wider dynamic range? High frame rate might give more posssible images, but if this is important then a camera than offers 100 frames/second, or 1000 frames/second, would be even more useful. Too many photographers think that holding down the motor drive button is the answer. No true at all. Remember Cartier-Bresson, and the decisive moment? His camera had no motor drive at all.
It’s very easy to make a series of shots taken with AWB have the exact same white balance, without using a gray card or color checker in the field. In Lightroom or ACR, take one of the images and set the WB to a different number. Let’s say AWB yields a temp of 5000. Change that by one digit (or more), and then synch all the other images. They now all have the exact same white balance.