Help me create a workflow for editing a ton of images for a portfolio.
October 10, 2016 12:37 AM Subscribe
I have to process a ton of jpeg images that will be used for a web portfolio of projects. I have been struggling to come up with a way to do this efficiently (as well as not shooting myself in the foot of the images need to be used in a different format down the road).
I have a huge backlog of images for projects our company has done over the past few years. I need to select and process the images to update our online portfolio. The images are all jpegs, so there is no need to do any sort of RAW conversion. I think it just might be the quantity that is overwhelming me, but I'm not sure if I'm doing this intelligently.
Each project might have a couple hundred rough images. Currently, I sift through these fairly quickly in Adobe Bridge to try and narrow them down to the dozen or so that I'll use for the project. I then copy those selected images to a "final selections" folder.
Then, I edit those selected images in either Adobe Photoshop or Affinity Photo (I just started using the latter because the touch-up tools and features are far better than my old version of Photoshop CS5). I am typically rotating and cropping the images, and then adjusting the levels/curves, colors and doing various touch-ups.
I have been trying to work in PSD or Affinity Photo file formats so that I can make non-destructive changes in case any tweaks are needed later. The images are all stored in folders by project on a shared server, so I can't really use a program like Apple Photos that will suck everything into a library. So, I create a PSD or Affinity file for each selected image. If I just open a jpeg and save it as a Photoshop or Affinity file, why is the file so large? For example, if I open a 2MB jpeg and save it as a Photoshop PSD with no edits, why is it suddenly 20MB?
After editing, I export the final jpeg from Photoshop or Affinity. However, I am having a hard time determining when I should do the resizing. Should I resize the image dimensions of the Photoshop file, or only the exported jpeg? In other words, if I wanted a 960 pixel wide image, would I resize things in Photoshop and export a jpeg that size, or would I export a jpeg that is larger and resize that so that I can always come back and export larger images from the Photoshop file if needed later on? If I scaled the PSD to 960 pixels wide and 72dpi and saved it and we need an image 1800 pixels wide for some other use, haven't I shot myself in the foot? I realize the file sizes would go down, but don't I lose image data?
Finally, it has been a while since I've made images for the web, so I'm not sure what a good size and resolution is these days. When I open a jpeg, it is usually 3264 x 2448. Our website currently has images going up to 960 pixels in width. And I recall from years ago that 72ppi was a typical resolution. Have retina screens changed these old rules of thumb? Should website images or resolutions be larger than that these days?
After all this, I have a folder of all the original rough images, a folder of the selected images, a folder of PSD files, and a folder of the final exported jpegs. Again, it might just be the quantity and our old machines, but I feel like there may be a better way to think about all this.
I have a huge backlog of images for projects our company has done over the past few years. I need to select and process the images to update our online portfolio. The images are all jpegs, so there is no need to do any sort of RAW conversion. I think it just might be the quantity that is overwhelming me, but I'm not sure if I'm doing this intelligently.
Each project might have a couple hundred rough images. Currently, I sift through these fairly quickly in Adobe Bridge to try and narrow them down to the dozen or so that I'll use for the project. I then copy those selected images to a "final selections" folder.
Then, I edit those selected images in either Adobe Photoshop or Affinity Photo (I just started using the latter because the touch-up tools and features are far better than my old version of Photoshop CS5). I am typically rotating and cropping the images, and then adjusting the levels/curves, colors and doing various touch-ups.
I have been trying to work in PSD or Affinity Photo file formats so that I can make non-destructive changes in case any tweaks are needed later. The images are all stored in folders by project on a shared server, so I can't really use a program like Apple Photos that will suck everything into a library. So, I create a PSD or Affinity file for each selected image. If I just open a jpeg and save it as a Photoshop or Affinity file, why is the file so large? For example, if I open a 2MB jpeg and save it as a Photoshop PSD with no edits, why is it suddenly 20MB?
After editing, I export the final jpeg from Photoshop or Affinity. However, I am having a hard time determining when I should do the resizing. Should I resize the image dimensions of the Photoshop file, or only the exported jpeg? In other words, if I wanted a 960 pixel wide image, would I resize things in Photoshop and export a jpeg that size, or would I export a jpeg that is larger and resize that so that I can always come back and export larger images from the Photoshop file if needed later on? If I scaled the PSD to 960 pixels wide and 72dpi and saved it and we need an image 1800 pixels wide for some other use, haven't I shot myself in the foot? I realize the file sizes would go down, but don't I lose image data?
Finally, it has been a while since I've made images for the web, so I'm not sure what a good size and resolution is these days. When I open a jpeg, it is usually 3264 x 2448. Our website currently has images going up to 960 pixels in width. And I recall from years ago that 72ppi was a typical resolution. Have retina screens changed these old rules of thumb? Should website images or resolutions be larger than that these days?
After all this, I have a folder of all the original rough images, a folder of the selected images, a folder of PSD files, and a folder of the final exported jpegs. Again, it might just be the quantity and our old machines, but I feel like there may be a better way to think about all this.
Best answer: So, I create a PSD or Affinity file for each selected image. If I just open a jpeg and save it as a Photoshop or Affinity file, why is the file so large? For example, if I open a 2MB jpeg and save it as a Photoshop PSD with no edits, why is it suddenly 20MB?
Because the image part of the PSD file is a straightforward pixel grid saved with lossless compression, not the lossy transformation that JPEG employs to reduce image file size by a factor of about 10. It's done that way so that you can do an endless sequence of open/edit/save operations on a PSD file without causing the progressive quality loss you'd get from recompressing to JPEG after each edit.
After editing, I export the final jpeg from Photoshop or Affinity. However, I am having a hard time determining when I should do the resizing. Should I resize the image dimensions of the Photoshop file, or only the exported jpeg?
I'd be inclined to keep the PSD in the highest resolution I had available, and resize any exports.
In other words, if I wanted a 960 pixel wide image, would I resize things in Photoshop and export a jpeg that size, or would I export a jpeg that is larger and resize that so that I can always come back and export larger images from the Photoshop file if needed later on?
The second one, more or less. If your image processing tool doesn't let you export a resized JPEG in one step, then make a temporary copy of your PSD, resize that, export as JPEG, then toss the temporary copy. That way, you keep your original full size PSD and your export suffers only one JPEG degradation instead of two.
If I scaled the PSD to 960 pixels wide and 72dpi and saved it and we need an image 1800 pixels wide for some other use, haven't I shot myself in the foot?
Yes.
I realize the file sizes would go down, but don't I lose image data?
Yes. The resized image would contain only (960/3264)2 = 9% of the original pixels. That's more than enough loss to ensure that any print materials you prepared from the shrunken version would look quite bad.
I recall from years ago that 72ppi was a typical resolution.
96ppi has been more typical for the last fifteen years or so.
Have retina screens changed these old rules of thumb?
Yes. Retina screens are typically around 300ppi, though most of them are also much smaller than the typical 96dpi desktop screen.
Should website images or resolutions be larger than that these days?
The best way to judge the coarsest resolution you can get away with for any particular device is to set up a web page, display it in that device's browser, and look at it. Needing a little fleet of test devices is a complete pain in the posterior, but that's web design in 2016.
I feel like there may be a better way to think about all this.
For what it's worth, the workflow you outlined strikes me as completely sound. The only change I'd be tempted to make is to copy shortcuts, rather than actual copies of the image files, when building the "final selections" folder. This would save disk space - you're not actually making any changes to the originals, just saving modified versions in PSD format, so you might not want to duplicate them.
If you keep on working the way you are already, then assuming your "final selections" folder is inside a wrapper folder named for the particular project, you can back up everything you need for that project just by copying the wrapper folder. If you went with my shortcuts suggestion, you'd need a backup of the entire pool of original source images as well, to avoid broken shortcuts. But if all of your work is already on a file server that gets backed up as a whole anyway, that might not be an issue.
posted by flabdablet at 9:15 AM on October 10, 2016
Because the image part of the PSD file is a straightforward pixel grid saved with lossless compression, not the lossy transformation that JPEG employs to reduce image file size by a factor of about 10. It's done that way so that you can do an endless sequence of open/edit/save operations on a PSD file without causing the progressive quality loss you'd get from recompressing to JPEG after each edit.
After editing, I export the final jpeg from Photoshop or Affinity. However, I am having a hard time determining when I should do the resizing. Should I resize the image dimensions of the Photoshop file, or only the exported jpeg?
I'd be inclined to keep the PSD in the highest resolution I had available, and resize any exports.
In other words, if I wanted a 960 pixel wide image, would I resize things in Photoshop and export a jpeg that size, or would I export a jpeg that is larger and resize that so that I can always come back and export larger images from the Photoshop file if needed later on?
The second one, more or less. If your image processing tool doesn't let you export a resized JPEG in one step, then make a temporary copy of your PSD, resize that, export as JPEG, then toss the temporary copy. That way, you keep your original full size PSD and your export suffers only one JPEG degradation instead of two.
If I scaled the PSD to 960 pixels wide and 72dpi and saved it and we need an image 1800 pixels wide for some other use, haven't I shot myself in the foot?
Yes.
I realize the file sizes would go down, but don't I lose image data?
Yes. The resized image would contain only (960/3264)2 = 9% of the original pixels. That's more than enough loss to ensure that any print materials you prepared from the shrunken version would look quite bad.
I recall from years ago that 72ppi was a typical resolution.
96ppi has been more typical for the last fifteen years or so.
Have retina screens changed these old rules of thumb?
Yes. Retina screens are typically around 300ppi, though most of them are also much smaller than the typical 96dpi desktop screen.
Should website images or resolutions be larger than that these days?
The best way to judge the coarsest resolution you can get away with for any particular device is to set up a web page, display it in that device's browser, and look at it. Needing a little fleet of test devices is a complete pain in the posterior, but that's web design in 2016.
I feel like there may be a better way to think about all this.
For what it's worth, the workflow you outlined strikes me as completely sound. The only change I'd be tempted to make is to copy shortcuts, rather than actual copies of the image files, when building the "final selections" folder. This would save disk space - you're not actually making any changes to the originals, just saving modified versions in PSD format, so you might not want to duplicate them.
If you keep on working the way you are already, then assuming your "final selections" folder is inside a wrapper folder named for the particular project, you can back up everything you need for that project just by copying the wrapper folder. If you went with my shortcuts suggestion, you'd need a backup of the entire pool of original source images as well, to avoid broken shortcuts. But if all of your work is already on a file server that gets backed up as a whole anyway, that might not be an issue.
posted by flabdablet at 9:15 AM on October 10, 2016
Response by poster: Thank you so much for such a thorough answer to my rambling question! This really helps. Now, I finally get why those files get so large. And I just discovered that I saving the history in Affinity Photo was adding another 10MB or so to each file.
And that all makes sense now for sizes and resolution -- I just wanted to make sure I was understanding things correctly, so this is all helpful.
As for automation, I have tried Hazel a few times, but I always end up accidentally creating these never-ending nested folder nightmares in Hazel. I now get by with Automator and Keyboard Maestro, and I do use the automation options available in Bridge and Photoshop for a lot of repetitive tasks like renaming and some saving/resizing (I haven't checked if you can automate Affinity yet).
posted by This_Will_Be_Good at 6:18 PM on October 10, 2016
And that all makes sense now for sizes and resolution -- I just wanted to make sure I was understanding things correctly, so this is all helpful.
As for automation, I have tried Hazel a few times, but I always end up accidentally creating these never-ending nested folder nightmares in Hazel. I now get by with Automator and Keyboard Maestro, and I do use the automation options available in Bridge and Photoshop for a lot of repetitive tasks like renaming and some saving/resizing (I haven't checked if you can automate Affinity yet).
posted by This_Will_Be_Good at 6:18 PM on October 10, 2016
« Older Wisdom teeth partially left in (or my molars are... | Need recommendatios for a low voltage (12v ideally... Newer »
This thread is closed to new comments.
posted by rippersid at 3:02 AM on October 10, 2016