Retrobatch and Image Optimization

I have some questions about batch processing in general and Retrobatch in particular.

I am new to batch processing. The only processor I have worked with so far is PhotoBulk. I was very impressed with how quickly and elegantly it resized photos. The version I am working with has the ability to resize only or resize AND Optimize.

I have also watched as many videos as I could find about Retrobatch.

As near as I can tell they are similar in function except that Retrobatch can create up to 8 different image sizes and additionally has some extra functionality with respect to cropping for device size and other composition tricks.

I may have this wrong but Retrobatch does not appear to have any built-in image optimization other than serving up the optimum size image for the device that is looking at it.

Is this observation accurate?
If so, is additional optimization necessary and where in the cycle would this occur?

I am getting ready to create a social media campaign for my company. We may embrace Instagram, Facebook, Twitter or whatever else comes along. Each of these platforms presumably have optimal size image content.

Could I use the results from Retrobatch to create a stack of images for each of these platforms?

I’m surprised there was no answers for this post so far…

Your observation is correct. I think it’s the right decision on the part of Retrobatch developer. That way, you can use the app of your choice for image optimization. Although it would be nice if there was a facility to automatically open such app. In my copy of Retrobatch ‘Open in App’ is grayed-out, for some reason. I just started using Retrobatch recently, so it’s possible I am missing something.

So, first, I use Lightroom to generate the largest size version of an image(s), put that image in Retrobatch ‘read’ folder, run the workflow, put the content of ‘write’ folder into ImageOptim and run all photos (either JPEGs or PNGs) through it. The whole process from Retrobatch to ImageOptim takes me a minute or so.

I can’t answer your question about Facebook, Instagram, et al. Personally, I am completely disconnected from these digital gangsters (that doesn’t mean I would like to persuade you to do the same).

P.S. Of course, you can set sizes of processed images in Retrobatch according to your specific needs.

Thanks Rob.

I really like your description “digital gangsters”.

I too have never participated with these people.
As I get older though I am starting to feel little like Ayn Rand.
(She wrote Atlas Shrugged but in the end signed up for Medicare like everybody else.)

I bought a copy of Retrobatch but have not yet spent any time with it. It was not as intuitive to work with as I thought so I just set it aside. (They did, however, very graciously offer to answer questions about how to run it.

The logic of serving the right size image to the device sounds really smart though I am not terribly certain how important image fidelity is to me. The customer for my website is someone who will stay up till midnight in their pajamas because they really really want to get their kitchen right. They are hungry for information and will endure a little pain to get it.

Right now I am using PhotoBulk for image optimization. It’s really fast and I like how it automatically puts the truncated images back into the folder from whence they came. To be fair I should do a comparison with ImageOptim to see which one I like best.

If you would like a jump-start in preparing Retrobatch for work, I could send you my workflow-template file, which I made for 3 different image sizes that work best with my family portal. You could just open that workflow file in Retrobatch and use it as is, or add more/different sizes and edit it to your own needs. You’d have to create on your computer a ‘read’ (input) folder and a ‘write’ (output) folder for Retrobatch.

I am a photographer by passion, so image quality is important to me. But, after some experimentation and comparison, I came to the conclusion that optimization at the level of 30%-50% of the original quality is something I can live with. And that saves a great amount of a file-weight. I use ImageOptim for that, because it proved to be the best for my needs and it is incredibly fast. Never had any problem with it.

Do you run the same amount of compression for all sizes of retrobatch output/

Do bigger images display at comparable fidelity to smaller images?

Yes, same amount of compression. That is, because bigger images used on desktops are viewed at larger distance from eyes than images on phones. So, the perceived sharpness is supposed to remain about the same at any device size. Even so, I can see a bit less sharpness on desktop than on smaller devices. On the other hand, I don’t want to make desktop-bound images less compressed, because they tend to be quite heavy even at 30% quality.

The above pertains mainly to full-width banners. All editorials that are made to the maximum of 980px (in my case) are less heavy and therefore could be compressed to a lesser degree.

All this is—of course—subject to experimenting and choosing your own formula…