No right-click allowed. © Mareike Keicher

Tag : RAW

Foto von Luis Quintero von Pexels

How to reduce file sizes – from RAW to PSD to JPEG

Large amounts of data are generated during post-production – previews, composites, layered and non-layered versions, revisions, etc. – also for the client. These files consume storage space and must be backed up. Ultimately, we have to keep an eye on our files and reduce the file size to a minimum – of course, without loss of quality.

Minimize RAW files

Dotphoton Raw (now Rawsie) offers lossless compression of RAW files with up to 80% file size reduction.
Check out their website to get an insight into the list with supported camera manufacturers. There is also the offer of 30 images a day for free, which is fantastic!
If you are a full member of Cherrydeck, you can also get an excellent deal with Dotphoton Raw from them: With the Full Membership, you get 50% off of their subscription.

One important thing to know: the files are converted to DNGs (= Digital Negative Image File; remain readable in Capture One). The DNG format was invented by Adobe to have a pseudo-RAW format. Some people argue that the DNG format is better because it allows for any future software to read it since it is an open standard.
If you carry out this step of conversion retrospectively, changes made in Capture One no longer fit 100%. It is therefore advisable to carry out this step immediately after importing the files – consequently, the first step. The files do not replace the RAW data. You can therefore compare directly where the differences are. Possible changes include lens distortion, image vignettes, skin tone changes, or the absence of .xmp changes.
After performing some tests, however, I conclude that the changes are positive – also apart from the significantly smaller file size.

There is also an option to optimize Lightroom catalog folders from Dotphoton Raw (also included in the free version), which we didn’t test because we use Capture One instead of Lightroom. If you made any experiences here – please feel free to comment in the comment section below.

Minimize PSD/PSB/TIFF files

Sometimes you wonder about the file sizes. Despite minimal, quick processing, the files can reach 20 to 50 times the size. So here are quick tips of our experience to help you out.

In general:

    • Try to work with setting levels and masks. If you create a mask and do little editing there, it will only take up a small space. Complex masks require complex space.
    • Avoids all techniques that make copies of the image (which you should always do for the non-destructive workflow). The same effect occurs when sharpening using a high-pass filter: a complete doubling of the memory requirement. If you use copies of layers, unused areas should be deleted.

Here are several options to minimize your PSD/PSB/TIFF:

    • Try to work with 8-bit if the file does not include many or large gradients (then, of course, use 16bit)
    • Use the Retouching Toolkit‘s actions “Delete identical pixels” and. “Remove help layers” regularly.
    • Try to do local and global Dodge together & local and global Burn together (advanced level)
    • Don’t use frequency separation on skin. And try to use it as less as possible on other surfaces in your file. It will enlarge your file very quickly and heavily.
    • Try to do as many corrections as possible in one adjustment layer instead of several ones (if you don’t need masking and it fits the image). E.g., in selective color,  adjust reds, greens, and yellow tones in one instead of 3 different adjustment layers.
    • Use as few smart-objects as possible: only where they make sense, i.e., if you have to make subsequent corrections or adaptations have to be taken back. Examples are Liquify or the Camera RAW Filter. In case of sharpening or noise, it makes sense to do it without a smart-object on a separate layer/layer duplicate and store the settings in the layer name as a reminder.
    • Use the integrated filter options of Photoshop to find and delete empty layers.
    • Apply masks when the job is approved.
    • Corrections in Capture One need less storage. If the workflow allows it and some corrections can even be better carried out in Capture One, you should choose this way.
    • Export TIFFs via Capture One will delete paths and channels, which reduces your file size. This is useful if your client doesn’t need them anymore.
    • In the beginning, work with PSDs instead of TIFFs for automatic recovery from Adobe in case your Photoshop has a crash. Then, compare how much the same image with layers as TIFF or PSD needs. Depending on the picture, very different results can be achieved here.

Composites – Working with linked Smart-objects:

    • If the same external file is required several times in the document,
    • It often makes sense to edit the elements in the image individually in different documents and to link them together in one document (e.g., compositings – this also makes working in a PSD significantly faster). Here you should make sure that the color profiles are identical.

Minimize JPEG files

We highly recommend buying JPEGmini at a one-time price (no subscription model). We have been using JPEGmini for a long time now and believe that their motto “Reduce file size, not quality” is 100% correct. At the moment, we don’t want to miss their service.
This is a handy tool, especially for discussing revisions with clients – fast upload times & fast download times – or for reducing file sizes to a minimum for Social Media and website exports.

They also have the option of Lightroom, Photoshop, and Capture One Plug-ins.

File structure

Overall, it is recommended to keep the following files at the end of a job:

    • The RAW files used (delete additional ones!)
    • An open file with non-destructive layers (working file): PSD / PSB
    • A final file without or merged layers: TIFF (lossless compression: 16-bit files -> ZIP compression; 8-bit -> ZIP / LZW)
    • JPEG exports of the final result (website, social media, specific customer exports as reference for future jobs)

All files should contain the version number.


Do you have any suggestions, additions, is this post out of date, or have you found any mistakes? Then we look forward to your comment.
You are welcome to share this post. We are very grateful for every recommendation.

2-Point Curve as best option to do Dodge&Burn

Dodge & Burn – Different methods compared

Dodge & Burn is a subject where opinions differ. There are many ways to make areas in the picture lighter or darker. We quickly discuss the different methods and test them for suitability – there are sky-wide differences.

Dodge & Burn Tools (DE: Abwedler & Nachbelichter)

Photoshop has always offered two tools for Dodge & Burn. Surely everyone has experienced this situation: take the Dodge tool and painted around in the picture, oops done too much, used the Burn tool to compensate, and the final result: light and dark patches in the image in which nothing would fit anymore. Many put these two tools in the “not to use” drawer.

It gets much better if you create an additional layer, fill it with neutral gray (50% gray) and switch it to the “Linear Light” blending mode:

50% gray layer

A 50% gray layer in blending mode “Linear Light” initially does not change the image. But if you make the layer lighter or darker — of course also partially — the picture also becomes lighter or darker. The advantage of the 50% gray layer is that we can also work with regular brushes, and if something went wrong, you could always paint over with 50% gray to get back to the original state.

Some people use the blending mode “Soft Light”, which has a more subtle effect (hard to push it really dark or light). Other than the layer has an impact on saturation, which should be taken into consideration.

Pro Tip: You can also leave the layer empty and paint with black or white – the effect is 100% the same and saves storage space.

Multiple RAW conversions

What is sometimes helpful is to use different RAW conversions.
One for the skin tones, one for the background, one for the hair — depending on what the picture requires. These individual RAW developments are loaded as layers in Photoshop and can then be masked accordingly. Of course, there are a lot of options here, because RAW development offers countless parameters — but this technique inflates the Photoshop file properly and can only be used at the very beginning. So you have to be sure of your decisions here.

Dodge & Burn Curves

Two curves – one for Dodge and one for Burn – are the optimal solution. Masking the effect in and out with white and black is very simple. Switching between layers is pretty easy with the following shortcut
Windows: Alt +, (down), or Alt +. (up)
Mac: Option +, (down), or Option +. (up)

However, the correct usage of curves for Dodge & Burn needs to be learned. There are many people out there, pushing and pulling their curves by their current mood. In addition to luminance, curves can also influence saturation and contrast. Tiny differences can make big differences.

Conny Wallström (a photographer, retouching-teacher, and software developer based in Sweden) tested various curves over a more extended time and compared them with the results of exposure levels of the raw converter Capture One.

The idea for that process was that a raw converter such as Capture One reproduces the most natural and realistic way of dealing with color with different exposures without simultaneously causing color problems and color shifts.

The result of the research was a 2-point gradation curve that he integrated into this retouching toolkit. So if we use the following Dodge & Burn curves, we can assume a natural, realistic result. At the same time, we work consistently and time-efficiently, since it is an action in Photoshop that always works in the same way.

But that’s not all. Many who already use this retouching toolkit know that this action creates two folders and that in addition to the special curves for Dodge and Burn, hue/saturation corrections are included.
Here is the reason for the hue/saturation corrections:
When we take an analytic look at a portrait picture, we find out that by increasing luminance, the saturation tends to decrease; with decreasing luminance, the saturation tends to increase. If we only lighten an image based on luminance, the proportion of saturation in the now lighter areas is suddenly too high (vise versa). The use of such hue/saturation corrections is, therefore, a measure against saturation problems caused by Dodge and Burn to save editing time in retouching.

Such hue/saturation corrections make the most sense when processing skin (skin build-up, blood flow in the skin).

Of course, this cannot be generalized 100%. Every image requires slight adjustments to the hue/saturation correction layers due to other camera manufacturers and raw formats. Therefore, a recommendation is to adjust those layers slightly for each image. However, one cannot avoid looking at the saturation in the image afterward (at least briefly).

Hidden Gems / Tips and Tricks

Pro tip 1: Separately adjusted curves for very granular, high-contrast structures such as make-up (see example below) are useful to avoid flattening the contrast.

(Photo by 𝐕𝐞𝐧𝐮𝐬 𝐇𝐃 𝐌𝐚𝐤𝐞- 𝐮𝐩 & 𝐏𝐞𝐫𝐟𝐮𝐦𝐞 from Pexels)

Pro tip 2: Dodge & Burn tools — these work perfectly for eyebrows. If you paint them on an empty layer, you can make individual hair lighter and darker for a realistic result. You will only affect the painted hair, which is very useful.

Don’t forget Dodge & Burn is much easier by using Wacom tablets.


Do you have any suggestions, additions, is this post out of date, or have you found any mistakes? Then we look forward to your comment.
You are welcome to share this post. We are very grateful for every recommendation.