I watched a Harry Durgin video lately where he exported an underexposed image out to jpeg then changed it to svg, placed it in the watermarks folder and then used it on another photo that was properly exposed for the foreground. Using a parametric mask with watermarks module, he blended the sky part of the svg with photo and made an excellent blended exposure. Like using gimp or ps with luminance masks and multiple exposures. It would be great if this could become a feature, i.e. export an image to svg into a dedicated folder for your blends. Then maybe even have a module for the svg blending with the image rather than using watermarks. Actually there is now 2 videos on his weeklyedit.com site. Anyone else think this is a good idea
Could you please post the link to the video?
http://weeklyedit.com/
The second video is the 1 I watched but the newest has the idea as well, i believe.
If it were an extra module there would be no need to limit it to svg, the only reason that it is used here is that the watermark module solely accepts svg. However, as far as I know (and I may be wrong), relying on more than the raw file is against dt’s philosophy, with the watermark module being a huge exception.
Edit: That being said, I would bet such a module would have a chance if it would be coded by somebody trusted* by the dt core devs.
*That means, somebody trusted to do the maintenance for it for a long time.
The processing skills of this guy are outstanding, at least to me! Thanks for sharing:)
Dario
Boucman once wrote such a module that lets you use other images. We rejected it because it is – as you already mentioned – against dt’s philosophy and posing too many questions and problems to be worth it. Use GIMP for advanced pixel pushing.
Guess I lost my bet .
But surely if the module used a secomd raw file then it wouldn’t go against the philosophy. Blend a second raw file into the first. Kindanlike using the HDR function but you would have more masking ability.
It would, since it would rely on a 2nd file. The HDR function generates a new input file, what is different.
As @chris said, it would. The problem is basically that there is no easy way to guarantee that the two files will be kept together for future processing, it would be hard to rerun this feature when files are copied or renamed, and probably more that I forgot. In the end it’s too much work for a very fragile feature that you can easily have when using GIMP or similar programs on your exported images.