To me Darktable is a video editing software that is used to edit photos. What is the difference between video editing and photo editing software (if any) in terms of grading tools and how they work?
And I see this only with Darktable, proprietary photo software is much more similar to each other while Darktable is more similar to DaVinci Resolve. Even the lingo we use is much more low level, akin to the video color correction and grading than photo editing.
I remember when I went into video editing from photo editing that there was a clear distinction to me that photo editing was very simple and video editing was very hard. Not so much now that I use Darktable. They are both kinda the same. Using the same concepts. Which might be good because it makes my life easier in terms of what I learn and do with Darktable can be easily applied on a lot of video and vfx software.
Should photo editing be a simpler work than video grading and how come the professional photo editing market didn’t ever demand a more powerful software with more control for editing raw files?
One might argue that professionals don’t even use something like Lr but just do their conversion in Camera Raw and edit in Photoshop. is that it?
Well pretty much everything that makes today’s Darktable how it is. The whole scene referred workflow and less abstractions. Much use of channel mixer, Splitting the tones into shadows, midtones and highlights instead of blacks, shadows, highlights and whites. Color balancing is done the same way.
Darktable is just more familiar to me as a video color correction and grading guy than a photo editing guy. I’m not that great in video grading tho but I know my way around a few apps.
they are not all this way, but the higher end suites are generally node-based.
you can branch off in as many directions as you want and sum all the results back together in anyway you like. Any given node can act on/in any colour space you wish. You can use Lift/Gamma/Gain wheels, curves, RGB mixers, contrast and pivot, temperature or other controls all in one node or separated across multiple nodes, or you can use nodes to simply host a plugin or to apply a LUT.
You probably mean the picture pipeline of such a video editor. I’d argue that programs such as Resolve, FCPX, Avid and maybe even Premiere are a tad more than that. All of them are asset managment systems foremost. Your need to cull huge amounts of material, make sequences and nested sequences, variants and have a versioning system for some if not all of the above. Some try to facilitate collaborative workflows with two or three editors working on one project. Resolve extends that to all steps of the production pipeline.
The reason for the complexity difference? I’d argue that you can very well be a solo photographer, but a solo-filmmaker is nigh impossible. But as soon as you can distribute tasks of filmmaking to different people, each of those people can dive into so much more detail. When you have to deal with all of those roles yourself as a photographer, there’s no real incentive to have the most fine grained control over every minute detail of your raw development.
I agree with your assessment. One basic complexity that has to be handled in video are the constant changes in wb and lighting from frame to frame or at least as the video progresses and characters or subjects move. When you edit a photo its a snapshot. You have the static data. You can play with it and of course change it but your base starting point is the same for that edit. In video you need to think about the flow of the scene. If a person turns or moves or enters a room things have to be adapted. I think the software and the need for skilled people to make this happen might exceed what is normally required to edit a single static frame with fixed conditions. Nevertheless in both cases on some level there are users that have both the vision and skill to apply the tools for this, however for some there may be a good technical knowledge but not a good artistic eye and others might have the vision to know what needs to be done however they don’t understand the technical aspects enough to demonstrate that fully. In the end, personally I think video is more complex for that reason, ie it takes a range of talents and the workload is much higher…
I have written both a photo editor and a video editor. (Neither is published.)
The video editor is all about asset management, maybe 90% of the code. True, it doesn’t have fancy features like follow-me masks (eg you want this part of the scene lighter, so we need to track this as it moves from frame to frame). But the code is mostly about: okay, you want these frames from that shot with a slow zoom-in, then those frames (and any shots from that camera need this pre-processing), synced with this sound, then fade up a caption while we play this music …
The image-processing parts of my video editor are comparatively simple.
To some extent, the same is true of the photo editor, but now the assets are mostly internal to the editing process: colour profiles, keeping track of the processing chain, different versions (sizes, crops, zoom-levels, colourspaces) of the image(s) currently being edited, and so on.
Again, the image-processing parts of my photo editor are comparatively simple. Loads of sliders and knobs tweak parameters to the various processes. Not a problem.
On the UI differences between video and still editors: many of these have evolved over time, and can easily be traced back to the (analogue) mixing desks and darkrooms I used nearly half a century ago. (Gulp.) For better or worse, we have to blame history (or established custom and practice) for many conceptual differences.
Focusing on the grading aspect, I would imagine only the math differs, and whatever modules the developer chooses to include, which is also the case with different photo editors. Aurelien has spoken before about his color balance vs video editing color balance - same idea, different math. Photography SHOULD have greater scope for complexity with masking and blending due to smaller file sizes (assuming raw v raw) and shorter rendering times, but that too varies by program. From the little I’ve used After Effects it seems you can perform quite similar operations there to Photoshop (layers, blend modes, masks), which is much different to the node based workflow of Da Vinci resolve. But for some reason unbeknownst to me video graders don’t seem to favour after effects as a grading tool. The language seems to be different. Video graders talk about tweaking the primaries, than the secondaries, which is something you hardly ever hear from photographers. All aspects of colour management are inherent in both.
Funny, as I went the other direction, video grading to photo editing, and video grading was WAY easier, just moving a few simple sliders, compared to layers and blend modes and so on of photo editing. Photo editing seemed much more complex and powerful. There is surely complex and powerful video grading tools out there, I just haven’t looked in a long time. Video editing had better scopes, but photo editing software is catching on to that and incorporating them now.
Well, that’s an interesting story! Would you be willing to share some details and how it looks, why they’re not published, personal projects or for some others that retain the copyright etc?.. I’m curious!
@KristijanZic That’s an interesting … opinion (I try to sound very neutral) of yours. I do not agree at all that photo post-prod is easy and video post prod is hard and can fully assure you that you are dead wrong assuming pros use Camera Raw plus PS and avoid LR.
I’d say that 80% of the professional photographers I know (people who pay their bills by invoicing photos to clients) use Lightroom as their main post-prod tool. 30% use Capture One and maybe 20% use something else. This adds up to 130% because not everyone attaches himself to one program exclusively, I know at least two pros, who use LR and C1 on a regular base.
PS is not used all that much (less than 50% in people photography). There are those who dig deep into the alteration of the subject (puppet-like skin) and those who rather document what they see. The latter get their job done in the Raw-converter without PS or Gimp. I am sure that things look VERY different in product photography, but that’s an area that I have no experience with.
Post-Production in photography and film are somewhat similar and photography has lead the way. I have seen the first digital picture treatment in 1990. It took a workstation worth half a year’s salary and a scanner worth two month on top of that to be able to edit skin imperfections or re-size the eyes. I bought my first Nikon Scanner (with a SCSI interface) in 1997 and could scan my color-slides in 3900 * 2500 resolution. That may be a reason why it took me until 2009 to swing over towards a digital camera. 6 MP (later 10 MP) and horrible noise at 400 ISO never really tempted me to change sides until we had 12 MP and usable quality at 1600 ISO in 2008.
OK, back to entertainment. As a photographer I can really alter every part of a picture today. Change details, add objects, erase stuff: add a different, more dramatic sky, change the color of the dress, erase the lamp-post in the background, add some snow on the mountains on the left, get more light on her face, apply a different color-grading and go softer on the contrast … any amateur can do that today. It will take between 20 minutes and half a day, depending on how well you master your software and how fast and stable hour hard-and software are, but after all it’s just standard stuff.
I’d like to see the video amateur do that.
Cut the stripes, assemble them to scenes, take care of the sound-track to be in sync and work on smooth or hard transitions are things no photographer has to care about. Then comes the part both do (color grading, contrast, general light) but no amateur changes a number of details on at least 24 pictures per second during any arbitrary number of five-second scenes. It’s impossible to do consistently by hand and a workstation capable of doing this in less than one week per minute of footage still costs the better part of a nice car.
Video takes a lot of time. When I come home from a kayak tour with my buddies, it takes me 20 minutes to download the 100 or so shots from the camera into my laptop, open dt, scan for the three or five good ones, develop them and send the pictures to the participants. This is incredible fast compared to the time it takes to download 90 or so minutes of footage from the GoPros, scan it (another 90 minutes) cut the interesting stripes, assemble them, add transitions, create titles, import those, adjust the volume of the sound-track(s), apply color-grading (that takes forever!), export the movie and upload it onto a video-platform. A day for a two or three minute flick, plus I need to send-out the same number of emails containing the link to our flick.
More hours, but not harder or more complicated. Just different work for a different result.
@beachbum thank you so much for the extensive reply. I find it quite interesting.
That’s what I do and that’s one definition or a professional, meaning you are a pro if you get payed.
But I think there is a difference between a “professional” photographer and a professional photographer (who knows his stuff).
And that’s what I’m saying, photography is easier than video editing in that regard. You cannot be a “professional” video grading guy, you can’t just push a few sliders, you have to think a bit.
I’d assume that the most photos sold to private customers by the single photo team get sold on cruise ships and amusement parks. And the people working a camera there are definitely not professional photographers. Those companies look for people without any photography or editing experience. Yet they make millions a year.
I don’t think those kind of “professional” photographers would like Dakrtable. They’d hate it. They would be much happier with Lightroom, just push a few sliders and you’re done.
So who is Dakrtable for then? I really don’t know. It’s not for amateurs. Just like Resolve is not for amateurs. But then again it’s not for “professionals” because they are just amateurs getting payed. But it’s not for professionals either because they may be using Lightroom to just profile the images and they just import it in Photoshop, just Camera Raw quickly and get down to business with layering, adjusting, smart objects, masking etc.
So who is Darktable made for now? Where will Darktable find its place in the industry?
I think we can forget about amateurs and “professionals” but we could potentially attract some professionals with the in dept knowledge of color grading and with big projects on their agenda. I’d also think many from the movie industry would be interested in Darktable to combine it with blender, autodesk software etc in the previz stage of the movie making. Someone might do the grading of the exported exr frames in Darktable to present the Director with the potential grade look.
I’m just not sure where Darktable will get it’s popularity first. Of course, if it ever get’s big popularity probably everyone will jump into Darktable, amateurs included.
Of course it’s similar to you, you came from the film. The only time I’ve developed a film was in school. Actually it wasn’t even a film it was a photo sensitive paper for DIY camera obscura. And even that only happened because my professor of photography and film was obsessed with camera obscura and basically dedicated his life to building and shooting with those.
He can’t, video is harder.
When I’m talking about video work requiring the grading editor to be more professional I don’t mean that he has harder time because of tracking. In today’s world that’s not an issue. If you have enough money you’ll buy BorisFX or something similar and track anything but I’m talking about video requiring the artist to be more knowledgeable and technical. You have to know where that will be shown. Is it for cinema, broadcasting or YouTube etc. And all of that requires you to make certain decisions and correctness. Hence the scene referred workflow. So that’s present in Darktable too, but why isn’t in other raw photo editors? Could it be that the photo industry is generally more laxed about everything and considered more amateurish so it’s ok to just abstract everything and users are perfectly happy with 10 sliders and that’s it?
Everything said above is obviously just my opinion. I’m just thinking who is Darktable for? It’s obviously a great software getting better every day. But where does it position itself? Will it ever achieve “industry standard” success or will photographers just reject it because of it’s complexity and because other software is “good enough”. Because grading feature wise it seems like it’s made like a video editor with the modern color spaces, pipeline etc. but it’s a raw photo editor.
I know this is not popular to think about but I’m trying to clear it up at least for myself.
edit: One note on the current Darktable master. While I personally disable this new feature, this might be just the thing that Darktable needs to cater to the Lightroom and Co. users. I’m talking about the new base tab where they’ve pulled out some of the sliders from other modules and made them accessible in one place.
Darktable is as far from a video editing tool as pretty much any other RAW converter, be it Rawtherapee, Aftershot, Lightroom, CaptureOne, et al.
They all leave out the time axis in their editing process.
Yes, we can copy/paste settings from one image to another or a bunch of them. But we can not define a range, transition, or both on a series of images for any arbitrary setting, filtered by any metadata or manual edit list or image analysis, computer vision and so on.
LR-Timelapse is the tool that is bringing LR into the range of a video editor - well, because it is meant to create videos, d’oh. But it’s not even close to what should be able.
Photographing a wedding over a day? Why not set whitebalance with a curve fitting the day over all images. Then split the indoor sequences - in moving pictures that’s a cut - and set a different whitebalance in a node or whatever. Create another node for that weird lens with the strong vignetting by setting a filter, adjust across all. And so on.
That’s the kind of tools us professional photographers are missing. How to edit hundreds of images efficiently and transparently, so that we can have our classic “nodes” setup beforehand. Or let an assistent set them up, letting us be the creative mind, not the clumsy copy-paste-monkey.
Let me win a multi-jackpot lottery and I will pay the right people the right money to get this going.