Comic creators: how is your workflow for rendering Scenes?
ameyn
Posts: 3
Hello,
after about 200 hours of videos about DAZ Studio, I am now at the point of hammering out the first steps to bring everything together: background, props, figures, poses, you name it.
So, as it seems, most information out there is about creating/rendering a single picture. But my goal is a (not animated) comic. So…10-30 pictures for one scene.
Now, I would create my scene, save is as “001 – Scene1”, render it, change it a little, save is as “002 – Scene1” and so on. With this method I would end up with a bunch of files and no overview at all.
Example:
First Scene is taking place in a kitchen.
First picture: one person is sitting, the other comes in, wide camera angle. “Hey Dave”
Second picture: close shot of person one, now standing at the fridge, “Aren’t you supposed to work now?”
Third picture: close shot of person sitting, finger pointing to person one, “I took the day of.”
And so on, and so on.
At the end you end up with 20 pictures, all from the kitchen, but figures, gestures and the camera change in every picture. The rest (kitchen/inventory) is fixed.
So, for all the creators of comics out there: how is your workflow to organize that? Do you end up with 20+ single DAZ-files or do you have another way to organize your workflow?
CU Albert
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
I don't do comics, but using the timeline seems like a good way to handle this situation. Set keyframes for each shot at intervals of 10, one for each shot. You can set up multiple cameras, or you can also add position keyframes to the camera as well. That can all be done in one file.
I am not sure if your saying the layout is going to be a video or traditional comic book or graphic novel style layout. If it's a video that's fine, if it's going to be paneled novel then there is a lot more to consider to do it correctly. A comic page with panels is a work in itself, not each individual panel, the direction and angle of the shots depending on where the panel is located and which page (left/right) determines the direction a character and scene should typically be facing. Typically all eyes on the page should generally point towards the inner seam of a traditional left/right page. There's also a lot about how text and and borders are done, borders can be a work of art in themselves. Way too much to cover here but one can find very detailed information online and many tutorials.
Good point. The result should be a "graphic novel style layout". And I'm aware of what you said.
For me a graphic novel is nothing less than an approach like the cutting of Alfred Hitchcock’s shower scene. I don’t know if I will ever reach this high standard, but a graphic novel is like a movie, but told in pictures. So, nothing less than this challenge coming up next. (I also watched about 50 hours of “Movies: Why Spielberg was using the ‘Axial Cut’ / How Alfred Hitchcock created tension”) So…challenge accepted! :-)
As for creating Visual Novels or 'Comics' as you said, I used to use the similar way just like what you're doing now. But a couple of months ago, I started to use the product Camera Magic and its add-ons. At least to me, it's been a better way, so far so good... I can manipulate the storyboard just in one scene file by adding 3 - 4 cameras and storing different poses, shapes and object shown/hidden status to different cameras just in one scene. In this way, I don't have to open many scene files at the same time for modification. Then well name the scene files as well as render files for easy organizing ,reviewing, post-work, etc...
I do the same, I set up the scene, save the scene as Pagexxpanel001, and render. Then set up the next scene based on that one, change pose, camera angle, etc, and resave the scene as Pagexxpanel002, ad infinitum.
So, if you forgot a prop you have to add and align it into all of your saved scenes.Dang... .
That's true but sometimes such a rework is unavoidable... esp. when you have some new ideas. But still, the less scene file you have, the less effort you're gonna make...
It's better to have 3 - 4 renders per scene file. That's my way...
More than a decade ago, I used to tell my stories on the timeline in Poser the way I think you mean, but after a break of some six years, when I started using Daz Studio, I tried using the same technique but the timeline in DS was just not built for it.
In DS one had to pin everything in a frame to keep it from changing with changes in other frames and that increased the size of savefiles to insane amounts, the biggest one I had was over 3GB's with just some 20 separate scenes. It took more than an hour to open that file and the same time to save it and when I lost the file because I let DS save it while closing and there wasn't enough space on the drive, I abandoned the idea and started saving each scene (frame) as individual scenes.
No big deal. Just throw it in the scene and re-render. Copy and paste the translations and rotations when adding it to the next scene.
If you forget something, throw it into the scene and use spot render to patch your existing image. Saves a ton of time. The number of times we've done it where we've spotted an intersection between a finger and some object after a 2 hour render has finished...
Open your Tools panel after selecting Spot Render and set the option to Render in New Window. You'll get a full sized canvas with just your defined area actually rendered which can be saved and then overlayed in Photoshop or image editor of your choice. It turns a minor mistake from a big wait into a roll of the eyes and a five minute job.
You might consider saving your frames as .pngs and editing them in Clip Studio Paint, as it was designed for comic creation. It helps alot if doing so to save your figures and set to seperate images/layers, if your doing a comic, your don't need to spend all the time needed for IRAY Renders, and their are comic/noir shaders in the DAZ store.
Thanks for the help everbody. I'll give the timneline a shot. If anything comes out, I will post ist here.
...and Clip Studio Paint looks good, too...
Here's how I make my comics. They're way too far outside the TOS to even link to from here, but here's a sample page
First, I sketch out how each page should be laid out, so I know how big each panel needs to be (I use a grid layout, so each panel is either full, two-thirds, half or one-third the height and/or width of the page, this gives a wide range of possible layouts, but I always know the exact numers I need).
Then I start work, making a sequence of duf files to complete the story. Let's suppose the stroy's called "The Day Off," I'll name the first scene dayoff01.duf, place the camera(s) and do the render(s), then move the characters around and save it as dayof02.duf with new camera positions.
I name my cameras Camera A, Camera B, etc. And the renders I make are called dayoff01a.png, dayoff01b.png and so on - so I know the scene and camera each shot came from. Where possible, I shoot multiple cameras in each scene (though there's no particular cost involved in having one camera per scene). So, for your example:
I set this up as dayoff01, and render that wide shot as dayoff01a
Change the position of the actors, and save as dayoff02. Camera A is pointed at person 1 for the first panel, and rendered as dayoff02a, Camera B is pointed at person 2 for dayoff02b. Sometimes I'll shoot things out of order if it's convenient to squeeze an extra camera in pointing somewhere not covered by the others.
Once I've got all the shots I need, I do whatever postwork I need, and then make them into a comic using Comic Life.
I use timeline for this. I basically use each animation frame for static posing.
I guess in the same way as bohemian3 suggested.
But to do this, you have to (or i'd recommend to) manually save -all keyframes- and use object propagation, (or what was it called), by clicking the plus sign on the timeline tool, for all the important models in the scene. On each frame separately.
Otherwise pose changes in one frame may influence other frames. Doing this the posing stays "static" on each frame.
e.g. in frame 1 in timeline the models may be walking in the room, in frame 2 in the timeline they're sitting on a couch, etc.
You also can save camera positions to each frame.
It's not perfect but to me its the most efficient way, if you work in the same "scene" otherwise as the scenes can easily take ages to load otherwise, etc..
If you could select "constant" as the default interpolation mode when setting keys (like you could with Poser) then sequential art would be so much easier!
As it is, you can lassoo-select the keys for each new frame you create and make them constant. I think it's also the case that if you set a key in a channel that hasn't been set before, you'll also need to go back to frame zero, lassoo select those keys and make them constant again (I think DS adds or updates a key at frame zero in the previously unused channel, and does a spline interpolation from there - so helpful!)
Note: you can "collapse all" before lassoo selecting so there's less stuff to deal with.
This is something I've been struggling a bit to understand as well that how it works exactly.
It's a shame the timeline doesnt easily show what the interpolation mode is around there.
I guess it always defaults to TCB?
I've been wondering if constant or linear is better for this kind of use.
My another tip would also to have all property types visible in the timeline (TRSOAH) so you are not left scratching your head that something changes and you dont see anything in the timeline.. hit my head on that for a bit, when something changed a "hidden property".
iirc. head propagating scale for example changes some hidden scaling properties that you do not see in timeline by default.
But to me it somewhat feels to work ok, if you dont care about the interpolation mode and just dutifully hit the "create keys" button on each frame for each model. With the "Object" key creation scope.
Though if you have blank frames in between, they might go super wild lol.
All in all with a bit of quality of life improvements the timeline would be just so much nicer to work with.
Hopefully it's improved in ds5.
The key icons show what sort of interpolation is in effect for that key.
At the lowest, per channel level, the keys do show the interpolation type - but it would be nice if the collapsed view could mirror that. So that if all the keys under a parent were constant, then the parent key in the collapsed view would be marked constant also (or, more importantly, if some key way down in the heiracrchy stops being constant, then the parent key would also stop being marked constant to reflect that mixed state).
I try to resist hitting "create keys" on figures with a lot of morph channels - you can end up with a lot of keys very quickly. Maybe if there was a way to create keys only on "currently used" parameters?
I don't think it is possible to key only "currently used" (which would of course miss those that were zero in the current frame due to interpolation), but you could make a feature request (through a Technical Suport ticket). It should be scriptable, I think - you might check to see if Casual has already done it.
Good points - thanks Richard! I was thinking "current scene" rather than "current frame" - I'll look into it. Cheers!
Wow. Didn't know there were so many options.
I just set up the scene the way I want it to appear, start it rendering, and while that's happening
go to Paint.net (It's free and very customizable) and edit the previous render, addng dialogue, etc.
Once the render is complete, I save it, then save the scene one last time so I can be certain that
it's the latest version and begin the next one.
Repeat as necessary.
Ayups, that was my issue as well and like some others already suggested: timeline can be a life saver for all this. However, it's not perfect and this will especially apply when you're in need of some minor changes "in between", but in general it just works like a charm.
As for your overview: keep in mind that you can easily (re)catagorize your assets, and you can also easily make new categories as you go along. For example my "Saved Files" category has subs like: "Actor/Story Sets", "Scenes/WIP", "Scenes/Fantasy", "Set/WIP", "Set/Characters". Where WIP stands for work in progress and where I basically differentiate between characters to be used in my scenes in their "vanilla state" (Actor category) or where I saved a full character setup (with assets (hair, clothes), optional props, etc. alike) like Set/Characters (which are basically sub-scenes).
Maybe also good to know: Daz Studio is like elastic and you can customize just about everything. Definitely press F3 and take a look around the 'Animation' section; these are the options which allow you to control the timeline and also assign keyboard shortcuts to that. Going back & forth frames? For me that's Alt + cursor key (left/right). Very easy. Speaking of the timeline... keep in mind that it filters action types by default, make sure you have all the relevant properties enabled before you mess around, otherwise you could be looking at making changes to your scene, only to see them stay out of the timeline. I basically work with TROAH. And you definitely want to set the view mode to 'Advanced'.
Maybe also interesting to know: you can assign keyboard shortcuts for activating individual panes as well. See the 'Panes' section in the previously mentioned customize screen. For example; I'm using control + alt + t to call up the timeline. As a result it will hover over my Daz program, allowing me to drag it onto my second monitor (where it's usually located by default), and that makes it really easy to keep an overview of both my scene and its timeline.
Keep in mind though that if you save your scenes in this way then the files can become quite larger than they normally are. So definitely make sure you got enough free memory space to work with. But this is by far my most preferred way of creating series of some sort.
Just to add my two cents worth... As far as actual workflow goes that is...
When creating comics we tend to follow a very 'traditional' methodology which goes like this:
STORY - A quick outline of the story/plot your going to tell.
SCRIPT - The actual writing out of the scenes (images) and dialog that will make up the pages. Usually, it's broken down by the issue and page number (Issue #1 Page 01) along with a notation on how many panels (6 panels). (HINT - always wait until the whole "issue" script is completed before you really dive into it, if you're doing a page at a time write/render/assemble you could really bog yourself down and fail to hit your goals or even get to a 'completed' place) (there is plenty for arts sake to do such as character design and main sets, such as principle characters homes or offices etc. where you already know a lot of the story is going to happen. We've often gone to the extent of creating a 'day time' and a 'night time' version of some of the sets so it's easier for us when the time comes to pull it up and use it. Also, it prevents you from having to re-light a complete set every time you want to use it.)
Since we are working digitally, and we already have the 'comic software' we're using to create our pages, the writer can help shorten the traditional process by doing a basic layout while they are writing. Once this is completed you have a page layout already as a template. Along with a description (as brief or detailed as needed) of what goes in each panel, and what is being said. So under the above information you carry on with something like.
Issue #1
Page 01 (6 panels)
Panel 1 - Fred and Ginger are dancing among the treetops. Smiling and happily in love with each other. In the far background is the Vogon Construction Fleet which they are unaware of.
GINGER: Freddy dear, you float so effortlessly!
FRED: Because your love makes me feel so weightless...
Yeah, or whatever... but you get the point.
Next is simple. The artist (or team) goes to work creating the scenes by creating the individual panels for each page. Create the full set (setting of the image for the panels/page) and place/pose the figures. Once it's all done RENDER and NAME - we use a simple Issue, Page, Panel so our internal names look like: 06-22-4 Move the figures and camera as required and save/render the next panel (image) 06-22-5 and so on.
Using this method allows for collaborations and for creating pages that have multiple locations without killing ourselves jumping back and forth for each render. We render EVERYTHING that's scripted for that scene/location (time frame in the story and whatnot so lighting and backgrounds should stay the same) and can have it crossing as many panels or pages as needed of course. Then we generate the next set. Simple example- Two people in different places on the phone. Do all of the first persons images for their panels in your comic, then do the next and voila you've completed multiple pages with individual sets (locations) without spending hours loading and re-loading the set up Hmm?
Now assemble and add your text and sound effects. For those that Photoshop everything, or need to do a bunch of post work, that can be done before you assemble your pages.
By making sure you have a script and layout ahead of time, you know the shapes your images need to fit and can render accordingly so as to waste as little of the image as possible and help control render times. Remember, if your doing 4 - 6 panels per page, all those renders add up pretty quickly so being efficient in your work is always a good thing.
Hope that helps a bit along with all the information and suggestions from the other folks.