I have two WIP drawings I want to finish with a lot of large scale stuff. Unfortunately, I don’t even know where to start. The context for the first one is a large (ish) battlefield with monkeys from BTD and plants from PvZ, because I’m bad at ideas. The second one is just a general scenery thing in a forest, because I’m bad at ideas. Also, I apologize in advance.
This sort of “Where am I even supposed to start with this?” thing happens every time I want to make a not-so-great thing.
Not sure what BTD (I assume it doesn’t stand for Biotinidase) and stuff is but a general approach to almost any painting is start with the big stuff and then work your way to the smaller things.
For this example, imagine you want to paint a carriage that travels through abandoned village ruins on a edge of a forest on the bottom of a mountain.
First you would start with the general landscape. Where is the horizon, the mountain, maybe a cliff and a river, where starts the forest? Don’t get into details yet, like rocks and trees or foliage, only rough shapes. If you use color from the start, you can use it to indicate changes in the landscape already (like where grass transits to rock), still keep the details for later. A good approach to not get lost into too much detail is to paint with a large brush that’s not much smaller than the whole thing you want to paint.
Next, when you have the general shape of the landscape ready, paint in the next smaller things. In this case, this could be trees or the houses of the village. Work your way from the background to the foreground. Don’t put too much work on the things at this point, rough shapes for houses are enough, no need to paint every window or tile on the roof.
(for quick landscape concepts, you are already finished at this point)
At this point it would probably a good idea to establish the general mood already, and the bigger lighting (like the sun or shadows of clouds maybe). If there is something very important to the setting or story, I would already roughly sketch it in. Like if I want to have some dudes waiting behind a house, ready to make an ambush I would already put them in now with a few lines to remember not accidentally ruining the set up of the scene. I would also block the area where the carriage should be.
So know the painting has a landscape and we know where houses and stuff go. I would start rendering a few details already but keep the brush large. Add bigger rocks and maybe windows, trees. establish the mood a bit more (are buildings destroyed? Maybe some of them ar still burning. Are trees growing out of them already?).
Now we get to human scale, pretty much the smallest scale I have in mind for this picture but for you that might be not the case depending on “how close the camera is”. Now getting a smaller brush that lets me draw the details on the humans, the carriage too, and the horse of course (except you want one of this modern self driving carriages ;3).
(for speed paintings or some concept art, our work would be pretty much done now)
At this point we have established were everything is, how big it is, all the lighting should be at least somehow be indicated. Now you can render the details. Put leafs on the trees (only indicate them, not needed to actually draw every leaf) add texture to buildings, ground. You know the stuff. A dagger for that bandits, all the juice. This time work you way from the foreground to the background (at least that’s how I prefer it at this point) because as things are farther away, less details are visible and things get smaller and kinda merge at some point (So we can also keep our “get smaller” rule).
Now it’s up to the taste of the artist how much detail they want to put into it. For big works with lots of stuff going on “less is more” is a good approach. Keep your focus on what’s important for the artwork (if everything is important nothing is). There is a place for super detailed artworks (that is usually photorealism) where viewers are expected to explore everything but usually it’s important to guide the viewers eyes (by color, lighting or detail).
So this is my painting approach, of course for drawing it’s a bit different but the general rule still works. Start with the big stuff and then work your way to the smaller things. Save the details until last.
BTD is Bloons Tower Defense.
I just realized how many layers and lag I need… I don’t think I can do this even with layer conservation.
Another problem I have with these things, are my details even gonna show up in the final thing once I share and resize it?
If the details don’t show up, then that’s usually a sign that they’re superfluous. The big picture is what you should focus on.
In another post in another topic you mentioned using document sizes of around 9000 pixels on each edge. this is a lot and unless you plan to make your work into a giant mural, it’s definitely too much. In comparison: the dimension for a document in DIN A4 size (a standard piece of paper) is (assuming 300ppi) 2480 px * 3508 px (209.97 mm * 297.01 mm). That’s roughly a quarter of what you use. In addition most screens nowadays rarely have greater resolutions than 1920 px * 1080 px, although higher resolution screens get more common (but they are still substantially smaller than 9000 pixels). This means every artwork that is bigger than that will be scaled down for viewing anyway, unless zoomed in to 100% were things that don’t fit the screen get cut of course. Does that mean you shouldn’t work on files bigger than that? No, but it means that at some point the overhead is just not worth it. I usually look what my target size is (piece of paper or canvas it gets printed on, screen size, etc.) and double it if it’s very small or increase it at least by 50%. This way you have more pixels to work with (which is a good thing to have when working with smudge brushes) and you don’t lose too much when downscalling for publishing (in fact it can even look better some times as small errors become unnoticeable after the scaling). And of course you don’t waste your time on details no one is ever going to see.
Now, with a reasonable canvas size, you can have more layers at the same time without the programs performance dropping too quickly. In addition the “merge early” approach helps too. This means you merge layers and groups as soon as you can to keep your existing layers count low, usually paired with Krita’s “Incremental Save” feature, in case one has to get back to a previous stage for any reason (you create an incremental save, then merge. the previous save serves as a backup).
Outside of a gaming forum people might have problems knowing what you are talking about if you just use these TLAs that are not common knowledge in an art context ;3.