How Much Is Enough?

I’m not asking about the usual RAM/CPU speed/CUDA cores etc, or the poly count or any of that. This is a question inspired by watching Ian Hubert videos, where he gets away with- I guess you’d call it Peak Laziness. And it works, beautifully, and makes him super productive. This is more like the question that traditional artists often ask about detail and where to apply it, and I know it’s all very subjective.

I am not particularly interested in the kind of realism that Ian Hubert pursues, and I really have very little use for the hyper-realist style that one sees in a lot of DAZ (and other) renders. It’s interesting on one level to see how well we can now render pores and peach fuzz, but I don’t see a lot of use for it- there are pretty much no circumstances in the real world in which we see people this way, and I’m more interested in storytelling than technical extremes at this point.

I am interested in style, though, and in what other people think about style issues.

Here and elsewhere around the net I see a lot of artist learning the craft, and a handful doing very stylish work. I don’t love all of the latter, but I admire the way the artists are able to use a very precise medium to create non-realistic visuals in a purposeful way (It’s easy to do in an accidental way- I make very crappy renders right now because I don’t know what I’m doing).

So, in a story telling context (and, when you dig down, most visual representation is about telling a story in one form or another), how much is enough? How much do you care about the sheer visual beauty of an image vs the story it’s telling? How much does the visual character of a comic, an animation or a game impact your enjoyment of it?

From what I suspect, the golden rule that triple-A studios follow is that use reference shots for everything, in every way possible. Game engines, on-location photos, live action reference, miniatures, animatics, comic books, everything goes. Lately triple-A studios (Fast Furious, Iron Man, Blade Runner) have literally shot the entire movie with a live actor (or a cosplay suit), instead of releasing the movie as it is, the source material serves as a reference to the CGI… :upside_down_face:

Another thing to consider is in the context of movies - many shots have been have been analyzed and calculated precisely. This is actually very clever approach to create these scenes by having a real clue about how are they composed and that way you can make better decisions regarding polycount, texture resolutions, etc.

I wonder how much of this is driven by costs rather than aesthetics, though. It is basically a plus-ultra form of roto-scoping, which has always had aesthetic implications, but which primarily saves a vast amount of time- which of course, CG in general is intended to do. I think it’s fairly recent that the technical limitations of CG have begun to have less influence on aesthetic design than ‘artistic decisions’.

I know that for indie producers, the technical restrictions- not the ‘pure’ technical restrictions, but the affordable technical solutions- play a big part in design. I certainly can’t afford to mocap everything (as a baseline for constructing scenes, for instance. So, I have to make some trade-offs between how I would like something to look and what it is practical, in terms of time, money, and the ability to master a wide variety of skills, to actually get something made. The huge number of people in the Blender community and elsewhere who keep creating and improving tools and assets that I can use for free, or nearly free, is what makes it possible for me to even think about incorporating 3D in my pipeline, but I am still capable of mastering only so much.

One thing that I’ve been thinking about recently is the difference (typically) between Visual Novels and Web Comics. There seems to be a really heavy focus on asset-recycling in Visual Novels, that may include a few high-end illustrations, but uses the exact same images over and over again. This is pretty much rejected in more Western-style web comics, but often at the expense of having a lot of unique but boring or poorly drawn panels.

Of course, the major aims of these formats is rather different at present, so perhaps one approach is ‘better’ or more sensible for some of those aims than others- a bit of apples and oranges- but a lot has to do with where the creators want to invest a relatively small ‘visual budget’ to support the story they are telling. A lot of web comic creators don’t seem to feel that the quality of the images they use is very important compared to whatever the story itself is. Japanese creators seem to feel they need to adhere to certain graphical standards, at the expense of using very few images overall.

It’s a bit like the difference between early Disney animation and Hanna-Barbera. As a kid, I simply could not watch Hanna-Barbera cartoons because of the low frame-rate, heavy use of static images and so on (all tricks taken to the max by anime animators…).

I can say for sure that choosing an art style, surely brings lots of pros and cons over the other. Most important:

  • Technically how is made, how easy is to made.
    As for example: Realistic is very easy by having access to a huge set of textures and materials, however NPR all backgrounds and textures need to be hand-painted from scratch.

  • Artistically how complex it is, how extensive the world building is.
    NPR scenes tend to be very simple and minimalistic (reduction of noisy and unwanted details) so they are much easier to compose. However realistic scenes need to be accurate and this goes mostly with extensive effort and research on why something has to look like this, and not the other way.

I can only consider that the criteria into producing these assets are relativistic, there is no one single universal standard, but lots of microscopic principles that exist in every type of domain.

But if you ask me personally, which is better, I can safely say that there is a combination between personal-interest (that will boost morale and maintain interest) and public-interest (things that work best for predefined and existing formats - standard ideas and techniques that the common consensus can handle and won’t be offended by).

Depends on the rest of the content.
If the story is good, I don’t need graphics at all, I’ll even read a book! :stuck_out_tongue:
On the other hand, I won’t play Call of Duty or watch another superhero movie unless it’s really damn beautiful to look at.
And then there’s zombie games and american drama which I won’t watch no matter how “beautiful” they are.

So I guess that means visuals do have some impact on enjoyment, but only as long as the rest falls within the middle band of what the viewer considers acceptable content. So in other words, they can make mediocre content better, but have virtually no impact on otherwise very good or very bad content.

I think this question is really hard to answer, as you hope to get sort of a concrete answer on something that infact has endless valid variations possible. Find your constraints first and set your key aspects which you want to maximize. It’s normally impossible to maximize all of them as they can be opposed, like “detail” and “time spent”. So the first thing is, accept that if you maximize a, you have to accept a reachable level of b within an optimization function. Some fields like realism have traps like the uncanny valley, thats why stylized options have a potential to save time here.

And in the end it’s all about your individual goals. If you set all of them really high, the goals are harder or impossible to reach. There is no rule what set is convincing enough. It eg also depends on your intended audience and on the thing as a whole. Some artists got famous by painting just rectangles and circles. Kids look at images and compositions differently than grownups and the readability of an image should respect that. You should first be clear on your goals. And then don’t see it in a cardinal but rather an ordinal way. What’s most important for you. If it’s the story then optimize that at the cost of other things. The limiting factor will likely be time.

Let me be clear, this question is about YOUR OPINION. I will not be grading, so go ahead and rant. I am not looking for a definitive answer, I’m trying to see what other people’s attitudes are, and how they go about deciding what’s “enough” for them.

I read books. No visuals needed- but then I require a certain amount of style in the writing, so the question remains. If something does have visuals, and especially if the format is fundamentally visual, such as comics or movies, then I want to see that the creator(s) have taken them seriously- but what that means is different in different stories, contexts, etc.

That being said, if I see a comic and the art on the first page is clumsy, I will not find out if the story is good or not. If I watch a movie and the acting in the first few minutes is wooden, I’m unlikely to find out if it has a great script- same kind of thing. If I open a book and page one has grammar and spelling errors, I’m not going to read the rest. Some things just wreck a project for me, and there is nothing that can make it okay. For anything, while the story might be the key element in one sense, the presentation of the story relative to the format it’s in is critical.

I am old enough and have read and watched and seen enough that I am not confident that there are any ‘new’ stories. Might have run out of those with Shakespeare, and most of his stuff was a re-do of older stuff. So style is important to me. I can, and do, watch and read things with the same old stories in them and get enjoyment out of the style, whether it’s visual or some other aspect.

Here’s an example: The Maltese Falcon with Humphrey Bogart is not the first movie based on the story. Nobody knows the first one, for good reason. Identical material, pretty much identical movie technology, made a few years apart, but worlds apart in terms of cinematographic style.

You were asking how it impacts enjoyment, not marketability. :stuck_out_tongue:
How you can convince people to consume your work in the first place is a different can of worms.

If you’re talking short term, for a one-off production like, say, a movie, then high production values will certainly help to draw in viewers. If you have a product that is serial, or better yet, never-ending in nature, then you can rely on things like word of mouth to gradually draw in an audience in spite of a low budget. Case in point: XKCD. As with everything, there’s a great deal of luck involved as well.

Oh, and regarding stories being all used up: It doesn’t matter if a story is old, as long as it’s new for the person viewing it. You have to redefine ‘original’ to mean ‘something my target audience probably hasn’t seen yet’. I think it’s perfectly fine to take Macbeth and repackage it for 12-year olds according to their sensibilities, or take Superman and make it palatable to older christian folks (they seem to enjoy superheroes like Jesus after all). But you have to be conscious who you’re targeting.

Well, for me personally, enjoyment and marketability are pretty closely tied. I’m not going to buy something I don’t enjoy. I may have different expectations than other people, but overall, people don’t buy stuff they don’t enjoy if there’s an alternative.

I saw a game on Steam the other day, a little indie production. It had a nice blurb, combining some elements I like in a way I haven’t really seen. I checked it out.
Unfortunately, it began with a lot of very poorly written copy, segueing into some really atrocious pixel graphics. I’m not a fan of pixel graphics to begin with, but these were serious eyesores, and it was only possible to tell what things on the screen were because there were so few of them. At this point I lose interest in novel game mechanics.

On the other hand, I came across a little gem, a simplified, somewhat chess-like game, that I didn’t think would be terribly interesting to play, but it had simple, pleasant graphics well calibrated to the nature of the gameplay. Because of this, I gave it enough of a chance to discover some real depth to the game play.

That’s just me, clearly a lot of people don’t much care what things look like, they are more focused on other things. I’m just curious to know where the balance is for different people.