Artists' view on proper human proportions
polycounter lvl 14
I'm a frequent lurker here on Polycount. I often stop by the stickies in Pimping and Previews and am astonished by the amazing art people make, it's a huge inspiration to me. During all the years I've been lurking on art forums, I've noticed a distinct pattern when it comes to great artists and anatomy which is making me concerned. To some, there is this notion that there is only one body type in the world. /The one seen in anatomy literature/
The golden ratio of faces and bodies seem to be the only acceptable type. If a character is asymmetrical, or has longer or shorter arms than normal, it is deemed 'incorrect' and should be adjusted or it's not good enough. I have seen this criticism arise, even with characters that has realistic proportions. This happens very frequently with beginner and intermediate artists. When an artist becomes really good at defining surfaces, renders skin very well and knows how to dress a person, other artists seem to accept that these proportions are actually correct. However, when a 'lesser artist' does exactly the same proportions but the art has lesser quality, they seem to receive improper feedback on character proportions.
Here's an example of two athletes (1) (2) with distinctly different proportions. Note the proportions around the pelvis. (hands, elbows, hips, thighs) This image has a great set of athletes showing how different people can be. I thought of making this thread when I saw a perfectly made character on a forum (not saying which) recently, which was critiqued by a really good artist for having incorrect proportions... This could be easily proven false with a 10 second image search of normal people on Google Images.
When I was younger and started drawing and got decent results in regards to proportions with my characters, I got the same type of feedback from artists better than me. I still remember I got feedback on one of my characters, which had a 'too long neck' despite me drawing from reference. I thought I had used my reference poorly, I disliked my piece and it's been deleted to the void ever since.
What I'm trying to get at is... I think this type of criticism can hurt artists. I deem it poor for the sole reason that beginners are getting the notion that only one body type is accepted. I have lost count on how many times female characters don't have "wide enough hips", "too short legs", "too big hands". The same for male characters, with their respective 'wrongs'. It's sadly the same thing I hear in real life when people aren't good looking, they aren't perfect and their nose is too big or whatever it might be.
I am humbly raising concern for this. I know many good artists are trying to be helpful with constructive feedback. Sometimes, it becomes destructive. I have thought of replying to these messages a few times but I have feared retaliation. I'm not a good artist myself. Being on a site like Polycount and trying to give feedback on a piece is hard when you don't have "the art" to back it up, that you're skilled enough to give such feedback. It is the same reason I have feared bringing this up on Polycount, that I am not a respected or skillful member of the community and hence my opinion is void. (I've seen it happen to other members before trying to raise a valid point)
I think the bottom line, of what I'm trying to say, is that I think we shouldn't jump the gun to give constructive feedback on anatomy when we're not perfectly sure ourselves of what is correct or not. Sometimes your impression of how a piece of art 'could be better', could be worse for someone else. It's the same reason why my impression of a beautiful or ugly person is highly likely to be different from yours. Think objective vs subjective feedback
Maybe we should rephrase our feedback so it doesn't say "I think this is incorrect ", when it's not by definition. Instead it could be "I think this would look better if you change this or that". This way we're not telling the artist that he or she did wrong, but rather are giving a suggestion on how to improve the piece. By doing so we're not taking a part of their toolbox, their artist repertoire, and telling them to throw it in the trash when it's perfectly fine for another occasion.
I hope this is a valid point worthy of discussion.
The golden ratio of faces and bodies seem to be the only acceptable type. If a character is asymmetrical, or has longer or shorter arms than normal, it is deemed 'incorrect' and should be adjusted or it's not good enough. I have seen this criticism arise, even with characters that has realistic proportions. This happens very frequently with beginner and intermediate artists. When an artist becomes really good at defining surfaces, renders skin very well and knows how to dress a person, other artists seem to accept that these proportions are actually correct. However, when a 'lesser artist' does exactly the same proportions but the art has lesser quality, they seem to receive improper feedback on character proportions.
Here's an example of two athletes (1) (2) with distinctly different proportions. Note the proportions around the pelvis. (hands, elbows, hips, thighs) This image has a great set of athletes showing how different people can be. I thought of making this thread when I saw a perfectly made character on a forum (not saying which) recently, which was critiqued by a really good artist for having incorrect proportions... This could be easily proven false with a 10 second image search of normal people on Google Images.
When I was younger and started drawing and got decent results in regards to proportions with my characters, I got the same type of feedback from artists better than me. I still remember I got feedback on one of my characters, which had a 'too long neck' despite me drawing from reference. I thought I had used my reference poorly, I disliked my piece and it's been deleted to the void ever since.
What I'm trying to get at is... I think this type of criticism can hurt artists. I deem it poor for the sole reason that beginners are getting the notion that only one body type is accepted. I have lost count on how many times female characters don't have "wide enough hips", "too short legs", "too big hands". The same for male characters, with their respective 'wrongs'. It's sadly the same thing I hear in real life when people aren't good looking, they aren't perfect and their nose is too big or whatever it might be.
I am humbly raising concern for this. I know many good artists are trying to be helpful with constructive feedback. Sometimes, it becomes destructive. I have thought of replying to these messages a few times but I have feared retaliation. I'm not a good artist myself. Being on a site like Polycount and trying to give feedback on a piece is hard when you don't have "the art" to back it up, that you're skilled enough to give such feedback. It is the same reason I have feared bringing this up on Polycount, that I am not a respected or skillful member of the community and hence my opinion is void. (I've seen it happen to other members before trying to raise a valid point)
I think the bottom line, of what I'm trying to say, is that I think we shouldn't jump the gun to give constructive feedback on anatomy when we're not perfectly sure ourselves of what is correct or not. Sometimes your impression of how a piece of art 'could be better', could be worse for someone else. It's the same reason why my impression of a beautiful or ugly person is highly likely to be different from yours. Think objective vs subjective feedback
Maybe we should rephrase our feedback so it doesn't say "I think this is incorrect ", when it's not by definition. Instead it could be "I think this would look better if you change this or that". This way we're not telling the artist that he or she did wrong, but rather are giving a suggestion on how to improve the piece. By doing so we're not taking a part of their toolbox, their artist repertoire, and telling them to throw it in the trash when it's perfectly fine for another occasion.
I hope this is a valid point worthy of discussion.
Replies
Some characters look creepy if their backs are arched a certain way that looks biologically impossible for example so I think there should be limits. That, or change the artstyle to that of a cartoon where realism could be optional.
1) It's an easy crit. You can look at a character, apply the proportion/ratio knowledge you have in your head and point out where the model differs from that. It's not an incorrect crit and unless the author is clear about the model being stylized, it's what you're going to get.
2) Some artists are too quick to jump to the "it's stylized!" defense, when in reality they just badly whiffed the anatomy and are compensating.
Newbies often can't get the proportions right and they don't see issues.
So 90% of time people assume that less typical proportions aren't deliberate, and point them out as errors.
Also, Did you just suggest that saying "I think this is incorrect" equals to telling people to completely scrap their work?
To be honest I don't see a big issue about the way feedback is given. If anatomy is based on actual photo(s) the author can point it out and people will shut up so what's the problem?
In my anatomy thread I was rigorously following the rules of proportion. Then a much better artist then I showed up to point out that every head I had produced followed the exact same proportional dimensions - eyes, chin, mouth, etc, were all in the same location.
I argued with him at first, as did others, because the proportions were accurate to idealism.
Then I got his point, and had to accept that after learning proportions, to move past the next plateau in the line I had to learn to break them, because holding rigidly to that single body type was holding me back.
So as with most rules: you have to learn the rule, before you learn to break the rule. Just my .02
I wouldn't say go for the realistic/ human approach, but make the character believable - does the head make sense with the body.
Damn right. I did this just the other day.
It's kinda the same with people x3 When drawing, we want to make something that's actually pleasant to look at for once, when we've spend all day being mind blown by nature's lack of respect for proportions. Still no reason not to try and break them too like ysalesaid.
This.
Great responses so far. I feel my main point kind of missed its target though, I didn't mean to make this into a beginner vs adept discussio . It's not that I'm against learning the rules/basics before breaking them, it's that any deviation from the ideal seem to be pointed out as wrong rather than less-than-ideal, even in pieces by good artists. I can see a really great character piece with uniqueness in proportions turn into a cookie cutter ideal from feedback sometimes. It's as if only beauty and ideals are good enough, otherwise the character has to be tweaked before being called done. The only exception to this is when people are doing an ugly character by choice. I guess I'm just getting irked from seeing most characters having to be photo models to get a flawless thumbs up.
I'll try to elaborate and respond more when I'm not on my phone.
I think this recent thread illustrates how stylized characters are done right. It's apparent that the artist knows his fundamental skills and can therefore advance to more stylized and exaggerated proportions etc.
It's not quite as simple as that. The question really ends up being whether the artist intended for the deviation to happen or not.
In the cases where people say something looks wrong, more often than not the case is that the artist simply messed up. Everyone does, it's no big deal, but lots of artists then don't admit it and simply chalk it up as the character being "stylized". If the style is crap, then I guess it makes sense. But as an artist you really should know when people are trying to help you and not take it personally. Otherwise you can't grow.
Now it gets a little more iffy in the case that the artist did intend for a proportional change, but people still say it looks wrong. Most of the time it's because the proportional change was done wrong. If you want a character with long arms, you can't simply make one with average arms, then go in and stretch them. The entire anatomy has to change to compensate for that. Likewise if you have a character with 4 arms, or a really long neck, or what have you.
It's possible that the artist intended for the proportional change, worked it into his piece wrong, but did a good enough job at it that most people know that it's wrong but can't articulate it beyond "it looks wrong" without specifics. That can be a problem too.
But at any case, the situation that you describe where someone wants an anatomy that differs from the norm, does a good job at it, and then people say it looks wrong still, that just doesn't happen very often. I can't recall ever seeing that happen actually. I remember the opposite, where I saw such a thing, thought it was wrong, but didn't say anything because nobody else seemed to have noticed.
If you are familiar with such a case, which I'm sure is what made you make this thread, I'd love to see it.
At the end of the day ideal proportions are bullshit. No human possesses those proportions.
BUT...
It is absolutely necessary to become intimately familiar with a system - any system - of proportions, and here is why. A proportional system is a measuring tool. When you get the tool out, you know full well that your reference is not going to be a perfect fit. This is the whole point. You use it to measure where your reference deviates from the ideal. All of the character and individuality of a subject lies within that deviation. If you have not mastered and internalized an ideal, how will you recognize where your subject deviates from that ideal?
There are many different methods of dividing the body into proportional units. Head heights, head widths, eye widths, cranial spheres, golden ratios... the list goes on. Some artists prefer a heroic 8 heads, some prefer a more realistic 7.5. I don't think any particular system is more correct than another, use whichever makes the most sense to you.
As several others in this thread have already mentioned, people often get into trouble (I'm sure I've been guilty of this myself) by using the concept of "style" as an excuse for poor construction. If several well respected artists took their time to point out that something looks off, there is probably something worth fixing. If the offending artists wants to invoke "style" as an excuse, that's all well and good, but then in all likelihood there is something wrong with that "style" that should be corrected.
I didn't refer to you before but I just want to say these are words of the wise which I fully respect!
Mischievous: I sure know what you mean about nature being fake sometimes.
MartinH: That sure is a great thread. Going straight to my inspiration folder.
Brandon.LaFrance, Bigjohn: I absolutely agree with much what has been said. It's still as if we're discussing two separate things. I think it is partly me being poor at communication, English is not my native language and I'm poor at conveying my point in my own language too.
Yes, learning ideal proportions is key to making great characters. Its part of core fundamentals every character artist should learn. However, does that justify that many great characters out there should be adjusted to ideal proportions because it gives a pleasing appearance? This is what I'm trying to get at. The hands of that woman isn't feminine enough... Make it feminine enough that even Ashly Covington becomes jealous! Doesn't matter if this woman is going to be a wartime space marine, she needs those ideal feminine hands of a specific size. What? A realistic woman can't possibly have big sturdy hands from a lot of work! Out of the question. (note I'm still speaking about realistic characters)
What if the deviation was unintended but yet works, why should it be changed to ideal proportions if the artist succeeded? Must every proportion of a character be perfect for an artist of any level to get thumbs up for how well a character is made?
I want to avoid posting examples which I have stumbled upon before, to avoid subjective discussions about specific artists and authors. Let me make examples from random Google photos instead and phrase "feedback" as if it were feedback of the sort I've stumbled upon. I don't mean to criticise the people in these photos in any way, I'm using them as examples for real life people and how their looks would likely be commented upon in a piece of art. These are portraits, the same thing can be seen with bodily proportions.
(bad feedback): "His/Her eyes are too wide apart, you need to narrow them more. You know about the one eye apart rule right?"
(better feedback): "I think you could make him/her better looking by narrowing the eyes. While it's not wrong right now, I believe it would look better that way. Placing the eyes one eye apart usually works."
Note how the first example states everything as a fact, referring to a standardized measurement seen in any anatomy or portrait drawing book. It is stated in such a way that there seems to be no other options for how eyes are placed on characters. Sure, it is unusual with eyes this wide apart and is a common mistake by beginners trying to make an ideal character. Isn't it better to phrase it similar to the second example if that is the aim of your feedback?
(make similar bad/better feedback about foreheads instead)
I'll try to phrase my concern in other words. Artists give feedback on a character and states that the non-ideal proportions are unrealistic and wrong, rather than the character could look better if you did this. This is a huge difference in communication. One is telling the other artist that he/she has failed in portraying a human, while the other statement says the human could look better.
Maybe this is all about how to phrase our feedback in the end... All I know is that when I see a great character on Polycount, or another forum, which looks pretty much finished, it gets feedback similar to the ones above. Nope, not good enough, it's not ideal yet so keep working on it. That character looks too short, that one has too long arms and that one has a too small ass. No girl have a flat ass, and most guys do right? In the end all characters end up with the same proportions it seems, unless someone goes cartoony or a certain style. Maybe I'm just tired of seeing perfect humans in games? I mean, can't an artist make a space marine with proportions similar to Usain Bolt or Micael Phelps without getting remarks about wrong proportions or that it appears stylized?
It seems that a lot of newer artists post their work and ask for feedback without showing what they've been referencing (some don't use any reference at all). Then when someone points out the features that look off, the artist yells "but I worked from a photo!". Solution - post your reference along with your work.
Yeah, people come in all shapes and sizes; they have all sorts of weird fucked up looks to their bodies. But just because it happens in real life doesn't mean it doesn't look out of place. When I see someone in public with arms that are crazy long, I still notice and think, "Damn, those arms are crazy long."
And with your art, you probably don't want someone looking at the character you put a bunch of time into and walk away with "Man those arms are crazy long", because (hopefully) you put a lot more into it than that. This is probably one big reason we've come to learn those "ideal" proportions. When you have a consistent, appealing base, you can ensure that the other aspects of your work get don't get overlooked.
I also think you can deviate more from the "ideal" when you're good enough to know how to work with it, when there's enough substance to the rest of it that it fits. "Man that droop to the eyes really works well with the world-weary feel this guy has", or "I kinda like the longer legs on this guy, looks like he could run the pants off normal people and he's got the outfit to help deliver that message", etc. Half the time you just get a body, and when you've just got the body...well, you don't have a whole lot else to critique.
It's just that there are certain assumptions most people work under. Such as that you want to make appealing and attractive characters, that the character should look like it can do the activities that it would on a daily basis, that its overall design communicates its history and uniqueness, etc.
So if you're working off of reference for a less-attractive person, you should just specify that so people know. Otherwise everyone is going to assume that you wanted to make an attractive character and failed. By far most people want to make attractive or idealized characters, so it's only natural that people would assume this.
Should we who give feedback, guide the artist to a physique matching that of a typical climber first, to stay true to his end goal, or should we get the artist to make a solid ideal base and then at the end deviate from it? What's most important in this case, should the artist learn ideal proportions first or should the artist focus on making a believable climber if we see the artist has the potential to pull it off? (as a rough example, thin defined legs, well defined back and upper arms)
Kot_Leopold, Bigjohn: As you said, we should absolutely encourage the artist to post references if he/she's using it.
Wouldn't they technically be doing both at once? If the artist is using reference for a climber then they should simultaneously be getting the correct proportions for a human and the correct proportions for a climber - they're not mutually exclusive.
The difficulties that an amateur faces tend to be endemic to their skill levels and untrained eye. While it's admirable to try and improve our critiquing, isn't it a bit "looking a gift-horse in the mouth"? If an artist is getting comments about anatomy it should be accepted and prompt at least a cursory analysis of their work to be sure it's being as realistic/stylised as they intended. If they don't like the critique then the work will either suffer or be better for it.
I feel that the more inexperienced you are and the closer to realism you try to get to, the more you fall into the Uncanny Valley. Which in turn will get you those critiques about ideal proportions.
Which bring us to what everyone here has been saying: post your reference, learn the ideal proportions first, understand and internalize it, than deviate from it.