04-21-2021, 10:43 AM
Hi Delorias, I was going to respond directly to your pm, but figured I'd post stuff here because it may help some others. Joseph's write up was a good rundown on various approaches to developing greater accuracy so good on you for taking it on.
I'll pop in a resource and an example of value grouping or value massing that you asked about.
This video from Dorien Iten is worth a careful watch. Explains much more in depth on some academic approaches, but very simply and well explained.
I recommend you also have a look at the free to download "Accuracy guide" also by Iten. It's a good beginner's look into some atelier methods for judging angles and proportions.
Your portrait attempt was nice, but I think a more structured approach can help you get to a better result.
The key idea, is simplifying the values into two "families" or groups of Shadow and Light (or Dark Values and Light Values) to begin with and finding where those shapes are and refining them before going too far with modelling with more values. Iten's video above has a great demo halfway or so through.
One way of helping you judge these irl is at first to squint the eyes which reduces and compresses similar values into blocks. The equivalent digitally would be by using a median filter. I see people use Posterise filter to do this but I think it makes for far too detailed shapes and loses all edge quality.
Drawing shapes that enclose the dark value family. Note that the shapes I've chosen are quite detailed, to get to the point quickly, but in practice, you really tend to start with more simple big angles and shapes and then refine as you get those relationships correct.
Filling in the dark value family into a two tone image. You can see how quickly the basic information describes her features.
How to go on from here, well you gradually add more halftones (values in the light) and dark tones as you need. Iten's video will go over that so I won't bother.
Finally checking the overlay ontop of yours, may be helpful to see where you went off. Handy if working digitally to check after the fact, but I wouldn't use overlays too much as a crutch as you work. You want to actually train your eye to become more accurate intuitively, and also, a reminder to keep checking and keep measuring as you work. Never assume any thing is correct without checking it if accuracy is your aim.
These kind of studies are great to do in traditional/pencil etc as well. Digital has a lot of crutches we can inadvertently use that make it easier than it actually is and as such we don't grow quite as much as we could.
So this is essentially a method of "flattening" of the form into basic shapes to help you see. How do we place those shapes accurately though? Well one way is to keep measuring angles and comparing proportional relationships until you get them right. The advantage of this method is it really trains you to observe patterns as they are, and get over the tendency to draw symbolically. For example your eyes are drawn symbolically, what you think an eye looks like, when actually the proportional shapes are very different. This happens a lot more with faces ofc, but good to keep in mind.
Another method we see a lot of is to use construction methods to lay a simple framework upon which you then start placing shapes. Any construction method can help. I tried a bunch for heads for example (Loomis, Reilly, etc) and mostly end up using a mix depending on the situation or just jump straight into trying to get the visual impression. It depends on many things, the medium, the subject, the lighting, just whatever helps me solve a particular problem. It's personal preference really.
Wrt to colour as you asked, I think your colour choices seem pretty good, so I think intuitive or preferential colour is absolutely a fine way to work. I did this for years, and still do. Colour is mostly intuitive for me. HOWEVER, once I started doing trad painting, I HAD to learn how to colour mix, and so all the theory starts to be necessary to really internalise it. When I was only doing digital, I still learned the most from traditional art focused blogs. So learning some theory or even doing some painting in trad will make your colour choices much more informed and deliberate imo. Again personal preference, but I think having theory under your belt is like a map that takes you towards certain places you want, rather than pure exploration, so definitely start to learn more about it if you like.
Hope that helps.
I'll pop in a resource and an example of value grouping or value massing that you asked about.
This video from Dorien Iten is worth a careful watch. Explains much more in depth on some academic approaches, but very simply and well explained.
I recommend you also have a look at the free to download "Accuracy guide" also by Iten. It's a good beginner's look into some atelier methods for judging angles and proportions.
![[+]](images/collapse_collapsed.png "[+]")
SpoilerYour portrait attempt was nice, but I think a more structured approach can help you get to a better result.
The key idea, is simplifying the values into two "families" or groups of Shadow and Light (or Dark Values and Light Values) to begin with and finding where those shapes are and refining them before going too far with modelling with more values. Iten's video above has a great demo halfway or so through.
One way of helping you judge these irl is at first to squint the eyes which reduces and compresses similar values into blocks. The equivalent digitally would be by using a median filter. I see people use Posterise filter to do this but I think it makes for far too detailed shapes and loses all edge quality.
Drawing shapes that enclose the dark value family. Note that the shapes I've chosen are quite detailed, to get to the point quickly, but in practice, you really tend to start with more simple big angles and shapes and then refine as you get those relationships correct.
Filling in the dark value family into a two tone image. You can see how quickly the basic information describes her features.
How to go on from here, well you gradually add more halftones (values in the light) and dark tones as you need. Iten's video will go over that so I won't bother.
Finally checking the overlay ontop of yours, may be helpful to see where you went off. Handy if working digitally to check after the fact, but I wouldn't use overlays too much as a crutch as you work. You want to actually train your eye to become more accurate intuitively, and also, a reminder to keep checking and keep measuring as you work. Never assume any thing is correct without checking it if accuracy is your aim.
These kind of studies are great to do in traditional/pencil etc as well. Digital has a lot of crutches we can inadvertently use that make it easier than it actually is and as such we don't grow quite as much as we could.
So this is essentially a method of "flattening" of the form into basic shapes to help you see. How do we place those shapes accurately though? Well one way is to keep measuring angles and comparing proportional relationships until you get them right. The advantage of this method is it really trains you to observe patterns as they are, and get over the tendency to draw symbolically. For example your eyes are drawn symbolically, what you think an eye looks like, when actually the proportional shapes are very different. This happens a lot more with faces ofc, but good to keep in mind.
Another method we see a lot of is to use construction methods to lay a simple framework upon which you then start placing shapes. Any construction method can help. I tried a bunch for heads for example (Loomis, Reilly, etc) and mostly end up using a mix depending on the situation or just jump straight into trying to get the visual impression. It depends on many things, the medium, the subject, the lighting, just whatever helps me solve a particular problem. It's personal preference really.
Wrt to colour as you asked, I think your colour choices seem pretty good, so I think intuitive or preferential colour is absolutely a fine way to work. I did this for years, and still do. Colour is mostly intuitive for me. HOWEVER, once I started doing trad painting, I HAD to learn how to colour mix, and so all the theory starts to be necessary to really internalise it. When I was only doing digital, I still learned the most from traditional art focused blogs. So learning some theory or even doing some painting in trad will make your colour choices much more informed and deliberate imo. Again personal preference, but I think having theory under your belt is like a map that takes you towards certain places you want, rather than pure exploration, so definitely start to learn more about it if you like.
Hope that helps.