04-17-2019, 11:56 PM
In regards to these recent images, they have some issues and I think the primary cause of them can be traced back to the images being more concerned with minor differences rather than large ones. This is a very common problem and something every artists should be very aware of because it is a tendency that can creep into the works of even the most experienced artists.
Our brain and eyes are very adept at discerning minor differences in things, communicating to our brain what the object in question would feel like to touch. The relationship between sight and touch is something very important to the way we see and interpret the world but it does give us a significant bias when we try to draw or paint something. We tend to overemphasize what touch-like information we want to represent, this is usually the material feel we’d pick up with our fingers, and neglect the larger visual impression and subsequently the larger form.
Historically, the artist’s education has significantly focused on counteracting our innate bias and training our abilities to see, conceptualise and represent the big impression of our subject matter. You’ve probably heard stories about how artists like Sargent and Mancini would look at their subject next to their paintings far away from the other side of the room, closing one eye and squinting to blur their vision to only be able to see the the big impression of their subject and painting, comparing the two on a very basic and abstract level. That would be one example of a historical practice aimed at counteracting our bias. There are other examples of this like three point value comparisons that were emphasised by Scandinavian artists like Krøyer.
Getting to details and texture is something you can generally figure out on your own. If you were to look at the painting manuals published in the 19th and very early part of the 20th century, they clearly omit any discussion on details in any drawing or painting. This is because details come very naturally to people and the big issues artists have to deal with is what is beneath all the textures. John Collier even going so far as to saying it’s the easiest thing an artist can do and therefore he won’t bother discussing it.
In general my point is that the historical focus in art has been on the big impression rather than the minor details and I would encourage you to continue in that tradition with your own learning. This isn’t to say that texture details are unnecessary, they have a very important place in art but they ought to be supported by an underlying impression.
I borrowed one of these images to illustrate this idea in practice. I compressed a lot of the surface information and instead put the emphasis on where light transitions into dark. I reduced the number of values to limit the amount of information I’d have to manage at the same time and to emphasise what things are similar and what’s different relative to the large impression. I reduced the “drawing” information into simpler straight lines to again reduce the amount of information I’m working with and to then allow me to put pressure on that limited information I do have to accurately express the drawing of the subject.
I would recommend doing this in the early stages of any image and something to always keep in mind when working. It happens to everyone that these very fundamental aspects of an drawing or painting may slip away while we’re working. It’s very easy to get tunnel vision and focus on something small that eventually harms the larger impression, and it’s a constant challenge to suck it up and correct yourself when you’ve found that you’ve made these mistakes.
There are certainly examples of artists throughout history front loading some of their paintings with texture and noise however these techniques are somewhat advanced and require a very abstract way of working (see Mancini as an example). It should certainly be studied by any ambitious student but probably after you’ve gotten a hang of things using more conventional methods.
I don’t mean to imply that you have to approach drawing and painting like a robot. I find that a kind of dialectic approach to art is very useful where you may switch between contradictory ideas/approaches to arrive and something satisfies both. Traditionally in art this has been switching between trying to express form and trying to express the light impression but we can also switch between a very restrained approach to painting and a more free one, to on one hand reason through our painting and on the other hand to knock the work forward whenever this approach gets “stiff”. Most people need to focus on the former rather than the latter.
Also, the color issue probably has something to do with the color profile of the image + how the web browser interprets the information. Look at what color profile you’re saving in. Set it to sRGB or something in photoshop and you’ll probably avoid any issues.
Anyway, if you need any help with art stuff, let me know.
Our brain and eyes are very adept at discerning minor differences in things, communicating to our brain what the object in question would feel like to touch. The relationship between sight and touch is something very important to the way we see and interpret the world but it does give us a significant bias when we try to draw or paint something. We tend to overemphasize what touch-like information we want to represent, this is usually the material feel we’d pick up with our fingers, and neglect the larger visual impression and subsequently the larger form.
Historically, the artist’s education has significantly focused on counteracting our innate bias and training our abilities to see, conceptualise and represent the big impression of our subject matter. You’ve probably heard stories about how artists like Sargent and Mancini would look at their subject next to their paintings far away from the other side of the room, closing one eye and squinting to blur their vision to only be able to see the the big impression of their subject and painting, comparing the two on a very basic and abstract level. That would be one example of a historical practice aimed at counteracting our bias. There are other examples of this like three point value comparisons that were emphasised by Scandinavian artists like Krøyer.
Getting to details and texture is something you can generally figure out on your own. If you were to look at the painting manuals published in the 19th and very early part of the 20th century, they clearly omit any discussion on details in any drawing or painting. This is because details come very naturally to people and the big issues artists have to deal with is what is beneath all the textures. John Collier even going so far as to saying it’s the easiest thing an artist can do and therefore he won’t bother discussing it.
In general my point is that the historical focus in art has been on the big impression rather than the minor details and I would encourage you to continue in that tradition with your own learning. This isn’t to say that texture details are unnecessary, they have a very important place in art but they ought to be supported by an underlying impression.
I borrowed one of these images to illustrate this idea in practice. I compressed a lot of the surface information and instead put the emphasis on where light transitions into dark. I reduced the number of values to limit the amount of information I’d have to manage at the same time and to emphasise what things are similar and what’s different relative to the large impression. I reduced the “drawing” information into simpler straight lines to again reduce the amount of information I’m working with and to then allow me to put pressure on that limited information I do have to accurately express the drawing of the subject.
I would recommend doing this in the early stages of any image and something to always keep in mind when working. It happens to everyone that these very fundamental aspects of an drawing or painting may slip away while we’re working. It’s very easy to get tunnel vision and focus on something small that eventually harms the larger impression, and it’s a constant challenge to suck it up and correct yourself when you’ve found that you’ve made these mistakes.
There are certainly examples of artists throughout history front loading some of their paintings with texture and noise however these techniques are somewhat advanced and require a very abstract way of working (see Mancini as an example). It should certainly be studied by any ambitious student but probably after you’ve gotten a hang of things using more conventional methods.
I don’t mean to imply that you have to approach drawing and painting like a robot. I find that a kind of dialectic approach to art is very useful where you may switch between contradictory ideas/approaches to arrive and something satisfies both. Traditionally in art this has been switching between trying to express form and trying to express the light impression but we can also switch between a very restrained approach to painting and a more free one, to on one hand reason through our painting and on the other hand to knock the work forward whenever this approach gets “stiff”. Most people need to focus on the former rather than the latter.
Also, the color issue probably has something to do with the color profile of the image + how the web browser interprets the information. Look at what color profile you’re saving in. Set it to sRGB or something in photoshop and you’ll probably avoid any issues.
Anyway, if you need any help with art stuff, let me know.
Discord - JetJaguar#8954