Close

Memories Of Old Reno: Pass 1

After writing yesterday’s post about painting from memory I sat down to do it. Actually I watched 3 hours of Grimm while I practiced drawing faces by pausing the show then drawing what I see. I chose Grimm because that show has become so completely stupid that it doesn’t matter if I’m not really paying attention to it, yet I want to see if the writers actually recover from the total disaster they created in this last season. Nope, that show’s junk. It does work great as a source of faces all lit with chiaroscuro though.

The First Memory Sketches

Once I finished annoying myself with a stupid TV show I setup the projector in my bedroom so I could study it and try to memorize it.

What The Projector Looks Like

Projected on my wall like that (but straight since the photo is so you can see it projected) I can lay in bed and take mental notes of what I need to remember. I’ll explain in a bit how you do that, but I should warn you the description of how to do it sounds batshit crazy because I’m trying to explain something I just…do. It’s weird. Like explaining blinking to someone who’s never had eyelids.

First thing I try to do is memorize the overall shapes of the whole thing, and to do this I focus my eyes on the edges of the scene (that’s the projected image) and look at it with my peripheral vision. I also point my eyes in the center but unfocus them so that the entire image fills my vision and I’m passively letting it all in. At the same time I’m clearing my mind so that I’m not thinking about anything and not internally talking, just observing and letting it all in. For lack of a better term I call this “gestalt sight”.

I do this for 6 minutes then get up and go into my living room where I’ve setup some newsprint paper and a stick of charcoal so I can draw a value sketch from memory. The first attempt I produced this:

First Sketch of Old Reno From Memory

Notice how I got the general big shapes, proportions are kind of alright, the big black shapes are there of the street and other places, but the relative size of the marquee and the upper portion of the building aren’t there.

With a first guess down on paper I need to then figure out the delta between what I remember and what’s actually in the scene. This is an important step because otherwise I won’t be improving my recall of images and correcting what I see in them. I take this first drawing to the scene and I make mental notes of what I need to focus on in the next memory round. Then put the newsprint pad back in the other room.

The next strategy is to use my sense of touch and my mind’s eye to help remember the scene. I set the timer for another 6 minutes and spend that cycling between “gestalt sight” and pretending I’m touching the scene. Yes, this is as weird as it sounds, but what I do is I first gestalt the scene for about 30 seconds or however long it takes until my mind starts to wander and I hear my mind talking. Then I take my pinky finger, close one eye, and I put my pinky into my vision of the scene and then trace it around the shapes but I imagine that I’m touching the shapes. As I do this, I’ll close my eye and try to hold the image in my mind and keep my finger going in the right direction for that edge, using just my imaginary sense of touch. Then I will open my eyes and peek to see if I got to the right part of the scene while my eyes were closed. If not I go back and keep my eyes open and keep touching it. Eventually this becomes boring and I then stop and switch back to gestalt seeing.

It’s bizarre, and really hard to describe but basically I alternate between absorbing the whole thing without thinking about it and tracing the image with my pinky while opening and closing my right eye in an attempt to feel the scene and test what my mind will remember of it. After 6 more minutes of this I produced this charcoal sketch:

Memory Sketch for Old Reno 2

Vast improvement in just two rounds of memorization. The advantage of using charcoal on newsprint for these sketches is you can do them very fast without having any setup, and you can get a good range of values. If you have to do cross hatching and changing pencils then it’ll be a friction on your recall and make it more difficult.

I felt I could have started painting from just that, but I wanted to do one more round and try to edit the composition. I don’t want to paint the photo exactly, not that I could from memory. I want to embellish what I see and edit something out to make it more interesting. For example, in the original photo:

The Basis Of A New Painting Project

There’s a bus stop on a light post I’ve already edited out. I want to change the building on the right, and maybe a few other parts so in round 3 of memorizing I’m going to try to memorize the scene, and then also remember what I want to change of it. Going through the same process of gestalt/tracing I’m able to produce this charcoal sketch:

Memory Drawing of Old Reno 3

It sounds impossible that someone can do this, and trust me I do not have a photographic memory. In fact, the skill of photographic memory hasn’t been proven to exist, and one of the major studies claiming it existed was of a single woman, who was married to the researcher, and who never demonstrated it again after the study. I’m just doing something anyone can practice and remember, I just think most people haven’t tried or even thought it was possible.

What I’m actually doing is remembering shapes and their relationship to each other, starting with about 5, then breaking them down into smaller ones inside those. It’s a trick called chunking where I’m not memorizing all the little squares and details in one giant mosaic. Instead I have a big gestalt framework that, if I used words, would be something like “big white trapezoid on top, dark black square on right, big medium triangle in the middle, another wider triangle on top, big square on right, two big squares below.”

Then when I have those down, I don’t need to keep them in memory. I kind of use a “graphic binary search tree” to use a programmer term. The next round I figure out what I need to remember details for, many times many shapes, and then I stare inside them. I navigate to the “big rectangle in the center” in my mental binary search tree, and start gestalt/tracing to remember what’s inside it, “vertical dark rectangle, trapezoid, big rectangle, words, etc.” once I think I have that I then have that piece chunked and can push it back onto my internal stack and do another thing.

I don’t literally use a binary search tree in my head. I can’t even use words to explain what I’m doing, but a binary search tree is the best description of what I’m doing. I don’t memorize thousands of features. I memorize 5 or so, then 5 or so inside those 5, then 5 more inside those, until I have enough.

I recommend getting Memory Drawing if you really want to learn this, as that book has progressive exercises you can attempt and more extensive explanations.

The Painting

With the scene mostly memorized I decided to do a simple monochrome block-in on the canvas I just bought. I wanted to attempt something large and involved, so I bought a 24″x36″ canvas with a black primed surface. Lately I’ve been into the black canvas to start out. Not sure why, but when I use a white canvas I start off by painting a color onto it so it’s not white anymore, so I figured why not just get one that’s black. Black canvas seems to make it easier to get values right since I just have to paint on the white shapes then push from there to the black.

Since this is just a first pass underpainting from memory, I decided to use simple black and white alkyd paints and thin them with turpentine so they dry fast. Alkyds dry in about 24 hours, so for this project I’ll be able to work for a night, then the next night I can work on it again without having to wait. A little bit of quick dry medium in the paints make them dry even quicker, and as I write this blog post I can see the painting is probably already dry.

With the scene in my mind, I just stood there and painted this:

Monochrome Block-in For Old Reno Painting

There’s some shapes that are totally crooked or just wrong, but I did it without looking at the projected image or at my sketches like I planned. I simply followed the binary search tree in my mind. I put down some lines for the big shapes, chose the first one that was clearest in my mind, developed it, then went to the next one that I could recall, then used those to guess about or improve the others.

Keep in mind that there’s also a lot of just guessing at what should there than some crystal image floating in my head. I’m painting what I remember, but I’m also stepping back and looking at the painting to say, “Hmmm, nah that can’t be the right shape, it’d look better like this.” That might make the painting wrong according to the scene, but if it makes the painting work better then I’ll go with that. Sometimes though I guess at what it should be, then I go look at the scene and say, “Aha! That’s why it looks screwed up, I remember this wrong.”

About the author zedshaw

I'm the author of the Learn The Hard Way series of books, a painter, a guitarist, and programmer.

All posts by zedshaw →

One Comment

Comments are closed.

%d bloggers like this: