The Elusive Zbukvic Touch

Every now and again I torture myself and try and get some insight into Joseph Zbukvic’s technique. I’ve had an inspirational printout of one of his wonderful Venice paintings on my wall for a while and I foolishly wondered if I could learn anything from trying a copy.

Now I wouldn’t recommend this if your ego is not built on rock solid foundations. I’ve tried before with varying degrees of success and didn’t really come away with any lessons learned. This time I made a little headway.

Firstly his values are surprisingly close together. The scene above (his is the bottom one of course) looks as though it has very dramatic value changes but actually they’re almost all about 3 steps apart. The sky especially near the horizon darkens significantly and helps to tie everything together.

Secondly his ‘colors’ are REALLY close to gray. Low chroma in other words. I got the Munsell chips out and had to go to my ‘nearly neutrals’ section to even come close to matching them. It’s really easy to slop far too much bright color in there even when you think you haven’t.

Once I’d got to grips with both those things I didn’t fair too badly. I till could have compressed my value range even more but on the whole this was one of my better attempts.

Here is a preliminary sketch I did before the bigger one.

This was definitely worth doing. Gave me confidence that at least my values were in the right ballpark.

A few Odds and Ends – Roosters and Zbukvic

As I was so cheesed off with my Harvard Yard fails I did a quick rooster and Zbukvic practice. Feel a bit better now.


Learning from Zbukvic and Chien Chung Wei – More Mini Paintings

Another set of Zbukvic and Chien Chung Wei copies today. Only 3 more as time was limited. Some successes (bottom left I was happy with) but struggled a lot.

Here they are right way up.

Learning from Zubukvic and Castagnet – 7 Mini Paintings

Every now and then I do some copies of paintings by people who really know their stuff. It’s a way of learning and also a measure of how good my skills are. It can be demoralizing as coming close to their style is close to impossible.

I did 7 mini paintings over a day and on the whole I came much closer to producing something I like than I ever have done before. The color especially I took care over and this really paid off.

Here they are in order.

Still a long way to go but a good day’s work and great fun.

Learning Time – Zbukvic Day 4

Well that was one hell of a day. Sometimes you just want to pack everything up, put it on ebay, and take up golf. And I hate golf.

Only 3 paintings done today and none went well. Muddy washes, dodgy backgrounds, clumsy finishing flourishes. Nothing worked.

But let’s go through them anyway. Number one :

This was probably the most successful. Nice light on the sunlit part of the building and the foreground shadows work well. Not much else to recommend it though.

Onto number 2 :

Blergh. Not much to say about this. Didn’t enjoy it one bit.

And number 3:

I’d been quite looking forward to doing this as lighter buildings with dark accents are generally easier. It was not to be however.

Takeaways :

  • I’m fed up.

  • I probably need to stop doing these before I lose all enjoyment.

  • I’m fed up.

The main thing that is so frustrating is that I have no idea how to improve these. I know they’re bad – I just can’t see what to change to make them better.


Learning Time – Zbukvic Day 3

Stepping it up a bit today and completed 4 paintings. All are from Zbukvic originals. I was hoping that a fast turnaround would speed up the learning process and consolidate some things into muscle memory.

Before the pictures a summary. None of these came out as paintings. They’re obviously sketches or studies but I started to soak in a few things.

  • Shapes. A few big shapes, more smaller shapes. Make sure the negative space is also an interesting shape.

  • Values. Keep the values even in the big shapes (with a little variation for interest) and you only really need 4 or 5 to make the picture work.

  • Color – ugh. Still need to work on this.

  • Texture. This was hard. Mr Zbukvic has lots of lovely slight variation in his washes which I really struggle to get.

  • Drawing – Surprisingly simple to draw out these paintings. Lots of simplification has already been done by Mr Z.

Anything else? I think I improved as the day went on and was making the paintings more mine than copying the original. No great breakthroughs but definitely worthwhile.

So to the paintings – in reverse order :

This was the last and I think the most successful. I’d got used to the sequence of washes – lightest in sky and the sunlit parts of the buildings. Second the background and shadows on the buildings. 3rd – cars and dark recesses. 4th and final – people and the foreground and other bits and pieces.

I’d also started to get used to some of the wet in wet for windows and background mistiness.

Next one – number 3 :

Struggled with this one. The original has lots of variation in the darks which I just wasn’t getting. The strong composition makes this one – especially the foreground shadow – which is of course all Mr Z’s doing.

Next one – number 2

This had quite a lot I liked. The color came out well. The combination of dull orange and the purple shadows worked without being too in your face. The center of interest has nice lots of choppy darks and lights which read well without specifically being anything. So not great but not too bad.

Finally – number 1

Again this didn’t come out too badly. I was feeling my way at this point and you can see the background washes are a bit muddy. Nice choppiness in the center of interest and the combination of darks and lights worked well.

So it was a busy day. As usual I’m too close to things to really assess whether it was worthwhile. It’s certainly a different beast copying paintings rather than scenes. The simplification and composition has already been done for you which are two things you don’t have to worry about. Once the drawing is done it’s a case of identifying which value wash goes where and trying not to get too fiddly. I think after another day of these (I have nine in total to have a go at) I’ll concentrate on extreme simplification of some of my own scenes.

Learning Time – Zbukvic Style – Day 2

I was eager to start another Zbukvic style copy today. This is one I’ve done sketches of before (see

) which were reasonably successful.

This time it didn’t really come together. I couldn’t feel the big washes properly and everything was a little bitty. One change was that this was done on rough paper and at a smaller size. I think this made a difference. I’ll go back to the original cold press and keep on soldiering on.

The Zbukvic original is this :

Learning Time – Zbukvic Style

Copy of a Joseph Zbukvic painting.

After yesterday’s rather glum conclusion it was back to basics again today. I was having a lot of trouble with pretty much everything. The values were either too different or too similar. The colors were not cohesive. The brushstrokes were too heavy handed. The only thing that really held up was the drawing which is one less thing to worry about I suppose.

So what to do? After spending an hour last evening riffling through pinterest I went back to one of the masters – Joseph Zbukvic. His style is deceptive. It looks like he just dashes things off but that masks a mastery of drawing, composition and above all value. Copying one of his paintings is not for the fainthearted but I was ready and had a plan.

Plan :

– Do a preliminary thumbnail sketch and work out the big shapes and values.

– Draw carefully but not too rigidly detailed. Make sure all the shapes work – especially the negative ones.

– Work out the large shapes ahead of time and what values they are.

* Lightest = sky

* 2nd = sunlit portions of the buildings.

* 3rd = shadow portions of buildings and parts of the cars

* 4th = foreground

* 5th = cars and middle portion of the picture

* 6th and darkest = foreground buildings at the sides and the poles/wires.

Phew – that turned into quite a lot of values. Was only really planning on 3 or 4.

– Keep the values fairly close. No stark changes and keep the highest contrast in the middle area of the painting.

– Keep an eye on color. Use a restricted palette (ultramarine, perylene maroon, yellow ochre).

– Don’t rush!!! Simplification doesn’t mean slapping the paint around willy nilly.

And here’s the original I was using to paint from :

Okay off we go.

First the thumbnail :

Well I’m not sure what I learned here. Was it worth it? It didn’t really feel like it at the time. Well actually it did help. It made me remember to keep the sunlit portions of the buildings pale and mark in where the shadows fell across the buildings. This got lost in the sketch and it shows.

Second the drawing and first washes :

Fairly happy at this point. Drawing is fine and the first washes are light but have some color. Frankly it’s hard to go wrong for this part.

Next is the tricky bit and I think I learned quite a bit here. The plan was to go to the next darkest value and put in large even areas of the same wash. At this stage this means the roofs and the shadow portions of the buildings. However there is a wrinkle. A lot of the sparkle in Mr Z’s paintings come from the twiddly bits. The little dots and dashes and also the small pieces of white left in between brush strokes. In addition his washes aren’t uniform – they have variation in color and texture and maybe have a couple of layers.

I had to remember all this so I went about it as follows :

  • Mix up a big purply wash and use a squirrel mop to block in the larger areas but leaving some gaps and not filling in the wash to the edges.

  • Using a smaller synthetic brush (escoda perla) use the same wash mix to put in the edges but giving them some interest and dottiness.

  • Use the small brush to also put in shadow pieces and windows on the sunlit parts of the windows.

  • Add in some interest to the wash areas but adding in some water.

  • Spray the whole area to keep things moist and add in some color variation. In this case is was a bluer version of the purply mix.

After the buildings I had a first pass at the cars. Things to remember about cars :

  • Keep those horizontals

    horizontal. No scrappiness there.

  • The highlit portions of cars need to be really quite small.

  • Don’t be too accurate about the rest of the car once the windows and the wheels are in.

  • Shadows under the cars are essential for the brain to read something as car.

  • Don’t go too dark too early. Good advice in general.

Things were looking fairly good so I then put in the foreground. It was too light to begin with and I had a couple of goes at it. It’s still a bit light but I felt I’d fiddled too much already. I made sure to get a rough brushstroke at the top where the shadow ended.

So at this point this is what I had :

Frankly I was pretty chuffed at this point. There’s reasonable variation in the washes. The brushstrokes nicely indicate the window and shadows on the buildings and the cars definitely read as cars. A little scrappy in the foreground but you can’t have everything.

Now I was getting worried I was going to screw everything up. All I had left to do was to put in the right hand side darker buildings, the people and the lampposts and telegraph poles. I almost went too dark with all of these but sponged off the worst of it. The final thing has a lot to recommend it :

So what did I learn? Let’s make a list :

  • Take your time over those twiddly bits. They look dashed off but they’re not. Why do I never remember this?

  • Remember the big shapes – keep the value within a small range in these. This is painting 101 – why do I have such a hard time remembering this?

  • Keep those large washes moist by spraying them and add in color/water to add some variation but not so much it changes the value too much.

  • Don’t go too dark too early. A small value change over a large fraction of the painting reads so much better than chopping and changing from light to dark all over the place. I think this is the thing that I mainly took away from this piece. I can also see the regions where I strayed from this concept and they suffer.

  • Horizontals are horizontal. Verticals are vertical. Should be covered by following point one really.

  • A restricted palette really helps.

Now all I need to do is to carry this over into my own paintings.

Life Class

I almost didn’t go as I was so tired but I decided it was worth it at the last minute. Very glad I did – nice to do some charcoal work for a change.