# Thread: [Archived Class 1] Week Two - Atmospheric Perspective

1. OK. Here's what I've learned about atmospheric perspective.
- The closer to the horizon the objects get, the more their colour becomes closer to that of the sky (which is not always blue)
- The closer to the horizon the object is, the less contrast it has (thus the brightest and darkest colours in the foreground which Gundersen mentioned).
- The farther from the viewer the object is, the softer its edges get. This combined with the merging of the colours towards the sky colour accounts for the less contrast in the distance.
Erm... I can't think of a number four and five. I guess I've learned three things only

Nice studies everyone Great even

Agustin: Thanks for the vids. They're really helpful!

D.Labruyere: Doing these studies really is fun, isnt't it? Good job on both of yours!

rvdtor: Here they are. The pictures I mean. I'm pleased with how the second one came out, but the first one isn't really good... Too bad I couldn't see that yesterday. Oh well...

Gundersen: About the colours getting darker in the background thing - I think that's most often because of shadows. Or just because things in the background are darker coloured. If you look at the images Form has provided, you'll see that the darker background is because of those two reasons.

2. - The farther from the viewer the object is, the softer its edges get. This combined with the merging of the colours towards the sky colour accounts for the less contrast in the distance.
I would like to add something to that and that is: the farther away from the point the viewer is focusing on the softer the edges.

rvdtor: I'm really slow, took around 30-40 minutes to knock one out. Never realised how hard it is to pick the right color. I had a huge problem with getting the right yellow/orange and I wasted a lot of time figuring out the colors of the shadows.

3. D.Labruyere I read something interesting regarding shadows in the latest issue of ImagineFX . They talked about colour theori and how to make shadows. Many people tend to use a black overlay as shadow, and that makes it a bit blurry very often.

What they talked about was that by useing the oposite colour in the "colour wheel", noticeably a very dark version of the oposite colour you get a more correctly shadow. I havent tested this yet, but it sounds interesting and strange at the same time.

Form you got any input on this?

4. that is actually correct Gundersen. Because in reality, when you look at an object you see all the colors, BUT the color of the object.

If we take for example a purely yellow object, and we would shine pure white light on it. It will "swallow" all the other colors, and reject the yellow beams which are sended towards the object. The yellow color will be send to our eyes, and that is what we will see.

When an object is in the shadow, we will see less of the yellow color, since less light, thus less yellow will fall on it, and see more of it's complementary purple.

and if I'm correct the complementair is always the color which comes out mixed by red, yellow and blue,

so:
red = green (blue and yellow)
yellow = purple (blue and red)
blue = orange (red and yellow)

Atleast I thought, that was how it sort off worked :s

5. blue is not the complementar of red, and so on...

what Labruyere said is right,

red - green
yellow - purple
blue - orange

that wheel is a Real Colour Wheel, something a guy made in wich the color darkens more like it does in reality like when you use tubes with pigment, and not like it happens in PS.... color pick a column of that wheel inside PS to see what i mean.

6. try to avoid deleting posts so the discussion makes sense later on...

7. I deleted it 2 seconds after i posted it ... didnt think anyone would be that quick Or maybe you refere to someone else that have deleted something...

Ill write down what gary Tonge says about Atmospheric perspective when it comes to "Getting Composition Right":

Atmospheric depth and occlusion are very important ingredients in solidifying the composition of a piece, be it large vista where the sheer amount of air between the viewer and horizon transform colouring and tonal contrast in the distance, or a smaller area, where light passes through fine particles in the air to create soft diffusion and subtle changes in colouring. Photonic bounce also adds to the relative atmosphere generating diffuse light, which shoots out from lit subjects bathing the surroundings with soft reflected light.

8. heres my first and second attempt

what i noticed while doing this:
as the object moves futher back it takes on the colour of the sky more and more and can actually become part of the sky.
objects closer have more noticeable value/hue range then further away
objects blur into the distance and plane changes become less noticeable

those are the main things i can think of and put into words when i did this. they seem pretty obvious i guess but i figured i should say them anyways
Last edited by DanJohnCox; October 30th, 2007 at 07:53 PM.

9. Post #2 D. Labruyere:

Like that, but push it a bit further. Your observations are getting there... but arent quite right yet. You need to include enough 'detail' to show the complex changes that happen in the receding space. Also watch your 'shapes' - reference your picture as often as you are painting yours if not more. Make sure you check shapes against other shapes again and again. Your different 'strips' of land are all too tall - they are thinner

10. Post #4 Gundersen:

Good first shot. You need to cool your colour down - push it to blue a bit in the background. You seem to have used one colour in different values? You need to pay attention to how the hue shifts - thats a big part of atmospheric perspective. As with Labruyere, you need to carefully check your shapes as your strips are too tall also.

Watch the shapes and values in front too. Your bottom end (darks) seems to be clipped in the foreground... where are those WARM almost blacks amongst the foliage? What about different levels of detail in space? Finally, you seem to have broken the foreground up into shapes of trees that arent really there Save the 'concept' art for your prod piece

Number two, pretty much same things applying. Your darks are clipped/missing, your range of hues is a bit monochrome, and you arent showing the different effects atmosphere has (contrast/temp shift/shape shift etc). And really watch your 'drawing' - your shape in the mid ground is very different to the ref!

11. Post #5 Windmaker:

Great first study. A few drawing niggles here and there - shapes, edges, values, but you got the atmosphere bang on. High 5s.

12. Post #11 Windmaker:

Again, you have captured the essence of the atmosphere here. As far as the contrast goes at least - and great work observing the effect of atmosphere (sky reflection) in the shadows! Some of your temp. changes, even though you have identified them, possibly arent strong enough?

As an experiment, focus your eyes on the edge between the ref and your painting, but observe with your peripheral vision the patch of forest in the left mid ground. You will notice yours is a lot lighter and more tinted to green...

All up though, you have a good way of keeping it loose but suggesting what needs to be suggested to get the atmosphere.

13. Post #12 Gundersen:

I generally start by filling the whole canvas with a gradient or solid fill of the sky colour, and then block in whatever jumps out. The trick to doing studies is when you are 'thinking' about it you need to know that the horizon is 20 miles away, that the big white shapes are buildings, etc etc. But when you are 'looking' and painting from that, you need to forget what the 'objects are' and just see shapes of different colours. Thats all a painting is after all, shapes of colour on a flat surface. So there should be no distinction between painting the far or near areas first - they are all just shapes of colour with equal importance. It is akin to painting a battle scene and painting one character first because he is the one to strike the first blow in the battle you know? I think its just an individual prioritisation.

14. post #14 rvdtor:

your first one is nice... did you double post the ref instead of your final? In either case your more final of the two paintings shows good skills observing shapes and colours. WHat stands out most is the harsh cool white you have highlighted on the rocks with. It really stands out more than it should - but thats a value niggle not an atmospheric perspective topic GW!

The flower painting is great - i think you really captured the essence there. Perhaps your forest floor has too sharp a 'jump' in detail level half way along so it looks slightly more like stage layers than the ref does. only other niggle is that the trees to the left and right of the twins in the middle are a bit too dark... you lose the nice composition of those two trees that the photographer achieved (slightly). Nice work though!

15. post #15: gundersen:

as you said, without the fog, the picture would look different. But in actuality, its fog that causes ALL atmospheric perspective. Atmospheric perspective is caused by the effect of the ATMOSPHERE (ie particles) between the viewer and the object. To really dumb it down, we could say the AMOUNT of these particles between the viewer and the object. So the case in that photo is that the dense fog means there is a similar amount of particles between us and the background as there is over a longer distance on a clearer day. Make sense?

16. post #16 rvdtor:

re: your comment to gunderson. Atm. persp. doesnt only happen when things are massive! see above your comments about the blur are correct though.

your study: watch out for temperature changes - warm greens in the light, cooler and blue shifted in the shadows. also note the detal around the edges of the path in the fg - those little bumps and grooves will help you achieve good atmos persp if you put them in. Mainly just watch the temp change. Cheers for showing process!

17. post #17 Gundersen:

the only thing wrong with that definition is The colours of the object also become less saturated and shift towards blue.
That doesnt always apply. Only under normal daytime blue sky conditions. How about at sunset when the sky is red?

18. post #18 windmaker:

looking good! I really like your confident brushwork and colour choices in this one!

lets hear some more info on the 3 c's if you still have it?

20. post #22: gundersen:

yep, but part of that is paying attention to edges and drawing - so be careful!

21. post #23 Agustin Poratti:

OK! nice... your first study is good. BG and sky a tiny bit too dark... some plane changes missing,... but good. Watch the saturation on the grass at the end of the road. And also the shift in values between the uproght trees and the shorter, bulkier shrubs (in the main vegetation area).

The second study is great. Again, too dark in the far BG. Nice colour choices in the midground.Ur vid is down for me?

22. post #25: gundersen:

good update on your first study of the chinese yellow mountains. I see you have fixed the contrast issue, and added some detail. GJ. But still value is too dark, and not enough blue in the far BG.

In the second study, look at the ref. The background mountains almost look like they have a faint blue GRADIENT laid over them. See how it fades to blue as it goes to the left? That gradual hue shift is what you want to use to achieve the at. persp. Also, keep pushing your OBSERVATION of shapes and colours in the foreground. The value jump from FG to MG on the right side is too steep/jumpy. And note where the path leaves the bottom of the page in the ref?

the final study looks too rushed for me... maybe spend some more time on it? Your drawing is off in a big way (look at the shapes of the distant mountain ranges?) and the values and shapes in the FG are off too (look at the castle wall - flick your eyes from the ref to the painting to notice what is most off). You are clearly getting the hang of the fading values/colours though.

23. post #26: D. Labruyere:

nice! thanks for posting process.. but i think they are out of order or something... no idea what order they go in? anyway assming the one to the right of the ref is the final... you need to watch values (compare the value of the tree line in the midground) and the clarity and shape...of your shapes . The colour isnt perfect (missing warms in the mid ground grass, and not enough blue vibrancy in the bg) but its achieving the effect. This could be quite a nice study - maybe push it a bit more?

24. post #28 D. Labruyere:

Seems like you are rushing these... drawing is way off (where did all the nice peoples' houses go? ) and where did that purple in the bottom left come from? BG is too saturated, hues of the trees are too green (wheres the warmth), top of sky needs more of a blue shift... are you taking enough time on these?

25. post #30 Gundersen:

rule 1: see above - not always blue. why?
rule 2: most contrast. not necessarily brightest colours... sometimes the sky will have a very light value, and the front will be darker... but it will have detail and contrast in its values.
rule 3: this one is more on the money.

26. post #31 Windmaker:

rule 1. what about images where you cant see the sky? Its not the sky then.. but something that reflects the colour of the sky in its particles - the atmosphere!
rule 2. yep.
rule 3. good! edges... edges become soft or lost... and especially when we are painting, we can exaggerate the loss of these edges to heighten the illusion of depth.

27. post #33 Gundersen:

What you are talking about here is colour theory...

In natural sunlight, we see warm light on the object from the direct sunlight, and in the shadows we see cool hues because blue is reflected from teh atmosphere onto the parts of the objetc not lit by direct sunlight. It isnt as simple as going 'opposite on the colour wheel'. Its all relative and depends on the environment as a whole... we will approach this in colour theory later on i think...

an yeah, thats the 'real colour wheel' which is more for subtractive colour (paint) than additive colour (light).

boom

good studies - both of them show what we need to see . Good work also on identifying that plane changes become less noticable. The interaction of planes in the foreground object of a painting serve as one of the easiest ways to direct the eyes to the focus... subordinating those in the background increases the sense of depth.

29. lying prophet, nathan house, jorge gecov, daldbaatar - please join in the discussion soon.

nathanhouse, you didnt complete any of the tasks last week, and you havnt shown up this week - im very close to giving your spot away to #1 on the wait list... get in touch man.

30. a little inspiration, and some more homework:

is this painting from our homeboy Ryan Church a good example of atmospheric perspective? Discuss

Hope its cool to use this image from Mr Church seeing as its in the public domain and all

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•