Comments: 94
nasshu31 [2021-05-05 10:01:15 +0000 UTC]
👍: 0 ⏩: 0
RENNYMEEL [2020-09-14 09:14:28 +0000 UTC]
👍: 0 ⏩: 0
PatternSeeker86 [2020-03-27 22:16:20 +0000 UTC]
Delightful. Thanks for sharing your beautiful work.
👍: 0 ⏩: 0
SCBart [2018-08-20 09:55:50 +0000 UTC]
your tutorials are great!
👍: 0 ⏩: 0
FantasyRebirth96 [2017-09-05 14:48:10 +0000 UTC]
Sorta kinda confused on this still. Maybe I should re read this. But I'll give practicing some perspective shoots a try and do my best to grasp the idea of it.
👍: 0 ⏩: 0
yezzzsir [2016-10-31 09:04:41 +0000 UTC]
Great job on this. Thanks for taking the time to put it together.
👍: 0 ⏩: 0
askoldnemirov [2016-10-02 18:32:59 +0000 UTC]
broke my brain a bit, but got it! Thanks for your work! Would really like to buy your book, if u will ever write one!
👍: 0 ⏩: 0
waltervan00 [2016-09-15 12:36:48 +0000 UTC]
Yo, great eye-opening perspective discussion. Upon reading it, I immediately wanted to try and do it myself.
And, only naturally, questions arise.
The orientation and placement of the beholder was really well explained. However, when it came to the objects with the scene, I find it a little hard to apply the logic to it. Say, you are looking at a cube from a 2-point perspective base, how will the guidelines bend and distort? I tried to identify the plane of a wall seen from an oblique angle and trying to apply distortion from it. It was stuck on whether the guidelines would bend up or down (making the wall bulge in, out, or warp around the image). If my question didn't make quite any sense, then I would happily link a crude diagram of my issue. (this applies to ground planes as well).
Another question I have is about you demonstration of wrapping up a panorama into what I assume is 4 Point perspective. I sometimes wonder whether that scenario could just be done in a 1 Point perspective. Also, about the demonstration with the worm's eye view of the room, may I ask why the guidelines are bent as such? Looking from the other examples, it could be related to the beholder's height in the scene.
I do understand that there are six vanishing points and distortion could be used to bring more life to the straight and stiff perspective scenarios. (which sounds very rewarding).
Cheers. I look forward to seeing you reply, once you have the free time to do so, of course.
👍: 0 ⏩: 1
Nsio In reply to waltervan00 [2016-09-17 07:41:24 +0000 UTC]
Note that in 2 point perspective, there are only 2 vanishing points and no more. Thus, you cannot distort the guidelines, because there are no other vanishing points for them. If you apply distortion here, it´s no longer 2 point perspective, because the distortion suggests that the beholder is turning her eyes and, as a result, revealing the influence of other vanishing points.
You can check how things distort in perspective by holding something (like a book) close to your eyes and then watch what happens in your peripheral vision. The larger the object is, the easier it is to see the distortion. Basically the closest part of the objects seems to buldge towards you, while the edges seem to bend away from you.
No, the panorama view cannot be done with 1 point perspective. If you don´t acknowledge the other vanishing points on the left and right (or sometimes behind you) and keep the lines straight, you get the effect as seen on the first example of "unwrapping 3D" part. If you force things into 1 point perspective but apply distortion, then it´s no longer 1 points perspective, and it may not be consructed correctly if you don´t know how the others vanishing points affects the view.
The room looks distorted because you normally wouldnçt see the scene like that. Your brain will treat the lines straight, it doesn´t need to know that the view is actually bending. The height of the eye level on that scene only makes the distortion above more pronounced, because the ceiling is farther away from the eye level. As I said before, try seeing what´s going on in your peripheral vision, and you ought to see the bending.
👍: 0 ⏩: 0
Tsiternish [2016-08-08 22:18:29 +0000 UTC]
I saw that effect the most clear when I got myself the new glasses. They made my vision sort of... closer to what you get with the wide-angle lens. When I first wore them, I realised that what I was taught about perspective was not completely valid. But I guess the job of an art teacher (or any teacher, perhaps) is to teach students basic things and shortcuts and encourage them to explore the more advanced material on their own.
👍: 0 ⏩: 0
merser24 [2016-07-31 06:53:21 +0000 UTC]
Wait so is this stuff supposed to be for everything? or drawings with distortion applied? First slide got me confused lol.
👍: 0 ⏩: 1
Nsio In reply to merser24 [2016-07-31 11:35:13 +0000 UTC]
Basically, it's supposed to be for everything, if the goal is to achieve real and natural perspective. However, in many cases the distortion is so insignificant it can be ignored. Sometimes the effect isn't desirable at all.
👍: 0 ⏩: 0
AoiKen [2016-07-30 12:59:55 +0000 UTC]
this is very very hard for me to understand.... sigh... I can never be able to understand this...
👍: 0 ⏩: 0
AsjJohnson [2016-07-13 04:18:31 +0000 UTC]
I have noticed that curve kind'a thing before, like when you look at a tile floor, and follow it to where you're standing, and behind you, where the vanishing point's at both ends. But I wasn't really sure how to incorporate it in drawings, and in the past I've often done that building/cube you have as the first, wrong, example.
It's nice that someone else has noticed that type of thing. (...haven't looked at the whole tutorial yet, but I'll get back to it when I have time. I thought I'd mention this now)
👍: 0 ⏩: 1
AsjJohnson In reply to Nsio [2016-07-13 15:05:58 +0000 UTC]
yeah, tile floors, and also tile(ish?) ceilings. Probably ceilings more so. Then, there's also the issue with cameras, where if you take a photo in a small bedroom, all the walls are bowed. Or that panorama example you have (those can turn out not-quite-right with cameras, maybe partly because of the other vanishing point).
It seems like having the slight arcs in pictures can really help make them look more real or organic. I noticed that even in that little building example where you do it both ways. It might make the difference between a cityscape or inside of a building that looks grid-like and harsh, and one that looks 'right' and more inviting.
...but, I wonder why no one's come up with a better perspective theory before. I could tell something was a little off about it for a long time, but there didn't seem to be anything about it academically. I suppose it would've been harder to tell before cameras were invented, though. Or maybe the perspective lines were always meant to just be a rough guide instead of taken literally.
👍: 0 ⏩: 1
Nsio In reply to AsjJohnson [2016-07-13 16:39:03 +0000 UTC]
The current theory of perspective just works so well and it's easy to construct. Since all lines are straight, it's easy to draw everything with a ruler, it's easy to set useful reference points and place the elements in perspective. When there are no straight lines, the benefits of perspective grids are lost. The perspective changes all the time so you can't use any ruler or define useful guidelines. From there on it's just about making decent guesses.
Anyway, the current theory isn't wrong by all means. It's just simplified and designed to work in certain situations where minor distortions don't really matter. The rules are the same with "real" perspective, it's just that the beholder acts more like a static camera with limited field of vision. All in all, perspective is just a tool and theory to simplify the reality and art is all about making convincing illusions, so it's not that important to follow everything literally (as long as the artist understand what she is doing).
👍: 0 ⏩: 0
poodlelover162 [2016-05-17 02:28:20 +0000 UTC]
Me: - prints this out - grabs hilighter - starts to take notes -
👍: 0 ⏩: 0
Holydiver79 [2016-04-23 12:00:58 +0000 UTC]
Thank you thank you ♡
Love your Nsio Explains. Great tutorials!
👍: 0 ⏩: 0
dreama-draws [2016-04-09 16:22:57 +0000 UTC]
any books you recommend?
👍: 0 ⏩: 1
Nsio In reply to dreama-draws [2016-04-10 17:15:30 +0000 UTC]
Not on this subject but "Making Comics" by Scott McCloud is pretty awesome book. Although it's aimed for comic artists, many things in it can be applied on any field of art.
👍: 0 ⏩: 0
birdarmyspy [2016-03-15 12:31:22 +0000 UTC]
Hey was wondering how to you determine which way the distortion guides bend? For the two examples on the right the distortion lines coming towards the viewer above the horizon bend in opposite directions. Why is this?
👍: 0 ⏩: 1
Nsio In reply to birdarmyspy [2016-03-15 15:38:27 +0000 UTC]
Very good question, I recall I wondered the same thing as you do now. The top right example (my profile pic) is actually faulty by the rules I presented. There the distortion is more about dynamism rather than distortion of perspective caused by eye movement. I think I wanted to "curl" the scene around the focus point for the sake of composition, but honestly I'm not sure what I was thinking when I drew that.
Another explanation could be that the eyes of the beholder move sideways along a linear path. So if you look at the vanishing point, the edge of the table is on the left. Once you look at the intersection of the table and canvas edge, your eyes has moved to the left. So the viewpoint slightly changes depending on which area you are looking at. It sounds weird and I don't think that was my original goal when I drew this.
So as a general rule, if you have a vanishing point in front of you (as you do), the guidelines bend away from you once you start moving your eyes around.
👍: 0 ⏩: 0
Musou-Enrai [2016-02-17 22:31:19 +0000 UTC]
b/a = c/d
Is this really true? For a specific pose the bottom part from the horizon down is equal no matter the depth? a=c?
👍: 0 ⏩: 1
Nsio In reply to Musou-Enrai [2016-02-18 06:08:31 +0000 UTC]
As long as the character height remains the same, the horizon cross the body at the same location, so yes, b/a = c/d. No matter how far the character is, she will still be human sized so c=b and d=a -> b/a = b/a. The height is only visually different in 2D. If you could measure the b and c or a and d, you would confirm them different (you just can't take in account the depth in your measures). The proportional relationship between the measures are still equal. Just like 8/4 = 4/2 = 2/1.
a isn't equal to c. This would be the case only if the horizon cross the body exactly from the middle, so a=b=c=d. 1=1=1=1. Of course, in 2D the measures are still not equal, just like in the case above.
👍: 0 ⏩: 1
Musou-Enrai In reply to Nsio [2016-02-18 22:36:33 +0000 UTC]
Oooh! This is terrible! But a fine example. I see what you mean.
I've found for myself that when I draw some perspective guidelines(horizon and lines from both ends that form a box for the character), the drawing turns out rather fine. Because I see how the figure sits in perspective and what parts of the body what directions are facing. But all that works when I don't mess up the guidelines and I manage to nail the figure somehow.
And when I draw just a figure without perspective guidelines, it turns out crappy 80% of the time. I guess with time and practice, artists get better and they know where the guides are and don't need to explicitly draw them. I remember seeing an artist drawing Chun-Li. He started with clean lines from the get-go. Almost no guides(if any at all). Damn...
👍: 0 ⏩: 0
MinatoLikesSmoothies [2016-02-13 00:40:56 +0000 UTC]
Very helpful. I didn't think of the lines bending like this, but I agree, the straight lines always seem too organized to be how people really see the world.
👍: 0 ⏩: 0
RoseJewel21 [2016-02-10 22:11:58 +0000 UTC]
This seems useful. I hope this will help me.
👍: 0 ⏩: 0
yuhsaku [2016-02-10 14:44:57 +0000 UTC]
Nice explanation on the distortion. There are several instances that perspectives needs to be better that we usually do and it gives that feel on how perceive it with our eyes.
👍: 0 ⏩: 0
TriadischArt [2015-12-17 06:53:06 +0000 UTC]
another bad ass tutorial niso!. thanks so much!
👍: 0 ⏩: 0
mayodha [2015-12-08 17:14:28 +0000 UTC]
Your Tutorials are the best ...helps a lot..
👍: 0 ⏩: 0
Howso-ARTlee [2015-12-01 05:16:49 +0000 UTC]
somehow I knew this subconsciously after trying to make my own perspective, and constructing spheres inside out. I just never knew why i knew.
👍: 0 ⏩: 1
Nsio In reply to Howso-ARTlee [2015-12-01 14:04:03 +0000 UTC]
That's how it is. Probably most people understand perspective in this way (subconsciously), but we are taught it differently. We cling to the thought "this is how I was taught, my own views must be wrong".
in order to become truly a great artist, one has to understand why there are rules and when it's okay to break them and devise your own. We are taught classical perspective because it's more useful tool in constructing perspective. I prefer slightly distorted perspective because it takes eye movement in account.
👍: 0 ⏩: 0
WhysoGurin [2015-11-30 22:16:59 +0000 UTC]
Great explanation! and you actually pointed out what most frustrates me.. thanks for this great tutorial bro...
👍: 0 ⏩: 0
Emperor-Koto [2015-11-18 20:26:22 +0000 UTC]
I've never even *heard* of this concept... Thanks so much for taking the time to create and share this!
👍: 0 ⏩: 1
Utrilus [2015-11-06 21:14:44 +0000 UTC]
Just as I was about to give up on looking for information on curvilinear perspective. A thought occurred. There is one person who knows EVERYTHING, and makes totorials.
👍: 0 ⏩: 0
Blirtt [2015-11-03 04:27:34 +0000 UTC]
yassss, finally someone gets it 0_0' Its actually impossible to draw something that covers a persons full range of vision such as a background with out curves, well acurately atleast.
👍: 0 ⏩: 0
everWinter [2015-11-02 09:54:47 +0000 UTC]
Thanks for the in depth explanation, I had not considered that before.
👍: 0 ⏩: 0
parameciumkid [2015-11-01 04:12:51 +0000 UTC]
Make your art better with SCIENCE!
👍: 0 ⏩: 0
| Next =>