Perspective Thread

>>3040221
There's literally a thread about this question once a month but it's never actually answered.

>>3040221
Get a basic 3d program and put a bunch of boxes everywhere and study them.

Can you show us your attempts at drawing boxes? Kinda confused on how to help

File: b954db31-33c6-4455-bedd-ea222ae41db1.jpg (57KB, 577x547px) Image search: [Google]

57KB, 577x547px

>>3040232

>>3040221
Do you mean like boxes that are falling? That don't line up with the horizon line?

>>3040245
OP here, ok wait a sec

>>3040221
the beauty of organic shapes is you don't need to be accurate

File: 20170628_124241.jpg (2MB, 1893x2136px) Image search: [Google]

2MB, 1893x2136px

>>3040245
What im trying to say is that with 2pp i cant get the boxes that represent the ribcage and the pelvis to tilt, so they will end up always parallel to each other

>>3040273
If you actually read up on perspective you'd know about vertical vanishing points. Usually used in creating stairs.

However, in the case of rotating a cube so that it's corner is only touching the ground, I would recommend creating a secondary grid. At the same height, lens distortion, center of vision, etc. Except that the horizon is rotated. That way when you go back to the original it still looks like it's in the same environment, but it's just not converging to that specific horizon.

Do it a couple times so you can get a feel for just freehanding it. It's not terribly difficult.

File: Capture.png (739KB, 969x732px) Image search: [Google]

739KB, 969x732px

>>3040273
For figures you can start by omitting vanishing points altogether to simplify things and focus on what planes are visible.

A surprisingly large amount of poses can be captured using simple cubes and cranial mass proportions.

>>3040291
Program?

File: i think.png (21KB, 1495x656px) Image search: [Google]

21KB, 1495x656px

>>3040280
>Except that the horizon is rotated
this is what i understood

i made two horizon lines, one is normal, the other is rotated, both of them are linked in the middle, not sure if its correct tho

File: c8ec613886a474c534874157b6f80723.jpg (35KB, 509x720px) Image search: [Google]

35KB, 509x720px

>>3040295
It's just some materials loaded into clip studio, made after following Krenz's tutorials and simplified a little more (for study purposes).

File: guide.gif (245KB, 1000x727px) Image search: [Google]

245KB, 1000x727px

>>3040273
Okay I think I get what you mean. I drew up a little gif, hopefully it helps? First time doing this, it might be wack.

>>3040329
Ah also should note I don't actually follow the method I drew up here cause I just freehand drawing bodies, but I think this should be pretty easy to follow.

>>3040310
>te I don't actually follow the method I drew up here cause I just freehand drawing bodies, but I think this should be pretty easy to follow.
This isn't how you study. Pic one pic and draw it 9 times with reff and 9 times from memory. You will improve much faster.

>>3040366
Ah I didn't mean it as a method of study, just as a method of doing it from imagination; OP seems to want to know how to freehand things.

Though you're right that I need to actually learn how to draw these things. I'm lazy as hell.

>>3040387
>Though you're right that I need to
It's all good man. What I found helps for stuff like that is if you find a drawing buddy to make sure you do you daily studies.

>>3040329
very interesting method

>>3040301
Yeah that's pretty much it!
As long as you make sure to keep the cone of vision the same, the camera height the same, the center of vision the same, and distance from the picture plane the same you'll be set!

But a quick way to do it is exactly like you did, having two horizons rotated around a fixed centerpoint.

>>3040613
> cone of vision the same, the camera height the same, the center of vision the same, and distance from the picture plane the same
How do you do that
>having two horizons rotated around a fixed centerpoint
And that
):

>>3040635
I'll have to dig around for the process shots of when I did two 3pt grids in one composition in "creative perspective" class.

>>3040740
Where was that class anon?

>>3040221

> "How do you draw boxes tilting, not parallel to the ground" its driving me crazy!


It's simple, you use a vanishing trace. Basically, assuming you are in two point perspective, you go to any of the perspective points and run a line in vertical crossing through it.


Now, you draw the tilted box, but continue in a straight line until you intersect the vertical line you just drew, congratulations. This is now your vanishing point for that specific tilt of a box.

File: Vanishing trace example.jpg (153KB, 1536x922px) Image search: [Google]

153KB, 1536x922px

>>3041552
>>3040221

Does it make sense now?

>>3040366
>This isn't how you study. Pic one pic and draw it 9 times with reff and 9 times from memory. You will improve much faster.
Proof?

>>3041560
The one with the tilted horizon looks wrong imo

I think you need to go perpendicular with that horizon line for your vertical lines

>>3041560
The verticals of the tilted cube will have a vanishing point somewhere above the left VP and along the perpindicular that passes through the left VP.

>>3041564
>>3041566

Ok, but does OP understand what the fuck vanishing tracers are now?

>>3041563
>you study. Pic one pic and draw it 9 times with reff and 9 times from memory. You will improve much faster.
It's what Dave Rapoza did. Watch his Bogwitch tutorial, the shortened 25 min one.

Start reading, scrubs.
https://www.handprint.com/HP/WCL/tech10.html

>>3041560
the fuck is this bs lol

>>3041575
the fuck is this bs lol

>>3041584
>>3041592

The Science of Art, which I am not in the mood to explaining. Go get yourself a copy of Gary Meyer's Perspective DVD or Vandruff's ''Science of Art''.

>>3041595
draw something in perspective if you actually read all of that crap

>>3041598

What would you like me to draw?

>>3041600
a building in 3 point

>>3041601

Any particular culture/style? Why a building?

>>3041603
demonstrates basic understanding of everything you posted. just draw a building man

File: Erevna.jpg (202KB, 1149x747px) Image search: [Google]

202KB, 1149x747px

>>3041607

You said basic, so here's basic.

>>3041618
There should be one more point going down at the top dude.

>>3041709

Draw me a building and show me.

What if a box's vertical lines are not perpendicular to the horizon line? How do you use perspective then?

File: cropped.jpg (548KB, 1495x1068px) Image search: [Google]

548KB, 1495x1068px

>>3041735

That's called three point perspective.

>>3041735
>>3041737

Actually, you can just use the vanishing trace for that.

>>3041743
wtf is a vanishing trace?

>>3041560
OP here, so each box followed ONE vanishing point from the horizon, and a different vp that represents the box tilting, am I correct?

File: 10-easy-perspectivedrawing-tutorial.png (12KB, 450x353px) Image search: [Google]

12KB, 450x353px

>>3041762

It's literally putting a vanishing point right above or below the vanishing points in the horizon line. If you run a line vertical either above or below a vanishing point in 2 point perspective, you get the vanishing trace. Pic related has an example of how to find the angle of a roof by running the roof up to a vanishing point, in the vanishing trace line.


You can use that to find the correct angles without much measurements, very useful for house roofs, ramps, stairs and other shit. Get Gary Meyers DVD or Vandruff ''Science of Art''.

>>3041781

Yes, you van draw things back into space into any angle of tilt by either going back towards your Horizon Line vanishing points, or your Vanishing Trace vanishing points. You can also make auxiliary vanishing points in the Horizon line if you feel like it, as it happens in real life since most things do not converge to the same vanishing point IRL.

>>3041618
dude
if the verticals are not parallel to the picture plane and are receding to a point below the eye level then you would be able to see the top plane of the building!

you have drawn your bottom and top planes such that this should be a 2pt perspective drawing

go back to the kindergarten stage! perspective made easy by ernest norling!

>>3041820

Hit me up with 3point perspective bares any significance at all in the industry. I have better things to study like light, values, anatomy and increasing my visual library than to spend time learning a system which is hardly used at all.


Feel free to post your buildings.

>>3041820
Not the Anon who drew it but I don't see why you would see the top plane of the building

The horizon line isn't below or above the building. The vertical vanishing point is irrelevant for that

anyone knows a book that teaches vanishing tracers,

File: 3point.jpg (135KB, 1300x792px) Image search: [Google]

135KB, 1300x792px

>>3041825
>>3041830
5 minutes in paint

imagine the box rotating towards you from the top end. as soon as it tilts the verticals cease to be parallel with the picture plane. this creates visual recession and thus 3-point.

another way to create 3-point would be to fly upwards in a helicopter then point your head down. this would give the exact same effect in a vacuum with no other environmental visual cues.

you can't stand with your eye level in the middle of a box/building looking straight ahead and have any of the verticals converge because they would be parallel to your picture plane. if you want them to converge as they were originally depicted then you would have to move the top plane so that you would see it.

>>3041737
I didn't mean 3 point. I meant like something that isn't on the ground. Like a bird flying at an angle. English isn't my first language so I don't know how to explain it that well.

how can i use 3pp in figures?

>>3041928
Sure

>>3041830
i'm not that anon, but i think...
what's wrong in that picture is the vertical receding lines are waaaay too steep and the height of the building is too short to be unable to not see the roof of it. The building has a very warped appearance to it.the way those vertical receding lines are you would most definitely see the top of that building according to it's height because the viewer is absolutely above the building based on how the angles of the building's base recede into the horizon lines.

The windows, the steps, and the door, make the building look small AF. If those were tiny the building would seem more correct, but because of how small it looks, the building appears warped.

>>3041881
yeah this guy gets it. unlike that other guy.

3-point means the viewer is standing above the ground looking at the building i think

>>3041881
the building would have to be tall as SHIT and the viewer would have to be waaaay above the ground at it's midpoint to get the receding lines on both the top and bottom creating a 4 point perspective maybe?? Where the top of the building is so far up above the viewer you not only can't see it's top it actually recedes from the viewer. And the bottom is so far bellow the viewer that it actually recedes from the viewer as well!

I bet anon could demonstrate

File: KRENZ 3.jpg (935KB, 650x4128px) Image search: [Google]

935KB, 650x4128px

Receding means it is extending away from the viewer into space. (this should be what is kept in mind whenever your vertical lines look like this \ | / instead of this | | |. )

if they look like this | | | that means you wont see the top or bottom, period.

If your verticles look like this \ | / you will see it's top

If they're like this / | \ it either means you are standing at it's base and it's a huge object, or you are underneath it as it recedes into space away from you above you.
This image might help a little bit

File: axvyqteq.jpg (150KB, 1536x922px) Image search: [Google]

150KB, 1536x922px

>>3040221
>>3041560
I worry about you sometimes, /ic/. Truly I do.

>>3042166
Birds eye or worms eye view

>>3042261
This! Where did you learn this??

File: 1477703941496s.jpg (2KB, 125x125px) Image search: [Google]

2KB, 125x125px

>>3042261
bump

how and where did you learn this?? is it in perspective made easy??

>>3042313
Why can't you just read it and find out? Don't skim through it and look for specific answers. Read the entire book in order and do the exercises.

You guys are being stumped by beginner-level, middle school drafting.

File: maybe.png (83KB, 839x681px) Image search: [Google]

83KB, 839x681px

>>3042261
This can't be it. Because as you move the vanishing points up and down the verticals. The distance between them increases and changes the distortion of the perspective.

Drawing tilted squares and keeping the same distortion, in this case keeping it "undistorted" would require a different method.

I don't know if pic related is correct, but the boxes are tilted and has the same distortion(vanishing points anchored to the same station point). I have simply tilted the horizon as if you tilted the paper, only they are all on the same paper.

The gray circle is the cone of vision. This way you can tilt the horizon, keep the station point the same distance from the center of the cone and move the vanishing points anchored to the station point and keep it undistorted(or whatever distortion you prefer) within the cone of vision.

>>3042493
None of the original pic is correct, I just wanted to show how why that particular box was most incorrect. Go take it up with him.

But as to your question, and I'm not entirely clear what you're trying to imply so I'm guessing by your drawing, but the horizon line of the box does *not* need to run through the viewer's focal point. The vertical vanishing point needs to rest on a line that is both perpendicular to the box's horizon (the line formed by the other two vanishing points) and run through the focal point. If the box's horizon runs through the focal point, then there is no vertical vanishing point.

All you've drawn are boxes in simple 2 point perspective, rotated.

Distortion has nothing to do with any of this.

>>3042313
>>3042289
The drawing >>3042261 made is just basic perspective. The verticals of the other 2 boxes appear as verticals to the observer since the zenith and nadir are at their furthest possible distance ("infinity"), so there's no vanishing point for them. Everytime you have verticals is due to this alignment between the plane and both vanishing points. If a box is tilted then this changes the zenith and nadir location, thus, they can't appear vertical anymore, but in slight perspective as >>3042261 did. It's not a new method. It doesn't involve new concepts. It's your basic Perspective Made Easy concepts, applied.

>>3042289
GARY MEYERS DVD

Does anyone have an examples of things tumbling? Like books or whatever rotated in many different ways.

File: persp1.jpg (187KB, 1200x538px) Image search: [Google]

187KB, 1200x538px

>>3042493
I think you guys are overcomplicating this.

You're not moving a VP "up and down a vertical, you're moving it around a sphere, and just as a sphere can't be flattened onto paper perfectly, you can't draw a whole vanishing trace on your paper.

Also - in outter space, where there is no 'up' or 'down', there's no difference between a trace and a horizon line, so you don't need to treat them differently. The sphere around you represents an infinite number of horizon lines, tilted in all possible directions.
We're only familiar with one because on Earth we only move on one plane.
Does this make sense?

File: persp2.jpg (317KB, 1200x932px) Image search: [Google]

317KB, 1200x932px

Here's just the left pic. I'd wanted to do this infographic for a while.

>>3043340
>>3043342
I'd pay you for a tutorial damn

>>3040291
That is very ugly perspective though.

am the only one here wondering what in the fuck this thread is even about

like you can't draw a box? how hard is that

>>3041618
You can barely draw basic shapes. Please, don't teach, it gives people a wrong idea.

File: persp2.jpg (235KB, 1000x777px) Image search: [Google]

235KB, 1000x777px

>>3043342
Update, ironically the perspective in mine was way off.

>>3043360
It might be a bit random, but perspective is quite safely the most important of the fundies, so if we can have a thread about gesture or color theory, why not.

>>3043346
Thanks m8, that might be the nicest reply I've received on this forum.

>>3043340
I'm trying to understand this.
>I thought I did
but after drawing some boxes with independent HL's I noticed that they seemed out of place. I guess it has something to do with the vanishing points being behind the SP, but how do you use that?

>>3043423
HOW DID YOU LEARN THIS?

>>3043572
Don't worry about it because its incorrect

>>3043423
What do you mean by spinning that axis? Also, you're blowing my fucking mind because I've never thought about how the second point in a set of VPs is out of our range of vision and goes through the viewer's eye line.

>>3043423
Too many acronyms for it to be useful to others maybe

What is sp? I mean It's supposed to be the camera from a humans eye right?

>>3044021
standing point

File: cubes2.jpg (471KB, 825x588px) Image search: [Google]

471KB, 825x588px

What's so difficul about drawing boxes? Just draw it them lmao. I never even think about vanishing points, vp's only really apply if everything is lined up perfectly over a large distance. You're not architects, you're artists

File: persp2_1.jpg (342KB, 1000x706px) Image search: [Google]

342KB, 1000x706px

>>3043572
All perspective principles are based on nature, so if you ask yourself the 'why' and 'how' questions, you should be able to figure it out logically.
I have to give some credit to Vandruff's perspective series though, that's where it started to click for me. Scott Robertson for practical use.

>>3043600
It (the purple line) doesn't necessarily have to go through the center (the camera), it just so happens because the box's front plane is facing the viewer directly.

About the axis - imagine the box, with a set of lines going back to a vanishing point. If you pierce its top and bottom plane with a needle as if it was a shashlik and spin the box around this new axis, the vanishing point will trace what we call a horizon line.
Do the same again, but put the needle through its side, and the vanishing point will draw a vanishing trace.
Spin the box on all x, y and z axes and the vanishing points will draw a sphere that you can see in >>3043423
Does it make sense now?

>>3043514
>pic related
The horizon lines aren't independent, but they get projected on top of a rounded surface (blue) rather than a flat plane (red, as in >>3042493).
Why rounded surface? Because the roundness of the lens in your eyes (smaller sphere) is what creates the distortion. Rectilinear lens in photography produce no such distortion because they're flat, while perfectly spherical lens in fish's eyes produce full, "fisheye" distortion.

cont.

File: Eye_fish_vs_human.jpg (52KB, 868x443px) Image search: [Google]

52KB, 868x443px

>>3044126
In picture A there's so little distortion we can dismiss it, hence linear perspective.
Pic B is your average curvilinear KJG drawing.
Pic C can't be captured by human eye, but only by an animal with spherical lens or a fisheye camera lens.

>>3044021
Sorry then, but if I were to explain all the stuff you can get from Norling's Perspective made easy, it'd be too cluttered.

File: persp2_2.jpg (428KB, 1000x911px) Image search: [Google]

428KB, 1000x911px

>>3044132
Last pic, full explanation.

If anything's unclear, I'll try my best to elaborate.

>>3044140
i think youre overcomplicating things.
fisheye is easy all you have to do is curve the horizon and draw within a cicle

Obligatory handprint perspective shill:
https://www.handprint.com/HP/WCL/tech10.html

Every one of your linear perspective questions are answered on that website, as long as you're willing to dig for them.

To answer OP's question:
You need to use 3-point perspective for any object that isn't parallel with the ground plane.

>>3044145
People asked me to elaborate on >>3043423 , so I did.
I don't really like the tutorials that tell you what to do, but not why. You're right though, I got carried away and that last post is super cluttered.

>>3043369

Any advice?

>>3044162

I think your post was very useful to me.

>>3044223
>writing a simple comment so that I don't feel like I've completely wasted the past 9 hours of my life
Thank you anon.

>>3044248

No, I meant it. Thinking about the VP's as being gripped to rails contouring a ball that can slide in all directions really makes it much easier for me to rotate boxes in my head.

>>3044269
Then it was worth it. Thanks anon.

File: Perspective.png (126KB, 1180x629px) Image search: [Google]

126KB, 1180x629px

>>3040221

You can draw tilted cubes with one edge being parallel to the ground, as this example. the wedge and the cube have the same angle (red mark).

If you want to draw a cube that it's not parallel to the picture plane and the ground, it's a 3 point perspective drawing.

Unfortunally I still can't find a good book that explains 3pp, without the " just draw 3 points and enjoy lel lel".

>i can eyeball it but it will never be accurate. ;-;

just for the human body it's more simple to estimate, because the body it's a very complex form, the objective is that your drawing must look believeble, not an autistic rendering.

File: Note1_0.png (233KB, 1080x1920px) Image search: [Google]

233KB, 1080x1920px

For a quick, dirty beginner method, I like to use "spings" or coils to help me define perspective. Sorry for the shit example. I'm on break at work, and all I have is my phone

File: Note1_0(3).jpg (142KB, 1080x1920px) Image search: [Google]

142KB, 1080x1920px

>>3044361
2nd example without shitty hand

What if you want to draw a box with vertical lines that are not perpendicular to the HL and horizontal lines that are not parallel to the HL/PP?

>>3044150
how would you use 3pp on a tilting box?

>>3044132
you could just make a small box in the corner where

sp=standing point

etc.

no need to go full explanation but make it so someone can google it. Your information is great but this would let it be a bit more approachable

>>3043340
>I think you guys are overcomplicating this.

I took a shot out of my ass and tried to keep it simple.

Anyway, I figured it out. OP wanted to draw these tilted boxes not flush with the horizontal ground plane. The thing is, technically a box no matter the tilt is always flush with "a horizontal ground plane".

Tilting a box in your hand vs putting the box on the table and tilting your head. It's exactly the same.

>>3044499
Just spin a new horizon line from the main one

>tfw brainlet and don't understand any of this

>>3040366
>you'll tell and convince yourself you're learning faster

>>3044976
Don't worry it's not you. About 70% of the posts ITT are malicious and littered with misinformation, be careful.

>>3044499
That's called 3 point perspective anon.

>>3043340
How on earth can a vanishing point be behind the viewer? That makes no sense to me at all.

How can two lines converge towards a point that isn't in your direction of view?

>>3044550
I literally linked to the most comprehensive and complete guide to perspective on the internet.
Here's the page on 3PP:
https://www.handprint.com/HP/WCL/perspect4.html
I'm not going to spoonfeed you how to do 3PP (that would take way too long)
Stop being lazy, read it and figure it out yourself.

File: persp3.jpg (160KB, 1100x572px) Image search: [Google]

160KB, 1100x572px

Hopefully pic related makes more sense.
The sphere is much, much larger in real life, so don't get distracted by that.
Notice how on the right, the vanishing trace is bending as it gets farther away from your cone of vision. If you were to look up, you'd see it converging to the top vanishing point in 3pp.

>>3044569
Yea you're right, somehow I forgot a tutorial is pointless if only I can understand it. Dumb mistake.

>>3044550
>pic related
You have to add a new vanishing point for every new set of tilted parallel lines.
>a box seen straight on has one set of lines going away from you,
hence 1 vanishing point.
>horizontally tilted box has 2 sets,
so 2 VPs.
>two boxes tilted both horizontally and vertically have 6 sets of lines,
so there should be 6 VPs in your construction, and so on.

Whether you label those scenarios as 1pp or 3pp is up to you.
Thinking only in 1/2/3/5 point perspective oversimplifies the issue, though.

>>3044976
Pick up Perspective made easy or even better, Vandruff's series, if you can get your hands on it. Drawabox has some good info too.
Afterwards, come back and we can discuss more specific problems that are unclear.

>>3045067
>Perspective is the illusion of objects appearing smaller as they turn away from the viewer.
Agreed? Well, then you have to agree that
>there has to be a point where they appear so small they vanish,
right? That's a vanishing point.
If an object, say, 3 miles IN FRONT of you appears as a VP, then same goes for an object 3 miles BEHIND you.
None of those rules are dependent on the position of your cone of vision (/head rotation). See pic related.
If you had you a third eye behind your head, you could see it all. Chameleons can, in fact.

By the way, I really like that you said "to me" in that sentence. It means you take responsibility before blaming others, kudos for that.

>>3045310
How long are Vandruff's videos?
I'll watch 'em either way since 12 bucks ain't much and I'm curious about what he'll have to say, but if the lectures are an hour or more each, I'd like to know so I have an idea of what I'm in for.

>>3045855
12 lessons, 30-40 mins on average, a chalkboard in front of a class. It's kinda comfy to listen to while you do something mundane, like line quality exercises or life drawing.
It's not a necessity, you can be a good artist with stuff Norling, Drawabox and Scott Robertson, but I like that he takes the chaos and complexity out of the subject.

File: deadbabies.gif (85KB, 560x560px) Image search: [Google]

85KB, 560x560px

>>3046139
Thanks for the recommendation. The way he talks makes it pretty comfy, so yeh.

Incidentally, I found these two doods a lil' while ago:

https://www.slideshare.net/MassimoMarrazzo/curvilinear-perspective-ofacuberotated45-degreesmassimomarrazzo

https://www.youtube.com/watch?v=cht57rv07tA

I'm a smidgen curious what anon thinks.

File: example.png (74KB, 814x425px) Image search: [Google]

74KB, 814x425px

Ok where are the vanishing points for the box on the right?
assuming that the box on the left has vanishing points connected to the horizon line that marks off the ground plane.

File: deaderbabies.jpg (907KB, 1680x1632px) Image search: [Google]

907KB, 1680x1632px

>>3046713
There's also these two things:

https://www.handprint.com/HP/WCL/perspect5.html

This one's on construction of ellipses:
https://www.uwgb.edu/dutchs/MATHALGO/Ellipses.HTM

Incidentally, I wanted to know if pic related is an accurate method if you wanted to approximate the angle of an ellipse, assuming the circle is parallel with the direction of sight.
Not trying to overcomplicate things, just wondering if this seems accurate if I wanted to use math to help visualize an ellipse.
Also I'm sorry my handwriting is terrible.

>>3046765
you guys aren't even drawing anymore

File: cxc.jpg (157KB, 1568x1168px) Image search: [Google]

157KB, 1568x1168px

>>3046752
This looks right to me, but i'm still studying perspective, I could be way off. Would appreciate some advice if I am.

>>3046813
Im not that anon but I don't understand this. If both boxes are on the same picture plane how are there 2 horizon lines?

Wow what the FUCK are you niggers doing, just draw a lot of boxes until you can rotate them in your head. What's all this bullshit fuck

>>3046819
Ahh ignore my post, I didn't take that into account.

>>3046813
that's how I used to solve that haha. thanks for trying.

>>3046838
How bad did I fuck up? any pointers?

What the fuck am I looking at? This >>3046765 >>3043423 is a fucking math lesson, with physics, advance polynometric calculus, and 4th dimensional trigonometry, is this necessary or am i never gonna make it?

>>3046950

>necessary

No, run (don't walk) from these

>>3046752
This is not possible.

>>3046950
Those are just bloviating by people who probably can't apply it. Follow the practical advice that actually fix the problems in the drawing.

File: 1498578026301.jpg (71KB, 604x531px) Image search: [Google]

71KB, 604x531px

>>3040329
Thank you so much anon! Your method helps me more than all those expensive books I bought, may the best come to you and your art

File: 1497242941715.png (973KB, 927x907px) Image search: [Google]

973KB, 927x907px

>Mfw this thread
Just when I thought I knew it all you fucks have to make things insanely complicated.

>>3042289
Its best to literally visualize and understand what you're drawing and it's relationship to other objects/space. If you go into a perspective grid to draw cubes and you don't think of every single line you draw is affected by every force--you are over-simplifying it and your shapes will be over-simplified and you'll see a crappy ... form.

>>3046765
>this nigga is literally over here calculating the circumference of the moon in order to draw a couple of boxes
this is advanced autism

File: INFINITE.jpg (932KB, 991x1470px) Image search: [Google]

932KB, 991x1470px

>>3046819

You can have as many horizon lines as you wish.

1 point, 2 point, 3 point all rotated however you want all in one go.

Pic related.

>>3047164
Thanks anon, I had been trying to look up how objects go to different horizon lines all night. What book is that from?

>>3047171
Perspective for comic book artists, by David Chelsea.

>>3047173
Many thanks!

>>3046765
If you actually showed what you're trying to do visually I could probably point out why it's retarded.

>accurate method if you wanted to approximate the angle of an ellipse
>accurately approximate
nigga wut

>>3046825
For anime doodles that might be fine, but if you want to draw mechs, vehicles, environments, extreme perspective or teach, you need to know why things work the way they do.

>>3046950
I think >>3043423 was a pretty good explanation on the problem it replied to. I'm surprised you're surprised perspective involves physics, when it's in its definition.

>>3047204
>there's 7 billion people in the world.
An example of accurate approximation for ya.

>>3047164
Is this the only book on this?
Everyone on /ic/ usually shits on this book.

>>3047164
Any books/vids/tuts that explain the center of vision well?

>>3047339
>>3047445

- Vanishing Point by Jason
- Sketching Manga Style Vol 4 - All About Perspective

Both PDFs can easily be found via google

>>3047339
Because it is a shit book. You will be mislead.

>>3046765
fuck this bs lol

File: unneccesary perspective .jpg (63KB, 600x468px) Image search: [Google]

63KB, 600x468px

>>3046950
It's unnecessary for everything but architectural drawings, some left brained people just like to turn drawing into a science

>>3047543
>turn fun into boring
sounds retarded, anon.

>>3047543
autism.jpg

>>3043423
What the fuck do you do if you want to spin on the x and y axis

>>3047977
You tilt the horizon lines. You can cover that entire sphere in horizon lines and vanishing points.

File: 1493327724640.jpg (33KB, 629x505px) Image search: [Google]

33KB, 629x505px

I understand none of this. I can't even draw a box. I'm never going to make it, am I?

>>3048037
you will make it anon, just stick to the sticky and you'll be fine

>>3045310
Do you have a blog?

>>3046765
I bet you had a nice laugh posting this.

>>3047977
You stay in place, the box rotates and its vanishing points draw a vanishing trace (/ horizon line) on the sphere. The sphere is made out of every possible vanishing point for every possible rotation of a box.

>>3048158
Thanks, but not yet. I'll stick around /ic/, draw porn for a few years, make it and write a book on perspective.

>>3048037
then I guess you have to
>drawabox.com

Don't worry bruh. It will all come naturally.

>>3047164
This doesn't make sense to me, my English is a bit slack. Anyone have any other tutorials?

>>3045310
maybe what you're talking about is curvilinear perspective which I don't know much about but a lot of what's in this thread is about linear perspective which is what one should use for most ordinary depictions of life. especially for beginners.

in linear perspective, as parallel lines move (not turn) away from the viewer they converge to a point. the distance between them APPEARS to get smaller even though it doesn't. this appearance/illusion has to do precisely with the cone of vision which is a function of the way in which light rays enter the focal point of the eye.

when the eye is born the point of the cone is born and all the rules of linear perspective are a function of this cone. perspective is entirely subjective. vanishing points are dependent on a unique point of view in space. they only exist to explain the terminal point of lines moving away from a subjective experiencer.

as parallel lines move AWAY from the vanishing point and TOWARD the picture plane they diverge. lines moving to the back of the viewer are never converging to a point in linear perspective - they only converge when extended in space. speaking of a vanishing point behind the viewer is a violation of linear perspective rules

>>3041568
Link?

>>3049055
Nvm it's on cgpeers

>>3049010
>rules of linear perspective are a function of your cone of vision
>speaking of a vanishing point behind the viewer is a violation of linear perspective rules
bruh
You might be surprised, but an object can move away from you even when you're not looking at it.
Any rules of perspective relying on your cone of vision is a faulty assumption.

The only thing light entering the focal point of your eye does is "focus" your vision, i.e. how blurry it is. Nothing to do with perspective.

>linear perspective
Everything you see is in curvilinear perspective.
We simplify it into linear perspective by assuming lines perpendicular to the line of sight are at a constant distance from the viewer, which isn't true and will become apparent once you move out of your cone of vision.

That being said, I have a good deal of respect to you for trying to understand the subject, so yea, keep making assumptions and keep asking questions, that's the way forward.

>>3049010
>>3049206
I just realized I have to admit being partially wrong.
Linear perspective doesn't apply outside of the cone of vision, therefore it IS a violation of its rules to talk about convergence behind the viewer.
However linear perspective itself is a violation of real world (curvilinear) perspective, so to understand how perspective works, rules of linear perspective have to be broken.

I don't understand any of this. Guess I'm never gonna make it.

>>3047472
Wtf. Don't listen to this guy. Those books fucking suck.

>>3049265
This
Perspective made easy is all you need you idiots

>>3040291

what is that program, looks like Clip but if so, where is that function?

>>3040291
Can you share those models/brushes please?.

>>3049265
>Those books fucking suck.

Great argument dude

This is the most quality thread on this board right now, goddamn.

>>3049874
Literally blind leading the blind: The thread

File: 1468551177065.png (229KB, 719x585px) Image search: [Google]

229KB, 719x585px

>Someone asks how to draw a box
>/ic/ has another bout of autism and manages to conjure up this fuckery >>3046765

Stay golden, /ic/

>>3040221
did anyone even give you the answer you were looking for?

>>3040221
Don't use perspective for organic subjects

>>3050304
Good advice dude. Enjoy flat drawings and paintings.

File: tumblr_op2y1gFRF81rnv6lfo1_500.gif (750KB, 500x388px) Image search: [Google]

750KB, 500x388px

guys what the fuck

>>3050358
The point is not to eyeball boxes in perspective. That's easy enough once know the basic rules, and that's all you need for nearly everything.

Will it be "accurate" though? Not at all. It will look believable, but the point was to answer how to draw 2 boxes at different tilts accurately in the same space.

Any good supplementary materials to go with Norling's book as an absolute beginner to perspective?

>>3050586
Drawabox. Vandruff's series if you have $12 or patience to torrent it.

>>3051333
>bumping this thread
>when 99% of /ic/ doesn't know what any of this means

>>3044073
These guys like reading books instead of actually drawing

>>3050358
is that supposed to be impressive or something?

>>3051413
Show us how well you can do that

>>3047543
Its... Beautiful

>>3047543
what is this? a pic for butterflies?