My friend and me are having an argument.
I see this picture and say that house roof is pictured parallel to the wall. However this house is drawn in 2point perspective (hill and other elements heavily imply so).
Thus, i say, house' roof shouldn't be pictured parallel, it should be pictured *vanishing*, as it is shown on pic2.
So i say this picture has perspective error. My friends disagrees with me.
Can you provide some insight please?
And a pic2
When the vanishing point is located in infinity the walls will be paralel
Sorry if my english is badly writen.
>>2314505
Yeah, but warped hill and warped wall of house imply that VP is not infinite.
Thus roof must share this VP and become warped too.
>>2314501
i dont understand your point at all, what are you trying to say exactly. What do you mean by the roof being pictured parallel?
Perspective looks fine
You're just setting your vanishing points too close together. The pic2 you drew looks ugly and distorted in comparison to the original picture.
The reason why its pictured 'parallel' is because its far away from the viewer. Heavier distortion like the one you have in pic2 is a result of picking a certain 'lens' or view point close to the object which fucks everything up (this can be good or bad).
Read How to Draw by Scott Robertson, he goes over this briefly.
>>2314516
Thats not what im talking about. Closest edge being too big due too close VP is another problem and it has no, absolutely no relation to the topic.
I just posted it to show that ROOF LINE and WALL LINE must be warping towards themselves in perspective drawing.
However in picture1 roof line is parallel to wall line like no perspective warping is in place.
This 2p perspective picture must have 2 VPS. And House roof and wall should be converging to it the same way the hill sides are convergint to them.
Another example. NOTE THE ANGLE OF ROOF TOWARDS WALL. They are clearly converging!
>>2314522
>>2314524
Ah yes you are right.
no idea what youre talking about OP but perspective is fine to me except for tiny changes and had the urge to make that light brighter
>>2314500
For the love of God, please kill yourself. Pic is fine. You wouldnt question it if it was a photograph.
>>2314524
that's not how it works at all, you have a weird rule in your head that you're trying to apply incorrectly
>>2314575
However your pic shows that roof converges not stays parallel to the wall/floor.
bruh not everything converges like that, that's way too extreme for the OP pic, it looks fine. you don't understand perspective at all, just use your head and think about it. Just think of it as a photo and camera angle, it's impossible for it to converge like you're saying. Dont think about horizon lines or whatever, yours are wrong, just use your head. How would a photo from this angle ever distort in the way youre talking about
>>2314606
Roof should converge to the same VP where wall and floor converges on your picture.
Roof shares VP with wall.
If you think this will be too extreme for a roof, change convergence level of wall and floor firsthand (and other converging parallel lines), because in 2p perspective picture there are 2 shared VPS for all sets of parallel lines.
Lines that are parallel to eacother IRL will converge to the same vanishing point o the picture plane. You will have as many vanishing points as you have sets of IRL parallel lines.
>>2314611
Roof-wall-floor are presumed to be *real-life* parallel thus must share one VP.
>>2314610
stop this thinking about Vanishing points, just think about the 3d space first, you're drawing it all wrong
>>2314626
This is another point to discuss.
A curve of a hill has Vp as well.
Shouldnt house be convergins to the same VPs the hill is converging?
A lot of manuals say that 2p perspective dictates that all objects in cone of vision are placed in two same VPs.
Meaning. In 2p perspective if the hill is curved house on the hill can not be uncurved (if he is not parallel to the horizon).
>>2314633
So? You voluntarily raised line-from-floor angle and lowered line-from-roof angle. What does it proof?
>>2314637
I didnt change anything that was just sloppy, it's the same point, that's how the house looks behind the hill
>>2314651
That's more like it. You are precisely extend the lines you'll see that floor and wall converge at one Vp while roof converges at another (if converges at all).
Considering how roof is implied to be parallel with wall (otherwise it will just float atop the wall on the corner) there is no reason for a roof to converge at a different point. So this is a perspective error.
>>2314657
>if you
>>2314657
you guys are confused.
>>2314626
is right.
you CANNOT see the bottom of the house. All three lines are paralell; there is no perspective. The bottom line just "appears" to be converging with the roof because you cannot actually see the bottom of the house, but it is being obscured by the hill.
Why is everyone putting the horizon line so high? We are below it looking up. The original drawing is right, and all these redlines are misrepresenting the perspective entirely.
>>2314708 here
just drew it in paint. As >>2314712 says, you guys aren't comprehending the bottom of the house being hidden under a curved hill.
>>2314717
Finally! Someone sane
Everybody who gave shit advice in this thread shouldn't post any advice for a year
It would depend on the lense of the Camera
>>2314717
so its an isometric house
>>2314730
No. Do you even know what isometric means? Just because the vanishing points are far away does not make it isometric.
>>2314734
except it is lol
https://en.wikipedia.org/wiki/Isometric_projection
>>2314742
the farther away you are from something, the less the effect of perspective distortion. The fact that it appears isometric proves that it is actually just far away, like it is drawn.
>>2314742
I think you need to actually read the article you just linked. The house is not isometric.
>>2314717
The problem is, this specific house CAN NOT have semi-parallel lines in 2p perspective. If you remember VPs are defined by parallels of edges from top view crossing View point (as on picture).
To have semi-parallel perspective - hence really really long VPs - we would need infinite distance between viewer and object.
You should look to the rest of op-pic - hill, its curves. Does this look like distance to viewer is infinite? Clearly no. Hill dictates medium distance to SP by its curve. House - infinite.
This leaves house out of perspective.
op is fucking retarded. there is nothing wrong with the painting, and even if there were. NO ONE WOULD FUCKING NOTICE!
>>2314791
such autism
>>2314792
much amaze
>>2314805
very retard
>>2314500
They are converging in pic 1. It's just not as drastic as you're thinking it should be.
>>2314979
wow
>>2314610
!!!!!!!!
Tis anon's got it. thes two upper lines converging to the right are parallel, they definitely should have been converging slightly more.
No there is nor rational excuse for parallel line when perspective is applied, unless you're look directly perpendicular.
>>2314500
I think the subtle unease you're feeling is intentional.
Maybe it's meant to portray the house as more sturdy or level than the surroundings. Maybe it makes you question the distance to the house. In any case it seems to be working. Otherwise why did it start such a long discussion?
1) the lines of the house do converge,
2) the two lines of the snow are not necessarily parallel -- the snow is read as uneven and therefore not parallel.
3) compositionally (probably most importantly) it's a very still and serene scene, if the VPs were closer, it would break this and be a more dramatic, tense scene.
