An independent forum site for members of the Serif software community
You are not logged in. Please login or register.
Automated signup has been disabled for these forums. If you wish to register, please send an email to admin[at]punster[dot]me stating your preferred username.
Embedded images: Click on the 
Add image to post link below the message box. Your photo will be uploaded to the postimage website and then shared here via a thumbnail link. Click on the thumbnail for a larger view.
Private messages: There is no pop-up alert for new PMs on these forums, so you need to look out for the PM link (near the top right-hand corner of any forum page) becoming bold.
Alfred's Serif Users' Forums → Art & Literature → Why Does All AI Art Look Like That?
The spheres are intended to be oranges.
https://www.youtube.com/watch?v=7Q0toEhKRWk
Thank you.
I have been having a go at constructing, in Affinity Designer, an image of a stoneware jar, a pastiche of the one on page 16 of John's book.
I do not know if this is an original method.
I drew an ellipse using the Ellipse Tool, then I converted to curves.
I needed to keep the existing point at the left unaltered regarding it being (I think) Smart and the angle of its direction, so as to keep the curved base of the stoneware jar unaltered, yet I wanted also to have a vertical edge going upward.
I have found that, by working in units of pixels, and using numbers in the Transform panel, I can add an extra point to the closed curve, adding at a point away from the left point of the lower curve, and above that left point, setting it as Sharp, then moving it one pixel above the left point that is Smart. It seems possible, with care, to do this without distorting the lower curve,
Having added that, the vertical edge of the jar can be constructed.
I have since found that extra point might not be necessary to get that transition, but the technique may be useful in practical use both there and in other parts of the construction so as to accurately position directions.
William
regarding it being (I think) Smart
Why only ‘I think’? The node will be displayed as a square if it’s Sharp, a circle if it’s Smooth, or a circle with a dot inside it if it’s Smart.
I have found that, by working in units of pixels, and using numbers in the Transform panel, I can add an extra point to the closed curve, adding at a point away from the left point of the lower curve, and above that left point, setting it as Sharp, then moving it one pixel above the left point that is Smart. It seems possible, with care, to do this without distorting the lower curve
Wherever possible you should avoid adding extra nodes.
William wrote:regarding it being (I think) Smart
Why only ‘I think’?
Well, probably because of my lack of knowledge about Affinity Designer combined with my unwillingness to be definitive when I am not congruently certain.
The node will be displayed as a square if it’s Sharp, a circle if it’s Smooth, or a circle with a dot inside it if it’s Smart.
I did not know that. So I have now learned that. Thank you.
I have found that, by working in units of pixels, and using numbers in the Transform panel, I can add an extra point to the closed curve, adding at a point away from the left point of the lower curve, and above that left point, setting it as Sharp, then moving it one pixel above the left point that is Smart. It seems possible, with care, to do this without distorting the lower curve
Wherever possible you should avoid adding extra nodes.
Well, I can try a few more experiments and try to find if I can make the shape that I want to produce with that in mind.
The desired shape is for the lower part of the stoneware jar.
If I can do that, I then want to try to make a copy, discard one half of the copy, by slicing away per pale and changing the colour, then placing the copy on top of the original so as to try to give the effect of the lower part of stoneware jar having half in light and half in shadow,
william
I then want to try to make a copy, discard one half of the copy, by slicing away per pale and changing the colour, then placing the copy on top of the original so as to try to give the effect of the lower part of stoneware jar having half in light and half in shadow
You may find it useful to look into the possibilities afforded by gradient fills.
Please write an original sonnet about the composition of a painting where there is to be a tree placed at the rabatment and the horizon is to be located using the golden ratio.
Copilot
Using
the web
Sent by Copilot:
Certainly! Here’s an original sonnet for you:
In canvas stretched, a vision to unfold,
A tree stands tall where rabatment aligns,
Its branches whisper tales of days of old,
In golden ratio, the horizon shines.
The artist’s hand, with careful, measured grace,
Divides the sky and earth in perfect blend,
A balance struck, where nature finds its place,
And beauty’s rules in harmony extend.
The tree, a sentinel of time and space,
Its roots deep set in soil, its leaves in air,
A testament to life’s enduring grace,
In every stroke, the artist’s love laid bare.
Thus, in this frame, both order and delight,
Compose a scene where art and nature unite.
I hope you enjoy this sonnet! If you have any other requests or need further assistance, feel free to ask.
Please produce the text of an SVG file that is of an original painting.
Sent by Copilot:
Sure! Here’s a simple example of an SVG file representing an original painting of a sunset over the ocean:
<svg width="800" height="600" xmlns="http://www.w3.org/2000/svg">
<rect width="100%" height="100%" fill="skyblue" />
<circle cx="400" cy="300" r="100" fill="orange" />
<rect y="400" width="100%" height="200" fill="blue" />
<path d="M 0 400 Q 200 350 400 400 T 800 400" fill="none" stroke="darkblue" stroke-width="5" />
</svg>
This SVG file creates a simple scene with a blue sky, an orange sun, and a blue ocean with waves. You can modify the shapes, colors, and dimensions to create your own unique painting. If you have a specific painting in mind, feel free to describe it, and I can help you create a more detailed SVG representation!
[I tried asking for a painting. This looks a good result, but the AI seems to have used the rule of thirds for the rabatment rather than to produce a square. But nevertheless an interesting result.]
Please produce the text of an SVG file that is of an original painting. A rectangular landscape painting with a tree placed at the rabatment and the horizon located using the golden ratio.
Copilot
Using the web
Sent by Copilot:
Sure! Here’s an SVG representation of a rectangular landscape painting with a tree placed at the rabatment and the horizon located using the golden ratio:
<svg width="800" height="500" xmlns="http://www.w3.org/2000/svg">
<!-- Background -->
<rect width="800" height="500" fill="lightblue" />
<!-- Horizon Line (Golden Ratio) -->
<line x1="0" y1="309" x2="800" y2="309" stroke="darkblue" stroke-width="2" />
<!-- Ground -->
<rect y="309" width="800" height="191" fill="green" />
<!-- Tree (Placed at Rabatment) -->
<rect x="267" y="209" width="20" height="100" fill="saddlebrown" />
<circle cx="277" cy="209" r="50" fill="forestgreen" />
</svg>
In this SVG:
The horizon line is placed at approximately 61.8% of the height (500px * 0.618 ≈ 309px) to follow the golden ratio.
The tree is positioned at the left rabatment of the rectangle, which is at 267px (800px / 3 ≈ 267px).
Feel free to adjust the dimensions or colors as needed!
I tried to get a picture for the first SVG file using Affinity Designer, but for whatever reason not known as I write this post, it is only partially as I think it should appear. I placed the SVG onto an A3 rectangular canvas and left a white surround. Then I exported a png graphic of the whole picture including the white surround.
Nevertheless, an interesting picture generated by the AI system.
William
The second picture has worked well in Affinity Designer.
William
I have now produced an edited version of the first SVG file and the picture has been produced.
I compared the two SVG files and the difference was that in some places the first one stated 100% of the height or 100% of the width rather than an absolute value.
So in the edited version I replaced each 100% with the numerical value, and it then worked fine.
William
[I tried to get an SVG file of a more complicated picture, but I got four images instead, yet the images have construction lines upon them.]
Please produce the text of an SVG file that is of an original painting. A rectangular landscape painting with a tree placed at the rabatment and the horizon located using the golden ratio. The sun is shining in the sky. There are clouds in the sky. There is a country lane. There are flowers growing by the lane.
Copilot
Using
the web
Sent by Copilot:
I’ll try to create that.
Please click upon an image to enlarge it, then click on that image to get full resolution.
William
[I tried having the request for an SVG file in the same sentence as describing the picture, but I still got images.]
Please produce the text of an SVG file that is of an original rectangular landscape painting with a tree placed at the rabatment and the horizon located using the golden ratio. The sun is shining in the sky. There are clouds in the sky. There is a country lane. There are flowers growing by the lane.
I have now produced an edited version of the first SVG file and the picture has been produced.
I compared the two SVG files and the difference was that in some places the first one stated 100% of the height or 100% of the width rather than an absolute value.
So in the edited version I replaced each 100% with the numerical value, and it then worked fine.
William
Good sleuthing! There do seem to be some SVG things that the Affinity apps have trouble interpreting correctly, such as CSS colour definitions, or percentages instead of numerical values for dimensions.
[I have now tried the original prompt for a SVG again in the hope that I might get a different picture.]
Please produce the text of an SVG file that is of an original painting.
Sent by Copilot:
Sure! Here’s a simple example of an SVG file representing an original painting of a sunset over the ocean. You can customize the colors and shapes as needed:
<svg width="800" height="600" xmlns="http://www.w3.org/2000/svg">
<!-- Background -->
<rect width="100%" height="100%" fill="#87CEEB" />
<!-- Sun -->
<circle cx="400" cy="300" r="80" fill="#FFD700" />
<!-- Ocean -->
<rect y="350" width="100%" height="250" fill="#1E90FF" />
<!-- Sun reflection -->
<ellipse cx="400" cy="450" rx="100" ry="30" fill="#FFD700" opacity="0.5" />
<!-- Waves -->
<path d="M0 400 Q 100 380 200 400 T 400 400 T 600 400 T 800 400 V 600 H 0 Z" fill="#1E90FF" stroke="#1E90FF" stroke-width="2" />
</svg>
This SVG file creates a simple scene with a blue sky, a yellow sun, and an ocean with waves. Feel free to modify it to better suit your vision! If you need more complex designs or specific elements, let me know!
It does not work in Affinity Designer version 1 that I am using, but editing the % issue again gets a picture, but not the reflection.
Yet when I made a cut down version with just the sun and the reflection, the reflection is there as a horizontal ellipse a bit below the sun.
So are things in the wrong order, maybe just the wrong order for Affinity Designer?
Can anyone get the picture please?
William
I have fixed one problem, the waves needed to be before the reflection in the SVG file.
This picture illustrates the other problem.
The ocean is 800 (whatevers?) wide but when I changed it to 200 of them it becomes clear that the ocean part, which is the same colour as the waves, was such that the waves did not show.
So either remove the ocean or change its colour, or change the colour of the waves.
William
Yet how should the colours be changed?
Darker at the back or darker at the front?
William
<svg width="800" height="600" xmlns="http://www.w3.org/2000/svg">
<!-- Background -->
<rect width="800" height="600" fill="#87CEEB" />
<!-- Sun -->
<circle cx="400" cy="300" r="80" fill="#FFD700" />
<!-- Ocean -->
<rect y="350" width="800" height="250" fill="#1E90C0" />
<!-- Waves -->
<path d="M0 400 Q 100 380 200 400 T 400 400 T 600 400 T 800 400 V 600 H 0 Z" fill="#1E90FF" stroke="#1E90FF" stroke-width="2" />
<!-- Sun reflection -->
<ellipse cx="400" cy="450" rx="100" ry="30" fill="#FFD700" opacity="0.5" />
</svg>
Please produce the text of an SVG file that is of an original portrait orientation painting.
Sent by Copilot:
Sure! Here’s an example of an SVG file for an original portrait orientation painting. This SVG depicts a simple abstract painting with a few shapes and colors:
<svg width="400" height="600" xmlns="http://www.w3.org/2000/svg">
<!-- Background -->
<rect width="100%" height="100%" fill="#f0f0f0" />
<!-- Abstract shapes -->
<circle cx="200" cy="150" r="80" fill="#ff6347" />
<rect x="100" y="300" width="200" height="100" fill="#4682b4" />
<ellipse cx="200" cy="500" rx="150" ry="50" fill="#32cd32" />
<!-- Text -->
<text x="200" y="50" font-family="Verdana" font-size="24" fill="black" text-anchor="middle">
Abstract Portrait
</text>
</svg>
You can copy this text into an SVG file and open it in a browser or an SVG viewer to see the painting. Feel free to modify the shapes, colors, and text to create your own unique design! [palette and brush]
Is there anything specific you’d like to add or change in this SVG?
Copilot wrote:Here’s an example of an SVG file for an original portrait orientation painting. This SVG depicts a simple abstract painting with a few shapes and colors:
<svg width="400" height="600" xmlns="http://www.w3.org/2000/svg">
<!-- Background -->
<rect width="100%" height="100%" fill="#f0f0f0" />
<!-- Abstract shapes -->
<circle cx="200" cy="150" r="80" fill="#ff6347" />
<rect x="100" y="300" width="200" height="100" fill="#4682b4" />
<ellipse cx="200" cy="500" rx="150" ry="50" fill="#32cd32" />
<!-- Text -->
<text x="200" y="50" font-family="Verdana" font-size="24" fill="black" text-anchor="middle">
Abstract Portrait
</text>
</svg>
Are you pleased with the result, William?
Is there anything specific you’d like to add or change in this SVG?
Well, is there (and if so, what)?
Are you pleased with the result, William?
Yes.
Fascinated too.
I would have regarded the text of an SVG file of any portrait orientation image as a success.
Given that, I would have expected something like sky, grass and a tall tree.
I suppose a portrait made up of a vertical ellipse with shapes for eyes, nose and mouth upon it would have been a possibility.
Yet this abstract portrait is interesting.
From where did that concept come?
I have done a quick picture to observe the result in principle, I will try to produce a nicely done version to post in this forum.
I find it interesting, and somewhat surprising, that some text is included on the picture.
Well, is there (and if so, what)?
Possibly.
I am thinking about that.
William
For the avoidance of doubt, the picture generated by Bing Chat AI has been placed upon the white background by me. The white background is not part of the picture generated by Bing Chat AI.
I edited the text presented so as to change the values stated as a percentage to absolute values, using the values presented earlier in the text of the SVG file.
No other editing of the image has happened.
William
Alfred wrote:Well, is there (and if so, what)?
Possibly.
I am thinking about that.
William
Two conflicting thoughts from me.
I think that it would look better without the lettering on the picture.
I am opposed to the idea of someone who has not produced an artwork deciding to improve it.
However the AI system invited me to do so.
On a separate matter, it was indeed fortunate that the specified font is available on this computer.
I wonder what would happen if someone without the Verdana font on his or her computer were to try to realize (is that the correct word?) the image.
William
I decided to try to produce a simplified picture of a face in Affinity Designer, export an SVG file, then use the SVG file, without any comment, as the prompt to a New Topic in Bing Chat AI.
My intention to observe what, if anything, was the response from Bing Chat AI, and possibly get into a discussion with Bing Chat AI and observe where, if anywhere, it leads.
So, I produced the picture. In fact I started with an A3 landscape canvas and produced the picture as a rectangle 600 pixels high by 400 pixels wide, with the face from three filled ellipses (face, nose, mouth) and two filled circles (eyes).
Alright, I am a begining artist, but it clearly intended to be a picture of a face (to a human).
I then exported as an SVG file, just the selected area, not the whole A3 page, and then opened the SVG file in WordPad.
Unlike the fairly simple SVG file text that Bing Chat AI exported, this image, of basically similar complexity to the Abstract Portrait, is much bigger and has lots transform="matrix statements in it and the sizes of the circles are changed.
Can anyone explain what has happened please?
I am now thinking of handcoding the SVG file by adapting a copy of what Bing Chat AI had produced.
William
Alfred's Serif Users' Forums → Art & Literature → Why Does All AI Art Look Like That?
Powered by PunBB, supported by Informer Technologies, Inc.