William Overington

I thanked Patrick because he replied and particularly because he stated his reasons for his decision. William

William

Over in the Share your work forum, is the following thread started by me. At present 3 replies and 255 views. That picture is made in Affinity Designer (version 1) using the Pixel Persona, and I made it 1024 pixels square, which is the size of the pictures output by Bing Chat AI, so that it is as compatible as possible due to this thread being active. Earlier this evening I tried loading a picture output from Bing Chat AI into Affinity Designer, aligning it precisely using the Transform Panel, which was straightforward as I am using pixels as the measurement units, then moving it to the back. The effect was quite impressive if I may say so, the face in front of the AI generated background image William

There are videos about Bauhaus on YouTube. Bauhaus - YouTube William

The picture that I posted yesterday used a technique that I had not used before, a technique that @Alfred had told me about in another thread, So I added a new Pixel layer and painted in black rgb = (0, 0, 0) with the layer opacity at 65%, which is dark enough to paint clearly but with the original showing through a little. Later I adjusted the opacity of the layer to give, in my opinion, the best effect available, intending this to be part of the face not so well lit, as if there is only light from a window at the left of the screen. Eventually I chose 8%. I then added another pixel layer and added a white area rgb = (255, 255, 255) on the right side of the subject's forehead, as my attempt at a highlight. Eventually I chose 7% as the opacity of the layer. I then exported the painting at its full size of 1024 pixels by 1024 pixels as a png file and posted it in this thread. William

Rather than hide them, could you possibly split them off into a new thread please? Like when Patrick split off some posts into the AI Discussion thread. I appreciate that the suggestion about the beer and the dancing and the video with the dancing could be seized upon, but they did brighten lives, but there was also a discussion about the way that discussions work in this forum and what people are wanting from the forum and so on and recently lots of restrictions keep getting added and they have had a bad effect on discussions, the one on AI discussion that Patrick split off from this thread ground to a halt due to a ban on posting AI-generated images in a thread on AI discussion and I cannot post an AI generated picture in another thread and thus limiting discussion. I regard a wide-ranging discussion, even if it drifts off-topic, as beneficial, like when one has a large lens to collect more light rather than just a tiny lens. Though it is your forum and you are a moderator, so you make the rules. So I am just politely asking please, not demanding or whatever. William

Yet hand weaving survives. As does letterpress printing with handset metal type. So mass production with machinery allows many people to have clothes and books and so on yet hands-on items in the same genres with traditional methods survive, It is a matter for each person to decide in relation to whether to be interested in what generative AI can produce, As a result of cave paintings being mentioned in this thread I looked at pages about cave paintings. Lascaux - Wikipedia I really do not understand the huge difference that is being made between applying an image such as a photograph that has been produced by someone else or from a clip art collection, and applying an image generated by an AI system. The source needs to be credited or made clear. There was a time when Serif used to sell discs of royalty-free photographs. So, in my opinion, making available a collection of images generated by an AI system seems reasonable, provided that the images included are a selection, not including the rubbish ones. William

Well, years ago I learned how to use a lathe. I used the lathe to produce brass screw threads that I could not have produced by hand myself, as a result of lathe settings set up by me. Where exactly is the dividing line between that being acceptable to some people yet those same people finding unacceptable paintings generated by an AI system in response to a text prompt from the human using the system? I suppose that the result expected from, and the result delivered by, the lathe are exactly the same. Yet the result from the AI is not precisely predictable from the text prompt as almost, (maybe always?), inevitably, some unspecified details need to be added by the AI to completely define the resulting painting. William

Following the post by @nickbatz Have you ever seen an astrolabe? The reason that I mention an astrolabe here is that they were made in the middle ages of brass or some such alloy, and what the makers really needed was a transparent material to carry the star map, but they had not got one suitable, so they improvised by using a brass disc with large holes in it, so they had a sort-of-effectively-transparent star map. So the makers knew what they wanted to do and did what they could with the technology available at the time. That seems comprehensible to modern minds. Harder to imagine is what people who produced cave paintings were wanting to achieve. What survives is all we have. Maybe they had other art that has not survived? William

I have added some effects to the picture. I used the technique that you wrote about. It is better than what I was doing as I can alter the opacity afterwards to get what seems the best result. William

William

I am finding that Bing Chat Ai can produce some beautiful paintings, yet can also produce rubbish. I can look at the results and choose those images that I like. I am not a salesperson for AI, I am a researcher and I am investigating what generative AI will do and will not do. This thread is about paintings and Affinity products, yet in another forum there are threads about poems and stories as well. Perhaps I may be allowed to post one link. Art & Literature (Page 1) — Alfred's Serif Users' Forums (punster.me) Bing Chat AI does not do all that I want, I have tried using it and I am impressed by some of its output. It is not as if the images are photographs of paintings painted by humans. The images are 1024 pixel by 1024 pixel electronic originals generated by a computer program based on my text input. The fact of the matter is that Bing Chat AI can produce paintings that are of much higher quality than what I can produce, so I want to try to improve my own abilities, I am not in any way saying that Bing Chat AI can produce paintings of a higher quality than can other users of Affinity products. I notice effects, produced using just pixels of various colours, and I would like to be able to produce such effects myself. It may be that I will not be able to do that, I can but try. William

@Alfred I had no idea that one could do that. Thank you. William

I have been experimenting with using at first black (0,0,0) at 20% opacity then 10% opacity then a grey (64, 64, 64) at 10% opacity on the right side of a copy of the image as we look at it to try to have that side of the face in shadow. Here is an idea, is it well-known or something new? When trying to go over an area with a colour at not 100% opacity, it is necessary to keep the mouse button or trackpad button down all the while so that the partly opaque effect is only applied once. A useful option on a computer running Windows is to use the mouse keys facility to tell the software that the mouse button or trackpad button is pressed and being held down, even though neither of them is actually physically pressed. Thus one may take time to draw all in effectively one go even though one may take one's hands away from time to time without needing to press down at all. William

It is no so much complaining as saying. I always use the Affinity programs on the light grey background with black lettering, and for a start I find tutorials that are black with white lettering difficult to read anyway, and the few that I have tried seem to be fast moving the pointer across the screen, displaying a menu and clicking before I can take in what is going on. I have only just started painting with pixels and I just looked for Affinity tutorials and they are all listed as for version 2. I have version 1. It seems it is like what I saw in a post somewhere once described as the great tsunami of only supporting the latest version. Well, it might appear like that to you and maybe others, mais c'est la vie! William

Thank you both. Which program or programs are those in please? William

How close is that to LNER Apple Green? Maybe the colours could be those of GWR, SR, LMS and LNER, giving them British style. William

I had not thought of doing that. I do not know how to do that. However, if I zoom in to the level of pixels being individually being clearly separate, the output produced by Bing Chat AI has lots of different colours, whereas the one pixel picture that I have done so far is flat with large areas all exactly the same colour. What I am trying to learn is how to get that "looks real" look in a pixel painting. William

I know that Welfare covers more situations that a healthy person being out of work, but in relation to unemployment, it seems to me that regardless of which party is in power, they can never decide whether to be kind and help people who are out of work or to be nasty and punish them. So they end up trying to do both simultaneously, For example, sending a standard letter inviting someone to a meeting where they will try to help and advise and so on, then at the end saying that if you cannot show evidence that you are making an effort to look for work that you could lose your benefit. So a mixed purpose meeting. William

I have produced a first picture and posted it in a thread in the Share your Work forum. I used rgb(255, 128, 128) then painted over most of it with rgb(255, 255, 255) at 20% opacity. Is there a formula known to end users of how, given those two rgb colours and the opacity of the second one, what is the rgb value of the thereby generated new colour please? William

It is not a need, it is a want. Not essential, just something that I want to do. Because the AI outputs a 1024 pixel by 1024 pixel images, and some of them, in my opinion, albeit not a professional art expert opinion, beautiful pictures. So as I can observe that the AI can produce such 1024 pixel by 1024 pixel images as electronic images, I wonder if I can do the same. For me, that is not the same as the alternative that you suggest. I have tried to make a start. Here are two threads. and William

But it does not need to be, it could be kind, empathic and eutopian. It is just policies and attitudes by people who were once new born babies. William PS No, that is not a spelling mistake, it is a separate word from the much more well-known one that has a similar, yet different, spelling and a different meaning.

A first attempt using the Pixel Persona in Affinity Designer @Alfred had kindly drawn my attention to the possibility of painting using less that 100% opacity. I started a new canvas by first setting measurement units to pixels and then generating a canvas 1024 pixels square. So I experimented, first of all by drawing the shape of a face all in red (255, 0, 0) and then using a Basic brush in white (255, 255, 255) at 20% and later 50%. This gave me experience of how it worked. I then started again, continuing to use the Basic brush, by drawing the shape of a face in a softer red (255, 128, 128) and then I painted over most of it, namely all except the lips and the nose, with white (255, 255, 255) at 20% opacity, having previously learned, and now applying that learning, to keep the trackpad down until I had completed doing that, so that the opacity effect would only be applied once. I then used various sizes of the Basic brush, using various colours at 100% opacity, to add the eyes. William

Those that came with version 1 of each program. I did buy a licence for some brushes from the Affinity store, but I might have got them on the old Lenovo computer that broke down. Maybe i can redownload them, I need to investigate that. I am not sure which program to use to start a pixel painting from a blank canvas. Up until now I have almost always done vector illustrations with solid colour areas. No. I had no idea of that being a possible technique to use. Can one paint individual pixels in an Affinity program? William