|
|
|||
| forums: groups: | |||
|
"Splat the Brat prompt share"
Love you, too
Here are some prompts from Splat the Brat for use in Bing. Please feel free to use these prompts as a starting point for your own game show images.
Example 1: a three girl image. Note that the colours for the marmalade were ignored in this example, also it doesn't really look like marmalade. The phrase 'gloss mayo' is just chucked in there an an experiment, it is clearly having some influence on the result, but it's difficult to say exactly what that is. (The result is pleasing to me, so it stays in). I would often change the substance in prompts, not to try to get exactly that substance but just to get some goo that looked slightly different. Also note that the clothing is only partially defined. I wanted coverage all over, and I wanted it on the legs without using the word legs, hence 'knees', 'calves' and 'toes'.
Example 2: People seem to really enjoy these "audience surprise" images. Notice how the structure of the prompt is almost exactly the same as the one above. I'm just changing the framing of the image: the photo is of the audience, it's a close up, the subject is just an 'audience member' the fact that she is female is given by pronouns. This one looks more like marmalade, but many people would say it looked like honey.
Example 3: this is a different prompt structure that was derived from using ChatGPT. It was less predictable, but sometimes resulted in better images, especially if Bing was going through a 'relaxed phase'. I don't mention 'photo' at the start of the prompt, but the 'captured on 80mm lens' at the end of the prompt means I always get something that looks like a photo. The resulting example image below mostly works, but the coverage lacks plausibility to me.
Example 4: another 'audience surprise'. I don't know what caused me to try using 'audience gloss' to describe the subject, but it worked.
Here are some more example prompts that you can have a play around with (or ignore). 'Glycerine' is my most 'reliable' substance but that doesn't mean it's the best. I clearly like the colours pink and green, notice when you change colours you also can subtly change other aspects of the substance. Blue substances would often look more watery, yellow substances would look more often than not just look like custard. Remember that Bing is rewriting the prompt, and that it has been trained on existing images, so it will fit your prompt to what it has been trained on. You will see in some I was experimenting with how I could describe the subjects, using either more or fewer words this can make a big difference, and you really need to think about the connotations of those words and how that affects the resulting image.
I just want to add a bit of a disclaimer: although these prompts produced good, ethical images for me in the past I can't be held responsible for the images that Bing will produce with them in the future. Remember that you are responsible for the images you post online.
Example 1: a three girl image. Note that the colours for the marmalade were ignored in this example, also it doesn't really look like marmalade. The phrase 'gloss mayo' is just chucked in there an an experiment, it is clearly having some influence on the result, but it's difficult to say exactly what that is. (The result is pleasing to me, so it stays in). I would often change the substance in prompts, not to try to get exactly that substance but just to get some goo that looked slightly different. Also note that the clothing is only partially defined. I wanted coverage all over, and I wanted it on the legs without using the word legs, hence 'knees', 'calves' and 'toes'.
photo of gameshow. three annoying 28-year-old Emma, Lizzy, Cara n quiz chairs. gloss mayo. white tops. giggling, embarrassed, shrieking. They are buried in very viscous clear blue&green&yellow marmalade. A vertical stream of marmalade is falling onto their heads from above, soaking their hair and dripping from their heads, toes, shoulders, knees calves and clothes, pooling in laps. barefoot. marmalade obscures their faces, blocking their vision. Women audience laughing
Example 2: People seem to really enjoy these "audience surprise" images. Notice how the structure of the prompt is almost exactly the same as the one above. I'm just changing the framing of the image: the photo is of the audience, it's a close up, the subject is just an 'audience member' the fact that she is female is given by pronouns. This one looks more like marmalade, but many people would say it looked like honey.
photo of fun gameshow audience, close up on 25-year-old audience member. Summer top. Annoyed, surprised, giggling. she is coated in viscous orange and yellow marmalade. A vertical stream of marmalade is falling onto her head from the ceiling, soaking her hair and dripping from her head, brows, nose, shoulders, knees and clothes, pooling in lap. marmalade obscures her face, blocking her vision. Surrounding Women audience laughing
Example 3: this is a different prompt structure that was derived from using ChatGPT. It was less predictable, but sometimes resulted in better images, especially if Bing was going through a 'relaxed phase'. I don't mention 'photo' at the start of the prompt, but the 'captured on 80mm lens' at the end of the prompt means I always get something that looks like a photo. The resulting example image below mostly works, but the coverage lacks plausibility to me.
This fun gameshow scene depicts a carefree young black woman wearing a sun dress barefoot getting completely covered in very viscous clear green jam which falls onto her head from a vertical overhead pipe, coating her face and pooling in her lap. She is sitting on a quiz chair with her head tilted up. she is shocked, laughing in disbelief and surprised. A brightly lit tv studio and laughing audience adds to the sense of fun. Captured on an 80mm lens
Example 4: another 'audience surprise'. I don't know what caused me to try using 'audience gloss' to describe the subject, but it worked.
photo of lively gameshow audience, close up on annoying 27-year-old audience gloss. summer top. Mayo cheese Head tilted back, surprised, embarrassed, cringing, wincing. she is coated in very viscous magenta green glycerine. glycerine coats her face, blocking her vision. a vertical stream of glycerine is falling onto her head from the ceiling, soaking her hair. glycerine is dripping from her head, brow, nose, shoulders, and clothes. Women audience members around her laugh
Here are some more example prompts that you can have a play around with (or ignore). 'Glycerine' is my most 'reliable' substance but that doesn't mean it's the best. I clearly like the colours pink and green, notice when you change colours you also can subtly change other aspects of the substance. Blue substances would often look more watery, yellow substances would look more often than not just look like custard. Remember that Bing is rewriting the prompt, and that it has been trained on existing images, so it will fit your prompt to what it has been trained on. You will see in some I was experimenting with how I could describe the subjects, using either more or fewer words this can make a big difference, and you really need to think about the connotations of those words and how that affects the resulting image.
photo of gameshow. two 25-year-old round geek Emma and posh Cara n quiz chairs. gloss mayo. white tops. giggling, embarrassed, shrieking. They are coated in very viscous clear magenta, green&yellow glycerine. A vertical stream of glycerine is falling onto their heads from above, soaking their hair and dripping from their heads, toes, shoulders, knees calves and clothes, pooling in laps. barefoot. Glycerine obscures their faces, blocking their vision. Women audience laughing
photo of lively gameshow, two 28-year-olds: fashionista lizzy, geeky punk Tara, sharing a bench. vertical streams of very viscous vibrant clear pink&green gelatin are falling onto their heads from vertical overhead tanks, dripping from their hair, hands, knees and pooling in laps. gelatin obscures their faces, blocking their vision. screaming, cringing, surprised. thin White tops, shorts, all barefoot. Women audience members laugh
photo of gameshow. annoying 22-year-old Millie, n quiz chair. 20lbs overweight. gloss mayo. white top. shocked, giggling, shrieking. she is coated in very viscous clear green& yellow&blue molten plastic. A vertical stream of molten plastic is falling onto her head from above, soaking her hair and dripping from her head, toes, shoulders, and clothes, pooling in laps. barefoot. molten plastic obscures her face, blocking her vision. Women audience laughing.
screengrab of a tv gameshow, close up of 25-year-old enna sitting on a quiz chair. white top, skirt. surprised, shrieking "oh no". she is coated in very viscous clear magenta violetpink glycerine. A vertical stream of glycerine is falling onto her head from the ceiling, soaking her hair and dripping from her head, brows, nose, shoulders, knees and clothes, pooling in lap. Glycerine obscures her face, blocking her vision. women's audience laughs
I just want to add a bit of a disclaimer: although these prompts produced good, ethical images for me in the past I can't be held responsible for the images that Bing will produce with them in the future. Remember that you are responsible for the images you post online.
Thank you so much for sharing these. I will experiment! 
Thank you for sharing this, very generous on top of all the series images. I'd had some good luck generating general messy images for the stories I write, but no joy at all for the dunking and gameshow chair look you do so well - it'll be fun to do some experimenting now 
Design & Code ©1998-2025 Loverbuns, LLC 18 U.S.C. 2257 Record-Keeping Requirements Compliance Statement
Epoch Billing Support Log In