//=time() ?>
One of my test cases on AI image comprehension. "an angry drunk computer tears apart a cabbage, hyperrealistic, covered in gravy, in a kitchen"
The new one is definitely "better," but still very wrong.
"WWII AFV, town plaza, Panhard aml"
HUGE improvements over ten weeks.
Still not *right*, but way better.
"descent into hell, fiery rock, the gods are dead, doom of mankind, apocalypse"
Again, one option is for framed art in the new version.
"detective in a trench coat with a Tommygun, comic-book style, bright colors, pulp adventure"
#Midjounrey still has no idea what a tommy gun is.
"Strong female warrior, clothed, in action, dark skin, green eyes, pulp cover style"
10 weeks ago (left), and today (right)
The new lower-left is one of the best hero shots I have gotten to date from #Midjounrey
Since I am trying to explore the capacities of AI image creators, I did "impressionist water lilies" 10 weeks ago (left). Here is the same prompt today (right).
"New York City as a feudal castle, urban setting, cyberpunk castles, gritty street-level cityscape, skyline"
10 weeks ago, and today.
"armored space knight" ten weeks ago and today.
"Skeletal warrior," from ten weeks ago, and today.