PSA: Found out the hard way that Midjourney v6 handles food+object blends way better than DALL-E 3

I spent like 3 hours last weekend trying to get DALL-E 3 to make me a picture of a waffle that looks like a vintage car with headlights and tires. Every result came out looking like a blob of syrup with wheels barely attached. Got frustrated and threw the same prompt into Midjourney v6 out of spite. First try came back with this crispy golden waffle that actually had grill slats made of butter pats and the wheels were perfectly formed. The detail on the car doors as slices of bacon was insane. I think DALL-E just struggles when you ask it to fuse objects with completely different textures like metal and batter. Has anyone else noticed one AI handles mashups way cleaner than the others for specific combos?

2 comments

2 Comments

grace50817d ago

Is it just me or does this kinda mirror how some people are naturally better at certain tasks than others? Like my friend can bake amazing cakes from scratch but burns toast every single time. Maybe these AIs have the same thing going on where their training data just clicks better with certain visual concepts. Dall-E feels like it sticks too literal to the prompt while Midjourney actually breaks the rules to make the mashup look believable.

murphy.linda17d ago

Right? It's like DALL-E is that kid who follows the recipe exactly and gets confused when you ask for a "creative interpretation" of chocolate chip cookies. Meanwhile Midjourney is the friend who sees "mashup" and immediately starts adding hot sauce and pickles because why not. My favorite is when DALL-E gives you a cat with a slice of pizza on its head because you said "cat wearing pizza hat" and it took that way too literally.