sssssssheldon :
The problem is in your prompting and in the lack of iterations. You give one lame prompt and let it run one time with no refinements. And that’s always the scenario with designers on here, but the reality is, as a designer, you *have* a concept in your mind already. Use that concept and then refine multiple times with generative AI, just as you would in the real world. Perhaps starting with a prompt like this would have yielded better results for you: “Design a hero logo for “Fusion Turf”, an artificial grass company. The logo is a custom wordmark stacked on two lines, with the lettering set at a slight upward tilt for energy. Use heavy, condensed, hand-cut letterforms with one or two quirky letter connections (for example, the tail of one letter flowing into the next). Pair it with a small mascot badge that tucks into the negative space between the words, drawn as a flat one-color line illustration with a retro cartoon feel: a simple character with a personality, wearing something that hints at what the business does.
Constraints: exactly two colors, one bold tangerine on a flat peach coloured background. No gradients, no shadows, no 3D effects, no photographic textures. Vector-style, clean edges, readable at small sizes. Mood: playful, confident, a little cheeky, with a vintage sports-team energy. Avoid generic icons, clip-art, and anything that looks like a stock template.” (That’s how I got this after a single round. Is it perfect? No. But that’s where iteration comes in next.)
2026-10-23 11:29:56