@gus_massa, Yes, ChatGPT does a good job of creating grids of same size. If you scroll to the end of the blog I have shared three versions of the game. You can use them and I also have shared the prompt to generate it.
In the "ice cream" example, the cones are taller than the others, but in the images you cut the tip to play.
In the "house" example, the labels are outside the cards, so it would be difficult to cut.
The "unicorn" example is perfect.
So my guess is that in the different examples you improved the prompt.
---
I really like that your examples have no numbers, so each family can use their preference. I imagine it would be very hard to convince my daughter that it's better to keep the "5-chocolate" and drop the "1-mint-with-chocolate-chips". (She would understand the problem and ragequit.)
@gus_massa, True, I improved prompt (by asking ChatGPT to rewrite the prompt) and even models have improved in last 2-3 weeks. Unicorn was created recently with ChatGPT 5.2 while others are from older version.
Incase you are interested, this is the prompt for Unicorn. You can modify it for others.
—-
Create a printable children’s game sheet with a 3 rows × 5 columns grid of equal-sized square tiles, arranged neatly with clear spacing.
Background: pure white, clean, uncluttered, printer-friendly.
Art style: soft watercolor pastel, cute, kid-friendly, gentle colors, soft hand-drawn outlines, minimal detail, calm and cheerful.
Row 1 – Unicorn Horns only:
Spiral horn, Crystal horn, Golden horn, Rainbow horn, Star horn
Row 2 – Unicorn Bodies only:
White unicorn body, Pink unicorn body, Blue unicorn body, Purple unicorn body, Golden unicorn body
Neural network was introduced in 1950s. However, the neural architecture, the compute and data required for them to be efficient has been only in last decade.
From perceptron to transformers (few hidden layers to 480B parameters), from multicore CPUs to distributed GPUs and WWW/social media has all contributed to the growth of Artificial Intelligence.
This has took almost 50+ years and so many iterations along the way.
I understood how LIGO works but one question I have is, how did scientists conclude the gravitational waves they detected are from a collision of 2 black holes that happened before 1.3 Billion/Million years ago.
A blog on how human learning has evolved over past decades, how we are learning actually the way computers are learning today i.e. using machine learning. Please read and share your thoughts.
CloudBoost offers a set of developer tool for parse developers to reduce their coding time by automatically generating boilerplate code and providing other developer tools.
As a part of alpha release, using CloudBoost you can visualize app structure (by drawing ER diagrams) and create all tables in Parse.com, generate CRUD cloudcode with just push of single button.
For next release, we have planned to work on better data browser, query analyzer, debugger etc...
Will such developer tools for Parse platform really help reduce your efforts. Kindly try and post your feedbacks.
