IROG'S PROFILE

The more efficient the tool is, the less human intervention it requires. What should get the credit: the tool or the human ?

AI is also able to generate programming code too. Soon it will be able to make games.

The usual all-or-nothing result of the RNG creates most of the player frustration.
A 5 damage attack at 80% hit chance deals, on average, the same as a 3-5 damage attack.

The 3-5 damage attack is 3 guaranteed damage + 0-2 random damage. So 60% of guaranteed damage + 40% of random damage.
The 5 damage attack at 80% hit chance is 0 guaranteed damage + 0-5 random damage. So 0% of guaranteed damage + 100% of random damage.

In the first case, bad luck can reduce attack efficiency by 40%, resulting in some setback to the player plan.
In the second case, bad luck can reduce attack efficiency by 100%, resulting in a sever failure of the player plan.

Moreover, unlucky players can fail that 80% hit tree times in a row leading to 0 total damage while the other deals 9 damage in the most unlucky scenario. Consecutive failures raise player frustration extremely quickly so the "adding an accuracy bonus if you miss two times in a row" helps.

I would recommand the following when using RNG:
* when possible, avoid the all-or-nothing and make the attack deal a minimum amount
* give bonus to limit the number of consecutive failures
* show the true % chance

You could look at the profile page of the user you want to follow, display the list of games he/she made. Open each game page and subscribe to each.

Thank you thesacredlobo for your constrictive review. You've made a lot of valid points about the flaws in my game.

The programming language I used to develop the game was C. And I used the SDL2 library.