Magic cards generated by neural networks

[Read the post]

1 Like

I have LOVED this account, but I can only read its timeline for a little bit before I start laughing too hard.

“Sacrifice a player? For only a +1/+1? That’s a little overcosted, don’t you think?”

1 Like

This is a bit of a trigger for me… I started playing that game when it was in Beta and rather than hold a job during my last year of college I just sold/traded cards as I’d burned out on it and, hahaha, wasn’t sure how long they would be worth more than the paper they’re printed on. If I still had all those cards at middle current prices I’d be well on my way to retirement. Filed under spilt milk.
Not a great example but having that be my first search certainly didn’t make me feel any better. The next one listed made me feel much better.
Yay regret.

I’d love to see a feedback loop added to these networks. If Gatherer had ratings for different aspects of cards (power, usefulness, coolness, etc.) and the community could give the ratings on those same aspects for the cards generated by a network, I think feeding the community ratings for cards previously generated by the networks would allow them to converge on useful and usable cards more quickly.

After all, there have been some reasonable cards and some that either don’t make sense, don’t work, or are broken (horribly overpowered) – if the community could “encourage” the networks to make more cards like the former I’d love to see where it could go.

This topic was automatically closed after 5 days. New replies are no longer allowed.