The picture that was actually random looked much more random to me from the start. However, it seems like it would be possible to get a random output that doesn't look random if you were to randomly place the dots enough times, so this seems more like a way to show a probability that something is random, but not determine definitively whether something is random.
Yeah I think the visual comparison thing works more on people who haven't much played with randomness and seen the unexpected 'patterns'. It more exploits people's naïve expectations of randomness.
It's actually pretty common for properly random images to look 'clumpy' like that, so much so that quite a lot of effort in video games which use procedural content goes into tweaking the distribution of 'random' data so it has the right degree of uniformity.
[–] varialus 0 points 2 points 2 points (+2|-0) ago
The picture that was actually random looked much more random to me from the start. However, it seems like it would be possible to get a random output that doesn't look random if you were to randomly place the dots enough times, so this seems more like a way to show a probability that something is random, but not determine definitively whether something is random.
[–] toats [S] 0 points 1 point 1 point (+1|-0) ago
Yeah I think the visual comparison thing works more on people who haven't much played with randomness and seen the unexpected 'patterns'. It more exploits people's naïve expectations of randomness.
[–] tame 0 points 2 points 2 points (+2|-0) ago
It's actually pretty common for properly random images to look 'clumpy' like that, so much so that quite a lot of effort in video games which use procedural content goes into tweaking the distribution of 'random' data so it has the right degree of uniformity.
[–] littul_kitton 0 points 1 point 1 point (+1|-0) ago
Printing too. If you want to represent gray using tiny black dots, you have to avoid unsightly clumping.