Avatar
Feβ‚‚πŸ¦€β‚ƒβ‹…Hβ‚‚πŸ¦€
26976d1f3f3270311c86a59315751d66bd770aa35a929529271fca721afee1af
:lainhair: :niggawind: "Why do people put the quotes of others in their own bio? Are you so empty that you must shovel their words into your own mouth and regurgitate them?" - pwm

nostr:npub1w7x9m8xutlezyzmu7d78d8ygzk6sdpncfuuds8jsz2wnwk8f52dqkrpc4w nostr:npub15fkerqqyp9mlh7n8xd6d5k9s27etuvaarvnp2vqed83dw9c603pqs5j9gr I will have to look at this but I was under the impression that the "high quality" versions in the C++ standards passed these tests and still produced bad results. I might just download and test it myself.

nostr:npub15fkerqqyp9mlh7n8xd6d5k9s27etuvaarvnp2vqed83dw9c603pqs5j9gr nostr:npub1w7x9m8xutlezyzmu7d78d8ygzk6sdpncfuuds8jsz2wnwk8f52dqkrpc4w yeah that was back when RNG was insanely expensive to do in realtime. It still is if you're doing it for graphics (noise functions). Ultimately though it's hard to verify that the precomputed list is any better. It MIGHT be but I wouldn't personally know how to test that in a way that isn't insanely placebo effecty

nostr:npub1w7x9m8xutlezyzmu7d78d8ygzk6sdpncfuuds8jsz2wnwk8f52dqkrpc4w nostr:npub15fkerqqyp9mlh7n8xd6d5k9s27etuvaarvnp2vqed83dw9c603pqs5j9gr that's interesting. I'll have to check it out. Even if that is accessible though it may be hard to verify that it works. Would likely still be a good idea to use streak resistance where you can to taste.

nostr:npub1w7x9m8xutlezyzmu7d78d8ygzk6sdpncfuuds8jsz2wnwk8f52dqkrpc4w nostr:npub15fkerqqyp9mlh7n8xd6d5k9s27etuvaarvnp2vqed83dw9c603pqs5j9gr well it's not just their imagination. Having a 10% chance that something will occur if you have bad luck, then having it occur 10 times in a row is a regular event in video games while being unheard of in real life. It doesn't just feel bad, it's extremely unfair.

Computer RNG seems to be extremely susceptible to clustering for reasons I don't understand.

nostr:npub15fkerqqyp9mlh7n8xd6d5k9s27etuvaarvnp2vqed83dw9c603pqs5j9gr I honestly don't even like high stakes RNG. It just encourages save scumming so it's pointless. Like if you're in fallout 4 and you have once chance to apply a game changing speech check, are you going to just say "aw shucks" when you fail it? Hell no.

nostr:npub15fkerqqyp9mlh7n8xd6d5k9s27etuvaarvnp2vqed83dw9c603pqs5j9gr symptoms of using the rand() call because they didn't know any better.

Tip though of you ever want to put RNG and stuff you do: you really need to put in anti-streak mechanisms. The way I do it is that I have the outcomes have counters and each time you get that outcome it gets some percentage less likely that you'll get that outcome, changing the weights each time.

So of you start with A/B with a 50/50 chance, of you hit A the offs next time will be 45/55, then hit A -> 40/60, then hit B -> 45/55. That makes it no longer true random but computer RNG does not at all resemble real life probability in the short term and this is a smoke and mirrors fix for it.

Note that this is bad for security it just feels good.

nostr:npub15fkerqqyp9mlh7n8xd6d5k9s27etuvaarvnp2vqed83dw9c603pqs5j9gr well now I know that I can just save scum and get it to happen in 3 hours in-game time thanks to the janky RNG system. I burned through just about all of my resources waiting for it because I didn't know this the first time.

nostr:npub107wnkjl6crsjs73ukvfjwd89wpvtfcwqdpwnf4xmqfts2pam9deqk89n4a I should spend all day watching these clips to feel better about myself this is the ultimate life hack

>drink coffee

>"fatigue reduced" perk

damn this game is unrealistic

nostr:npub1zhj8u3ap67ufls4rq44kt9u5mlq46ja6xjel32za2ajk8g75efnsevkmgu sleep has been around for billions of years what makes you think you need to fix it