Yes, there is a degeneration of replies, the longer a conversation goes. Maybe this student kind of hit the jackpot by triggering a fiction writer reply inside the dataset. It is reproducible in a similar way as the student did, by asking many questions and at a certain point you’ll notice that even simple facts get wrong. I personally have observed this with chatgpt multiple times. It’s easier to trigger by using multiple similar but non related questions, as if the AI tries to push the wider context and chat history into the same LLM training “paths” but burns them out, blocks them that way and then tries to find a different direction, similar to the path electricity from a lightning strike can take.

Yeah that’s pretty bad. We all know you can bait LLMs to spit out some evil stuff, but that they do it on their own is scary.

On its way.

Yeah, also with physical games, it might be possible to save money, unless they all want to play the same game at the same time.

Stop hitting yourself, argument.

Rumors say they are already dead.

The cat and mouse game is especially why I no longer root my phone. Ain’t nobody got time for dis.

Innovation for the innovation sake. I don’t ever feel my notification miss anything. Not once.

The switch only had Super Mario Odyssey. I don’t count rehash with a short new bonus level. SMO was amazing, but where is SMO2?

Zelda Totk is basically Botw and you just need the later to have the same map. Also both Zelda were no traditional Zelda’s, they were mostly sandbox games.

Kirby was largely seen as too easy, one really has to be okay with that. I was hyped but didn’t expect it to be that easy. Left me kind of disappointed.

Metroid Dread, I wish it hadn’t been a 2D platformer as there are so many of them. Most interchangeable. Looking forward M4 still.

Megaman, I’m no fan of so I can’t say much about it.

Splatoon is amazing and a big selling point. Same with smash, but you could just own one of the dozen different versions on a different system and would not notice.

Pokémon has always been my selling point, but everyone knows the issue with those. I literally have more fun playing old DS Pokémon games, even though challenge never was their strong point.

Nintendo Switch 2 is just Nintendo. While with a portable device you could play so many indie games on the go that either have expensive Switch ports or don’t exist for the Switch at all. Switch 2 is not going to change that. And yeah, I asked myself a lot if I have just outgrown Nintendo games, but truth is Nintendo changed a lot and so have their other publisher releasing for Nintendo systems. I’d not have as much fun with old games, who I have never played before otherwise.

PC is the only console I’m excited about. Switch 1 was already collecting dust and I’m still waiting for that Metroid game that was supposed to be on it. However portable devices are on the rise, like Steam Deck for instance. I’d currently rather get that, than a Switch 2. we don’t even know what games S2 will have.

I hope enough people will buy it in EA, so they can get the required money. But I’ll 100% wait til the game is complete for my first run.

Dead internet, it will be drowned in bots.

Lol, another one of those “players expect perfection”. No, they don’t, but buggy games like Cities Skylines 2 should’ve never been released. Games like Cyberpunk 2077 should’ve been delayed on consoles and not released on old consoles at all, to fix critical PC bugs first. I can understand the issue of infinite delay and finite money, but some games should’ve been delayed just a bit more to not release in a mess.

Starfield however isn’t that buggy, it’s the typically Bethesda-Jank, but what is much worse is the boring game. Not bugs are the big issue. Starfield should’ve never been released or been much cheaper to cover some of the cost, but instead they hype it and sell ultimate editions not worth the price. It’s lying.

$5 alarm clock for $100 because it says Nintendo, nice. Even with its extra features.

For $100 you can buy a phone with android 13, 70€ in a sale.

Jota Text Editor, Connect for Lemmy, google translate for live translation

One can’t comment on your thread. My app says “post not found” was it deleted? Even though I can click the URL and read the comments etc. Some Fediverse thing going on.

Thanks, it’s a complicated read because of the math approximation, but it’s not a bad study per se. I noted my critiques in the original comment.

I wish I could access this study. From the summary alone I couldn’t tell how one would compare Denuvo vs non Denuvo, when only large publisher use Denuvo, which are more likely to be mainstream and more known and therefore more in total pirated compared to an indie title. How would one measure that a DRM is preventing this revenue loss? We barely have two exactly similar games release at the same time. Gaming is fluctuating, game scores are and therefore the sold numbers, some are brand loyal while for others it counts less etc. How has anyone found a way to calculate that? Some publisher are much more likely to use Denuvo but are also more likely to be pirated because their game releases are much more likely to be buggy, boring or mtx infested.

Edit: After reading the study. Some, in my eye, major issues popped up immediately in my head when reading it.

1. The calculations of sale numbers going by reviews.
2. The low number of cracked games within two weeks is only 7 18 in total.
3. The usage of those 7 18 games (8% of the data) cracked in two weeks, to calculate and predict a logarithmic hypothetical number for less than two weeks.
4. it acknowledged the possibility of zero revenue loss but not which week those zero revenue loss games were at. Even though the frequency of 0 revenue loss is higher. Compared to the 20 percent ones on page 10 figure 6.
5. the numbers combine third world countries with very high amount of piracy with first world countries, in all numbers and calculations, while a third world county very likely does provide a much smaller part of the revenue in total per sold copy but gets more significant in using the amount of reviews weighted the same. Also approximating that by cinema visited numbers is kind of far fetched as no good cinema platforms exist in my eye that aren’t as manipulate as IMDb. Which again leads to the necessity to read the other study and conclusions on how this number emerged to use it as base to calculate the number of games sold.
6. page 11 also states that using new reviews as proxy does overstate the fall of revenue.
7. games that are more expensive are more likely to be bought via key store or on a sale, which are not possible to calculate into this as reviews again have a delay depending on type of game and average playtime.
8. going by 18 games this makes 1 and 2 significant as it could very well be online single players with long playtime etc. Making it even more relevant to differentiate what type of game etc.
9. We had, in my eyes, big shifts in game quality and games being patched later on. We have more often than not big disparity between reviewer score and player score effecting both sales of <2 weeks and >2 weeks. Over the years and the study wasn’t short.
10. fair enough games that were released early on Epic and then on Steam were removed from the study, but what about simultaneously releases or releases with or without gamepass especially later games are affected by this a lot, gamepass is cheaper than a key or a sale, some of those games release there on day one by now.

1 and 2 combined in my eyes, picking different games in a different time might lead to a complete different conclusion as we can’t tell if a game of those was of significant hype or disdain. While a review score does effect the probability of a game being cracked, does this work for shitstorms or streamer hypes? I think especially the later will significantly alter the calculations especially going by numbers on a low number of games.

20% of revenue loss is the highest, estimated into day one of less than two weeks cracked, which no game was. Realistic numbers start with larger than two weeks and even then it’s difficult to go by just 7 18 games, which would then go close to 7%, rapidly going down to 5% or less, the more significant the number of games get to make a conclusion.

If you then keep in mind that Denuvo isn’t free for a publisher, we get close into insignificant territory.

My most interesting takeaway is that we didn’t have had any cracked games since a while and the author makes a side jab “are there any groups left?” I can’t find the quote anymore, it was somewhere early on.

I want to add that I in no way see myself as superior or unfailable, if anything this showed me once again, how difficult it is to read a study and my respect for people who can do this better than me. My thesis was easier than this study.

Creating an economic incentive is known for sane country leaders to hesitate to attack their business partner. Sadly Putin isn’t sane.

Heliboard, if you hit that little “>”. Highly recommended as it works well and one doesn’t need the google keyboard. Can also do multiple languages at once.