Yeah guilty gear strive and ofc the fgc is absoulte ass cancer and i've seen mfers get groomed into non binary or troons cause of strive having those characters. fucking munchausen retards.
I swear the Ye Olde English stuff is annoying, it feels like i am reading Peter Pan and then getting goofy words like Dopey, Blueblood and then Piss, it feels forced onto these characters because the Translator finds it to be bland dialog even though it's standard for the rest of the world. Hopefully a lite retranslation patch comes out for this because I do worry that these are not going to be the worst examples.
And already this game, has had the fucking level of insanity that any atlus game these days has. Its all fucking tiresome and beyond pointless altus/sega deseves all the shit that they get for cultivating and rewarding bad behavior with their fandoms. :/
They have no skill to make simple speech sound natural so they overcompensate with something that ironically makes them look even less competent. Hell, a lot of the time they don't even know JP...
Judging by the Cygames Localization Computer Screen it's clear they use DeepL and try to retranslate it back to make it sound like they did it or they can write when they should just make it sound clean and normal. They do not know the language hardly at all. Maybe that's why they keep telling everyone to learn Japanese because they know they are scam artists and living a lie.
That twitter bio is enough telling already. If the demo are already like that, it's easy to imagine the "colorful thesaurus" and quipfest rewrites xhe's going to put in the main game.
“queer jewish”
Localizers are awful when they do this. I mean the fact the original ended up just translating it as is and this tries to "punch up the text because Japanese is boring". I do find it funny how sometimes some users do end up defending localizations and going against AI Translations. Here is an example of one where Falcom is looking into AI Translations.
Since Netflix provides Japanese subs as well, someone translated the Japanese script and provided an alternative track using Claude AI to translate the entire Ranma 1/2 episode 1. The results surprised me a bit because the subs are surprisingly usable.
Beware that these subs are not without flaws. There is a bit of inaccuracy in them, some lines could be written better, but these things could be corrected quickly with an editor who knows Japanese. Prompts used:32.7 KB file on MEGA
mega.nz
Note: Because the first episode has a twist about a boy transformed into a girl. It got the context right without telling it in the prompt, so it's even more surprising how good LLMs already are.
Comparison:
Netflix <=> AI-kun
The fuck is this person on about? People are still on FE Fate’s arse for that. They’re the ones that chose to ignore it and went on about how it was our “only example”, but now suddenly it’s relevant and a good example because THEY brought it up?
Eh its old but im whatevz bout it :p i personally don't really give much of a shit, bout my old posts being replied too sides i have more pressing matters at hand these days with taking care of the household stuff.
-"Literal translations would not be sufficient"
Even if the AI omitted all honorifcs, if you watch enough anime, you should be able to tell which one is the accurate TL and which one is localized. I haven't watched the remake, but I can already tell that the localizers are taking a piss with the unnecessary quips from your screenshot alone.Since Netflix provides Japanese subs as well, someone translated the Japanese script and provided an alternative track using Claude AI to translate the entire Ranma 1/2 episode 1. The results surprised me a bit because the subs are surprisingly usable.
Beware that these subs are not without flaws. There is a bit of inaccuracy in them, some lines could be written better, but these things could be corrected quickly with an editor who knows Japanese. Prompts used:32.7 KB file on MEGA
Note: Because the first episode has a twist about a boy transformed into a girl. It got the context right without telling it in the prompt, so it's even more surprising how good LLMs already are.
Comparison:
Netflix <=> AI-kun