I'm interested in the balance of this new edition's classes. Are casters still noticeably more effective in almost all situations than non casters?
I’m going to answer this the same way I answered a similar question about Starfinder: we just don’t know yet. Pathfinder 2e is a far larger change from Pathfinder 1e than Pathfinder 1e was from D&D 3.5e, and at this stage in Pathfinder 1e’s life, we didn’t know for that yet either. So for a system that has changed far more than the previous case, we certainly don’t know either.
Determining overall balance truths in a system takes a lot of playing it. In particular, since games are typically supposed to be balanced, you cannot rely on the descriptions of things to determine if they’re actually powerful—you have to try them. And you have to try a lot of things, and people have to come together and compare experiences, and you have to try yet more things. You have to try rather out-of-the-box things, too.
And for a real consensus to form, you need a critical mass of players who have played enough different things to be able to form an informed opinion about things—and then you need still more time for them to go back and try different things when they learn others had different experiences. Maybe the campaign you played just happened to miss some important weaknesses in the class you tried, or maybe some of the things you chose for your character weren’t the best choices and the result wasn’t as good as it could have been. That leaves you with a particular impression—which may, due to the specific instances of the character and the game, run contrary to popular opinion. But you have to verify that for yourself in most cases—sometimes it’s as easy as “oh, yeah, I could see how in most campaigns that’d be a bigger problem” or “hm, yeah, if I’d had that I wouldn’t have struggled so much.” But usually it’s “really? I’m skeptical. I’ll have to try that myself.”
That’s a ton of playtime necessary across many, many players in order for a consensus to grow. Pathfinder 2e simply hasn’t been out long enough.
We won’t likely have an answer to this question for years. This is one of the things that makes RPG design so fiendishly difficult, why balance in particular is such an elusive goal—just information alone is hard to come by, and takes a long time to collect. The only real approach to avoiding any problems, if you’re really dedicated to it, is regular balance patches and updates—which are vastly harder to disseminate for pen ‘n’ paper RPGs than they are for video games, and even in video games cause a lot of strife and gnashing of teeth. It’d be much worse in the RPG industry, where the community 1. doesn’t even agree that balance itself is valuable, and 2. isn’t used to such disruptive practices.
There's a pretty strong balance between the classes in the game, particularly compared to D&D 3.5/Pathfinder 1e, such that dividing them into tiers like this answer really don't make sense.
With the mandatory skill ranks and skill feats that every character has as well as the various multiclass options to get up to 8th-level spellcasting on any character, there's no class that can't achieve just about anything if they want to.
Even a barbarian with just 14 Intelligence, Wisdom, or Charisma can eventually learn spells like earthquake, chain lightning, or teleport after spending 4 of their guaranteed 10 class feats to multiclass and pick up whichever class's Spellcasting Feats.
With 4 Ability Boosts to different scores every 5 levels, it's actually impossible to not qualify for some spellcasting archetype on a character by 20th level (as long as they aren't employing the Voluntary Flaws optional rule to reduce their stats). Most characters meet the requirements at level 5 or on creation.
And with the core retraining mechanics, our barbarian could learn how to do all of this in about a month of downtime with sufficient levels. It would cost them whatever class feats they had originally chosen, but they could always retrain back if the situation called for it.
In essence, a character's class doesn't determine whether they could handle a particular challenge or how versatile they are. For all characters it depends on their specific build choices, and whether they have time to change those choices if needed.