As many people have said around here, my population growth became excruciatingly slow around 2300, with ~600 pops. It became difficult to meaningfully grow new colonies, often after significant player investment into them (such as building habitats or terraforming), or to get a good number of pops on new Ecumenopoli (also at a significant player investment). That was very much a letdown, and I had a realization that I had set the crisis settings far too high for what I would be able to produce with my economy stagnating as it was, so I quit that game soon afterward.
There were some moments that were great fun - the maneuvering to become Custodian, having a lot more to do with envoys with the addition of espionage, and a greater sense of the unknown when dealing with other empires. Those are things that were done well.
I'm not entirely opposed to the empire-wide penalty; it could be looked at as modeling demographic transition, and it does encourage keeping / creating more vassals versus being a single monolithic entity, more like the HRE, and it did get me to seriously consider a feudal play through, now that I've learned from the mistakes of my first attempted 3.0 run. Shared Destiny is now a great perk (+2 envoys is very flexible, since that's two more spies, +20% weight in the galactic community, leveling up your federation faster, or integrating your vassals faster when you need pops yourself). Previously, it felt like there was pretty much no reason at all to create vassals... I certainly haven't really done so since the Core System limit was removed.
The downside is that I feel like I'm pushed too strongly toward creating vassals now and then abusing Corvee system to resettle their pops to my core worlds, where I have enough infrastructure to make them truly productive. Also, I now have a perverse incentive to intentionally balkanize any empires that I conquer, so that each smaller vassal has higher growth.
It would obviously be at the expense of some of the performance gains, but a lower rate of the empire-wide penalty would alleviate these issues. Obviously, they will still exist and there will be cheesy ways to continue to expand the player's growth and economy as long as the penalty remains in effect, but it wasn't really a serious issue for me in the first 100 years of the game. If that phase can be expanded, where overall population growth is still lower than 2.8, but not so aggravatingly low that it becomes frustrating for the player, that would be great. I can't really say what a good middle ground would be though,not without testing it myself. I can say though that I was still getting frustrated even when I activated a mod that set it to 0.25, since it was still hard to meaningfully grow new colonies at midgame.
Also of note is how badly robots got nerfed. They were dominant in 2.8, and they're still worth building in the early game in 3.0. But I reached a point where I found myself tearing down Robot Assembly Plants, as it was taking too long to get a reasonable return on investment for them, much like the issues that Gene Clinics had (and still have). On my core worlds, additional labs, strongholds, or other buildings would actually give me meaningful benefits, and on new planets, the Assembly Plants were basically useless (since both organic growth and pop assembly had slowed to a crawl).
Given that Organic pops can now grow logistically, allowing them to compensate enough for the higher growth requirements that it doesn't feel frustrating to the player (at least until around 2300). Allowing organic empires to upgrade assembly plants to get more roboticist jobs wouldn't really solve the problem, since, like most buildings, you would likely need an advanced capital to do so, and one of their key uses in the mid/late game is getting new colonies off the ground so that you can upgrade the capital in the first place, a catch-22. I saw it suggested elsewhere to lower the growth penalty for robots somewhat if their assembly speed remains fixed (unlike organic pops and their S-curve), but if not handled properly that could favor robotic pops too much again.
Wow... wrote a lot more than I intended to. But I felt it was important to share my thoughts. I love Stellaris, have played more hours on it than I care to say, and welcome reduced lag, though I think the current attempt at fixing it erred in causing excessive frustration to the player in mid-game, which means many might not even continue to the endgame, where a lot of the new Nemesis content is.