First, improvements.

Antimatter missiles are great, antimatter structures not so much.  If you have a full ring of improvements around a central structure, you get a combined bonus, from that central structure's presence, of 60%.  One to each of the six, and one from each of the six, at 5% each.  If you use the power plant instead of a regular research structure, you get 60%, two to each of the six from itself, and none from the six.  You then lack the base improvements of the structure, making it an absolute loss.  If you're using one for production instead of a durantium refinery, or a solar power plant, you are taking a loss worth multiple structures.

The only time an antimatter power plant can possibly be better is when you have an utterly trashed build plan with multiple different structure types between it, and even then it will have to be multiple types to come out better in the math, all in return for a limited resource far better spent improving missile damage.

A structure with no base stats, that can't be improved by it's neighbors, has to be substantially better than ones that do simply to be viable.  When it requires a resource, it needs to actually be better than viable.  This structure would make far more sense if it were like slave pits, which are improved by multiple structures as well and give multiple bonuses with extra levels.  Slave pits are simply amazing

Then there is tourism.  The port essentially has no purpose.  The only circumstance where it ever makes sense is if you surround it with influence structures on a low population planet not making any money.  On a planet focused on money, you can easily bring in hundreds, with a raw production in the 20's even on moderate planets.  Tourism is essentially static, it doesn't increase with raw production, or population.  No matter how powerful you make your planet, it will never amount to anything significant until you've all but won the game, so the port can only be viable when your planet literally amounts to nothing.  Trade is much the same, disconnected from the value of the planet, but multiple trade routes to a single planet can bring hundreds, and 10% of 250 is a lot better than 10% of 5.  Once you've gained enough influence in the galaxy for a bonus to tourism to really amount to anything on the typical planet, you've long since won the game.

The research is great, unlocking that initial income boost is huge, but building improvements for it are a real tragedy in decision making.

Map generation.

Can we say precursor worlds?  I'm seeing map after map that is essentially populated by precursor worlds.  They're nice, but I thought they were supposed to be special?  They're basically the only thing, and they're not even that good.  Since they halve populations, they can lose far more raw production than they gain.  You can easily gain 40 raw production from populations with the right build on a good planet, a precursor world getting a 50% research bonus and 10 raw production, will halve that population bonus, leaving them way down the line even with a research focus.  They're very good early, but somewhat anemic as terraforming turns even Mars into a 20 population planet.  Their frequency greatly reduces the maximum potential of a civilization, slowing down the end game significantly

There's also the matter of planet distribution in general.  On an utterly enormous map, clustered planets aren't too bad, on a large and smaller, they're typically just blobbed together in one or two spots.  The clusters are the same size irrespective of map size, which makes them not really clustered at all.  On the smallest map setting, it's just one big cluster, there is little to no difference regardless of whether you pick scattered, tight clusters, or loose clusters until you have a very large map.

I reloaded yesterday, over and over, trying to get a large map that had separate clusters, I gave up and played that extremely easy Snathi map instead.  I've had multiple maps without a single extreme planet, and the only ones that required atmospheric cleansing were a couple more precursor worlds, between the two issues, it's a major dud at present.

The game has other problems that really irk me, like not having sub directories to particular ship types in the build list, or how epic the death furnace is, but I'm sticking to the basics here.