Fewer fights make them more interesting.
A random example - the goblins in the village.
If you set them on flat ground, the fight would be as interesting as fighting the kobolds in Kingsmaker.
Due to the fact that they are placed on the rooftops, the fight is much more interesting and more difficult because your melee have to get to the place.
If the game were adapted to a larger number of encounters, it would have to be much simpler. Which would end up with trash mobs at the WotR level. I don't feel like fighting a group of kobolds in TB.
Let's take the fight, for example, that the Spectator. If he had the standard amount of hp, the fight would be a joke and he would be killed in one turn.
Also, spamming cantrips in combat is not an interesting project.
Tougher fights with more intelligent enemies are always better.
There is no point in comparing a computer game to PnP because it is completely different.
Look. I'm obviously not going to convince you differently. You are dead set on 2-3 fights is the sweet spot for a video game.
I will say this, however, perhaps for anyone else who might buy into your belief that it MUST be 2-3 fights to create meaningful encounters per adventuring day. There is an in-between. I understand you've played games with boring trash mobs and you think that if they have 10 battles in a game that they must all be trash mobs and therefore boring and sucky.
But there is an in-between. It doesn't have to be either "2-3 Meaning Fights" or "10+ Trash Mob Fights with 1 Meaningful Fight." If that was the case, D&D wouldn't exist at all.
There are 300 monsters in the D&D 5e monster manual. 300! They all have unique stories and backgrounds and abilities. A video game could utilize more monsters instead of having repeat encounters over and over again.
You used kobolds as an example. Yes. Fighting 6 kobolds in this room, 6 more in that room, 6 more in this room, and 6 more in that room, all with the same stats and so forth is certainly not all that exciting. But this is a cookie-cutter approach to encounter building.
And that's why I said that the prologue in BG3 would be MUCH better if they didn't use imps but instead used monsters more suited for level 1 players. While players are moving through the nautiloid, first encounter could be maybe 3-6 lemures - baby devils that are like grimers from pokemon. 3-6 lemures against 2 level 1 characters and maybe an Ustilagor (Us) would be a relatively challenging encounter, and the monsters wouldn't need to be nerfed. It would be an exciting fight because you are outnumbered by these nasty blob devils. They might have low AC and do low damage, but that's the point. You don't WANT monsters who can literally kill you in 1 hit.
Unlike what they went with. 3 imps. 3 IMPS! Resistance means half damage when you hit, so even 7 HP is 14 and would require possibly 3-6 hits to kill just 1 depending on how well you roll for damage. And then, to top it off, they are supposed to have Stings with their tails. Poisonous stings. 1d4+3 physical damage and 3d6 poison. Even a fighter at level 1 only has maybe 12-13 HP. A single sting from an imp could fell Lae'zel. If they crit you? Dead. Insta-kill. Meanwhile, monsters like lemures only do 1d4 damage. Period. So, if you get hit, NOT a potential insta-kill or insta-incapacitate. You might have two or three of them attacking you, and they might all even hit you, but you don't fall because they're literally only taking 1-4 damage each time. Even baby wizard PC can handle a few lemures, but a single imp could 1-Hit KO a wizard without hardly batting an eye.
But again, mix it up. First fight is lemures. Second up on the deck is a couple of thralls. They're baby thralls who only get +2 to hit and maybe 1d6 damage. Slightly tougher, but nothing that will possibly kill a PC in one hit. Maybe 2 or three hits will fell a PC, but not a 1-Hit KO. Like the lemures, they might only have 7 HP, so YOU can possibly 1-Hit KO them, but not the other way around.
The fights are still tough, mind you. Why? Because you can't just full restore each time. First fight may have reduced your PCs to 3/4 health. Second fight reduces to 1/2. Ah! But you got some healing potions. You use them and restore to full. Next fight reduces to half health. Eek! Now you're hurting. Hope you find something soon. Ah! More potions from the generous DM who doesn't want his PCs to die.
You see, the true way encounters are supposed to be built is that each fight has the potential to totally wipe the party, but that potential is not so high that one bad RNG session totally kills them without giving them even a chance to survive. Again, I point out the imps versus the lemures. Lemures can be very challenging and even deadly if in large enough numbers. A DM can use them to create a VERY scary encounter that brings the PCs to death's door. However, lemures even in greater numbers are more appropriate for level 1 characters because their damage potential is much lower as individuals. There is a greater chance that characters will survive against a horde of lemures than a single imp because again, a single imp can 1-Hit KO a PC and even kill them with that single hit while there is very little chance that a single lemure will 1-Hit KO a PC. Only 2 or 3 working together against a single PC will likely take them down if they get extremely lucky. Thus, MUCH more appropriate.
Next fight, mix up the combat again. Larian created hellsboars. Have them be the next fight. Maybe 3 lemures and a hellsboar. Oooh! Fight is interesting and yet still very appropriate because you are fighting only a few lemures (boring) but also a new monster (interesting). New monster has new stats and abilities to learn.
So, in conclusion and summary of that point, the only reason lots of fights are boring in video games is because devs fashion the encounters with cookie-cutter trash mobs (the same encounters over and over again). There is nothing wrong with fighting kobolds. They can be very interesting enemies. But instead of using what makes them interesting, and mixing it up with other interesting monsters, etc. most devs just take a single monster and slap it all over the map. Understandably so, mind you, because creating unique monsters and encounters is a lot more work than simply taking 1 monster and copy/pasting it everywhere. But that's not the point. We're talking what CAN be done.
And, on that note, what even CAN be done in BG3 is Larian could take what monsters they did create and mix them up throughout at more appropriate levels. Instead of 3 intellect devourers in the nautiloid, why not some fishermen infected by intellect devourers. You fight the fishermen and kill them, an easy to moderate challenge, only to have the devourers spring out. However, not wanting to face you, because they sense the parasites in you, and because they are ambush predators by nature, they jump out of their hosts and run away. Hmmm. Not a super tough fight with a high chance of dying, but interesting because you immediately ask yourself, "Why did they run away?" Also interesting because they TOOK PEOPLE OVER. So, immediately, the player is thinking, "Dang. I can't trust anyone. Anyone I encounter could be actually an intellect devourer.
And I absolutely do not believe that PnP is not able to be translated into a video game. You CAN compare the two because they are NOT completely different. Do you know why they are thought to be completely different? People make them completely different and refuse to believe that they are not different.
The core mechanics in D&D 5e can be translated into a video game with obvious exceptions being minor ones. Certain spells, naturally, would not translate into a video game, for example. But the core mechanics can. And, what Larian HAS proven, is that a DM CAN be translated into a video game as well. How have they proved this? The sheer volume of dialogue options they are providing players and the sheer freedom they provide players to go wherever they want (within reason). They have proven that one of the most challenging aspects of a video game translation from a TT RPG can be done and done well. This is one of Larian's strongest points. Solasta has proven that the core combat mechanics for D&D 5e can also be implemented well, even implementing spells like flying and levitate. Sure, you may not like them all, but that's why Tactical Adventures created all those nifty options so players could turn off the rules they don't like.
You are kidding yourself when you say you can't compare the two. I absolutely cannot accept that because it has literally been done not only in video games like Solasta, but in TT programs like Fantasy Grounds, apps like the 5e companion I have on my tablet and phone, apps that allow you to create 3d maps and allow you to put pawns on the maps. All these things are computer-controlled apps that allow you to literally play an RPG 100% on computers and apps. The ONLY thing players add is themselves.