I never played this game but I watched gameplays on youtube and I must say without insulting someone who like it that I don’t like it because it looks like unfinished game to me compared to other games in terms of graphics.
I firmly believe that publishers, in an attempt to cut costs, tell the game studio to not prioritize performance
So, I agree there’s some amount of that. You also have things like Dice (the studio that makes Battlefield) where they lost their veteran development team to poor internal management.
There are also some (now fairly large) studios that are just absolutely terrible at game performance like Studio Wildcard (makers of the Ark games).
while trying to rely on software like super resolution algorithms, to make their games run.
There’s definitely some of this too. I believe the bigger issue is that games have gotten so much bigger and more expensive to develop. Making and shipping a game that runs with 4k textures, dynamic (possibly ray traced) lighting, variable rate shading (instead of manual level-of-detail systems), etc is a lot to get right.
A common thing with any software development is to take advantage of newer abstractions that make your life easier. For instance, I’m fairly confident Hunt Showdown 1896 has moved to some form of variable rate shading instead of level-of-detail (in pre-1986 when you zoomed in on some of the trees they’d literally change shape when they flipped between the models in the worst case; I’ve yet to see that post-1896). Not having to make a bunch of models and having the software “just figure out” good lower-poly models for things that are sufficiently far away is presumably a huge productivity boost. Similarly, when ray-traced lighting becomes the standard a lot of game development will get easier because setting up lighting won’t (per my understanding) require as many tricks. In both cases, it’s both less work for developers and a better result for players with the hardware to run it.
In some instances they reused old game Engines for a new and bigger game, for example with Cyberpunk, Stellaris and Elden Ring.
Old engines aren’t necessarily a bad thing (if they’re appropriately updated) and I think people focus too much on the engine vs the game play. Take Starfield, I’ve heard a lot of people complain about it on forums for copying a similar formula as some of Bethesda’s past titles.
The issue almost certainly isn’t the engine used, but the design choices associated with using that engine (and the decision to not make new things work).
Linux, Darwin (MacOS), Windows, Chrome, Firefox, etc are all long running software projects (as are Unreal Engine, Unity, Source Engine, CryEngine, etc). Occasionally, someone throws out their current product entirely and replaces it, but normally there are incremental upgrades made to provide the new functionality that’s desired.
Smaller developers are doing everything they can to make a game run smoothly. The best example for this is Factorio.
The performance profile of something like Factorio vs Cyberpunk, Elden Ring, or Hunt Showdown is extremely different.