I mean, companies that are spending hundreds of millions of dollars on a single game, and can't make it profitable, can only blame themselves. Nobody is forcing them to use ultra realistic high end graphics on every game. Plenty of games look amazing in different art styles.

Besides, with diminishing returns, the best graphics of today aren't that much better than from like 10 years ago - like oh the resolution on textures is ultra ultra high instead of just ultra high and there's fancy ray tracing lighting that the game can barely run and there's 15 million polygons on screen instead of 10 million. And all that costs tens of millions of dollars extra for the man-hours to make these small changes to look slightly better and most people wouldn't care one bit. The obsession with ultra realistic graphics not being good enough and we need extra ultra realistic graphics is what is killing these companies. Back in the day people cared about graphics because the tech was rapidly advancing, now from PS4 to PS5 it's like okay things look slightly better and it makes no difference to the gameplay so who cares, but it made the game cost an extra $50 million for all that fine detail work.

The reason why Nintendo is making gobs of profit hand over fist is because they don't go after ultra realistic graphics that adds tens of millions or even more to a game's budget. And their games are no worse from it. There are probably lot of games that could be taken from $100M to $50M budget and not much would change about the game other than the graphics looking great instead of extra great.

Yeah I've been playing a fair few PS4 games lately (didn't own one back in the day) and it often strikes me how some games from like 10 years ago still look better than a lot of games today, yet they cost a lot less to make.

Indeed. The thing about shorter games is that I'm more likely to replay them. I've replayed many old NES, SNES, & Genesis games many times over, which is easy to do when you can breeze through them in an hour or two. I probably replay Halo CE once every month or two, and I can beat it on Legendary in like 7 or 8 hours. The RE2 and RE4 remakes I've played several times each since they can be beaten in a handful of sittings. Most of my favorite games of all time are games I can reasonably beat in less than a week's worth of sessions, many of them in one to three afternoons. Meanwhile, Horizon Forbidden West was the first open-world game I replayed, mainly just to experience the remastered version. Before that, the only long games I replayed were some of the old Final Fantasy games.

Replay value gets overlooked in discussions of a game's value. Time investment is an important factor in replay value, at least for me. Like, I'd prefer a 6-hour game I'll play at least ten times compared to a 60-hour game I might play once or twice. It's also why it's easier for me to get motivated to rewatch a movie rather than an entire TV show. I only have so much time, and I have to decide whether I want to sink that time into a single big thing or a bunch of smaller things, especially these days when there's so much more stuff competing for my attention and my ADHD-addled mind starts to get overwhelmed by too many choices.

Going through the HD-2D remake of Dragon Quest I and despite the added touches, I'm surprised at how good the overall package is for ultimately a 15h-25h hours experience. 

It is definitely possible to keep some of these games runtime down on meandering and keep the overall experience quite plentiful.

The problem is a lot of devs and executives don't seem to know how to make these longer form games engaging in the same ways a fulfilling 10-20 hours game can be, hence why Ubisoft and other similar publishers have struggled a lot with their models recently. 

Yeah I also find shorter games more approachable and often better quality than a lot of really long games, which are often padded out with a ton of filler.

I feel like this effect really kicked in last gen, with so many franchises going open world, and enthusiast gamers insisting on stuff like "If I'm gonna pay full price I want at least x hours of content".

Every big publishers tried to outdo each other with bigger and bigger games with better and better graphics, and budgets exploded.

One of the biggest problems devs face with building a huge game world is "What do we fill it with?" Sadly, Ubisoft template is what caught on. Divide the map up into sectors. Give the player a bunch of repetitive checklist objectives in each sector. Have them climb towers to unlock the map. Have them beat up criminals/liberate bases/rescue hostages over and over and over. Et cetera. The game is a mile wide and a foot deep.

Halo was fine when it was mostly linear levels that occasionally branch out to offer multiple routes or even allowing players to go explore a bit (esp. in the first game). While each level is self-contained and had a definite beginning and end point, they rarely felt restrictive (several Halo 4 levels notwithstanding), and the levels often felt big in terms of scope even if they were collectively a fraction of the size of your typical modern open-world game. Meanwhile, Halo Infinite introduced open world to the series, and the overworld wasn't bad from a technical perspective, but it was boring. Not only is the overworld the same biome in a game series known for varied biomes, but it's filled with the same repeating objectives we've all seen in some variation in other games (free captive Marines, liberate UNSC bases, defeat unique Banished mini-bosses, clear out Banished outposts). None of it actually moves the game forward, and instead serves the primary purpose of populating the huge map and the secondary purpose of doing things like unlocking fast-travel points or giving you a slightly better variation of a gun you find lying around everywhere. The best parts of that game were the linear story missions outside of the overworld, which felt more like classic Halo. 

Even when developers try their best to ensure there's as much unique content as possible, they still run into the same problems with repetitive objectives. For example, Elden Ring is probably one of the best-designed open worlds I've played, but even it has dungeons that are just slight variations on previous dungeons, and multiple repeat bosses. Also, Breath of the Wild fell back on "Ubisoft towers" for revealing parts of the map screen, and the multitudinous mini-shrines were some variation on the same three things. At some point, the developer runs out of ideas yet still has to cram something into every corner of the world to make both the size of the map and the exploration of it feel "worth it" and to give the player 40-60+ hours of "content." There's just no way around it. Even some games that aren't open world have a problem with artificial padding. Alien: Isolation was a 15-20 hour game that could have easily been an 8-12 hour game and still felt satisfying. It's gameplay loop didn't justify a whole 19 levels and at least one false ending. It's one of the few games that I can recall actually being criticized for being too long.

Sometimes developers will try to make those fundamentally repetitive tasks in big open worlds different, like what Guerilla did with the Cauldrons and Tallnecks in Horizon Forbidden West, which at least shows that some developers are aware that shit starts to be repetitive after a while and that they need to do better. But it doesn't change the facts that it's fundamentally the same basic objectives (unlocking machine overrides and revealing parts of the map, respectively, in the aforementioned side objectives in HFW), and that it was all there in the first place to fill in space and pad out the game's length.

So many open world games could have been better served with worlds that were smaller, with each story mission its own self-contained level, and all the extraneous "checklist objectives" done away with. To go back to Halo, all the genuinely cool sights and structures in Infinite's overworld could have easily been part of a regular linear level, and they could have told the same story with a smaller, more linear game.

Even games that aren't truly linear and that encourage exploration can still do so in a structured way, like A Link to the Past or Super Metroid. Those games were in a sense single interconnected worlds instead of a linear progression of discrete levels that you may or may not be able to revisit. But they also didn't give players absolutely free reign over the whole map. You were free to explore to a point, and through your poking and prodding you'd find useful things, not just basic upgrades like health or new weapons, but also tools that expand your ability to explore further into the game world. And the game worlds themselves were just large enough to tell a story and give the player enough unique things to do and places to visit within a reasonable time frame. I never got to the point where I was like "Get on with it." They weren't a short breeze through like Mega Man, but neither did they overstay their welcome. I think between that and their timeless 16-bit pixel graphics is why they still hold up so well decades later, and why for me they're still among my top ten games of all time.

Why does it need to be filled? One of my favourite games of all time is SotC, and partly why I like it so much (next to the awesome colossus of course) is the empty world.

Most open world games I just get annoyed at all the stuff popping up. Can't walk 2 steps without something trying to get your attention. There is enough of that in real life. If I set out to do something like in SotC, then don't bother me with your missing cat.

I've played the biggest open world games possible NMS and Elite Dangerous. Of the two Elite Dangerous is my favorite since I can roam around the galaxy without meeting anyone :) It doesn't feel like 'discovery' when someone already lives there.

Of course Elite Dangerous is also full of repetitive tasks, but you can at least escape that and just go wander around the galaxy. Which I also loved to do in FS2020, land somewhere without quests popping up :) 

Anyway all the open world games I remember are for the geography, art style, world building (The little stuff you find everywhere). Not for the quests and repetitive tasks. Like in RDR2, the best was setting out on my own in a Kayak, or wandering around hunting big game and coming up on little structures with their own story to tell. 

Open world games should focus more on world building rather than todo lists, and empty space is fine. Give the player some time to reflect, plan, prepare while running around.

But it's easier to fill it with repetitive tasks and then sell it as a 100+ hour rpg, worth your money :/ Players did that to themselves. Just today someone replied to Hotel Infinity with, it's only 90 minutes, $20 is a bit steep for that. 90 minutes of mind bending ideas is worth more than any 100 hour repetathon. 

Yeah the obsession with game length is stupid, that's how we ended up with bloated 100 hour slogs that cost $300 million to make and so are packed with predatory microtransactions and such.

8 hours of quality beats a hundred hours of repetitive padding any day.

RayTracing is computationally heavy but it saves development time and cost when the game is designed around it.

People like to assume Nintendo's games are cheap to make, and it's probably true with their less revered games and I guess Pokemon (Nintendo's worst popular title). But do we actually have any proof that the likes of Tears of the Kingdom, Mario Odyssey, Xenoblade, and Metroid Prime 4 are "cheap"? These games take too long to make (maybe not Xenoblade. Monolith is quite productive). And we don't know how much Nintendo spends on marketing, do we?

Lower game developer wages in Japan and the weak yen likely bring the dollar cost down more than graphical fidelity. Nintendo's best games are probably quite expensive still, especially in a Japan context, but it doesn't matter at the end because they tend to sell like hotcakes and at high prices. Other areas where I think Nintendo often saves money are voice acting and cutscene/cinematics/mocap.

Certain aspects about improved graphics increase development budgets, but many of them cost next to nothing, and some of them (like RayTraced lighting) can actually reduce costs.

If Ghost of Yotei cost similar to the OG (60m)... It's definitely possible to make AAA experiences at a reasonable budget.