Advertisement

my experience with game-dev (why most games suck)

Started by March 24, 2024 06:43 PM
26 comments, last by JoeJ 8 months, 2 weeks ago

But: The issue was much worse in the 2000s than it is now. They're improving overall. Most games get it right already i would say.

if you look at moto racer then compare it to Ride series

/watch?v=g-aplaGHQPc

the graphics of moto racer look a lot nicer and more fun to play, the modern game Ride is not very enjoyable (and i assure you i played it on next gen xbox series x hardware.)

A more fair comparison would be this vs. Ride 5, since Ride 5 actually has appealing graphics compared to Ride. But the gameplay of Ride 5 looks to be bogstandard and boring.

__

“We must increase the draw distance” “More draw distance is better gfx”

Why? By having PS1 draw distance you gain several advantages that AAA games do not have. Firstly it creates a way to blend the background skybox into the game atmosphere. With a large draw distance it occludes most of the skybox. Second it allows for more mystery into the game, the game level feels more mysterious and mystical. With AAA games the entire level is predictable once you see the entire level by just making 4 90degree turns. Third it requires more skill since you cannot plan that far ahead.

Some psuedoretro games try to be both next-gen and retro, by having a “fade-in” type draw distance but it doesn't work as well because adding fog or transparency doesn't match the art of the background. Additionally, the instant object appearances is very 90s, chunky and sharp, all next-gen games try to be very soft and mild

__

looks like quake is better and here is why

when you play a board game do the designers put real photos for the gfx or do they do handpainted art style? why would 3d games benefit from extremely realistic graphics then? In a fps you lack depth perception and the computer screen is only a small fraction of the FOV of real life so you would just be squinting at NPCs, even if the graphics were as good as real life it would be not the same, at most it would be like 30% as good as real life so 3/10. The only games that benefit from extremely realistic gfx are VR

read the comments of every non-shovelware retro game and i assure you it will be dominantly positive with a lot of comments saying “oh how great those days were” “the good ol days” “what a great childhood” etc. Wait a couple decades for the current next-gen games to have videos, i assure you most of those types of comments will not be there

None

I remember thinking Quake was hideously ugly when it came out, and I still think so. I blamed the move to full polygon models at the time, because Doom wasn't ugly and Hexen (built with the Doom engine) was downright pretty. Now we've reached the point where we can use modern technology to make games as pretty as Hexen, but with full polygon models.

Not all modern 3D games look as good as Hexen, but almost all of them look better than Quake.

Advertisement

ReignOnU said:
A more fair comparison would be this vs. Ride 5, since Ride 5 actually has appealing graphics compared to Ride. But the gameplay of Ride 5 looks to be bogstandard and boring.

I have tried Ride 5 because of the realistic viral video int he rain some years ago.
But it's too much of a serious realistic simulation for me. Can't enjoy that - i prefer Trackmania.
Comparing this to some old Moto Racer is much more sim vs. arcade than new vs. old, imo.

ReignOnU said:
Second it allows for more mystery into the game, the game level feels more mysterious and mystical. With AAA games the entire level is predictable once you see the entire level by just making 4 90degree turns.

It's true new games lack mystery, psychedelics, trippy or flashy stuff. There was more of that in the 90's. But now AAA does not dare to serve such niche, so it's on us. Their failure is our opportunity.

Regarding ‘all the same in every direction’, that's the price for open worlds. Even if they have different biomes, due to size and slow traversal, the transition is too smooth to notice the difference, like in the real world.
For me that's really the main reason to doubt the open world idea. I prefer levels, where a new level means new environments and just new stuff. Seeing this stuff is my primary motivation to play games at all.

But i do not share any love for retro graphics. Vertex snapping, object pop ups, 2D sprite enemies in boomer shooters - no thanks. That's all just technical limitations i'm happy we got rid of. Contrary, pixel art is still nice. Here the limitations created a new art form. But with 3D all became ugly, and only technical progress could help it.

ReignOnU said:
why would 3d games benefit from extremely realistic graphics then? In a fps you lack depth perception and the computer screen is only a small fraction of the FOV of real life so you would just be squinting at NPCs, even if the graphics were as good as real life it would be not the same, at most it would be like 30% as good as real life so 3/10. The only games that benefit from extremely realistic gfx are VR

Nah - that's like saying TVs are (or were initially) small, so all they should show is toon animations.
Realistic gfx are necessary for depth perception and immersion in general. I don't want to see real world content in games, but the lighting and detail should convince me about the fantasy content i want to be real, ideally. I mean, that's the whole point - create some imaginary reality, but do it so well it's convincing to be real.
But there is a point of diminishing returns. Even if ‘path traced’ games would look like real, which they do not, i would not pay 2000 bucks for a data center GPU just for that.
And it's interesting to see that many people even prefer retro over modern gfx. Again - an opportunity for us.

Regarding VR, no - they can't solve their problems with realistic gfx. Even less so imo. Their primary visual problem is vergence accommodation conflict, which can't be solved using current display tech. Holodeck is out of reach, Zuckerberg is a fool.
But on a flat screen, we can achieve realism in the same way TV does. It's an image, but eventually showing something which can be believed to be real.

ReignOnU said:
read the comments of every non-shovelware retro game and i assure you it will be dominantly positive with a lot of comments saying “oh how great those days were” “the good ol days” “what a great childhood” etc. Wait a couple decades for the current next-gen games to have videos, i assure you most of those types of comments will not be there

We will see.
I say: In 10 years, games might be mostly AI generated, and they will be very, very, really very bad.
Current gen Z will remember CP 2077, and they will say ‘Remember? This was a good game!’. lol.

Anyway. You know - it's easy to figure what's bad about games. But there is nothing to learn from that.
It's much harder to realize what's actually good about them. What works, and how. And there is a lot to learn from that.
So if we don't like modern games, we are free to ignore them completely, focusing on the older ones we like, and attempting to bring back what was eventually lost with time.
This works for many indie devs currently. But they do not compete AAA this way, so there is no point to focus so much on criticizing AAA. It's bad marketing anyway. You want to tell why your creation is good, not why others are bad.

JoeJ said:

I say: In 10 years, games might be mostly AI generated, and they will be very, very, really very bad.

This seems like an unlikely combination. People want to create games, people can create good games, and people prefer playing good games over bad games. Therefore if the new AI games are truly that bad, the first indie developer that comes along is going to completely destroy their market.

More likely dystopian scenarios:

  • AI-generated games turn out to be pretty good, and we'll all be too busy playing them to make our own games.
  • The flood of bad games is going to kill games as a cultural phenomenon completely.
  • Locked down platforms are going to kill the indie game scene.
  • The new games will be fine, but I'll turn into a bitter old man ranting about how things were better in the good old days.
  • Global warming will wipe out all life on the planet.

a light breeze said:
Not all modern 3D games look as good as Hexen, but almost all of them look better than Quake.

If you haven't tried Amid Evil, it might be good for you too. : )

But to me Quake looked better than any 2.5d game. It has destroyed them, and they were forgotten in a second.

But i understood my reason only many years later. It's the lightmaps. I did not realize what this adds at all, initially.

But what still puzzles me to this day is: Why were those old games so dull? Why was there never a pixel with a brightness higher than 50%? Why? Why just grey, brown, and some darker shades of that?

I did not like that. So games like MDK which had no lighting at all looked much better to me.
When i worked on a software rasterized game for early smart phones myself, i also got this dull look. I have used lightmaps, and color palettes. The shades of colors needed for the lightmaps have reduced the amount of available colors (R4G4B4). It looked like Quake. But then the iPhone with GPU came out, and i stopped working on the game before trying to fix the dullness. I think it would have been possible by using more interesting gradients for color shades, containing brighter, more saturated colors as well.

So if it's not tech limits, why the dullness, which still dominated so many UE3 games in the PS3 era?
Hell, and i even see it coming back! Many modern games are dull again! Similar to Quake.

Do people like this? They do it, because they want to have it this way???

I really want to know! : )

a light breeze said:
Therefore if the new AI games are truly that bad, the first indie developer that comes along is going to completely destroy their market.

Which market? People won't have a job, so no market. They will eat rats, and shitty AI games will be good enough. \:D/

Advertisement

@JoeJ Unfortunately, everyone is not using Unity anymore because it established a greedy policy:

“We're increasing this eligibility limit to $200,000 in annual revenue and funding, which will apply to organizations that use and accept the terms of Unity 6, the next LTS to be released in 2024. For example, if you have $150,000 in combined revenue and funding, you can use Unity Personal starting with Unity 6.”

No one is happy with it. Everyone is aiming for Godot now.

<void> -Cato or some other dude idk

ReignOnU said:
Maybe they can make an “unreal n00b version" for indies lacking AAA features they don't need.

Beat me to it with this. I still think they should put some care into small scale projects, such as Paper2D games. As long as i know, blueprint style is the easiest form of Unreal Engine to handle. But still, it feels too spaghetti-fied and it requires just too many actions and nodes to make a little key-press action work. I have seen UE 5 is despised because it feels like a downgrade and Paper2D is crappy. Wish they thought of doing it right with their 2D engine. I still don't know if making key-press combos is even possible.

<void> -Cato or some other dude idk

@a light breeze ok what about quake3 though?

Regarding ‘all the same in every direction’, that's the price for open worlds. Even if they have different biomes, due to size and slow traversal, the transition is too smooth to notice the difference, like in the real world.
For me that's really the main reason to doubt the open world idea. I prefer levels, where a new level means new environments and just new stuff. Seeing this stuff is my primary motivation to play games at all.

open world is boring, if i want an open world i touch grass IRL. i want hand curated levels not bland digital open worlds.

Contrary, pixel art is still nice. Here the limitations created a new art form. But with 3D all became ugly, and only technical progress could help it.

i like true retro pixel art but psuedo retro pixel art does not feel right to me

Nah - that's like saying TVs are (or were initially) small, so all they should show is toon animations.

Not an analogy. Movies and tv are carefully directed whereas in games are just free roaming around. And in a movie the action hero just does stuff automatically, it is not required to detect distant NPCs during the heat of battle.


Realistic gfx are necessary for depth perception and immersion in general. I don't want to see real world content in games, but the lighting and detail should convince me about the fantasy content i want to be real,

quake wins again in that regard. A lot of the next gen AAA games have shadow pop in from several meters, quake just has baked shadows that are more immersive.

And it's interesting to see that many people even prefer retro over modern gfx. Again - an opportunity for us.

i wish this were so, but for some reason AAA always will make much more big bux. People seem to gravitate to AAA slop for no reason

Regarding VR, no - they can't solve their problems with realistic gfx. Even less so imo. Their primary visual problem is vergence accommodation conflict, which can't be solved using current display tech. Holodeck is out of reach,

what is that

We will see.
I say: In 10 years, games might be mostly AI generated, and they will be very, very, really very bad.
Current gen Z will remember CP 2077, and they will say ‘Remember? This was a good game!’. lol.

haha

Anyway. You know - it's easy to figure what's bad about games. But there is nothing to learn from that.
It's much harder to realize what's actually good about them. What works, and how. And there is a lot to learn from that.
So if we don't like modern games, we are free to ignore them completely, focusing on the older ones we like, and attempting to bring back what was eventually lost with time.
This works for many indie devs currently. But they do not compete AAA this way, so there is no point to focus so much on criticizing AAA. It's bad marketing anyway. You want to tell why your creation is good, not why others are bad.

both are valid. the one you speak of is copying from others, which is valid as well.

But what still puzzles me to this day is: Why were those old games so dull? Why was there never a pixel with a brightness higher than 50%? Why? Why just grey, brown, and some darker shades of that?

artistic choice

some ask why is fortnite so bright

I did not like that.

isn't there a contrast setting? There is in quake 3. the inital gfx of quake3 were too dull for me so i had to increase the brightness and contrast.

also the devs of quake might have tested it on different monitor settings so it looked different on theirs

Which market? People won't have a job, so no market. They will eat rats, and shitty AI games will be good enough. \:D/

if the ai games suck there will be a market for indie and even AAA games

Catomax26 said:
Unfortunately, everyone is not using Unity anymore because it established a greedy policy: “We're increasing this eligibility limit to 200,000inannualrevenueandfunding,whichwillapplytoorganizationsthatuseandacceptthetermsofUnity6,thenextLTStobereleasedin2024.Forexample,ifyouhave200,000 in annual revenue and funding, which will apply to organizations that use and accept the terms of Unity 6, the next LTS to be released in 2024. For example, if you have 200,000inannualrevenueandfunding,whichwillapplytoorganizationsthatuseandacceptthetermsofUnity6,thenextLTStobereleasedin2024.Forexample,ifyouhave150,000 in combined revenue and funding, you can use Unity Personal starting with Unity 6.” No one is happy with it. Everyone is aiming for Godot now.

unity is good objectively as a game engine. it has a lot of flaws though. one flaw is that they can change their financial license agreement arbitrarily at any time.

one thing i find interesting is that gms2's ide seems more confusing than unity or unreal 4's ide but it is marketed as an easy game dev engine, almost feels like they lost sight of overmar's vision.

Beat me to it with this. I still think they should put some care into small scale projects, such as Paper2D games. As long as i know, blueprint style is the easiest form of Unreal Engine to handle. But still, it feels too spaghetti-fied and it requires just too many actions and nodes to make a little key-press action work. I have seen UE 5 is despised because it feels like a downgrade and Paper2D is crappy. Wish they thought of doing it right with their 2D engine. I still don't know if making key-press combos is even possible.

its like unreal could have kicked gm in the nose with blueprints but chose not to. in theory blueprints could have been epic, when i first saw it it i though it would be like gm. but as i mentioned earlier, for example i try to put keyboard input in it and it does nothing. and they give excuses such as too much input polling. in gm you can have 10,000 objects with keyboard listening, and there are no excuses such as “we can't do that because too much input polling”.

None

ReignOnU said:
quake wins again in that regard. A lot of the next gen AAA games have shadow pop in from several meters, quake just has baked shadows that are more immersive.

Baked stuff fails at dynamics, so it fails at immersion as well.
But i would say Quake 3 marks a sweet spot, and the diminishing returns started here.
Doom 3 was super impressive after that. Since that, i never was impressed from gfx so much again, til yet.
Currently i'm very impressed from UE5. It's real progress and good work. But sadly, no games i like anymore, so it comes too late for me.

ReignOnU said:
i wish this were so, but for some reason AAA always will make much more big bux. People seem to gravitate to AAA slop for no reason

Well no. AAA companies invest half a billion maybe for a game. They have 1000 workers, investors, etc.
You as a single person invest very little on production and marketing. After that there is Steam, taxes… but you might 50 % from the cake.
In the end, it might not matter if you work on GTA 6 or a successful indie game. Financially, a similar outcome is quite likely i guess.

ReignOnU said:
what is that

Your eyes and brain want to focus differently on distant stuff than on close up stuff.
A photorealistic render is not enough. For proper VR, we need display tech with true depth.
In theory it can be done using kind of layers (i'm no expert), but afaik no success yet, and nothing practical in sight.

If you want to make a VR game, the advise is to avoid close up stuff, or if so avoid a large depth range behind it, to avoid headache.
Which is quite funny, since without close up stuff, there is no more point for a stereo display at all.

Thus: 3D TVs failed, 3D glasses failed, VR games remain a gimmick, and Zuck must be from another planet assuming people would wear VR/AR goggles for longer periods of time.

ReignOnU said:
some ask why is fortnite so bright

Because it's for kids. Kids like saturated colors, olds like desaturated colors.
Saturated colors are only possible at higher brightness, but desaturated colors can use the whole brightness range.
So my question stands: Why did they use only half of the range?
I never dared to ask. But i think there must be a reason beyond artistic choice. Otherwise it's a really bad choice.
(Brightness and contrast settings do not really help. You can use then to make it worse, but it won't become good.)

ReignOnU said:
both are valid. the one you speak of is copying from others, which is valid as well.

No, only one way is valid.
What i really mean is attitude, again. See the reactions to your other post. Just is i had predicted, no?

This topic is closed to new replies.

Advertisement