This is true of literally any technology. There are so many things that can be improved in the early stages that progress seems very fast. Over time, the industry finds most of the optimal ways of doing things and starts hitting diminishing returns on research & development.
The only way to break out of this cycle is to discover a paradigm shift that changes the overall structure of the industry and forces a rethinking of existing solutions.
The automobile is a very mature technology and is thus a great example of these trends. Cars have achieved optimal design and slowed to incremental progress multiple times, only to have the cycle broken by paradigm shifts. The most recent one is electrification.
path tracing is a paradigm shift, a completely different way of showing a scene to that normally done, it’s just a slow and expensive one (that has existed for many years but only started to become possible in real time recently due to advancing gpu hardware)
Yes, usually the improvement is minimal. That is because games are designed around rasterization and have path tracing as an afterthought. The quality of path tracing still isn’t great because a bunch of tricks are currently needed to make it run faster.
You could say the same about EVs actually, they have existed since like the 1920s but only are becoming useful for actual driving because of advancing battery technology.
Yea, it’s doing that. RT is getting cheaper, and PT is not really used outside of things like cyberpunk “rt overdrive” which are basically just for show.
Except it’s being forced on us and we have to buy more and more powerful GPUs just to handle the minimums. And the new stuff isn’t stable anyways. So we get the ability to see the peach fuzz on a character’s face if we have a water-cooled $5,000 spaceship. But the guy rocking solid GPU tech from 2 years ago has to deal with stuttering and crashes.
This is insane, and we shouldn’t be buying into this.
It’s not really about detail, it’s about basic lighting especially in dynamic situations
(Sometimes it is used to provide more detail in shadows I guess, but that is also usually a pretty big visual improvement)
I think there’s currently a single popular game where rt is required? And I honestly doubt a card old enough to not support ray tracing would be fast enough for any alternate minimum setting it would have had instead. Maybe the people with 1080 ti-s are missing out, but there’s not that many of them honestly. I haven’t played that game and don’t know all that much about it, it might be a pointless requirement for all I know.
Nowadays budget cards support rt, even integrated gpus do (at probably unusable levels of speed, but still)
I don’t think every game needs rt or that rt should be required, but it’s currently the only way to get the best graphics, and it has the potential to completely change what is possible with the visual style of games in the future.
Edit: also the vast majority of new solid gpus started supporting rt 6 years ago, with the 20 series from nvidia
That’s my point though, the minimums are jacked up well beyond where they need to be in order to cram new tech in and get 1 percent better graphics even without RT. There’s not been any significant upgrade to graphics in the last 5 years, but try playing a 2025 AAA with a 2020 graphics card. It might work, but it’s certainly not supported and some games are actually locking out old GPUs.
Often the lighting systems used require some minimum amount of processing power, and to create a lower graphics setting you would need a whole separate lighting technique
If you think about it, the gaming GPUs have been in a state of crisis for over half a decade. First shortages because everybody used them to mine bitcoins, then the covid chip shortages happened and now AI is killing cheaper GPUs. Therefore many people are stuck with older hardware, SteamDecks, consoles and haven’t upgrades their systems and those highly flammable $1000+ GPUs will not lead to everyone upgrading their PCs. So games are using older GPUs as target
These are Gaussian Splats; you take a bunch of photos of a scene from different angles, recording position and orientation (usually in the metadata), and an algorithm tries to match pixels across these independent images to build a 3D virtual scene of pixel density clouds that you can traverse through.
There are even plans to make it 4D, by making the scenes change with time, by constructing a scene from independent videos of the same object.
The reason I find this next-gen tech, is that when you navigate these scenes yourself and rotate to angles that were never truly captured the scene begins to “shard” apart and it’s like reality itself falls apart, almost like our own reality is this fleeting illusion that we cannot see past.
I can imagine highly immersive videogames being built like this, whilst always being just one dexter angle away from these sharding artefacts.
Kind of like smartphones. They all kind of blew up into this rectangular slab, and…
Nothing. It’s all the same shit. I’m using a OnePlus 6T from 2018, and I think I’ll have it easily for another 3 years. Things eventually just stagnate.
I was hoping that eventually smartphones would evolve to do everything. Especially when things like Samsung Dex were intorduced, it looked to me like maybe in the future phones could replace desktops, running a full desktop OS when docked and some simplified mobile UI + power saving when in mobile mode.
Yeah whatever happened to that? That was such a good idea and could have been absolutely game changing if it was actually marketed to the people who would benefit the most from it
I used it for a while when I worked two jobs. Is clock out of job 1 and had an agreement with them to be allowed to use the screen and input devices at my desk for job 2. Then I’d plug in my Tab S8 and get to work, instead of having to carry to chunky laptops.
So it still exists! What I noticed is that a Snapdragon 8 Gen 1 feels underpowered and that Android, and this is the bigger issue, does not have a single browser that works as a full fledged desktop version. All browser I tested has some shortcomings, especially with drag and drop or context menus or whatever. Like things work but you’re constantly reminded that you’re running a mobile os. Like weird behavior or oversized context menus or whatever.
I wish you could lunch into a Linux vm instead of Dex UI. Or for Samsung to double down on the concept. The Motorola Atrix was so ahead of it’s time. Like your phone transforming into your tablet, into your laptop, into your desktop. How fucking cool is that?
Apple would be in a prime position, they’re entire ecosystem is now ARM based and they have the chips with enough power. But it’s not their style to do something cool to threaten their bottom line. Why sell one phone when you can sell phone, laptop, tablet, desktop separately?
It’s super easy to forget but Ubuntu tried to do it back in the day with Convergence as well, and amusingly this article also compares it to Microsoft’s solution on Windows Phone. It’s a brilliant idea but apparently no corporation with the ecosystem to make it actually happen has the will to risk actually changing the world despite every company talking about wanting an “iPhone moment”
Apple would be in a prime position, they’re entire ecosystem is now ARM based and they have the chips with enough power. But it’s not their style to do something cool to threaten their bottom line. Why sell one phone when you can sell phone, laptop, tablet, desktop separately?
Let’s be real, Apple’s biggest risk would be losing the entire student and young professional market by actually demonstrating that they don’t need a Mac Book Pro to use the same 5 webapps that would work just as well on a decent Chromebook (if such a thing existed)
Or just something like Termux, a terminal emulator for Android. Example screenshot (XFCE desktop over VNC server), I didn’t know what to fit in there: https://files.catbox.moe/zr7kem.png
Full desktop apps, running natively under Android. For better compatibility Termux also has proot-distro (similar to chroot) where you can have… let me copy-paste
Though there is apparently some performance hit. I just prefer Android, but maybe you could run even full LibreOffice under some distro this way.
If it can be done by Termux, then someone like Samsung could definitely make something like that too, but integrated with the system and with more software available in their repos.
What’s missing from the picture but is interesting too is NGINX server (reverse proxy, lazy file sharing, wget mirrored static website serving), kiwix-serve (serving ZIM files including the entire Wikipedia from SD card) and Navidrome (music server).
And brought to any internet-connected computer via Cloudflare QuickTunnel (because it doesn’t need account nor domain name). The mobile data upload speed will finally matter, a lot.
You get the idea, GNU+Linux. And Android already has the Linux kernel part.
Yeah, I remember trying it and while it works the performance hit was too big for my use case. But it’s been a while!
Fortunately I’m in a position where I don’t have to juggle two jobs anymore so I barely use Dex these days.
Which in reverse is also why Samsung isn’t investing a lot into it I suppose - it’s a niche use case. I would guess that generally people with a desktop setup would want something with more performance than a mobile chip.
there is an official android desktop mode, I tried it and it isn’t great ofc but my phone manufacturer (oneplus) has clearly put no work into making it functional
I would love to have a smaller phone. Not thinner, smaller. I don’t care if it’s a bit thick, but I do care if the screen is so big I can’t reach across it with one hand.
One company put a stupid fucking notch in their screen and everyone bought that phone, so now every company has to put a stupid fucking notch in the screen
I just got my tax refund. If someone can show me a modern phone with a 9:16 aspect ratio and no notch, I will buy it right now
OnePlus 6 line of phones are one of the very few with good Linux support, I mean, GNU/Linux support. If custom ROMs no longer cut it you can get even more years with Linux. I had an iPhone, was eventually fed up, got an Android aaand I realized I am done with smartphones lol. Gimme a laptop with phone stuff (push notifications w/o killing battery, VoLTE) and my money is yours, but no such product exists.
I don’t understand why developers and publishers aren’t prioritizing spectacle games with simple graphics like TABS, mount and blade, or similar. Use modern processing power to just throw tons of shit on screen, make it totally chaotic and confusing. Huge battles are super entertaining.
The dream of the '10s/20s game industry was VR. Hyper-realistic settings were supposed to supplant the real world. Ready Player One was what big development studios genuinely thought they were aiming for.
They lost sight of video games as an abstraction and drank too much of their own cyberpunk kool-aid. So we had this fixation on Ray Tracing and AI-driven NPC interactions that gradually lost sight of the gameplay loop and the broader iterative social dynamics of online play.
That hasn’t eliminated development in these spheres, but it has bifricated the space between game novelty and game immersion. If you want the next Starcraft or Earthbound or Counterstrike, you need to look towards the indie studios and their low-graphics / highly experimental dev studios (where games like Stardew Valley and Undertale and Balatro live). The AAA studios are just turning out 100 hour long movies with a few obnoxious gameplay elements sprinkled in.
The generational leap from PS3 -> PS4 wasn’t that significant already, and that happened more than 10 years ago. The biggest difference seem to be lights/shadows and texture size, the latter of which balloons game size and can tank performance
Ironically, Zelda Link to the Past ran at 60fps, and Ocarina of Time ran at 20fps.
The same framerates are probably in the Horizon pictures below lol.
Now, Ocarina of Time had to run at 20fps because it had one of the biggest draw distances of any N64 game at the time. This was so the player could see to the other end of Hyrule Field, or other large spaces. They had to sacrifice framerate, but for the time it was totally worth the sacrifice.
Modern games sacrifice performance for an improvement so tiny that most people would not be able to tell unless they are sitting 2 feet from a large 4k screen.
Had to, as in “they didn’t have enough experience to optimize the games”. Same for Super Mario 64. Some programmers decompiled the code and made it run like a dream on original hardware.
The programming knowledge did not exist at the time. Its not that they did not have the experience, it was impossible for them to have the knowledge because it did not exist at the time. You can’t really count that against them.
Kaze optimizing Mario 64 is amazing, but it would have been impossible for Nintendo to have programmed the game like that because Kaze is able to use programming technique and knowledge that literally did not exist at the time the N64 was new. Its like saying that the NASA engineers that designed the Atlas LV-3B spacecraft were bad engineers or incapable of making a good rocket design just because of what NASA engineers could design today with the knowledge that did not exist in the 50s.
One of the reasons I skipped the other consoles but got a GameCube was because all the first party stuff was buttery smooth. Meanwhile trying to play shit like MechAssault on Xbox was painful.
I never had trouble with MechAssault, because the fun far outweighed infrequent performance drops.
I am a big proponent of 60fps minimum, but I make an exception for consoles from the 5th and 6th generations. The amount of technical leap and improvement, both in graphics technology and in gameplay innovation, far outweighs any performance dips as a cost of such improvement. 7th generation is on a game by game basis, and personally 8th generation (Xbox One, Switch, and PS4) is where it became completely unacceptable to run even just a single frame below 60fps. There is no reason that target could not have been met by then, definitely now. Switch was especially disappointing with this, since Nintendo made basically a 2015 mid-range smartphone but then they tried to make games for a real game console, with performance massively suffering as a result. 11fps, docked, in Breath of the Wild’s Korok Forest or Age of Calamity (anyehwere in the game, take your pick,) is totally unacceptable, even if it only happened one time ever rather than consistently.
I’m usually tolerant of frame drops, especially when they make hard games easier (like on the N64), but I agree it has gotten much worse on recent consoles. Looking at you, Control on PS4 (seems like it should just have been a PS5 game with all the frame drops; even just unpausing freezes the game for multiple seconds).
They can’t. It’s too sweet of a deal when there is an already existing fanbase ready to give them their money for much less work and risk than building a totally new game.
Why? You have to login to Xbox for Xbox published games. You have to login into blizzard to play Diablo or overwatch fron steam, you have to log into EA launching from steam or Microsoft.
So why does a PSN account make you “untrusting”? You just want to cheat on those games or what?
Sony has stopped releasing games on GoG ever since they required login to the PSN. For example, you can buy God of War and Horizon Zero Dawn on GoG, but you can’t buy Ragnarok or Forbidden West.
They lifted the requirement a few weeks ago, but the games still have not been released on GoG, and at this point it’s doubtful they will.
I buy games on epic. I have never purchased a game or have installed gog. Idk get a boner over denuvo removal as denuvo doesn’t effect the games I play
Even if I wanted to cheat in some old God of War games (I don’t), so what? Who does that hurt?
What I don’t trust, for good reason, is that that server will always be there to authenticate my game. Allegedly it requires talking to their server at first install, and that works now in the year 2025, but who’s to say it will be there in 2035?
Splinter Cell: Blacklist came out in 2013. A friend of mine bought it this past winter sale. The UPlay launcher that existed when that game came out has now been renamed and reworked, and the launcher that comes up when he tries to play it asks him for a product key that he was not provided (there is a function for this in the Steam overlay, and we checked, and it was not available for this game). Now I’m sure that he could eventually get it working if he had the patience to wait through Ubisoft support, but A) he shouldn’t have to, and B) what if Ubisoft goes out of business in the next couple of years? That’s not an unlikely scenario at this point, and all the online requirement did was introduce an additional point of failure in the thing that he paid money for. I’m old enough and have been playing games long enough to see these points of failure rear their heads plenty of times now.
The login requirement for the likes of Diablo 3 and 4 are exactly why I’m not buying Diablo 3 and 4. Honestly, even Steam’s DRM, which isn’t present on every game and usually works seamlessly, has still caused some friction for me lately, and every time it annoys me, I get that much closer to only buying games on GOG. The threat of these games getting an online requirement patched in after the fact is enough to make me rather emulate them than deal with that nonsense, if I was so inclined. If they put their games on GOG, I don’t have to trust them, because it’s impossible for them to do that to me.
Oh my god, do you think this is some kind of hypocrisy? If I have internet now and download a GOG game’s installer to my hard drive now, I have it forever, even if I’m in a place without internet access like on a train. Even if GOG goes out of business. Even if Sony goes out of business. Even if the internet ceases to exist. When Blizzard turns the lights off on Diablo 4, that game is gone due to no fault of your own.
The history of sony data breaches. Also nobody wants any of those other accounts either, and diablo and overwatch look boring and manipulative as fuck and not everyone plays those. I haven’t touched an ea game since nhl 98 on snes. I wanted to play horizon, god of war, and spiderman and there is no justification for enforcing log in for any of those.
They lifted it on four games but omitted some others, which isn’t great, and they’ve shown a willingness to patch this in after the fact, so I still don’t trust them. EA did the same thing with the likes of Jedi: Fallen Order and such, so they’re on the same shit list.
I’m with you most of the time, but at least in this case the games are older and they’re still releasing fantastic new installments in the series. (looking at you gta trilogy or whatever that hzd Remaster is)
Making brand new content costs a ton of money, and Sony clearly made colossal missteps in regards to that recently. With this, they’re hoping for what’s essentially free cash
I think the worry is that pushing out so many remasters depletes resources that could be used to make other games.
Not actually sure if that’s true, tbh. I think remasters are often handled by different companies/studios/teams rather than the teams working on all new projects. Although if Sony is tight on cash, funding both teams at once could prove really difficult.
Odd take in this context. The last game in this trilogy was released almost 20 years ago—if they're not gonna remaster this, what are remasters for exactly?
The improvement levels are the same amount they used to be. It’s just that adding 100mhz to a 100mhz processor doubles your performance, adding 100mhz to a modern processor adds little in comparison as a for instance.
Well, that’s what Moore’s Law was for. The processing power does increase massively over each generation. It’s just that at this point better graphics are less noticeable. There is not much difference to the eye between 100.000 and a million or more polygons.
We’ve basically reached the top. Graphics fidelity is just down to what the artists do with it.
Go watch a high budget animated movie (think Pixar or Disney) and come back when real time rendered graphics look like that.
Yea games look good, but real time rendering is still not as good as pre rendered (and likely will never be). Modern games are rife with clipping, and fakery.
If you watch the horizon forbidden West intro scene (as an example), and look at the details, how hair falls on characters shoulders, how clothing moves in relation to bodies, etc, and compare it to something like inside out 2, it’s a world of difference.
If we can pre render it, then in theory it’s only a matter of time before we can real time render it.
If we can pre render it, then in theory it’s only a matter of time before we can real time render it.
Not really, because pre renders are often optimized to only look good from one side. If you try to make a 3D model out of it and render that in real time in the game world, it might look ugly or weird from another angle.
Any given frame is just looking at something from one side though, this is the case for video games as well and it’s part of the reason why real time rendering is so much slower. It’s an art and game direction challenge to make things look good however you want to not a technical limitation (in the sense of, you can make a video game look like a Pixar movie does today, it’s just going to render at days per frame instead of frames per second)
There isn’t really a conceptual difference between rendering a frame with the intent to save it and later play it back, and rendering a frame with the intent to display it as soon as it’s ready and dispose of it.
Toy story 1 took days to render a single frame, now it could be rendered on a single home GPU at 24 fps no problem, which would be real time rendering.
To clarify my first paragraph. The challenge is not that it is impossible to render a video game with movie like graphics it’s that the level of effort is higher because you don’t have the optimizations, and so art direction needs to account for that.
As far as considering unexpected behaviors, that is technically only a concern in psuedo-nondeterministic environments (e.g. dynamic physics rendering) where the complexity and amount of potential outcomes is very high and hard to account for. This is a related issue but not really the same one, and it is effectively solved with more horsepower, the same as rendering.
I think the point you were making is that potentially, artistic choices that are deliberately made can’t always be done in real time, which I could agree with. Something like ‘oh this characters hair looks weird the way it falls, let’s try it again and tweak this or that.’ That is awarded by the benefit of trial and error, and can only be replicated real time by more robust physics systems.
Ultimately the medium is different, and while they are both technically deterministic, something like a game has potential for unwanted side effects. However, psuedo-nondeterminism isn’t a prerequisite for a game. The example that comes to mind are real time rendered cutscenes. They aren’t fundamentally different from a movie in that regard, and most oddities in them are the result of bugs in the rendering engine rather than technical impossibilities. Similar bugs exist in 3d animation software, it’s just that Hollywood movies have the budget and attention to detail to fix them, or the luxury to try again.
I’ll end with, if we have the Pixar server farm sufficient hardware, there is nothing that says they couldn’t render Luca or whatever in real time or even faster than real time.
lemmy.world
Aktywne