As far as excitement about graphics goes...how much further can it be pushed beyond current gen or even last gen GPU capability? And what would be the point as far as player experience goes? Performance, certainly, that's a reasonable concern.
Graphically, style lasts forever, photorealism gets outclassed by future graphical capabilities, limits buyers (not going to buy if I can't play it smoothly on my system), and I think can cause a kind of graphical stagnation where creativity gets optimized out (am I in the forests of Skyrim or the forests of God of War?). Indie devs see all this and going for style is often a better use of their limited resources. I've been playing a lot of Against the Storm and it embraces the style originally inflicted upon ol' Warcraft 3 due to technical limits. It doesn't hit the insane shooting-for-photorealism detail found in, say, Total Warhammer, but it doesn't need to in order to be pleasant to look at.
The topic of "how much fidelity is too much fidelity" is another one I ought to cover at some point as well!
While movies like Avatar 2 show us that there are still interesting problems in the space of photorealism, I think we really have reached a point where the ROI on advanced graphics has dramatically lowered in real-time applications. But, we've also seen a dramatic loss in "responsiveness" in high-fidelity games as a result of the techniques used to achieve them.
Every millisecond the CPU spends managing reflection probes, shadow volumes, and the like is a millisecond it could have spent calculating the physics or game rules of a sandbox interaction. Like, what is more important to a Battlefield game? Photorealism or being able to shoot a physically accurate hole in the wall? I'd say the latter.
There's also the question of how this all effects developer workflows. The standard lighting pipeline these days is absurdly complicated with all of the metadata you have to put into a scene to make it look good... only for it to break the second the player goes into a corner of the map they weren't expected to go into.
This is a long way of saying: I think unlocking the full performance of the hardware we have—and taking advantage of techniques like path tracing which have "ceilings" to how much processing they need—could shorten game development times dramatically and make them more interactive. 😂
I wonder how much discussions this sort of thing is seeing in the industry. Dev pipelines need streamlining badly on every level; you can pour millions of dollars into dev but you won't see a dime until release. And it needs to be in a state fit for release, or else you're just setting yourself up to spend more money and time fixing it while earning customer ire.
I generally don't understand how game dev is getting approached these days by the movers and shakers. It doesn't seem to be following any kind of best practice for IT or product development.
I've been asking the same question about all of software dev recently... For games I see at least one big factor being the attempt to run like a 9-5 white collar business while also budgeting and team building like they're making grandiose blockbusters. It seems like an attempt to mimic the Asian mobile market's results without understanding how those teams are structured and budgeted to make it effective.
There's a lot of delusion going on in the industry at every level. I just read a thing where Bioware won't be releasing Legendary editions of the old DA games because they don't have enough people at the company who currently understand the Eclipse engine and, I guess, aren't interested in investing resources in teaching anyone. Thing is, you know people would buy those. But the industry and even investors don't seem to think even in terms of most likely profit, or if they do, it's an idea of it from someone's drug trip. And yet they're spending money as if they are, like you said, certain to make Asian mobile market cash hand over fist.
I'm trained (for my sins, hah) in Agile, and the number one thing I get out of it is user focus. You need to be thinking about what the end product is to the player, in this case, not what it looks like to shareholders or your peers or the media or you. Developing a game no one buys is a self-indulgent hobby.
As far as excitement about graphics goes...how much further can it be pushed beyond current gen or even last gen GPU capability? And what would be the point as far as player experience goes? Performance, certainly, that's a reasonable concern.
Graphically, style lasts forever, photorealism gets outclassed by future graphical capabilities, limits buyers (not going to buy if I can't play it smoothly on my system), and I think can cause a kind of graphical stagnation where creativity gets optimized out (am I in the forests of Skyrim or the forests of God of War?). Indie devs see all this and going for style is often a better use of their limited resources. I've been playing a lot of Against the Storm and it embraces the style originally inflicted upon ol' Warcraft 3 due to technical limits. It doesn't hit the insane shooting-for-photorealism detail found in, say, Total Warhammer, but it doesn't need to in order to be pleasant to look at.
The topic of "how much fidelity is too much fidelity" is another one I ought to cover at some point as well!
While movies like Avatar 2 show us that there are still interesting problems in the space of photorealism, I think we really have reached a point where the ROI on advanced graphics has dramatically lowered in real-time applications. But, we've also seen a dramatic loss in "responsiveness" in high-fidelity games as a result of the techniques used to achieve them.
Every millisecond the CPU spends managing reflection probes, shadow volumes, and the like is a millisecond it could have spent calculating the physics or game rules of a sandbox interaction. Like, what is more important to a Battlefield game? Photorealism or being able to shoot a physically accurate hole in the wall? I'd say the latter.
There's also the question of how this all effects developer workflows. The standard lighting pipeline these days is absurdly complicated with all of the metadata you have to put into a scene to make it look good... only for it to break the second the player goes into a corner of the map they weren't expected to go into.
This is a long way of saying: I think unlocking the full performance of the hardware we have—and taking advantage of techniques like path tracing which have "ceilings" to how much processing they need—could shorten game development times dramatically and make them more interactive. 😂
I wonder how much discussions this sort of thing is seeing in the industry. Dev pipelines need streamlining badly on every level; you can pour millions of dollars into dev but you won't see a dime until release. And it needs to be in a state fit for release, or else you're just setting yourself up to spend more money and time fixing it while earning customer ire.
I generally don't understand how game dev is getting approached these days by the movers and shakers. It doesn't seem to be following any kind of best practice for IT or product development.
I've been asking the same question about all of software dev recently... For games I see at least one big factor being the attempt to run like a 9-5 white collar business while also budgeting and team building like they're making grandiose blockbusters. It seems like an attempt to mimic the Asian mobile market's results without understanding how those teams are structured and budgeted to make it effective.
There's a lot of delusion going on in the industry at every level. I just read a thing where Bioware won't be releasing Legendary editions of the old DA games because they don't have enough people at the company who currently understand the Eclipse engine and, I guess, aren't interested in investing resources in teaching anyone. Thing is, you know people would buy those. But the industry and even investors don't seem to think even in terms of most likely profit, or if they do, it's an idea of it from someone's drug trip. And yet they're spending money as if they are, like you said, certain to make Asian mobile market cash hand over fist.
I'm trained (for my sins, hah) in Agile, and the number one thing I get out of it is user focus. You need to be thinking about what the end product is to the player, in this case, not what it looks like to shareholders or your peers or the media or you. Developing a game no one buys is a self-indulgent hobby.