• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Has Unreal Engine 5 (lumen, nanite, metahuman) really helped reduce development time/costs?

Has Unreal Engine 5 really reduced development time/cost

  • Yes

    Votes: 29 19.0%
  • No

    Votes: 124 81.0%

  • Total voters
    153

The Cockatrice

Gold Member
so if developers are not correctly forecasting
THey never do, and will almost always use the max amount. I see this where I work, corporation. Something that would take a month at best, they triple it to 3 months, mostly because if they were super efficient, companies wouldnt need them anymore. Thats how reality is these days. Society has become lazier overall, not just in the gaming industry, but overall. Managers only care about short term profit, as soon as possible.
 

Clear

CliffyB's Cock Holster
I dont. If anything if you've seen my posts in the PS5 threads I'm against getting new systems and whatnot, because developers used to pull insane things with engines such as TLOU2 and HFW running on a fucking PS4 despite these games looking next-gen. UE5 was also supposed to be the savior, an engine that could simulate RT even on lower-end hardware, lumen and all that marketing BS and so far, it's been a fucking mess. Even to this day devs still refused to do shader precomp which is as basic as shit and even when they do, games still stutter as fuck. No, I dont want to upgrade my GPU every year because lazy devs cant optimize their games. They spend 8 fucking years making a game that runs like ass and ure telling me in all those years they they didnt have time for optimizations? We're getting a million fixes post releases. Safe to say they are lazy as shit, dev times tripled in the last years. You seem offfended so Im assuming u might be one, not saying all of them are in the same boat, but yeah.

Crazy thought: Maybe you shouldn't believe marketing hype? Especially when the marketing hype isn't directed specifically at you.

The reality is that you aren't the target client for the product "Unreal Engine version X". Devs are.

The product is pitched on function, relative ease-of-use and performance. None of which promises perfect results in every circumstance and deployment. In fact, the only thing that's more or less guaranteed is function. Lumen, Nanite, etc. Are just features that are pitched on the basis that they are faster methods for doing GI and geometry virtualization, techniques that aren't new, but tend to be used sparingly due to cost.

Which isn't to say they are offering "free lunches", because nothing in computing is cost free in performance and manpower terms.

The problem with your argument is that you don't appreciate the gains, only the defecits. And in your ignorance, claim lazyness because you aren't getting the "free lunch" that was never promised in the first place.
 

The Cockatrice

Gold Member
Maybe you shouldn't believe marketing hype?

I dont but heres a crazy thought, should we allow them to lie? If we can;t complain then how are things going to improve, with like anything?

The reality is that you aren't the target client for the product "Unreal Engine version X". Devs are

Everyone is the target. If consumers arent satisfied with how the game run, companies dont make money. Have you seen what the majority of people have as hardware on steam surveys? Low towards mid-end stuff.

The problem with your argument is that you don't appreciate the gains, only the defecits.

Ofc, I'm eastern european. We are are impossible to please and we rarely take shit. Diminishing returns is a real thing and while techniques have improved, more realistic shadows and whatnot, at the end of the day, no one is going to stop and stare at them when they shoot stuff on screen. I hardly care for raytracing even though I appreciate what it does, especially in CP2077 for example, the light bounces and how each tiny useless shit u will never notice casts a shadow that halves your framerate. Work smarter is all they have to do.
 

i_mean

Banned
its clear from the latest batch of delays that no dev gives two shit about the xs or even the xsx enough to optimize for it.

I highly doubt xs is the reason for how long its taken for us to get UE5 games. It's mostly epic. They didnt release the fucking thing until 2022. And then when they did it was extremely poor performance wise. It wasnt until 5.2 when foliage finally got nanite support and it wasnt until earlier this year when they finally improved CPU performance by 50%.

A LOT of games like Immortal, Lord of the Fallen, Robocop and even Black Myth to some extent were fucked over by Epic. Not XS.

It's been obvious for a long time now that no dev, not even epic, gives shit two shits about the xs. MS had to send Coalition to get the matrix demo working on the xs because Epic didnt want to optimize for it. It didnt hold back the Matrix demo, it didnt hold back Wukong and certainly not hellblade 2. devs just downgrade the shit out of the xs version and call it a day or simply refuse to ship it like we have seen in black myth and dune recently. even indie devs are like fuck this console.

UE5 just like UE4 has taken some years to mature. Remember, Batman AK also shipped on UE3 instead of UE4 despite UE4 releasing a year prior to its release. this delay was expected. we saw some good results last year in Robocop and Remnant 2 at times, and this year Hellblade 2 and Wukong are finally delivering on the promise of UE5. Marvel's 1943 is just going to take it to the next level next year.
What? none of these games were "fucked" over by Epic, they CHOSE to use UE5 with FULL knowledge of it's current state. I worked on Immortals of AVEUM (i can get this verified, who did i contact? so some weirdo named Toots doesn't harass me), it was a UE4 game first and the management made the decision to move over the UE5 after they did some initial performance tests/tech debt tests and deemed we should wait, but they wanted the marketing points. The pipeline wasn't built for UE5, i.e lots WPOs and PDOs in the materials which Nanite didn't support at the time (UE5 5.1)...VSM was in beta etc....all of this was known PLUS they have access to UDN and can ask Epic directly about anything engine related, even send them your scenes if you can't get around a particular issue.
 

Clear

CliffyB's Cock Holster
I dont but heres a crazy thought, should we allow them to lie? If we can;t complain then how are things going to improve, with like anything?

They aren't lying though; like I said the features exist and noone wrecks their sales pitch by highlighting limitations and shortcomings.


Everyone is the target. If consumers arent satisfied with how the game run, companies dont make money. Have you seen what the majority of people have as hardware on steam surveys? Low towards mid-end stuff.

Untrue. Epic want their product's usage highlighted because its good for their business, and why showing their logos is part of their licensing arrangement. That being said they cannot take responsibility for the quality of every implementation.

Buying a Ferrari won't make you a better driver than you were already! What it does offer though is the potential to drive real fast and look cool while doing it!

Ofc, I'm eastern european. We are are impossible to please and we rarely take shit. Diminishing returns is a real thing and while techniques have improved, more realistic shadows and whatnot, at the end of the day, no one is going to stop and stare at them when they shoot stuff on screen. I hardly care for raytracing even though I appreciate what it does, especially in CP2077 for example, the light bounces and how each tiny useless shit u will never notice casts a shadow that halves your framerate. Work smarter is all they have to do.

Your nationality is irrelevant. The problem is that you probably have spent way too much time listening to shysters like DF, who pretend to know everything but in actuality understand remarkably little and merely parrot buzz-words.

One of the most toxic consequences of their rise is the popularisation of feature-sets as a metric for quality; they emphasize stuff like RT being implemented (and the quality of its implementation) as being somehow really valuable and important, when the real question is to what degree it is beneficial to the presentation as a whole.

There's some screenshots earlier in this thread used to argue the case how little things have improved; when all what I see is yet another example of the importance of good art-direction and ingenuity versus sheer tech grunt.

My point being that Art and Tech are different, and should not be used interchangeably. DF is toxic specifically because more than anyone else they have driven this idea that the tech is somehow more instrumental than the art in creating an impressive result.

The consequence of this -to get back on point- is that a lot of the time with modern games, devs get hung up chasing feature sets in order to keep-up with the more accomplished teams because the media influence encourages them to think along those lines.

The end result is that they spend so much time and effort ramping up their workload so they can boast using Nanite, or Lumen, or RT or whatever, as part of their competitive pitch, that the smaller stuff simply does not have time to be polished.

I hope you're following me on this because if you have you'll see why I was offended by the "lazy devs" line. They are not being lazy, they are in fact trying to do more then ever technically in many cases because they feel like they have to.

In the real-world, the upshot of all this is the end-user seeing flaws and issues in the final product that simply weren't there using older, more limited tech*, resulting in them calling the devs "lazy" despite them working harder than ever to please the audience.


*Shader comp stutter is a classic example of this. It only became an issue after shader-usage and variety reached a point where the compiled scripts could not be kept permanently memory resident, and needed to be generated on-the-fly. It being a problem borne not out of lazyness, but out of making efforts to do more, and more work.
 

FireFly

Member
THey never do, and will almost always use the max amount. I see this where I work, corporation. Something that would take a month at best, they triple it to 3 months, mostly because if they were super efficient, companies wouldnt need them anymore. Thats how reality is these days. Society has become lazier overall, not just in the gaming industry, but overall. Managers only care about short term profit, as soon as possible.
If the lazy developer argument was "the development team was given X months to optimize the game with these technological constraints and this much technical resource. Equivalent but "best in class" teams using similar technology have achieved Y results in the same time period", it would be a lot more reasonable. Since then it would be based on actual evidence about what happened during development. But the lazy devs argument seems to be used simply after seeing the end product with no knowledge about how long the developer was actually given to optimize the title, what constraints they were under, and what technical resources they had. (Lots of the UE5 games seem to be made by very small developers trying to punch above their weight).
 
Last edited:

The Cockatrice

Gold Member
They aren't lying though; like I said the features exist

Bending the truth, is still lying lol.


It's not. Consumers are everything. We live in a consumerism world.

What it does offer though is the potential

It's been many years since they've shown the potential. At this point, in regards to UE5, I no longer see any potential, even in the eyes of a dev. Whether that will change, who knows, Will see how Orion turns out for example, which is prolly the biggest UE5 project as of now.

They are not being lazy, they are in fact trying to do more then ever technically in many cases because they feel like they have to.

I've yet to see concrete examples of this where art and technicality is used in perfect sync. It's always one or the other. Projects dont need 8 years to develop, no matter how big they are and what you're essentially saying is that they actually need more time, which would mean a decade per game and if you cant see how ridiculous that sounds, I got nothing else to say.

Agreed on DF tho. They focus on the wrong things and in turn makes their followers do as well.
 
a close up of a man 's face with a beard and a bald head .'s face with a beard and a bald head .

The Coalition demo is even falling short compared to the Unreal Engine 5 demo.

There are various scenes in that Unreal Engine 5 demo that is showcasing a substantial more amount of details being drawn compared to current games (to me it looks obvious). I’ll be here waiting ‘till I see anything else that is drawing like a monster in scenes.
You wanted to talk about the light, so stay on the subject. Do you have a feeling that the lumen lighting has been downgraded? If so what properties have been downgraded? Do you see more light leaks? Be specific, because your assumptions are not enough to prove that the lumen lighting was downgraded.




OdGDqEU.jpeg
ivi7U3H.jpeg
sAHk5Q0.jpeg
ZUZK7J5.jpeg
 

SlimySnake

Flashless at the Golden Globes
What? none of these games were "fucked" over by Epic, they CHOSE to use UE5 with FULL knowledge of it's current state. I worked on Immortals of AVEUM (i can get this verified, who did i contact? so some weirdo named Toots doesn't harass me), it was a UE4 game first and the management made the decision to move over the UE5 after they did some initial performance tests/tech debt tests and deemed we should wait, but they wanted the marketing points. The pipeline wasn't built for UE5, i.e lots WPOs and PDOs in the materials which Nanite didn't support at the time (UE5 5.1)...VSM was in beta etc....all of this was known PLUS they have access to UDN and can ask Epic directly about anything engine related, even send them your scenes if you can't get around a particular issue.
Just because they chose the engine Epic created for next gen games doesnt mean Epic is absolved from all the blame. Epic shouldve had the engine ready on time and even if they didnt, it shouldve come with all the performance improvements and features they have added since.
 

i_mean

Banned
Just because they chose the engine Epic created for next gen games doesnt mean Epic is absolved from all the blame. Epic shouldve had the engine ready on time and even if they didnt, it shouldve come with all the performance improvements and features they have added since.
What blame? Ready on time for what??? They literally stated that this engine is IN DEVELOPMENT when they first introduced it, they introduced it at GDC i believe...and that's for DEVELOPERS not for the average person. The average person has no idea what development is like or even what engines do...just look at the replies when this topic pops up, it's filled with non sense and oh well DF said this and this guy from youtube said that.... they make an engine and it's ready when it's ready lol, you decide whether you want to use it or not. Devs will have a team of engineers that will make changes to the engine to align with their projects needs..that often includes rewriting portions of the engine...this is expected.

That isn't how any of this works or has worked since middleware engines like UE were a thing. It takes years and a lot of time for performance improvements to happen and it takes real world data, putting it into other developers hands and getting feedback, putting games out etc etc,,,this is why Unreal and other engines have their designations. I.e

Experimental = Do not use in a production environment, features might completely die or get retooled
Beta = Use at your own risk..plenty of bugs and things may change in subsequent releases since it's in active development
Production ready = This is the standard has been battle tested, won't change dramatically

If you use Unreal or unity for development you know this. It's just middleware..you test it out and see if it's right for your game or not. If Epic says XYZ thing is great and you test it and deem XYZ to be not so great for your game...then you don't use it. If you go ahead and use it...the blame it on you.
 

SlimySnake

Flashless at the Golden Globes
What blame? Ready on time for what??? They literally stated that this engine is IN DEVELOPMENT when they first introduced it, they introduced it at GDC i believe...and that's for DEVELOPERS not for the average person. The average person has no idea what development is like or even what engines do...just look at the replies when this topic pops up, it's filled with non sense and oh well DF said this and this guy from youtube said that.... they make an engine and it's ready when it's ready lol, you decide whether you want to use it or not. Devs will have a team of engineers that will make changes to the engine to align with their projects needs..that often includes rewriting portions of the engine...this is expected.

That isn't how any of this works or has worked since middleware engines like UE were a thing. It takes years and a lot of time for performance improvements to happen and it takes real world data, putting it into other developers hands and getting feedback, putting games out etc etc,,,this is why Unreal and other engines have their designations. I.e

Experimental = Do not use in a production environment, features might completely die or get retooled
Beta = Use at your own risk..plenty of bugs and things may change in subsequent releases since it's in active development
Production ready = This is the standard has been battle tested, won't change dramatically

If you use Unreal or unity for development you know this. It's just middleware..you test it out and see if it's right for your game or not. If Epic says XYZ thing is great and you test it and deem XYZ to be not so great for your game...then you don't use it. If you go ahead and use it...the blame it on you.
UE5 was released in spring of 2022. Immortals released in august 2023. This is before Epic had made performance improvements or added support for nanite foliage.

Their 'production ready' system had performance improvements upwards of 50%. Sounds pretty dramatic to me.
hkVgR6a.jpeg
hkVgR6a.jpeg
 
Last edited:

i_mean

Banned
UE5 was released in spring of 2022. Immortals released in august 2023. This is before Epic had made performance improvements or added support for nanite foliage.

Their 'production ready' system had performance improvements upwards of 50%. Sounds pretty dramatic to me.
hkVgR6a.jpeg
hkVgR6a.jpeg

Forgive me, i don't get what you're trying to say here. As i stated i worked on immortals..it was released on UE 5.1 while VSM and Nanite were still in Beta...the images you're showing me there is mainly due to Parallelization which was introduced in unreal 5.4. Parallelization has been a huge issue from unreal since UE3....they finally addressed it and will continue to address multi-threading in 5.5.
 
The problem is that developer ambition tends to increase with technology. If something becomes easier, it just means you have more time for something else. Development time will likely increase.
 
I haven’t seen any game thus far that has the capability the Unreal Engine 5 demo showcased, I shared one scenario earlier. This isn’t new to me, stuff like this happens all the time, I seen it with Unreal Engine 4 and Unreal Engine 3. It’s not that big of deal to me, games like Hellblade 2, Wukong ect. are still great looking games.
Black Myth Wukong looks amazing overall like you say, but it still has average looking assets here and there. UE5 tech demo running on the PS5 used movie like assets, and I doubt developers are going to use similar assets quality in their game (I'm guessing game size would be an issue probably).
 

CamHostage

Member
Just because they chose the engine Epic created for next gen games doesnt mean Epic is absolved from all the blame. Epic shouldve had the engine ready on time...

Epic did not miss any publicly-announced milestones AFAIK in the release timeline of UE5? They laid out some of the roadmap when they showed it in 2020, then they released the Early Release in May 2021 to the public, and they released the 5.0 version in April 2022.

(I think the Fortnite conversion proof-of-concept project was a little later than expected? If I remember right they said fall but it came out in Dec. I don't know that I'd call that a milestone on the product timeline though, plus they had already released Matrix Awakens by that point.)

If you're saying Epic is to blame for showing UE5 too early and setting expectations of the new gen too high for the susceptible general public, okay, that's one opinion... many would say that they would rather know things and deal with the abundance of info than be left out of the loop, but surprise and mystery and realistic tablesetting is also of value I suppose.

(Of course, I don't know how Epic could have kept UE5 "developer-eyes-only" through 5.0 release when they offered an open Early Access and also released Matrix/Fortnite in 2021, but maybe some could accuse Epic of spending too much time making it available and losing speed on making it done. Maybe?)

...and even if they didnt, it shouldve come with all the performance improvements and features they have added since.

...How?

You have a habit of being angry that things which get done only now didn't finish/release two or three years prior when you wanted them to, but that's not how technology development works. Or how anything works. If it's not done, if it doesn't exist yet, you can't have it.

WIP showings and demos are cool, but these are not done things. A demo is built to avoid showing all the unfinished work that actually needs to be done on the project; sometimes, they even have mock solutions which work in the carefully-arranged demo but must be replaced ultimately because they won't work at full scale in the finished product. If games could be made modularly, where the vertical slice really was one finished room on a finished foundation and all the rest of the work was building out every other room just like that one, it'd be great, but unfortunately that's not a reality designers work with.
 
Last edited:

simpatico

Member
Absolutely. It doesn’t stop bad studios from milking it and taking 7 years, but look at Trepang2. A project like that would be impossible 10 years ago.
 

Fafalada

Fafracer forever
TLDR: Nanite introduce a lot of overdraw, which means you are wasting GPU cycles, LOD even generated one is far better for optimisation. Given what has been said in this video and by own account, while playing with it extensively, I think Nanite will be something like a ID Tech mega texture down the line. Not the mention the strain on CPU, which isn't handled in a way ID Tech 7 is for example.
CPU issue is just implementation detail that's specific to Nanite though - not the approach itself.
The thing is - virtualized geometry has been 'just around the corner' for over 15 years - nanite wasn't the first by a longshot - but it was the first properly funded push (and Epic could afford it since it's not like they make money off of Engine licensing anyway).

That being said - the field is expanding - and more GPU centric/friendly alternatives that aren't tied to proprietary tech stack are available now:

If I was a betting person - I'd predict most proprietary stacks will have one of their own (or just adapt the above) within 5 years from now. So TLDR - I don't think virtualized geometry is going away like textures did, but it also won't be a tech-differentiator - just a standardized workflow like regular meshes have been for past few decades.
 
Last edited:

samoilaaa

Member
every day you see a new UE5 game announced and it looks incredible but they lack in terms of level design , game mechanics , npc AI , story , all of these make the difference between a game and a tech demo

sure the water looks great , the sky is amazing , the forest is dense but its not enough
 
EA (new Mass Effect), CD Project Red, and a few others have switched to it.

They aren’t doing that because it’s more expensive.
 
Last edited:
Movie asset? These are movie assets (published in 2014)

One of those robots probably have more moving parts than an entire retail game today, none of these developers would be able to afford it with today’s business model.

I made a mistake on my analysis earlier, looking at it again it was one point of entry from the light source even though the point still remains.

ei0HJl9.jpeg

Epic said:

"Much of what you see was built with quixel megascan assets, but instead of using the game versions we use the cinematic versions which would typically only be used in film. There are around a million triangles each. Thanks to virtual texturing they all use 8K textures as well."



I cannot imagine a developer shipping a game with similar assets. Black Myth Wukong isnt using as detailed assets and it already takes up 128GB of disk space.
 
Last edited:

hussar16

Member
This is a great question and i would say it offloads the next gen detail graphics by doing everything faster ij one part so its spent oj more detail everywhere else making it closer to a cgi
 
Epic said:

"Much of what you see was built with quixel megascan assets, but instead of using the game versions we use the cinematic versions which would typically only be used in film. There are around a million triangles each. Thanks to virtual texturing they all use 8K textures as well."



I cannot imagine a developer shipping a game with similar assets. Black Myth Wukong isnt using as detailed assets and it already takes up 128GB of disk space.

I haven’t seen any retail game using anything close to film assets.
 
Last edited:

i_mean

Banned
I haven’t seen any retail game using anything close to film assets.
they are referring to the environment scans which is like 99 percent of that demo and are indeed film quality assets and some of those assets are available on megascans and have been used in retail games. So yes, you have seen film quality assets used in games :)
 

Dane

Member
These features aren't adopted overnight, they require adpation as expected, the issue is when they take too long, for example, on PC Gaming we're still having games who leans towards single thread despite multi threads being the norm for years. IMO there has been a lot of disinvestment on the programming side, features like DLSS and FSR are quickly adopted for cheap optimization while the code to the metal one is avoided, making the former a remedy to poort optimization in most cases today.

Controversially, the Xbox Series S would be seen as a good thing to keep studios in check to focus on optimization, games like Matrix Tech Demo and Hellblade 2 ran with similar performance and not much of a visual downgrade.
id's mega texture feature was, effectively, adopted by hardware vendors and manifests today as the progenitor of things like partially resident textures that we see in the PlayStation and Xbox. Carmack, as usual, was a generation head of everyone else. After Rage, Carmack stated he believed the next barrier - after virtualised textures - was virtualised geometry. He moved over to Oculus after that, so we never really got to see his work on that front. UE5's Nanite in merely an implementation of REYES, which is a geometry virtualisation approach. Offloading the overhead to the hardware is the next logical step, given how it contributes to both image quality, performance, and asset production - the trifecta of things hardware vendors can sell new hardware on. In addition to NPUs, for things like advanced image upscaling, I imagine we'll see hardware support for virtualised geometry in the next generation of hardware.
Issue is that Carmack tought the next gen would be powered by SSDs, which was indeed a thing expected on PCs, then the costs took way longer to be down.
 
Last edited:
they are referring to the environment scans which is like 99 percent of that demo and are indeed film quality assets and some of those assets are available on megascans and have been used in retail games. So yes, you have seen film quality assets used in games :)
Bullshit. Megascans have been around before Unreal Engine 5, so I’m assuming those were the “game versions” using the technology? Come on man, if that’s the case I’m expecting retail games to just use the “cinematic versions” from this point on and give us the quality we expect from films, go and model every piece of object in the scene using pixel sized polygons.
 

i_mean

Banned
Bullshit. Megascans have been around before Unreal Engine 5, so I’m assuming those were the “game versions” using the technology? Come on man, if that’s the case I’m expecting retail games to just use the “cinematic versions” from this point on and give us the quality we expect from films, go and model every piece of object in the scene using pixel sized polygons.
Goddamn! do you guys even look at or understand the things you talk about? Megascans comes in Variety of versions...you have the Original scan as a zbrush file if you want to alter it, you have the Nanite version..which is pretty much the original scan but decimated a bit, you Have LOD 0 to 5 which you can use if you're aiming for lower end hardware or a production that still wants to use megascans but aren't using virtual geometry . Before speaking...go to the site download the shit..try it yourself and THEN SPEAK...not before. It's all right there for all to see yet y'all choose not too but pretend like you know what you're talking about just the same.
 
Goddamn! do you guys even look at or understand the things you talk about? Megascans comes in Variety of versions...you have the Original scan as a zbrush file if you want to alter it, you have the Nanite version..which is pretty much the original scan but decimated a bit, you Have LOD 0 to 5 which you can use if you're aiming for lower end hardware or a production that still wants to use megascans but aren't using virtual geometry . Before speaking...go to the site download the shit..try it yourself and THEN SPEAK...not before. It's all right there for all to see yet y'all choose not too but pretend like you know what you're talking about just the same.
Which Megascan version was Immortals of Aveum using?
 

i_mean

Banned
There are probably only a few people on this board that can speak to this, the rest is just speculation.
True,,,but it's an enthusiast forum...i'd expect a bit more knowledge but video game enthusiast forums are a bit weird...if you go to a basketball enthusiast forum like realgm...you have people who know basketball..they've played it, coached it etc..the know ball! you go to a car enthusiast forum, they know how to build them, fix em etc..film enthusiast forum..they know composition, how to write screenplays...they've filmed, done photography etc.

Video game forums however are a bit unique as the vast majority have no idea how the sausage is made and no idea how to make em,,,yet have extremely strong opinions about them, even insulting and calling those who do lazy and all that despite having all the tools available to them for free even...ironic.
 

SHA

Member
Excellent video

It seems like these effort saving features are a bit too resource heavy resulting in a lot of the benefits being cancelled out on current-gen hardware.

Would you say a PS6 with a spec bump similar to the PS4>PS5 transition be enough to allow these kinds of features and engines to become ubiquitous?
But it can't be wrong %100, there is some truth to it and future tech will be based on this theory.
 

Fafalada

Fafracer forever
My point being that Art and Tech are different, and should not be used interchangeably. DF is toxic specifically because more than anyone else they have driven this idea that the tech is somehow more instrumental than the art in creating an impressive result.
The consequence of this -to get back on point- is that a lot of the time with modern games, devs get hung up chasing feature sets in order to keep-up with the more accomplished teams because the media influence encourages them to think along those lines.
This one you really can't put on DF, in any form.
My professional experience started just before PS2 launch - and even back then everything about the industry was chasing tech featureset trends.
There was a rather fun quote I recall from Tom Forysth back in the day when he first saw ICO - where he lamented the disbelief the game even got made / past the publishers shouting demands to emulate the popular trends (eg. Unreal tech checklists and all). I can't find it anymore - but it ended with something like 'more shadows, more polygons, do whatever Tim Sweeney Does!'. TLDR - game industry defining characteristic - for as long as I can remember it on the inside - has been a rather self-indulgent obsession with all this, hell a lot of what makes AAA what it is is born out of the same chase.

And don't even get me started on how bad this place was in the early 00s - tech discussion nowadays are downright civilized in comparison.
 
Last edited:

Clear

CliffyB's Cock Holster
This one you really can't put on DF, in any form.
My professional experience started just before PS2 launch - and even back then everything about the industry was chasing tech featureset trends.
There was a rather fun quote I recall from Tom Forysth back in the day when he first saw ICO - where he lamented the disbelief the game even got made / past the publishers shouting demands to emulate the popular trends (eg. Unreal tech checklists and all). I can't find it anymore - but it ended with something like 'more shadows, more polygons, do whatever Tim Sweeney Does!'. TLDR - game industry defining characteristic - for as long as I can remember it on the inside - has been a rather self-indulgent obsession with all this, hell a lot of what makes AAA what it is is born out of the same chase.

And don't even get me started on how bad this place was in the early 00s - tech discussion nowadays are downright civilized in comparison.

Maybe I'm being unduly harsh, but I truly feel like DF have pushed "graphics whoring" (which you're right, goes back to the beginnings of gaming) to a whole new level by miseducating the audience, in the sense that they highlight and metricize all the wrong stuff. Specifically what I dislike about their approach is their insistence on taking a junk science approach, over a more holistic appreciation of the quality of the created illusion. Its why I much prefer MIchael's stuff with NXG - with his work I really get the sense that he appreciates the "art" of crafting strong visuals, as opposed to simply describing whether the latest tech is deployed as described in vendor-provided acronyms and buzz-phrases.

From an industry end, feature-set has always been good for bonus pts in marketing/pitching. I've written plenty of documentation for that exact purpose over the years, knowing full-well its just smoke-and-mirrors in the service of actual smoke-and-mirrors!
 

YeulEmeralda

Linux User
True,,,but it's an enthusiast forum...i'd expect a bit more knowledge but video game enthusiast forums are a bit weird...if you go to a basketball enthusiast forum like realgm...you have people who know basketball..they've played it, coached it etc..the know ball! you go to a car enthusiast forum, they know how to build them, fix em etc..film enthusiast forum..they know composition, how to write screenplays...they've filmed, done photography etc.

Video game forums however are a bit unique as the vast majority have no idea how the sausage is made and no idea how to make em,,,yet have extremely strong opinions about them, even insulting and calling those who do lazy and all that despite having all the tools available to them for free even...ironic.
I don't want to know how videogames are made and I certainly don't give a shit about who makes them. I spend my time playing them.
Sesame Street Idk GIF
 

i_mean

Banned
I don't want to know how videogames are made and I certainly don't give a shit about who makes them. I spend my time playing them.
Sesame Street Idk GIF
and thats fine Elmo no one asked you to care but since you're in a thread that clearly will touch upon how video games are made..why are you here? Elmo lost?....you're likely too old to play the cool kid card...i mean you're clearly interested in the topic if you're in here.
 
I’ve searched online and found nothing, but is there anything documented from devs regarding Metahuman?

How are they finding it, have they managed to significantly reduce modelling time using it?
 
Top Bottom