Why Are Modern PC Games Using So Much VRAM?

[Aquinus]

Lazy game devs that refuse to optimize bad ports are the ones with most to lose, and we are seeing that, sales of shit are getting incredible hard to go through distracted buyers.

I take offense to this statement. Most engineers would like to do stuff as well as they can but there is a caveat, doing things right tends to take more time and depending on management, the engineers might not have a choice to do it the correct way in order to hit deadlines. I'm not this sort of manager, but I know of those who are and it comes at the cost of quality. However, in a business where you're trying to plan out business objectives for a year or two, knowing how long it's going to take to get stuff done is important and one could argue that estimating effort for work to be done is often a lot harder than just doing it.

Don't blame engineers for things that are out of their control. It's not like these are people doing it out of the kindness of their heart then half-assing it. The SDLC is a bit more complicated than just writing code correctly and people who don't do it professionally tend to have loud voices and an impoverished idea of how these things work in a corporate setting.

 

[Vayra86]

I take offense to this statement. Most engineers would like to do stuff as well as they can but there is a caveat, doing things right tends to take more time and depending on management, the engineers might not have a choice to do it the correct way in order to hit deadlines. I'm not this sort of manager, but I know of those who are and it comes at the cost of quality. However, in a business where you're trying to plan out business objectives for a year or two, knowing how long it's going to take to get stuff done is important and one could argue that estimating effort for work to be done is often a lot harder than just doing it.

Don't blame engineers for things that are out of their control. It's not like these are people doing it out of the kindness of their heart then half-assing it. The SDLC is a bit more complicated than just writing code correctly and people who don't do it professionally tend to have loud voices and an impoverished idea of how these things work in a corporate setting.

And then there's the consumer POV: offer me a product that works as expected, else I hate it.

The problem is the corporate setting, clearly in the way of two parties who kinda want the same thing - and are willing to invest time in it / pay for invested time.

 

[bug]

I take offense to this statement. Most engineers would like to do stuff as well as they can but there is a caveat, doing things right tends to take more time and depending on management, the engineers might not have a choice to do it the correct way in order to hit deadlines. I'm not this sort of manager, but I know of those who are and it comes at the cost of quality. However, in a business where you're trying to plan out business objectives for a year or two, knowing how long it's going to take to get stuff done is important and one could argue that estimating effort for work to be done is often a lot harder than just doing it.

Don't blame engineers for things that are out of their control. It's not like these are people doing it out of the kindness of their heart then half-assing it. The SDLC is a bit more complicated than just writing code correctly and people who don't do it professionally tend to have loud voices and an impoverished idea of how these things work in a corporate setting.

It depends on the level of competence. Good engineers will try to do a good job, as you have noted. But if the project doesn't have enough competent engineers (because the budget is tight or whatever), they'll just happily eat your budget and deliver garbage.

Either way, not my problem

 

[Shihab]

I take offense to this statement. Most engineers would like to do stuff as well as they can but there is a caveat, doing things right tends to take more time and depending on management, the engineers might not have a choice to do it the correct way in order to hit deadlines. I'm not this sort of manager, but I know of those who are and it comes at the cost of quality. However, in a business where you're trying to plan out business objectives for a year or two, knowing how long it's going to take to get stuff done is important and one could argue that estimating effort for work to be done is often a lot harder than just doing it.

Don't blame engineers for things that are out of their control. It's not like these are people doing it out of the kindness of their heart then half-assing it. The SDLC is a bit more complicated than just writing code correctly and people who don't do it professionally tend to have loud voices and an impoverished idea of how these things work in a corporate setting.

Thank you!
I was writing something along those lines when your post popped up.

I dare say the majority of technical issues in video games (and commercial software in general) are management-related. Cutting back on QA, poor resource allocation, unattainable goals at ridiculous time frames, etc. Sad really you don't see anyone mentioning this outside disgruntled engineers circles and a few decent management literature.

And to be fair, part of this is the market's own fault. The demand obviously favors rapid release cycles with bleeding edge features. Wait 8~10 years for a stable, well tested game with dated graphics? No sir!
The entire software industry has morphed to include the first years post release as part of the development cycle itself.

 

[Vayra86]

Thank you!
I was writing something along those lines when your post popped up.

I dare say the majority of technical issues in video games (and commercial software in general) are management-related. Cutting back on QA, poor resource allocation, unattainable goals at ridiculous time frames, etc. Sad really you don't see anyone mentioning this outside disgruntled engineers circles and a few decent management literature.

And to be fair, part of this is the market's own fault. The demand obviously favors rapid release cycles with bleeding edge features. Wait 8~10 years for a stable, well tested game with dated graphics? No sir!
The entire software industry has morphed to include the first years post release as part of the development cycle itself.

'Agile development'...
Fancy way of saying 'we really have too much work in given time, but we'll manage it - somehow'

 

[Bomby569]

I take offense to this statement. Most engineers would like to do stuff as well as they can but there is a caveat, doing things right tends to take more time and depending on management, the engineers might not have a choice to do it the correct way in order to hit deadlines. I'm not this sort of manager, but I know of those who are and it comes at the cost of quality. However, in a business where you're trying to plan out business objectives for a year or two, knowing how long it's going to take to get stuff done is important and one could argue that estimating effort for work to be done is often a lot harder than just doing it.

Don't blame engineers for things that are out of their control. It's not like these are people doing it out of the kindness of their heart then half-assing it. The SDLC is a bit more complicated than just writing code correctly and people who don't do it professionally tend to have loud voices and an impoverished idea of how these things work in a corporate setting.

where did i mentioned engineers in any way shape or form?! i clearly talked about developers.

But i got to say managing projects doesn't seem to a thing most anyone involved on the game development industry can do these days, it's basically target for the moon, unrealistic time management for the allocated resources, unrealistic expectations about own coding capacities, projects keep overpromissing and underdelivering time and time again.

I have no idea what part is who's fault inside the industry, but i do see a lot of projects with really unrealistic visions for it. From the absolute start, don't just blame the corporate suits.

 

[Vayra86]

where did i mentioned engineers in any way shape or form?! i clearly talked about developers.

But i got to say managing projects doesn't seem to a thing most anyone involved on the game development industry can do these days, it's basically target for the moon, unrealistic time management for the allocated resources, unrealistic expectations about own coding capacities, projects keep overpromissing and underdelivering time and time again.

I have no idea what part is who's fault inside the industry, but i do see a lot of projects with really unrealistic visions for it. From the absolute start, don't just blame the corporate suits.

Engineers are part of the development team, its one and the same thing in the current context AFAIK.

Project management, especially in a corporate setting and with many stakeholders involved, is incredibly difficult though, the project manager often isn't the problem, but rather the structure/mandate he has to work with. I've had that experience just now, finished off a couple phases on a project with one left to go, and you can't begin to imagine how many pawns are moving around the chess board all the time.

 

[Shihab]

'Agile development'...
Fancy way of saying 'we really have too much work in given time, but we'll manage it somehow - somehow'

Ironically, Agile was meant to include developers in the management process and have projects go at manageable paces with emphasis on producing good software. But it seems every manager out there just took the "early release" part, and skipped everything else.

 

[Bomby569]

Engineers are part of the development team, its one and the same thing in the current context AFAIK.

Project management, especially in a corporate setting and with many stakeholders involved, is incredibly difficult though, the project manager often isn't the problem, but rather the structure/mandate he has to work with. I've had that experience just now, finished off a couple phases on a project with one left to go, and you can't begin to imagine how many pawns are moving around the chess board all the time.

if i mention the all, i'm clearly not singling out the part, so i didn't talk about engineers but all the team involved.

But like i said before i think it's very easy to shift all the blame to the suits when we see project after project shooting for the moon in the promises department. Same of the blame has to fall with the creative teams for sure. No doubt for me. No company can bankroll engineers teams dreams forever, and some would really go forever if they let them to.

The industry as a all has a problem.

 

[bug]

'Agile development'...
Fancy way of saying 'we really have too much work in given time, but we'll manage it - somehow'

Well, agile development also says "it will be done when it will be done". Agile with a strict deadline isn't agile at all. Unless you start throwing features overboard.

where did i mentioned engineers in any way shape or form?! i clearly talked about developers.

But i got to say managing projects doesn't seem to a thing most anyone involved on the game development industry can do these days, it's basically target for the moon, unrealistic time management for the allocated resources, unrealistic expectations about own coding capacities, projects keep overpromissing and underdelivering time and time again.

I have no idea what part is who's fault inside the industry, but i do see a lot of projects with really unrealistic visions for it. From the absolute start, don't just blame the corporate suits.

Developers are software engineers. Or at least the good ones are...

 

[Bomby569]

Well, agile development also says "it will be done when it will be done". Agile with a strict deadline isn't agile at all. Unless you start throwing features overboard.

Developers are software engineers. Or at least the good ones are...

Developers as in usually mentioned the company developing the game not the profession. As opposed to the publishers. That's the way i meant it anyway.
Individual coders would also not make sense in this context to be blames or in the context of my answer. Salary men usually don't make decisions in this context. Unless it's a one man indie i guess.

 

[80-watt Hamster]

Developers as in usually mentioned the company developing the game not the profession. As opposed to the publishers. That's the way i meant it anyway.
Individual coders would also not make sense in this context to be blames or in the context of my answer. Salary men usually don't make decisions in this context. Unless it's a one man indie i guess.

You're getting too hung up on verbiage (yes, this is a pot/kettle situation). "Engineer", at least as I interpreted it in context, referred to the technical muscle of the team. Not necessarily those with the title of Engineer, or an engineering degree, but those doing the development work on a nuts-and-bolts mechanical level.

 

[Bomby569]

You're getting too hung up on verbiage (yes, this is a pot/kettle situation). "Engineer", at least as I interpreted it in context, referred to the technical muscle of the team. Not necessarily those with the title of Engineer, or an engineering degree, but those doing the development work on a nuts-and-bolts mechanical level.

i never mentioned engineers, i was quoted as blaming engineers.

 

[bug]

i never mentioned engineers, i was quoted as blaming engineers.

Again, what's the distinction between (software) developers and (software) engineers?

 

[dirtyferret]

I apologize for such a long letter post, I didn't have time to write a short one.

I think a lot of people are having different arguments and trying to make one factual thing lead into another thing. I created a new post for this topic because the numerous "X amount of Video RAM is just not enough" have derailed into the typical fan boy ego arguments of my specs are right and your specs are wrong.

Its like boiling a frog in hot water right.

Not to be off topic but I am the OP; I hate this saying mainly because it's not true. If you put a frog in a pot of water and slowly turn up the heat, the frog will jump out once the temp becomes uncomfortable and the frog has access to jump out. That's really the morale of the story, not turning up the heat slowly but slowly preventing people (or frogs) from having an option to leave.

I then look at studios like Larian (Divinity OS, Baldurs Gate 3) and I see something the way its meant to be done. You can just feel the love there if you watch any content they put out.

Yet Larian, Blizzard, and CD Projekt Red build their games primarily for the PC first. It's when CDPR decided to make Cyberpunk playable on two generations of console at the same time as PC that it was clear they had bitten off more than they can chew and problems arose.

you really don't see a problem with games requiring increasingly more VRAM while looking no better than games 3-5 years old?

I mean, I don't suggest we should sit on 8GB forever, but why demand 16GB just to do the job that was done by 8GB a few years ago?

I agree with @bug on this main principle. Let's toss out 8GB and use a 20GB as an example which we can all agree is a high vram number for today. If I play Ultra Epic Bejewel on Ultra settings and it uses 24 GB of RAM making my 20GB card a stuttering mess and at very high settings the game plays just fine yet visually the game looks identical in both settings than we have a right to ask "what's going on?". And if the answer is, well at Ultra settings with an Nvidia card the double secret ray tracing is turned on and it eats up RAM than I have a right to call BS on the publisher/developer for capitulation towards Nvidia for implementing something that shows no visual improvement and for Nvidia for creating something simply to make their previous cards outdated, sell more cards, and have their army of idiot fan boys posts on every forum "24GB is the new entry level for video cards".

but if you haven't got that 16GB bottom line on a new PC GPU of some calibre these days, you simply have a shit piece of hardware that offers less than the mainstream requirement.

But Nvidia tells me on their own web site the XX60/ti series offers "incredible performance" and have been historically their mid tier /mainstream cards. I mean shouldn't they at least change their web site to say these cards offer "shit performance but cost $400"?

I'm not disagreeing, I think right now 12GB is what all mid tier cards should offer with an eye of 16GB+ in another generation or two.

i never mentioned engineers, i was quoted as blaming engineers.

you said "Lazy game devs that refuse to optimize bad ports" and software engineers are usually near the top of game development look more on the game big picture. The lead designer is usually a software engineer. It's like saying I blame a lazy kitchen crew for the bad food they put out but I never mentioned the chefs so I don't blame them.

 

[Bomby569]

you said "Lazy game devs that refuse to optimize bad ports" and software engineers are usually near the top of game development look more on the game big picture. The lead designer is usually a software engineer. It's like saying I blame a lazy kitchen crew for the bad food they put out but I never mentioned the chefs so I don't blame them.

Again game devs as in the company, i can't blame the salary men that don't make decisions. But if they do, if the software engineers make the decisions in the development company so yes, they are to blame not the publisher. What is the doubt here? someone that makes decisions at the devs company is f'up game after game on PC, lazy port after lazy port, excuses after excuses, i'm sorry's one after the other.

TLdr: if the engineer makes the decisions, they are lazy and should worker harder for PC gamers money.

potato, potahto, semantics are irrelevant.

 

[bug]

Again game devs as in the company, i can't blame the salary men that don't make decisions. But if they do, if the software engineers make the decisions in the development company so yes, they are to blame not the publisher. What is the doubt here? someone that makes decisions at the devs company is f'up game after game on PC, lazy port after lazy port, excuses after excuses, i'm sorry's one after the other.

TLdr: if the engineer makes the decisions, they are lazy and should worker harder for PC gamers money.

potato, potahto, semantics are irrelevant.

If you meant the company, perhaps it would have been a little more clear if you wrote "the developer" instead of "game devs". Just a suggestion, English isn't my first language.

 

[dirtyferret]

If you meant the company, perhaps it would have been a little more clear if you wrote "the developer" instead of "game devs". Just a suggestion, English isn't my first language.

I see what he is trying to say but he's going back and forth between developers and publishers. On smaller games they are one and the same and your AAA titles the publisher actually owns the development studios. If we are talking about your large franchise games than all the major business decisions (budget, launch window, etc.,) are actually made by or heavily influenced by the publisher.

 

[Bomby569]

If you meant the company, perhaps it would have been a little more clear if you wrote "the developer" instead of "game devs". Just a suggestion, English isn't my first language.

sorry about the confusion

"game devs" or "game developers" don't just mean the people involved, just as "game publishers" don't mean any individual, but the companies, i think. In Redfall the game developer is Arkane (not some code monkey) and the Publisher is Bethesda, it's on the steam page for the game.

Redfall on Steam

The town of Redfall is under siege by a legion of vampires who have cut the island off from the outside world. Trapped with a handful of survivors, slay alone or squad up in open-world FPS action.

store.steampowered.com

 

[dirtyferret]

sorry about the confusion

"game devs" or "game developers" don't just mean the people involved, just as "game publishers" don't mean any individual, but the companies, i think. In Redfall the game developer is Arkane (not some code monkey) and the Publisher is Bethesda, it's on the steam page for the game.

Redfall on Steam

The town of Redfall is under siege by a legion of vampires who have cut the island off from the outside world. Trapped with a handful of survivors, slay alone or squad up in open-world FPS action.

store.steampowered.com

I'm not picking on you just explaining how this becomes confusing and why it's not just a "blame the dev" situation

Redfall is developed by Arkane Austin
published by Bethesda
both Arkane and Bethesda are subsidiaries of Zenimax
Zenimax is a subsidiary of Xbox game studios
Xbox game studios is a division of Microsoft

That's a lot of people looking down on the developer, Arkane, saying what's going on? Why are you delayed? Are you going to hit the launch window we set for you? Why are you over budget?

 

[Bomby569]

I'm not picking on you just explaining how this becomes confusing and why it's not just a "blame the dev" situation

Redfall is developed by Arkane Austin
published by Bethesda
both Arkane and Bethesda are subsidiaries of Zenimax
Zenimax is a subsidiary of Xbox game studios
Xbox game studios is a division of Microsoft

That's a lot of people looking down on the developer, Arkane, saying what's going on? Why are you delayed? Are you going to hit the launch window we set for you? Why are you over budget?

fair enough, lots of people, lots of blame to go around, still we buy redfall because of Arkane not because they are owned by Bethesda, Zenimax or Microsoft. As a consumer internal disputes are not my concern.

But as a developer company you have to manage customers expectations and corporate overloads money, or just don't sell your soul to them.
I'm sure there is a budget and a timelime for all these projects. Just manage work to fit them, all i see is overpromises tbh, most games would be in development until my grandkids are old enough for retirement. How can i blame the suits for this?

Few companies i assume can pull a Rockstar, it's done when it's done, 10 years if we want to

 

[64K]

I'm not picking on you just explaining how this becomes confusing and why it's not just a "blame the dev" situation

Redfall is developed by Arkane Austin
published by Bethesda
both Arkane and Bethesda are subsidiaries of Zenimax
Zenimax is a subsidiary of Xbox game studios
Xbox game studios is a division of Microsoft

That's a lot of people looking down on the developer, Arkane, saying what's going on? Why are you delayed? Are you going to hit the launch window we set for you? Why are you over budget?

The old saying comes to mind:

"Too many chiefs and not enough Indians."

 

[dirtyferret]

still we buy redfall because of Arkane not because they are owned by Bethesda, Zenimax or Microsoft.

absolutely

As a consumer internal disputes are not my concern.

But as a developer company you have to manage customers expectations

You are a consumer but are you the customer in that scenario I just posted?

Now I'm not saying you can't rant, in fact I already stated you every right to do so

I don't believe the hostility (personally) is towards those people who create games, I still believe there is a strong level of gratitude towards them. I believe the hostility is towards the publishers and GPU tech companies that increasingly treat the consumer as a product that they deliver to their shareholders (their one true consumer). You are seeing rushed products go out the door where consumers almost act as beta testers for these products and hardware that costs a pretty penny, positioned as high performance, and is now outdated in 24 months. The response from these companies is just buy more, just pay more, but consumers have the right to say what are we getting for our money and we want better.

but to paraphrase my 64k friend, it's the industry that has created this mess not some guy/girl just doing their job so they can pay off their mortgage this month.

 

[Vayra86]

fair enough, lots of people, lots of blame to go around, still we buy redfall because of Arkane not because they are owned by Bethesda, Zenimax or Microsoft. As a consumer internal disputes are not my concern.

But as a developer company you have to manage customers expectations and corporate overloads money, or just don't sell your soul to them.
I'm sure there is a budget and a timelime for all these projects. Just manage work to fit them, all i see is overpromises tbh, most games would be in development until my grandkids are old enough for retirement. How can i blame the suits for this?

Few companies i assume can pull a Rockstar, it's done when it's done, 10 years if we want to

Well I'm not so sure I buy a game because of studio A or B.

I do know for damn sure I'm NOT buying games because they are published by EA, Ubisoft, Activision or some shitty Asian gacha game publisher. OTOH, I might STILL be triggered to buy a game if the game itself is interesting enough. Nowhere, ever, do I consider the studio. All those great tales about how this or that guy or girl worked on game X or Y or was in the team that made Z have never, and I do mean never, been the cause of more great games. At best they provide input, but not a single time is it pivotal in a game's development. What IS pivotal is how a studio is managed. And I also believe that if you place people developing games in their strength and let them execute their creative vision (to clarify: the team) and if that team can work together as professionals, 9 out of 10 times you really do get a game that has some idea of greatness, or at least has its own defined character, and is not a soulless piece of shit that feels like rehash/copy pasta. Why - because professionals can pool their ideas and knowledge and have the skillset to come out with the best possible execution.

That works irrespective of the type of game it is, too. I don't know about you but I'm quick to recognize when something is really good, even if it doesn't tick all the usual boxes. Some things just stand out as their own thing. And frankly we do get a lot of those games especially in the single player narrative-type that is emerging fast. And in the slipstream of a lot of truly original titles, there's an endless wave of copies and spinoffs that barely matter and are made by untalented studios and individuals. They might all do their damnest best to do it right, but, yeah, individual qualities on all levels differ.

It all comes down to a difference in vision on what games ought to be. Some publishers look at games as products that have a brief lifecycle and serve to sell upfront and then rake in as much cash in that lifecycle, the whole project is managed around that idea. Other publishers look at games as their chance to make a mark, to make something unique, set a new bar and stand out above the crowd - the uniqueness of the title is in fact its business model. Again other publishers, and a good example of those are the likes of Guerrilla Games that build first party system sellers for Sony, kinda exist to do both, or, succeed best when they do both. The finite lifecycle then doesn't matter much, the project itself is unique enough and might even build its own franchise to become a game of the first category later (e.g. Killzone, Horizon).

Absolutely none of the above is attributable to individual engineers/developers or even just a dev team. Its all down to the entire project management, whether in corporate setting or not.

 

[dirtyferret]

And yet another article on the subject although more from a consumer take than a tech take, this one is from digital trends

It pains me to say, but it’s officially a bad time to be a PC gamer

A string over disastrous releases in 2023, combined with expensive hardware, has made this one of the worst times to be a PC gamer.

www.digitaltrends.com