AMD FSR 3.0 has exposed the ugly truth about most PC gamers

Klo · 30 Dec 2023 at 20:56

Grim5 said:
About half the users on this forum got very worked about about frame gen

And they tend to own Nvidia and post constant threads about it.

Dicehunter · 30 Dec 2023 at 21:12

I always laugh at the whole fake frames thing, It's all fake frames as none of it is real

sparkymark75 · 30 Dec 2023 at 22:24

rare said:
I don’t think it’s cheating, it all comes down to how it’s marketed. It’s a clever feature with obvious benefits just like so many others than have come before it.

It's literally in the name. Frame "generation". It's making up extra frames that wouldn't otherwise be there. So nVidia have the cheek to vastly increase prices for their hardware and then have you rely on software to get the performance you should be getting out the hardware at the price you are paying. And you all fell for it. nVidia marketing is genius.

rare · 30 Dec 2023 at 22:36

sparkymark75 said:
It's literally in the name. Frame "generation". It's making up extra frames that wouldn't otherwise be there. So nVidia have the cheek to vastly increase prices for their hardware and then have you rely on software to get the performance you should be getting out the hardware at the price you are paying. And you all fell for it. nVidia marketing is genius.

I know how it works. Fell for what exactly? I know how it works and it seems like a pretty useful technology if you take in to account what it actually is and its limitations. Not sure anyone fell for anything to be honest, I would suggest the opposite, pretty much anyone interested in hardware is well aware of what it is and what it isn’t.

Nexus18 · 30 Dec 2023 at 22:41

sparkymark75 said:
It's literally in the name. Frame "generation". It's making up extra frames that wouldn't otherwise be there. So nVidia have the cheek to vastly increase prices for their hardware and then have you rely on software to get the performance you should be getting out the hardware at the price you are paying. And you all fell for it. nVidia marketing is genius.

And amd followed and it's being claimed as the next best thing now....

Regardless of who does it, what does it matter if the positives outweigh the negatives?

Making software is not cheap/free, something has to fund R&D and the money isn't just being pulled from nowhere to pay for the developers and so on working on these technologies.....

GPU prices are a joke though and shouldn't be quite what they are even when accounting for inflation, that goes for both nvidia and amd and we can thank the mining boom and people paying stupid silly money for gpus back then for setting the precedent.

chris85oc · 31 Dec 2023 at 07:58

It’s the usual VERY LOUD and annoying vocal minority. You can see it everywhere in every facet of our society.

Grim5 · 31 Dec 2023 at 08:08

chris85oc said:
It’s the usual VERY LOUD and annoying vocal minority. You can see it everywhere in every facet of our society.

That's certainly one way to describe amd owners, as a vocal minority of the GPU market

0h_no_not_again · 31 Dec 2023 at 08:46

This is a bit of a weird take. The article has made some large generalising statements about PC gamers, moaning that PC gamers won't state the real reason they're not going and buying overpriced hardware to access new features.

DLSS3 and 3.5 have been reviewed excellently. FSR3 received poor reviews on release, and average reviews on recent update.
Lots of reviewers pointed out that only 4000 series GPU's having DLSS3 was annoying, but it was a function of the hardware required to make it work.

I think everyone is ****** off that you need to buy overpriced hardware to access a new feature, and it is all designed to lock you and developers into the ever growing ecosystem. Soooo, yeah, NV deserve some hate for building a closed ecosystem then over-pricing entry, but the tech was review accurately as being fantastic.

ANd I think that's where this article loses the plot. It cant seperate the tech from the access to the tech and price.

Dicehunter · 31 Dec 2023 at 08:50

chris85oc said:
It’s the usual VERY LOUD and annoying vocal minority. You can see it everywhere in every facet of our society.

Sadly true, I see it in everything, Gaming, Gyms, Work, Always a loud minority that have "main character syndrome" and cannot keep their traps shut for once in their lives.

Freddie1980 · 31 Dec 2023 at 08:51

K1LaW8tt5 said:
maybe this gamer always dreamed of using fake frames, But couldn't afford an RTX4000 GPU?? Or maybe it's a different person??

The whole piece seems very contrived.

tommybhoy · 31 Dec 2023 at 14:10

Freddie1980 said:
The whole piece seems very contrived.

It's only page 2 and just like the linked article, there's that much whining about the original FG whining, which makes it harder to guess who's getting whined at this time.

Is the DSO article author simply whining at everyone that didn't get sent a free 4090 from NV?

If I got sent a free 4090 I'd tell you FG was that good it improved my sex life too.

Or is he just whining at 20/30 series Nv users that complained they were locked out from FG by Nv as Nv want you to pay for the pleasure of FG?

Or is this thread just whining that it's AMD users fault again that there was a FG pushback despite the majority complaining were RTX users that got locked out of FG with Nv's even more ramped up planned obsolescence step up program?

At this point I can't read between the lines of who's whining louder.

Better to have than not, but I run native myself(which aligns with the majority in the forum)-unless I'm away from home and break out the old G-Sync 1060 laptop that 100% needs FSR to keep it playing newer titles.

Which brings me to the conversation of Nv's track record of planned obsolescence be it lack of vram or software/ hardware features(when they claim a hardware requirement for features), for example, my old 1060 G-Sync laptop that doesn't require the G-Sync hardware module for working G-Sync, despite the fact that Nv dug their heals in and 100% said a module was needed, was in actual fact a bear faced lie as it's not required for VRR.

Dicehunter · 31 Dec 2023 at 14:12

Do you have enough spaces between your sentences ?

tommybhoy · 31 Dec 2023 at 14:33

Dicehunter said:
Do you have enough spaces between your sentences ?

It's Frame Gen prep, need space for the insertion of fake sentences.

Dicehunter · 31 Dec 2023 at 21:39

tommybhoy said:
It's Frame Gen prep, need space for the insertion of fake sentences.

You're a wizard Harry

drakulton · 31 Dec 2023 at 21:49

So, are they saying that the truth is, most PC gamers are ugly?

Journey · 31 Dec 2023 at 22:05

Pretty sure if they just allowed fake frames on 3000 series cards (and maybe even AMD cards) they would have been less drama in general, but why give something away that you can charge for?

muon · 31 Dec 2023 at 22:07

Journey said:
Pretty sure if they just allowed fake frames on 3000 series cards (and maybe even AMD cards) they would have been less drama in general, but why give something away that you can charge for?

It wouldn't have worked on other cards anyway. It's like how they eventually turned on RT for Pascal cards and it was awful.

The frame generation nvidia is using, uses hardware designed specifically for it and only exists on Ada cards onwards. It can probably be ported back, but will it perform well enough for anyone to use?

Harnessing the NVIDIA Ada Architecture for Frame-Rate Up-Conversion in the NVIDIA Optical Flow SDK | NVIDIA Technical Blog

The NVIDIA Optical Flow SDK 4.0 is now available, enabling you to fully harness the new NVIDIA Optical Flow Accelerator on the NVIDIA Ada architecture with NvOFFRUC. Starting from the NVIDIA Turing…

developer.nvidia.com

The Optical Flow algorithm requires certain pre– and post-processing steps to improve the quality of the flow vectors.

In the NVIDIA Turing and NVIDIA Ampere architecture generation GPUs, most of these algorithms use a compute engine to perform the required tasks. As a result, when the compute engine workload is high, the performance of the NVIDIA Optical Flow Accelerator (NVOFA) could be affected.

On NVIDIA Ada-generation GPUs, most of these algorithms are moved to dedicated hardware within the NVOFA, reducing the dependency on the compute engine significantly.

In addition, NVIDIA Ada-generation GPUs bring several other optimizations related to reducing the overhead of interaction between driver and hardware. This increases the overall performance and context switches between various hardware engines on the GPU.

With these changes, the speed of the NVIDIA Ada Lovelace architecture NVOFA is improved ~2x compared to the NVIDIA Ampere architecture NVOFA.

Joxeon · 1 Jan 2024 at 09:01

muon said:
It wouldn't have worked on other cards anyway. It's like how they eventually turned on RT for Pascal cards and it was awful.

I think the fact that they never enabled FG for 2000/3000 like they did RT for pascal shows it would have worked perfectly fine.

Take away FG and only 2 ADA cards are faster than ampere and only 1 by over 15% which is why most of Nvidia’s slides shown performance with FG enabled.

chris85oc · 1 Jan 2024 at 09:08

Joxeon said:
I think the fact that they never enabled FG for 2000/3000 like they did RT for pascal shows it would have worked perfectly fine.

Wrong. You’d have FSR FG just as good as DLSSFG then. Which is not the case. Just like with every amd vs nv feature, amd’s is the ‘we have x at home’ one. DLSSFG is would have been either a) not generating enough frames to matter because of the overhead b) ridden with artifacting to make it fast enough to work on 2x/3x series. Even if you’d have a separate version for 2x/3x ( which might happen in the future ), the feature would have its name stained by the lower common denominator with the worse quality.

Nvidia’s features are built to certain standards. And that’s why most people continue to buy Nv and completely ignore Amd. The numbers don’t lie. Wish things were different but, maybe intel can shake things up a bit?

Dicehunter · 1 Jan 2024 at 09:11

Joxeon said:
I think the fact that they never enabled FG for 2000/3000 like they did RT for pascal shows it would have worked perfectly fine.

Nvidia's FG is hardware based, An Nvidia engineer near the beginning of Ada's launch explained it pretty well that the OFA, Optical flow accelerator, Inside Ada is up to 4-5 times faster than the OFA inside Ampere and even with the 4-5 time increase in performance latency is still increased.

Now get that same latency on Ada and multiply it by 5 and you would have your result on Ampere, It would not be a nice experience, I'd rather use AMD's FG over that.