Support NeoGAF

[llien]

Nice review. A good example of time at CES well spent.

Not Linus' levels of "I was the first to play on x090" impressive, but pretty good.

Does this confirm that FSR 4.0 that appears to be much better than 3.1 is actually PSSR and DLSS5 mix?

Does it use neural intraconstruction radionics? Any hint of transformers, or optimus himself?

No neural thingies no buy, AMD!

 

[Wolzard]

Off topic, what AI program do you use to summarize?

This extension is interesting because you can use ChatGPT, Claude, Mistral, Gemini, with just one button.

YouTube Summary with ChatGPT & Claude - Chrome Web Store

Summarize YouTube videos, web articles, and PDFs to save time, powered by ChatGPT, Claude, Mistral AI, and Gemini.

chromewebstore.google.com

 

[Kenpachii]

It's looking pretty good!



Here's an AI summary of the vid:

Overview of AMD's FSR 4 Technology

- The video begins with a demonstration of AMD's FSR 4 upscaling technology, which was not prominently featured in their presentation.
- Two prototype RX 970 XT monitors are used to showcase the performance of FSR 3.1 and FSR 4 in the game Ratchet and Clank: Rift Apart at 4K resolution.
- The focus is on comparing the visual quality between FSR 3.1 and FSR 4, particularly in performance mode.

Visual Quality Comparisons

- Initial observations highlight significant improvements in image quality with FSR 4 compared to FSR 3.1, especially in handling transparent effects and particle details.
- The FSR 3.1 performance mode exhibits blurred and garbled visuals in areas with transparency, whereas FSR 4 presents clearer definitions of individual particles.
- The video illustrates how FSR 4 resolves issues related to garbling in effects such as fire and particle movements, leading to a smoother visual experience.

Performance Mode Insights

- The discussion emphasizes that FSR 3.1 struggles in performance mode, resulting in lower quality visuals, particularly in fast-moving or complex scenes.
- In contrast, FSR 4 demonstrates a marked improvement, providing a cleaner and more defined appearance even when upscaling from 1080p to 4K.
- The presenter notes that while FSR 4 shows promise, it is still not without limitations, particularly in maintaining high detail across all elements.

Detailed Observations of Specific Effects

- The video showcases specific examples of improvements in particle effects, noting that FSR 4 handles confetti and other small details with less ghosting and streaking.
- The analysis includes a comparison of how fur details are rendered, revealing that FSR 4 reduces pixelation artifacts significantly compared to FSR 3.1.
- The presenter highlights the overall stability of images with FSR 4, especially in scenes where fine textures are critical, such as carpet details in the game.

General Impressions and Future Considerations

- Overall, the presenter expresses a positive impression of FSR 4, noting it as a substantial upgrade over its predecessor, particularly in visual fidelity.
- Despite some remaining issues, the technology shows potential for enhancing gameplay experiences in visually demanding titles.
- The video concludes with a mention of the broader implications of FSR 4 for future gaming, suggesting further exploration and testing will be necessary to fully assess its capabilities.

Looks actually really good.

 

[Sanepar]

479 dollars, better than 4080 super, better than 5070, more vram, almost on par with 5070 ti in raster and now with good upscalling.
I would go with that if I was buying a new gpu. Ray tracing 10-20% more perf doesn't worth $300 more.

 

[Lootlord]

479 dollars, better than 4080 super, better than 5070, more vram, almost on par with 5070 ti in raster and now with good upscalling.
I would go with that if I was buying a new gpu. Ray tracing 10-20% more perf doesn't worth $300 more.

BUUUT where is muh 4x FrameGen?!

 

[Kazekage1981]

I cant wait for RDNA 4 APU's with FSR 4.0!

 

[Gaiff]

479 dollars, better than 4080 super, better than 5070, more vram, almost on par with 5070 ti in raster and now with good upscalling.
I would go with that if I was buying a new gpu. Ray tracing 10-20% more perf doesn't worth $300 more.

$480 for this? I find this unlikely. Something closer to $600 sounds more plausible.

 

[kevboard]

Ray tracing 10-20% more perf doesn't worth $300 more.

at a time where raytracing is slowly starting to become the default, that's a very shortsighted thing to say

 

[Zathalus]

$480 for the 9070 XT is highly optimistic if the 4080 performance claims are true. It would be great but the last time AMD did a performance undercut of Nvidia like that was ages ago.

 

[winjer]

$480 for the 9070 XT is highly optimistic if the 4080 performance claims are true. It would be great but the last time AMD did a performance undercut of Nvidia like that was ages ago.

I agree with you. But when AMD has a market share under 10%, doing something like this would be a good way to start improving things.

 

[lachesis]

$480 for the 9070 XT is highly optimistic if the 4080 performance claims are true. It would be great but the last time AMD did a performance undercut of Nvidia like that was ages ago.

But AMD do have to fight the fact that 5070 is $550, and the power of Nvidia's exaggerated marketing word of 5070 performing similar to 4090, which can be very significant for folks who are in the market for mid-range GPU.

If it's really competing 4070S - 4070Ti (which may be little bit weaker than 5070, or around the ball park) - then 5070 sets the pricing of $550, and AMD has to undercut that price in order to gain any traction... so my prediction would fall to $499 for XT, $459 for non-XT maybe.

 

[llien]

But AMD do have to fight the fact that 5070 is $550, and the power of Nvidia's exaggerated marketing word of 5070 performing similar to 4090, which can be very significant for folks who are in the market for mid-range GPU.

You can't fight unhinged marketing bazinga with price cuts.

 

[Crayon]

You can't fight unhinged marketing bazinga with price cuts.

This. That 4070 = 4090 line went over REALLY WELL. My theory on amd cancelling was they had to figure out how to answer that. Or at least give some time for the truth to spread a bit.

 

[Zuzu]

DF’s video:



AI summary:

CES Expectations and AMD's Presentation

- The discussion begins with the presenters, Alex Battalia and his colleague, expressing their high expectations for AMD at CES, anticipating the unveiling of new RDNA 4 GPUs and FSR 4 technologies.
- Despite some journalists being briefed on these products, AMD did not showcase them in a formal presentation due to alleged timing issues.
- Instead, AMD revealed some technologies in a less conventional manner that surprised both presenters.

Visit to the AMD Booth

- The presenters recount their experience at the AMD booth, which was actually a suite in a hotel rather than a traditional booth setup.
- They entered with low expectations as AMD had not provided much information about the RDNA 4 GPUs or FSR 4 technologies prior to the event.
- Upon arrival, they found various displays, including a Black Ops stand, but lacked context about the technologies being showcased.
- During their exploration, they encountered familiar AMD representatives who introduced them to a research project involving FSR 4 technology.

FSR 4 Research Project Overview

- The research project featured two PCs, one running FSR 3.1 and the other presumably showcasing a work-in-progress version of FSR 4, which utilized machine learning-based upscaling.
- They observed gameplay from Ratchet and Clank, comparing the performance of both versions in real-time.
- The presenters noted that the right PC was equipped with an engineering sample of the latest RDNA 4 GPU, which had not been publicly announced.

Image Quality Comparisons

- The presenters conducted a detailed comparison of image quality between FSR 3.1 and the new research version, noting significant improvements in various aspects.
- They highlighted that the research project demonstrated better image quality than FSR 3.1, particularly in handling fine textures like carpet patterns without introducing moiré artifacts.
- The improved image quality was especially noticeable in background objects and effects, with fewer flickering issues compared to FSR 3.1.

Technical Observations and Artifacts

- The presenters noted that while both versions exhibited some artifacts due to the offscreen capture method, the research project appeared cleaner and more stable in person.
- They also commented on the overall sharpness of the image produced by the research project, which was not soft and maintained a high level of detail.
- Specific improvements were noted in particle effects and animation fluidity, particularly concerning confetti and background animations, which had previously suffered from issues in earlier FSR versions.

Conclusions and Future Implications

- The presenters concluded that the new upscaling solution from AMD, although still a research project, showed promising advancements in image quality compared to previous iterations.
- They speculated about the potential relationship between AMD's new upscaler and Sony's PSR technology, noting that while both involve machine learning, they did not observe direct similarities in performance.
- The discussion ended with a light-hearted note about their experience at CES and the anticipation of more announcements and videos related to the event.

 

[SolidQ]

From PCWorld Q&A (link)

Maybe to talk a little bit about FSR specifically — FSR4 is ML super resolution, and it is built for… as we bring it to market, it will be built for our RDNA 4 architecture. RDNA 4 will bring a pretty massive increase in terms of ML [operations] and compute capability in the shader unit itself. So it is kind of fine-tuned for RDNA 4.

Bringing that to other product families is certainly a possibility for the future, but not something we’re talking about right now, nor committing to a timeline of when that will be available. But as we launch it, it’ll be RDNA 4-focused.

Translation: RDNA3 will get FSR4 but later