Support NeoGAF

[DeepEnigma]

I thought DLSS 2 was horrible. Am I wrong here?

It was a big step up over 1.0, but suffered many of the issues people would rake PSSR over the coals with. Yet, some are going to pretend it did not suffer those very same flaws.

Let's ask AI:

While DLSS 2.0 significantly improved over its predecessor, some users find it imperfect, citing issues like ghosting/smearing in fast motion, image softness, and artifacts, especially at lower quality presets or in poorly optimized games, though it offers huge FPS gains and can even look better than native TAA in many titles. Its effectiveness varies by game and settings (Quality vs. Balanced), with newer DLSS versions and specific game updates often improving quality, making it a valuable tool for performance but not always flawless.

Common criticisms and issues:

• Ghosting/Smearing: More noticeable in fast-paced games, creating artifacts around moving objects, say Reddit users.
• Softness/Blur: The upscaling can sometimes make the image look less sharp or softer than native resolution, sometimes requiring Nvidia's sharpening filter.
• Not Always Better Than Native: While often compared favorably to native TAA, it doesn't always beat native rendering in quality and can introduce its own set of visual quirks, as discussed on Tom's Hardware forums, notes this forum post.
• Game-Dependent: Performance and image quality differ significantly between games; some handle it beautifully (like Doom Eternal), while others struggle (like early Warzone), note users on Reddit.

Why some still find it good (or even great):

• Massive FPS Boosts: Delivers significant frame rate improvements, especially at higher resolutions, making demanding games playable.
• Improved Over Time: DLSS 2.0 was a huge leap from 1.0, and newer versions (like DLSS 3 with Frame Generation) and game updates (like MW2 v2.4.12) have addressed many flaws.
• Better Anti-Aliasing: Often provides good anti-aliasing, which can reduce shimmering seen in native TAA, according to users on Quora and Steam Community.
• Quality vs. Performance: Choosing the "Quality" preset often yields results very close to native, while "Balanced" or "Performance" modes introduce more noticeable upscaling artifacts, say users on Facebook.

 

[Bojji]

I thought DLSS 2 was horrible. Am I wrong here?

DLSS1 was horrible. DLSS2 was great in 2020:

I'm renting the game from gamefly right now and I can't tell any difference. Only problem that stands out to me is water. Looks like shit in both versions.

They did some updates few months ago for Avatar and Outlaws, so it probably is better now.

It was a big step up over 1.0, but suffered many of the issues people would rake PSSR over the coals with. Yet, some are going to pretend it did not suffer those very same flaws.

Let's ask AI:

While DLSS 2.0 significantly improved over its predecessor, some users find it imperfect, citing issues like ghosting/smearing in fast motion, image softness, and artifacts, especially at lower quality presets or in poorly optimized games, though it offers huge FPS gains and can even look better than native TAA in many titles. Its effectiveness varies by game and settings (Quality vs. Balanced), with newer DLSS versions and specific game updates often improving quality, making it a valuable tool for performance but not always flawless.

Common criticisms and issues:

• Ghosting/Smearing: More noticeable in fast-paced games, creating artifacts around moving objects, say Reddit users.
• Softness/Blur: The upscaling can sometimes make the image look less sharp or softer than native resolution, sometimes requiring Nvidia's sharpening filter.
• Not Always Better Than Native: While often compared favorably to native TAA, it doesn't always beat native rendering in quality and can introduce its own set of visual quirks, as discussed on Tom's Hardware forums, notes this forum post.
• Game-Dependent: Performance and image quality differ significantly between games; some handle it beautifully (like Doom Eternal), while others struggle (like early Warzone), note users on Reddit.

Why some still find it good (or even great):

• Massive FPS Boosts: Delivers significant frame rate improvements, especially at higher resolutions, making demanding games playable.
• Improved Over Time: DLSS 2.0 was a huge leap from 1.0, and newer versions (like DLSS 3 with Frame Generation) and game updates (like MW2 v2.4.12) have addressed many flaws.
• Better Anti-Aliasing: Often provides good anti-aliasing, which can reduce shimmering seen in native TAA, according to users on Quora and Steam Community.
• Quality vs. Performance: Choosing the "Quality" preset often yields results very close to native, while "Balanced" or "Performance" modes introduce more noticeable upscaling artifacts, say users on Facebook.

Most of those things happen with TAA as well, even native 4K + TAA.

 

[mckmas8808]

It was a big step up over 1.0, but suffered many of the issues people would rake PSSR over the coals with. Yet, some are going to pretend it did not suffer those very same flaws.

Let's ask AI:

While DLSS 2.0 significantly improved over its predecessor, some users find it imperfect, citing issues like ghosting/smearing in fast motion, image softness, and artifacts, especially at lower quality presets or in poorly optimized games, though it offers huge FPS gains and can even look better than native TAA in many titles. Its effectiveness varies by game and settings (Quality vs. Balanced), with newer DLSS versions and specific game updates often improving quality, making it a valuable tool for performance but not always flawless.

Common criticisms and issues:

• Ghosting/Smearing: More noticeable in fast-paced games, creating artifacts around moving objects, say Reddit users.
• Softness/Blur: The upscaling can sometimes make the image look less sharp or softer than native resolution, sometimes requiring Nvidia's sharpening filter.
• Not Always Better Than Native: While often compared favorably to native TAA, it doesn't always beat native rendering in quality and can introduce its own set of visual quirks, as discussed on Tom's Hardware forums, notes this forum post.
• Game-Dependent: Performance and image quality differ significantly between games; some handle it beautifully (like Doom Eternal), while others struggle (like early Warzone), note users on Reddit.

Why some still find it good (or even great):

• Massive FPS Boosts: Delivers significant frame rate improvements, especially at higher resolutions, making demanding games playable.
• Improved Over Time: DLSS 2.0 was a huge leap from 1.0, and newer versions (like DLSS 3 with Frame Generation) and game updates (like MW2 v2.4.12) have addressed many flaws.
• Better Anti-Aliasing: Often provides good anti-aliasing, which can reduce shimmering seen in native TAA, according to users on Quora and Steam Community.
• Quality vs. Performance: Choosing the "Quality" preset often yields results very close to native, while "Balanced" or "Performance" modes introduce more noticeable upscaling artifacts, say users on Facebook.

That's very similar to PSSR today LOL! So, why is Bojji tripping? And if PSSR 2 comes out this year, that's a pretty good pace of improvement. Especially considering it's been less than 18 months since PSSR came out. PSSR2 coming out in less than 2 years would be AMAZING (If....big IF......it is similar to a full fat version of FSR4)!

 

[Bojji]

That's very similar to PSSR today LOL! So, why is Bojji tripping? And if PSSR 2 comes out this year, that's a pretty good pace of improvement. Especially considering it's been less than 18 months since PSSR came out. PSSR2 coming out in less than 2 years would be AMAZING (If....big IF......it is similar to a full fat version of FSR4)!

PSSR 1.0 is worse in most aspects than DLSS 2.0.

 

[DeepEnigma]

PSSR 1.0 is worse in most aspects than DLSS 2.0.

No. It suffered the very same fate of lower than a certain resolution threshold upscaling.

Like I said, PSSR 1.0 is pretty decent for their first foray.

 

[Bojji]

No.

Yes.

 

[yogaflame]

If PSSR 2.0 can fix the issues with shimmering and artifacts with games using Ue5, that will be a win.

 

[mckmas8808]

PSSR 1.0 is worse in most aspects than DLSS 2.0.

If you remove the UE5 games, would you still be saying this?

 

[Clear]

Yep. We literally have the plethora of threads still on here.

PS5 Pro and PSSR specifically has been the subject of some the most blatant gate-keeping I can ever recall seeing.

Its pretty comical.

 

[Topher]

PSSR 1.0 is worse in most aspects than DLSS 3.0.

ftfy

 

[Senua]

PSSR (PlayStation Spectral Super Resolution) on the PS5 Pro is Sony's AI upscaler, generally softer and less stable in motion than Nvidia's DLSS 2
, often showing shimmering or fizzling, especially on moving geometry, whereas DLSS 2 provides sharper images, better anti-aliasing for moving objects, and superior temporal stability, though DLSS can struggle with certain ray-traced reflections (like checkerboarding) that PSSR handles better in specific instances. PSSR is considered a promising V1.0 tech that improves over time, while DLSS is more mature, but PSSR can offer better edge gradients on rounded surfaces and has unique strengths in handling ray tracing reflections compared to DLSS, despite its overall softer look and motion artifacts.

PSSR (PlayStation Spectral Super Resolution)

• Pros: Can offer better edge gradients on rounded surfaces, unique stability with some ray-traced reflections (like checkerboarding), and shows potential for future improvement as a V1.0 tech.
• Cons: Tends to be softer than DLSS, exhibits more shimmering/fizzling on moving objects, and has less effective anti-aliasing for rapidly moving elements.
• Best For: Specific ray-traced scenarios where it outperforms DLSS, with potential to mature beyond initial limitations.

 

[Bojji]

If you remove the UE5 games, would you still be saying this?

A lot of games use UE5.

Many UE5 games avoided using PSSR because it sucks ass in that engine. Plus you have some Ubisoft games showing the same issues, Dragons Dogma 2, Jedi Survivor (UE4) and few others.

 

[DeepEnigma]

PS5 Pro and PSSR specifically has been the subject of some the most blatant gate-keeping I can ever recall seeing.

Its pretty comical.

 

[Karim]

I guess it's a matter of expectations. For that price tag, I expected much more and I have to say I'm very disapointed in that machine. PSSR is broken in half the games I've played and we're still not getting the promised fidelity + framerate in one mode. (Base PS5 fidelity mode is often higher res than the pro performance version). We basically get a resolution boost with the pro, and that's pretty much it.
Some games definitely look better but they do not represent the majority imo.

Broken in half games? Where it coming this nonsense statics? And broken for the fuck sake with FSR3 on base hardware? Why everytime this stupid bullshit? Most of the games with "broken" PSSR coming from minor studios using UE5 from my personal experience. I understood the disappointment and I don't have the presumptiom to change the other feeling but no need to be overdramatic and irrational. Most of the times PSSR works decently. Not incredibly but did it's job. It's not even true ps5 pro offer just a resolution boost, I found more higher setting on ps5 pro games enhanced than on ps4 pro but eh.

 

[Bojji]

PSSR (PlayStation Spectral Super Resolution) on the PS5 Pro is Sony's AI upscaler, generally softer and less stable in motion than Nvidia's DLSS 2
, often showing shimmering or fizzling, especially on moving geometry, whereas DLSS 2 provides sharper images, better anti-aliasing for moving objects, and superior temporal stability, though DLSS can struggle with certain ray-traced reflections (like checkerboarding) that PSSR handles better in specific instances. PSSR is considered a promising V1.0 tech that improves over time, while DLSS is more mature, but PSSR can offer better edge gradients on rounded surfaces and has unique strengths in handling ray tracing reflections compared to DLSS, despite its overall softer look and motion artifacts.

PSSR (PlayStation Spectral Super Resolution)

• Pros: Can offer better edge gradients on rounded surfaces, unique stability with some ray-traced reflections (like checkerboarding), and shows potential for future improvement as a V1.0 tech.
• Cons: Tends to be softer than DLSS, exhibits more shimmering/fizzling on moving objects, and has less effective anti-aliasing for rapidly moving elements.
• Best For: Specific ray-traced scenarios where it outperforms DLSS, with potential to mature beyond initial limitations.

And PSSR was only better in Ratchet with reflections. It sucks in most other games using Ray Tracing (or software RT).

 

[DeepEnigma]

PSSR (PlayStation Spectral Super Resolution) on the PS5 Pro is Sony's AI upscaler, generally softer and less stable in motion than Nvidia's DLSS 2
, often showing shimmering or fizzling, especially on moving geometry, whereas DLSS 2 provides sharper images, better anti-aliasing for moving objects, and superior temporal stability, though DLSS can struggle with certain ray-traced reflections (like checkerboarding) that PSSR handles better in specific instances. PSSR is considered a promising V1.0 tech that improves over time, while DLSS is more mature, but PSSR can offer better edge gradients on rounded surfaces and has unique strengths in handling ray tracing reflections compared to DLSS, despite its overall softer look and motion artifacts.

PSSR (PlayStation Spectral Super Resolution)

• Pros: Can offer better edge gradients on rounded surfaces, unique stability with some ray-traced reflections (like checkerboarding), and shows potential for future improvement as a V1.0 tech.
• Cons: Tends to be softer than DLSS, exhibits more shimmering/fizzling on moving objects, and has less effective anti-aliasing for rapidly moving elements.
• Best For: Specific ray-traced scenarios where it outperforms DLSS, with potential to mature beyond initial limitations.

Wow, AI agrees about the potential due to the decent 1.0 outing.

And DLSS did have a half step overhaul with 2.5, which is important to differentiate between 2.0 and 2.5.1.

And PSSR was only better in Ratchet with reflections. It sucks in most other games using Ray Tracing (or software RT).

Works great in GT7 with RT reflections, shadows and lighting.

 

[Bojji]

Wow, AI agrees about the potential due to the decent 1.0 outing.

And DLSS did have a half step overhaul with 2.5, which is important to differentiate between 2.0 and 2.5.1.


Works great in GT7 with RT reflections, shadows and lighting.

GT7 isn't exacly issue free with PSSR:



Timestamped sections shows boiling lighting in cockpit. Is it PSSR issue or just low ray count? Who knows...

 

[Topher]

lol.....I asked Gemini to compare PSSR to DLSS 2 and it gave me a summary of differences, but every source that it used referenced DLSS 3, not DLSS 2.

Gemini is saying the upscaler of DLSS 3 is the same as 2.

 

[Senua]

lol.....I asked Gemini to compare PSSR to DLSS 2 and it gave me a summary of differences, but every source that it used referenced DLSS 3, not DLSS 2.

Wasn't DLSS3 just framegen ontop of 2? Maybe there were some improvements I don't remember

 

[DeepEnigma]

GT7 isn't exacly issue free with PSSR:



Timestamped sections shows boiling lighting in cockpit. Is it PSSR issue or just low ray count? Who knows...

So you share DF for everything, but find a non-DF video and not theirs for GT7 with the Pro update? Cherry pick your sources now, eh?

I never said it wasn't without issue, none of them are, it works very well, especially for a 1.0.

 

[FUBARx89]

It was a big step up over 1.0, but suffered many of the issues people would rake PSSR over the coals with. Yet, some are going to pretend it did not suffer those very same flaws.

Let's ask AI:

While DLSS 2.0 significantly improved over its predecessor, some users find it imperfect, citing issues like ghosting/smearing in fast motion, image softness, and artifacts, especially at lower quality presets or in poorly optimized games, though it offers huge FPS gains and can even look better than native TAA in many titles. Its effectiveness varies by game and settings (Quality vs. Balanced), with newer DLSS versions and specific game updates often improving quality, making it a valuable tool for performance but not always flawless.

Common criticisms and issues:

• Ghosting/Smearing: More noticeable in fast-paced games, creating artifacts around moving objects, say Reddit users.
• Softness/Blur: The upscaling can sometimes make the image look less sharp or softer than native resolution, sometimes requiring Nvidia's sharpening filter..

Is this right? I fucking hate ghosting. FSR 2 and Unreal engine upscaler ghosting is hideous on console and the PC master race has acted like it's never been a problem on PC.

Probably the best thing about PSSR right now. Getting rid of ghosting.

The only reason I want a Pro patch for Cyberpunk 2077 is to get rid of the disgusting ghosting in it.

 

[DeepEnigma]

lol.....I asked Gemini to compare PSSR to DLSS 2 and it gave me a summary of differences, but every source that it used referenced DLSS 3, not DLSS 2.

Gemini is saying the upscaler of DLSS 3 is the same as 2.

Which is why it's important to distinguish 2.0 release. Not 2.5.1 that fixed a lot of issues. But 2.0 itself.

A lot of comparisons will be like "DLSS 2" but are using 2.5.1.

 

[Bojji]

lol.....I asked Gemini to compare PSSR to DLSS 2 and it gave me a summary of differences, but every source that it used referenced DLSS 3, not DLSS 2.

Gemini is saying the upscaler of DLSS 3 is the same as 2.

That's why when PSSR launched we were in late DLSS3 version already (4.0 launched just two months later).

So you share DF for everything, but find a non-DF video and not theirs for GT7 with the Pro update? Cherry pick your sources now, eh?

I never said it wasn't without issue, none of them are, it works very well, especially for a 1.0.

From DF (timestamped):

 

[DeepEnigma]

That's why when PSSR launched we were in late DLSS3 version already (4.0 launched just two months later).



From DF (timestamped):

Looks fantastic for 1.0.

You had that timestamp quick AF like a catalogue of saved links or spreadsheets

 

[Senua]

Is this right? I fucking hate ghosting. FSR 2 and Unreal engine upscaler ghosting is hideous on console and the PC master race has acted like it's never been a problem on PC.

Probably the best thing about PSSR right now. Getting rid of ghosting.

The only reason I want a Pro patch for Cyberpunk 2077 is to get rid of the disgusting ghosting in it.

Any of these TAA solutions will have ghosting, it's just a matter of how much it can be minimised. AI upscalers are doing a great job and will only get better as time goes on. DLSS is the best we have by some margin, but none of them are perfect.

 

[Bojji]

Looks fantastic for 1.0.

PSSR off produces better results in this game (more stable, less visual issues), DLSS2 would also look better.

 

[DeepEnigma]

PSSR off produces better results in this game (more stable, less visual issues), DLSS2 would also look better.

Looks fantastic for 1.0.

 

[winjer]

Remember that temporal upscalers, including PSSR and DLSS, depend on frame rate as well as base resolution, for it's quality.
Digital Foundry does not normalize FPS when they do their comparisons. This means that PSSR running at 30 fps will be at a significant disadvantage if compared to DLSS on PC, running at 100+ FPS. Even if the base resolution is the same.
I'm not saying DLSS 3 and especially 4, are not better. Just that PSSR has a handicap inherent to consoles.

 

[Topher]

Wasn't DLSS3 just framegen ontop of 2? Maybe there were some improvements I don't remember

Yeah, I just edited. That's what Gemini is saying after I asked why it used articles referencing DLSS 3. Pretty much said the upscaler of 3 is the same as 2.

Which is why it's important to distinguish 2.0 release. Not 2.5.1 that fixed a lot of issues. But 2.0 itself.

A lot of comparisons will be like "DLSS 2" but are using 2.5.1.

Here's how Gemini explains it:

---

NVIDIA updates their DLSS upscaler versioning independently of the "DLSS 3" marketing name. For example, a game might support "DLSS 3.7," but 3.7 is just the latest version of the DLSS 2 upscaler.

The Bottom Line:Even without the "Frame Gen" magic of DLSS 3, the core DLSS 2 upscaling technology remains more refined and stable than PSSR's first iteration. PSSR is effectively at the "DLSS 1.9" stage—it's way better than standard spatial upscalers (like FSR), but it still needs more "training" time to match NVIDIA's current level of polish.

---

To be fair, the only comparisons I ever did between PSSR and DLSS on my own were with DLSS 3.x which is why I've been adamant that PSSR wasn't far behind 3.x.

 

[Felessan]

DLSS improvements were made for models that already were good - best in the business even. PSSR is worst ML upscaler.

It doesn't matter
You argue that Cerny improve PSSR not because it can be better, but because it's bad (your opinion).
Same logic should be applied to DLSS. If new better version released - it's because previous version was shit. No excuses.

Or you are riding double standards - Nvidia do it for improvement and Cerny because he fucked up, which is purely your judgement and you can't prove this point besides repeating "It's so because I'm telling you so"

 

[Bojji]

Remember that temporal upscalers, including PSSR and DLSS, depend on frame rate as well as base resolution, for it's quality.
Digital Foundry does not normalize FPS when they do their comparisons. This means that PSSR running at 30 fps will be at a significant disadvantage if compared to DLSS on PC, running at 100+ FPS. Even if the base resolution is the same.
I'm not saying DLSS 3 and especially 4, are not better. Just that PSSR has a handicap inherent to consoles.

Most PSSR games are 60fps. There are very few 30fps locked releases this gen.

It doesn't matter
You argue that Cerny improve PSSR not because it can be better, but because it's bad (your opinion).
Same logic should be applied to DLSS. If new better version released - it's because previous version was shit. No excuses.

Or you are riding double standards - Nvidia do it for improvement and Cerny because he fucked up, which is purely your judgement and you can't prove this point besides repeating "It's so because I'm telling you so"

Just sucking ass in the most popular engine should be proof to you that something is wrong with it. There are tons of examples where PSSR fails, it's not just my opinion.

Yeah, I just edited. That's what Gemini is saying after I asked why it used articles referencing DLSS 3. Pretty much said the upscaler of 3 is the same as 2.



Here's how Gemini explains it:

---

NVIDIA updates their DLSS upscaler versioning independently of the "DLSS 3" marketing name. For example, a game might support "DLSS 3.7," but 3.7 is just the latest version of the DLSS 2 upscaler.

The Bottom Line:Even without the "Frame Gen" magic of DLSS 3, the core DLSS 2 upscaling technology remains more refined and stable than PSSR's first iteration. PSSR is effectively at the "DLSS 1.9" stage—it's way better than standard spatial upscalers (like FSR), but it still needs more "training" time to match NVIDIA's current level of polish.

---

To be fair, the only comparisons I ever did between PSSR and DLSS on my own were with DLSS 3.x which is why I've been adamant that PSSR wasn't far behind 3.x.

Improvements between 2.0 and 3.0 for super resolution weren't big. Biggest jump was with 3.7 version and introduction of E preset.

 

[MALpractice311]

All I'm going to say is, let's hope this is true. I'm not certain it would convince me to shell out almost $800 for the Pro with a drive, but it'll effectively be beta testing for the PS6.

 

[DeepEnigma]

Yeah, I just edited. That's what Gemini is saying after I asked why it used articles referencing DLSS 3. Pretty much said the upscaler of 3 is the same as 2.



Here's how Gemini explains it:

---

NVIDIA updates their DLSS upscaler versioning independently of the "DLSS 3" marketing name. For example, a game might support "DLSS 3.7," but 3.7 is just the latest version of the DLSS 2 upscaler.

The Bottom Line:Even without the "Frame Gen" magic of DLSS 3, the core DLSS 2 upscaling technology remains more refined and stable than PSSR's first iteration. PSSR is effectively at the "DLSS 1.9" stage—it's way better than standard spatial upscalers (like FSR), but it still needs more "training" time to match NVIDIA's current level of polish.

---

To be fair, the only comparisons I ever did between PSSR and DLSS on my own were with DLSS 3.x which is why I've been adamant that PSSR wasn't far behind 3.x.

Right. DLSS 3 is not based on 2.0 though, but 2.5.1+ which fixed a lot over initial release.

Per AI:

---

DLSS 2.0 was a major leap in AI upscaling, but DLSS 2.5.1, released later, significantly refined image quality by improving fine detail stability, reducing ghosting, and providing better edge smoothness, often by disabling problematic in-game sharpening filters, making 2.5.1 generally superior and a favored manual update for many users seeking cleaner visuals and better anti-aliasing than earlier DLSS 2 versions.

---

 

[winjer]

Most PSSR games are 60fps. There are very few 30fps locked releases this gen.

But the base resolution is lowered.
And even then, at 60 FPS, there is les frames to accumulate per second, than at 100+ FPS.

 

[Bojji]

But the base resolution is lowered.

DF tested games with the same internal res (they even adjusted DLSS to use 864p for AW2).

 

[DeepEnigma]

It doesn't matter
You argue that Cerny improve PSSR not because it can be better, but because it's bad (your opinion).
Same logic should be applied to DLSS. If new better version released - it's because previous version was shit. No excuses.

Or you are riding double standards - Nvidia do it for improvement and Cerny because he fucked up, which is purely your judgement and you can't prove this point besides repeating "It's so because I'm telling you so"

Gatekeeping as Clear mentioned.

Can we just go back the more simple console war times? Scaler wars are next level nerd cringe we're getting sucked into.

I'd even take grassgate and puddlegate over this.

 

[winjer]

DF tested games with the same internal res (they even adjusted DLSS to use 864p for AW2).

But they never normalize for frame rate.
It's one of the flaws in their methodology, when comparing upscalers.

 

[Bojji]

Gatekeeping as Clear mentioned.

Can we just go back the more simple console war times? Scaler wars are next level nerd cringe we're getting sucked into.

I'd even take grassgate and puddlegate over this.

In the end I want PSS2 to be as good as possible, and I said this everywhere so far.

I just don't get why people are so defensive about PSSR1, tech that showed issues from day 1. Why defend it when it will be forgotten by history in the next few months?

 

[Felessan]

Just sucking ass in the most popular engine should be proof to you that something is wrong with it. There are tons of examples where PSSR fails, it's not just my opinion.

You are just keep bitching over nothing. And find any number of bullshit excuses to continue to push your agenda.
Yes, PSSR is not perfect, and what? It just a first version, some issues bound to happen, and it ain't completely failure, issues often non-noticeable.

 

[DeepEnigma]

In the end I want PSS2 to be as good as possible, and I said this everywhere so far.

I just don't get why people are so defensive about PSSR1, tech that showed issues from day 1. Why defend it when it will be forgotten by history in the next few months?

People are pointing out the nitpick hypocrisy and hyperbole, nothing more. We all want it to be better, it makes it better for all of us, PC and consoles.

 

[Topher]

I just don't get why people are so defensive about PSSR1, tech that showed issues from day 1. Why defend it when it will be forgotten by history in the next few months?

That's works both ways. Why are you so aggressively attacking PSSR in every thread that it is the topic? It makes a lot more sense for people to defend something they bought and see value in than someone attacking something where they have no skin in the game.

 

[Senua]

In the end I want PSS2 to be as good as possible, and I said this everywhere so far.

I just don't get why people are so defensive about PSSR1, tech that showed issues from day 1. Why defend it when it will be forgotten by history in the next few months?

Because it's the solution the Pro has atm and people get defensive over things they've bought and enjoyed with their hard earned money. It's human nature.

Not me though, my 10gb vram 3080 can suck a dick

 

[Bojji]

You are just keep bitching over nothing. And find any number of bullshit excuses to continue to push your agenda.
Yes, PSSR is not perfect, and what? It just a first version, some issues bound to happen, and it ain't completely failure, issues often non-noticeable.

Non noticeable issues (timestamped):

People are pointing out the nitpick hypocrisy and hyperbole, nothing more. We all want it to be better, it makes it better for all of us, PC and consoles.

It's not nitpicking when issues are in your face.

PSS2 will deliver good results, I have no doubts about it. But what is sad is that most of games with PSS1 issues will still have them without developers interventions.

That's works both ways. Why are you so aggressively attacking PSSR in every thread that it is the topic? It makes a lot more sense for people to defend something they bought and see value in than someone attacking something where they have no skin in the game.

But I have PS5 Pro and want it to be as good as possible. Me (and few other Pro owners) are called haters for simly wanting things to improve. Fanboys are happy no matter what shit their favorite company will serve them.

 

[mckmas8808]

Wow, AI agrees about the potential due to the decent 1.0 outing.

And DLSS did have a half step overhaul with 2.5, which is important to differentiate between 2.0 and 2.5.1.


Works great in GT7 with RT reflections, shadows and lighting.

I think he's confusing DLSS 2.0 versus 2.5

 

[DeepEnigma]

But what is sad is that most of games with PSS1 issues will still have them without developers interventions.

Just like going from DLSS1.0 to 2.0

And most other games included a .dll file which will more than likely be the case here as well but on a firmware/driver level.

---

Yes, a few games that initially launched with DLSS 1.0 were later updated by their developers to support the significantly improved DLSS 2.0.
DLSS 1.0 required per-game training, which was a barrier to implementation and resulted in variable image quality. DLSS 2.0 introduced a generalized AI network, making it much easier for developers to implement and update their games with consistently better results.

Games Updated to DLSS 2.0
Notable examples of games that received patches from the developers to upgrade from DLSS 1.0 to 2.0 include:

• Control: Received a patch on March 26, 2020, to add DLSS 2.0 support, which improved image quality and handled motion artifacts more effectively than the original implementation.
• Shadow of the Tomb Raider: The developers updated the game to DLSS 2.0 support, which significantly improved the visual fidelity and performance over the initial, often blurry, DLSS 1.0 version.
• Wolfenstein: Youngblood and Deliver Us The Moon: These were among the first games to launch with or be quickly updated to DLSS 2.0 around its release.

Developer Action Was Required
It's important to note that these updates were not automatic for all DLSS 1.0 games; the developers had to integrate the new DLSS 2.0 SDK into their games. Some games like Metro Exodus and Battlefield V never received an official developer update to DLSS 2.0.

---

 

[squidilix]

I mean, PS5 Pro with PSSR can achieve 100fps on some scene at Gran Turismo 7 with RT and VRR.

Meanwhile Nvidia-tard needed 4x FrameGen to achieves a fake 140fps because RT is fking heavy? Come on....

 

[Bojji]

Just like going from DLSS1.0 to 2.0

And most other games included a .dll file which will more than likely be the case here as well.

DLSS1 had excuse for being first of its kind. PSSR launched 4 years after DLSS2 and 2 years after XeSS.

 

[Senua]

I mean, PS5 Pro with PSSR can achieve 100fps on some scene at Gran Turismo 7 with RT and VRR.

Meanwhile Nvidia-tard needed 4x FrameGen to achieves a fake 140fps because RT is fking heavy? Come on....

 

[DeepEnigma]

DLSS1 had excuse for being first of its kind. PSSR launched 4 years after DLSS2 and 2 years after XeSS.

OMFG.

THEY HAD TO TRAIN FROM SOMEWHERE ON A PER GAME BASIS LIKE ALL OF THESE AI SCALERS STARTING OFF.

 

[Felessan]

Non noticeable issues (timestamped):

Cherry-picking at it's finest, not that I did not expect it from you, you can't really just relax
I don't play DF video with x8 magnification, I play games (on quality btw, so the issues with upscaling from really low res much less pronounced) - so I don't care about your bullshit