Support NeoGAF

[Karim]

DF praises Sony far, far more than any other developer. They were the ones praising GoY and DS2 graphics last year, when people dismissed GoY outright on this very forum. They are the ones that get exclusive interviews with Mark Cerny, who, in his own words, is a fan of them. Not only that they praise PSSR in a lot of titles. Pointing out the bad as well isn't a crime. Also, I'm not sure why you brought DF up? I certainly didn't mention them, nor are they are hardly the only one who have pointed out issues with PSSR noise in some titles.


"latency hiding 5ms run-time"? I'm not even sure what that means. 1500 TOPs? I can only assume you mean the 5090? Which DF doesn't usually use in PS5 Pro comparisons. The first batch of comparisons they did in 2024 was before the 5090 released, and the AW2 comparison you seem to have a problem with was done on a 4070, which has less TOPs than a PS5 Pro.


Some odd things here:

1. AW2 is by far not the worst offending game. Off the top of my head, Avatar, Star Wars Outlaws, Silent Hill 2, Silent Hill F, Rise of the Ronin, Metal Gear Solid 3 are all worse. Star Wars Outlaws and Avatar got patched and look decent now, although Outlaws could use some work.
2. AW1 XB1 exclusive roots? AW1 was a 360 game and on a completely separate engine. AW2 is based on the Northlight engine, which is one of only four that uses mesh/primitive shaders on current gen. The engine had roots on XB1, but it works really well on PS5, albeit it very demanding on both PS5/XSX. Control uses the same engine and it is categorically better on the Pro vs the XSX.
3. Any comparisons to PSSR has always been just with the regular DLSS versions and near as possible matched settings. Never with RR, unless it is a separate video showing off the benefits of RR, but PSSR never enters that discussion.


Demon Souls is a good example of PSSR. Its also not on PC so I'm not sure what a 5090 or MFG has got to do with anything.


Even Sony is replacing it with a newer version based off FSR4. They clearly understand it is lacking in certain respects vs other upscalers. That is not to say PSSR can't look good, it absolutely can, but it can be better. And it will be better. Unless you think Sony and Mark Cerny are liers?

DF are the same guys who claimed variable frequency cause bottlenecks to the gpu and the cpu when maximised and other bullshit not worthy even to mention eh but yeah lately their narrative has surely changed because you know without the xbox the only left which care a bit to push the console tech is only the playstation brand and not surely Nintendo, so they haven't much choice.

 

[Karim]

After the main issues became more evident, they should have acknowledged it and say they were working on those. It's ok to select a few best cases for release, but then the bad cases came and they still acted like everything was ok for too long.
They should showcase improvements made instead of forbidding devs to talk about PSSR versions. I don't understand their reasons, unless work on PSSR was already just maintenance until Amethyst/PSSR 2.0 was announced.

What you not understand? An AI upscaler can't surely work at his best the first period, it's quite expectable raw edges especially without any kinda of previous experience to put in use. They cant surely use the same method used by nvidia or AMD well not until the deal.

 

[omegasc]

DF are the same guys who claimed variable frequency cause bottlenecks to the gpu and the cpu when maximised and other bullshit not worthy even to mention eh but yeah lately their narrative has surely changed because you know without the xbox the only left which care a bit to push the console tech is only the playstation brand and not surely Nintendo, so they haven't much choice.

There's no narrative. They theorize based on what they know (even if it might have been MS that planted the seed), but change their minds when proven incorrect.

What you not understand? An AI upscaler can't surely work at his best the first period, it's quite expectable raw edges without any kinda of previous experience to use.

You say that as if they had no games to test before the Pro was released.
There's nothing wrong with issues in the first release. But why block devs from talking about versions? Would it harm it? If anything it would show their progress. Maybe even put a little pressure on devs to update their games with the best version?

 

[Karim]

There's no narrative. They theorize based on what they know (even if it might have been MS that planted the seed), but change their minds when proven incorrect.

Their theory were based to ignorance and net discourse eh. They are supposed to be expert.

 

[RSMan]

This reminds me — despite the Digital Foundry misinformation, RE4 Remake on PS5 Pro actually uses PSSR and has higher visual fidelity when High Frame Rate mode is enabled with ray tracing off.

It's a rare case where disabling High Frame Rate mode actually downgrades the visuals on the Pro.

Try it yourself.

 

[Karim]

This reminds me — despite the Digital Foundry misinformation, RE4 Remake on PS5 Pro actually uses PSSR and has higher visual fidelity when High Frame Rate mode is enabled with ray tracing off.

It's a rare case where disabling High Frame Rate mode actually downgrades the visuals on the Pro.

Try it yourself.

That's funny because the game with pssr run at 800p and I found the IQ infinitely better than their shitty cbr solution even selecting the quality mode. And it's the first pssr iteration for sure.

 

[omegasc]

Their theory were based to ignorance and net discourse eh. They are supposed to be expert.

I take them as people who do way more research than most, maybe not exactly experts. But yes, their theories can be flawed, either due to incorrect inference, old/bad sources, or new developments in the area.

This reminds me — despite the Digital Foundry misinformation, RE4 Remake on PS5 Pro actually uses PSSR and has higher visual fidelity when High Frame Rate mode is enabled with ray tracing off.

It's a rare case where disabling High Frame Rate mode actually downgrades the visuals on the Pro.

Try it yourself.

They did acknowledge that in a DF direct of something. I don't know if YT allows them to edit published videos. And to be fair that was kinda obvious to me when I played it, I don't know how they missed that.
In Ghost of Yotei they suspected PSSR was in use only in some modes, which the dev clarified it was used in all modes.
PSSR is really good in high resolutions. It's when it goes below 840p or something that it starts to fall apart.

 

[Zathalus]

DF are the same guys who claimed variable frequency cause bottlenecks to the gpu and the cpu when maximised eh but yeah lately their narrative has changed because you know without the xbox the only left with push a decent tech lead is just the playstation brand and not surely Nintendo, so they haven't much choice if they want to survive.

Wasn't the exact claim from them the following:

"Several developers speaking to Digital Foundry have stated that their current PS5 work sees them throttling back the CPU in order to ensure a sustained 2.23GHz clock on the graphics core."

Which they were pretty much correct about? PS5 average gaming frequency is 2170Mhz, not 2230Mhz as confirmed by Mark Cerny himself in the recent PS5 Pro talks he did. Hence, the "official" PS5 Pro clock is 2170Mhz, even though it can boost up to 2350Mhz.

 

[winjer]

Wasn't the exact claim from them the following:

"Several developers speaking to Digital Foundry have stated that their current PS5 work sees them throttling back the CPU in order to ensure a sustained 2.23GHz clock on the graphics core."

Which they were pretty much correct about? PS5 average gaming frequency is 2170Mhz, not 2230Mhz as confirmed by Mark Cerny himself in the recent PS5 Pro talks he did. Hence, the "official" PS5 Pro clock is 2170Mhz, even though it can boost up to 2350Mhz.

This is also why when some people modded their PS5 with a better cooler, performance in games increased by a small, but notable margin.

 

[Three]

They did acknowledge that in a DF direct of something. I don't know if YT allows them to edit published videos.

What? Of course youtube allows them to edit a video.

 

[Bojji]

Nobody ever stated it was, Although compared to DLSS3 in games that use both (you did not say 4.5) PSSR tends to fight ghosting artefacts in motion better (this was evident in screenshots of the videos DF did even though DF did not always call it out).

This is the release where they traded some performance too in order to improve IQ in motion and fight off against ghosting artefacts… who'd is what PSSR1 prioritised too so it is a bit fun if so hehe (as well as have variable Frame Gen which is a big deal for those that want to use Frame Gen).

You are probably talking about infamous preset D, the worst DLSS3 preset that introduced heavy ghosting to elemnents on the screen that didn't have motion vectorts. That where this infamous TLOU1 gif done by mazinger dude comes from, you can change it to preset C or update DLSS3 file and get preset E (they eliminated preset D in newer .dll files entirely):



With DLSS4 you get much sharper and stable image vs. DLSS3 but there were some regresses when it comes to ghosting (vs. preset E), with 4.5 you get even more sharpness and details but this time ghosting is eliminated.

At the same time it shows more artifacts in some games. And it kinda makes sense when you think about, DF speculated that PSSR artifacts comes from very low accumulation time, it doesn't have ghosting thanks to that but at the same time it can introduce fizzling in some games (and this is exactly what we saw). DLSS4.5 seems to reduce accumulation time as well in order to combat ghosting and it seems to show similar effects in some games.

 

[Cyborg]

Who could have imagined that this thread would be about shitting on DF.... (by some)

 

[omegasc]

What? Of course youtube allows them to edit a video.

AI just told me what can be done after publishing a video, and it's not much. Can't add new content, so not really useful in this case.

 

[Panajev2001a]

You are probably talking about infamous preset D, the worst DLSS3 preset that introduced heavy ghosting to elemnents on the screen that didn't have motion vectorts. That where this infamous TLOU1 gif done by mazinger dude comes from, you can change it to preset C or update DLSS3 file and get preset E (they eliminated preset D in newer .dll files entirely):



With DLSS4 you get much sharper and stable image vs. DLSS3 but there were some regresses when it comes to ghosting (vs. preset E), with 4.5 you get even more sharpness and details but this time ghosting is eliminated.

At the same time it shows more artifacts in some games. And it kinda makes sense when you think about, DF speculated that PSSR artifacts comes from very low accumulation time, it doesn't have ghosting thanks to that but at the same time it can introduce fizzling in some games (and this is exactly what we saw). DLSS4.5 seems to reduce accumulation time as well in order to combat ghosting and it seems to show similar effects in some games.

Considering how ghosting / temporal accumulation artifacts already annoy me in UE5 games I would trade fizzing for no ghosting in motion.

 

[Bojji]

Considering how ghosting / temporal accumulation artifacts already annoy me in UE5 games I would trade fizzing for no ghosting in motion.

And that's exactly what you can do by changing DLSS versions and presets in games.

I prefer some bit of ghosting but with stable details vs. shimmering in foliage and things like that. And ghosting you can see in (for example) DLSS4 is not close to natural ghosting TSR produces in UE5 games, it only looks bad compared to other versions of DLSS but still better than what TAA or TSR can do.

 

[Panajev2001a]

And that's exactly what you can do by changing DLSS versions and presets in games.

I prefer some bit of ghosting but with stable details vs. shimmering in foliage and things like that. And ghosting you can see in (for example) DLSS4 is not close to natural ghosting TSR produces in UE5 games, it only looks bad compared to other versions of DLSS but still better than what TAA or TSR can do.

I prefer zero added ghosting, too many games pay for unoptimised features with temporal accumulation artefact inducing stuff (including The Matrix demo by Epic), do not need to add anymore guesses.

On consoles, at the price you pay for a PS5 Pro you are going to be more limited and PSSR 1.x reached users in this more limited cheaper HW before FSR4… so I call it a win instead of trying to troll it in as many threads as possible by magnifying every possible issue to oblivion.

 

[Three]

Considering how ghosting / temporal accumulation artifacts already annoy me in UE5 games I would trade fizzing for no ghosting in motion.

One of the worst cases of this was when Bojji was calling PSSR "unsuitable" for Ratchet and clank because it had some fizzle, when the accumulation artifacts on DLSS removed the moving light visual cue for ammo crates and this went completely unnoticed by himself and DF:

Unsuitability of PSSR is very apparent..

As if I should be overly concerned about standing still and concentrating to see this minor fizzle over not seeing this gameplay visual cue that there is in fact an ammo crate there and not a wall light. Now we learn that whrn this accumulation is reduced on DLSS there is some of the same visual artifacts as PSSR.

 

[Bojji]

I prefer zero added ghosting, too many games pay for unoptimised features with temporal accumulation artefact inducing stuff (including The Matrix demo by Epic), do not need to add anymore guesses.

On consoles, at the price you pay for a PS5 Pro you are going to be more limited and PSSR 1.x reached users in this more limited cheaper HW before FSR4… so I call it a win instead of trying to troll it in as many threads as possible by magnifying every possible issue to oblivion.

You don't have to zoom in to see issues in UE5 games to see obvious issues on the entire screen, this shit is in your face...

And I'm glad your preference is in line what PSSR1 is doing, but what will you say when PSSR2 shows more ghosting but is way more stable thanks to more accumulation? You rarely have any choice in console games so you will have to like whatever is served.

And PS5 Pro ML hardware is fine, it should be able to run more complicated models just fine, even full FSR4 int8 faster than most RDNA3 GPUs.

One of the worst cases of this was when Bojji was calling PSSR "unsuitable" for Ratchet and clank because it had some fizzle, when the accumulation artifacts on DLSS removed the moving light visual cue for ammo crates and this went completely unnoticed by himself and DF:



As if I should be overly concerned about standing still and concentrating to see this minor fizzle over not seeing this gameplay visual cue that there is in fact an ammo crate there and not a wall light. Now we learn that this accumulation is reduced on DLSS there is some some of the same visual artifacts as PSSR.

I tested the game when this video showed up and that pulsating light wasn't removed by DLSS.

And Ratchet is one of the best uses of PSSR (minus that ground fizzling seen on gif), I don't know why you concentrate on that when you can compare AW2 for example...

 

[Three]

I tested the game when this video showed up and that pulsating light wasn't removed by DLSS.

And Ratchet is one of the best uses of PSSR (minus that ground fizzling seen on gif), I don't know why you concentrate on that when you can compare AW2 for example...

What are you talking about? What do you think it was removed by then? I even remember you were talking about using different presets to try and fix it. So if R&C use of PSSR is one of the best why did you call PSSR "unsuitable"?

 

[Bojji]

What are you talking about? What do you think it was removed by then? I even remember you were talking about using different presets to rry and fix it. So if it's the best why did you call PSSR "unsuitable"?

I don't remember that but ok. Light is in the game, with in game DLSS version. Why it didn't show up on Alex video is beyond me...

 

[Three]

I don't remember that but ok. Light is in the game, with in game DLSS version. Why it didn't show up on Alex video is beyond me...

Probably because much like that time we talked about somebody seeing shimmering with DLSS on their gun (in cyberpunk) and you said they don't exist you're not using the exact same DLSS version/preset and settings as Alex was at the time. That light effect was removed by DLSS ghosting though.

 

[Bojji]

Probably because much like that time we talked about somebody seeing shimmering with DLSS on their gun (in cyberpunk) and you said they don't exist you're not using the exact same DLSS version/preset and settings as Alex was at the time. That light effect was removed by DLSS ghosting though.

That shimmering was caused by DLSS Ray reconstruction and not DLSS Super resolution if I remember correctly, two separate things. Let's not have this discussion again...

 

[Zathalus]

One of the worst cases of this was when Bojji was calling PSSR "unsuitable" for Ratchet and clank because it had some fizzle, when the accumulation artifacts on DLSS removed the moving light visual cue for ammo crates and this went completely unnoticed by himself and DF:



As if I should be overly concerned about standing still and concentrating to see this minor fizzle over not seeing this gameplay visual cue that there is in fact an ammo crate there and not a wall light. Now we learn that whrn this accumulation is reduced on DLSS there is some of the same visual artifacts as PSSR.

I just tested it myself, default DLSS, forcing E,K,L,M. The strobing light is present in all instances.

 

[Gaiff]

One of the worst cases of this was when Bojji was calling PSSR "unsuitable" for Ratchet and clank because it had some fizzle, when the accumulation artifacts on DLSS removed the moving light visual cue for ammo crates and this went completely unnoticed by himself and DF:



As if I should be overly concerned about standing still and concentrating to see this minor fizzle over not seeing this gameplay visual cue that there is in fact an ammo crate there and not a wall light. Now we learn that whrn this accumulation is reduced on DLSS there is some of the same visual artifacts as PSSR.

Cause this isn't caused by DLSS? It's a bug of the PC version.

 

[Panajev2001a]

And I'm glad your preference is in line what PSSR1 is doing, but what will you say when PSSR2 shows more ghosting but is way more stable thanks to more accumulation? You rarely have any choice in console games so you will have to like whatever is served.

…

I will have to hope it is not as crappy and can replace TAA, but I might not be happy . I would hope the ghosting increase is not too noticeable and on average the benefit to games is positive (lower runtime cost and easier adoption for titles that already support DLSS and FSR could bring enough pros to offset what could be hopefully a small regression)… but yes, sigh .

 

[Three]

Cause this isn't caused by DLSS? It's a bug of the PC version.

It was caused by DLSS. When you disabled DLSS it worked fine. When you used FSR it worked fine. Exact same video, on the exact same version of the game, exact same time that comparison was done, notice the lights worked on FSR:


Now fast forward to DLSS at 11:49.
Not sure what this deflection is about.

I just tested it myself, default DLSS, forcing E,K,L,M. The strobing light is present in all instances.

So what are you suggesting happened exactly in Alex's comparison?
DLSS caused this issue because it was accumulating frames and just ghosting into a single colour. At the time when DLSS was disabled it worked fine. I'm not sure which version Alex used but I would guess 3.7 and not the default 3.1. He tested both in that video but did not specify version in individual clips. This was 100% a DLSS ghosting issue though. Not a game bug as you can see above.

 

[Zathalus]

So what are you suggesting happened exactly in Alex's comparison?
DLSS caused this issue because it was accumulating frames and just ghosting into a single colour. At the time when DLSS was disabled it worked fine. I'm not sure which version Alex used but I would guess 3.7 and not the default 3.1. He tested both in that video but did not specify version in individual clips. This was 100% a DLSS ghosting issue though. Not a game bug as you can see above.

No idea. He was forcing a custom scale for DLSS so maybe something went wrong? Or it's a bug that randomly happens? FSR was done with the default quality preset. I tested both DLAA and DLSS Quality and both seemed fine. Strobing effect on all the boxes in that scene were present.

 

[Karim]

Wasn't the exact claim from them the following:

"Several developers speaking to Digital Foundry have stated that their current PS5 work sees them throttling back the CPU in order to ensure a sustained 2.23GHz clock on the graphics core."

Which they were pretty much correct about? PS5 average gaming frequency is 2170Mhz, not 2230Mhz as confirmed by Mark Cerny himself in the recent PS5 Pro talks he did. Hence, the "official" PS5 Pro clock is 2170Mhz, even though it can boost up to 2350Mhz.

I remembered they said quite insisting there was a catch to have the variables frequency and when cpu and gpu were overloaded something we have to "sacrifice" in terms of perfomance which is frankly an absurd hyperbole. And they repeat this meme for a very long period. It's funny how there is always someone who tries to excuse their theories because they really like flame some territory for their audience and they are imprecise and vague many many times at the ps5 launch; to the otherside the series X seemed the second coming of Christ hearing them and they full supported the true RDNA2 narrative and when something underperformed on xbox, it was a mystery, tools or whatever, to retreat years later casually when the xbox start is fall in the disgrace. The coincidence.

 

[Three]

No idea. He was forcing a custom scale for DLSS so maybe something went wrong? Or it's a bug that randomly happens? FSR was done with the default quality preset. I tested both DLAA and DLSS Quality and both seemed fine. Strobing effect on all the boxes in that scene were present.

He was using the newer 3.7 in the comparison. You used 3.1. I'm just letting you know that this was 100% a DLSS issue though because I know it was. Don't really care about what preset/version you used. I already stated that when Boji tried to say the same "works for me" along with the other cyberpunk example you don't have a clue what was used exactly, game updates can even add motion vectors. I'm just saying in those examples where 3.7 was likely used and compared with whatever preset it had less fizzle but that ghosting issue which was caused by DLSS.

 

[Zathalus]

I remembered they said quite insisting there was a catch to have the variables frequency and when cpu and gpu were overloaded something we have to "sacrifice" in terms of perfomance which is frankly an absurd hyperbole. And they repeat this meme for a very long period. It's funny how some of you try always to find an excuse to their theory because they really like flame some territory for their audience and they are imprecise and vague many many times at the ps5 launch where series X seemed the second coming of Christ hearing them and when something underperformed it was a mystery and unexplainable to retreat years later casually when the xbox start is fall in the disgrace. The coincidence.

They have been praising the PS5 since the very first review they did of the system. They constantly have PS5 games in the best looking games for each year. The prosecution complex some of you have on behalf of Sony is funny. Mark Cerny himself is a fan and they have been far more critical of MS over the past two years. Give this nonsense a rest already.

He was using the newer 3.7 in the comparison. You used 3.1. I'm just letting you know that this was 100% a DLSS issue though because I know it was. Don't really care about what preset/version you used. I already stated that when Boji tried to say the same "works for me" along with the other cyberpunk example you don't have a clue what was used exactly, game updates can even add motion vectors. I'm just saying in those examples where 3.7 was likely used and compared with whatever preset it had less fizzle but that ghosting issue which was caused by DLSS.

Ratchet and Clank ships with 3.7. That's what I tested with.

 

[Bojji]

He was using the newer 3.7 in the comparison. You used 3.1. I'm just letting you know that this was 100% a DLSS issue though because I know it was. Don't really care about what preset/version you used. I already stated that when Boji tried to say the same "works for me" along with the other cyberpunk example you don't have a clue what was used exactly, game updates can even add motion vectors. I'm just saying in those examples where 3.7 was likely used and compared with whatever preset it had less fizzle but that ghosting issue which was caused by DLSS.

In game DLSS



vs. 4 and 4.5



Now I will try 3.7 and 3.8...

Edit:

 

[Gaiff]

It was caused by DLSS. When you disabled DLSS it worked fine. When you used FSR it worked fine. Exact same video, on the exact same version of the game, exact same time that comparison was done, notice the lights worked on FSR:

The light also works with DLSS.

 

[Karim]

They have been praising the PS5 since the very first review they did of the system. They constantly have PS5 games in the best looking games for each year. The prosecution complex some of you have on behalf of Sony is funny. Mark Cerny himself is a fan and they have been far more critical of MS over the past two years. Give this nonsense a rest already.


Ratchet and Clank ships with 3.7. That's what I tested with.

I don't follow you they have to say ps5 is a shitty hardware and despise the ps5 exclusive? I just reported outside John they were always the MS bitches until the thing start to going wrong, they make like 6 videos analises for Forza releases either and if I'm not wrong also for the other MS ip, are you for real? Nothing wrong to prefer a brand but I thought I stated the obvious, they never were exactly impartial and they used a very different tone for the 2 brands.

 

[Senua]

I don't follow you they have to say ps5 is a shitty hardware and despise the ps5 exclusive? I just reported outside John they were always the MS bitches until the thing start to going wrong, they make like 6 videos for Forza I don't remind which chapter as many other ip, are you for real?

 

[Three]

Ratchet and Clank ships with 3.7. That's what I tested with.

The game released in 2023. It shipped with 3.1.13. DLSS 3.7 released in 2024. You may have updated to whatever version through the game or yourself. the version you used isn't important. It's that when that testing was done with 3.1 or 3.7 or whatever inbetween that issue was caused by DLSS 100%. Not sure why you're adamant to suggest it wasn't. I even ran into the issue myself and the fix was to not use DLSS. Boji himself was suggesting the fix is to use a different preset. Whatever version they were using in that comparison that accumulated frames for stability suffered from ghosting and that light issue occurred.

 

[Three]

The light also works with DLSS.

Read replies above, we've discussed this. I'm not interested in "works for me" on whatever version of DLSS, preset or settings is being used outside of that comparison back then. I'm saying at the time of that test DLSS was causing a ghosting/accumulated frames issue with that effect. (BTW whatever that image is doesn't load for me)

 

[Bojji]

The game released in 2023. It shipped with 3.1.13. DLSS 3.7 released in 2024. You may have updated to whatever version through the game or yourself. the version you used isn't important. It's that when that testing was done with 3.1 or 3.7 or whatever inbetween that issue was caused by DLSS 100%. Not sure why you're adamant to suggest it wasn't. I even ran into the issue myself and the fix was to not use DLSS. Boji himself was suggesting the fix is to use a different preset. Whatever version they were using in that comparison that accumulated frames for stability suffered from ghosting and that light issue occurred.

Rumor Hardware  Post in thread 'PS5 Pro is getting PSSR 2.0 between January and March 2026'

Jan 27, 2026

He was using the newer 3.7 in the comparison. You used 3.1. I'm just letting you know that this was 100% a DLSS issue though because I know it was. Don't really care about what preset/version you used. I already stated that when Boji tried to say the same "works for me" along with the other cyberpunk example you don't have a clue what was used exactly, game updates can even add motion vectors. I'm just saying in those examples where 3.7 was likely used and compared with whatever preset it had less fizzle but that ghosting issue which was caused by DLSS.

In game DLSS...

• Bojji

• 

 

[Gaiff]

Read replies above, we've discussed this. I'm not interested in "works for me" on whatever version of DLSS, preset or settings is being used outside of that comparison back then. I'm saying at the time of that test DLSS was causing a ghosting/accumulated frames issue with that effect.

And I don't care one bit about you using a single isolated example to push an agenda. We can't reproduce it under the exact same parameters, it's not the cause. End of fucking story.

 

[Kazdi]

Check this article from DF on PSSR2. A lot of great info straight from Mark Cerny: https://www.digitalfoundry.net/arti...-part-in-the-next-evolution-of-pssr-upscaling.

Our target is to have something very similar to FSR 4's upscaler available on PS5 Pro for 2026 titles as the next evolution of PSSR; it should take the same inputs and produce essentially the same outputs. Doing that implementation is rather ambitious and time consuming, which is why you haven't already seen this new upscaler on PS5 Pro."

 

[Three]

And I don't care one bit about you using a single isolated example to push an agenda. We can't reproduce it under the exact same parameters, it's not the cause. End of fucking story.

You don't know the exact same parameters. All I know is that disabling DLSS, or using FSR fixed this issue back then. This issue was only present with DLSS enabled. End of fucking story. The only person with an agenda is you. I showed you clear evidence that at the time the issue was not present in the game with FSR. You tried to pass it off as a game bug unrelated to DLSS.

 

[Gaiff]

You don't know the exact same parameters. All I know is that disabling DLSS, or using FSR fixed this issue back then. This issue was only present with DLSS enabled. End of fucking story. The only person with an agenda is you. I showed you clear evidence that at the time the issue was not present in the game with FSR. You tried to pass it off as a game bug unrelated to DLSS.

You used a single example. I not only used another example, but your own was also used and we couldn't reproduce it. So how the fuck do you conclude it was DLSS with that paltry amount of data? You need thorough testing, not one instance. Your conclusions is therefore hogwash.

 

[Bojji]

You don't know the exact same parameters. All I know is that disabling DLSS, or using FSR fixed this issue back then. This issue was only present with DLSS enabled. End of fucking story. The only person with an agenda is you. I showed you clear evidence that at the time the issue was not present in the game with FSR. You tried to pass it off as a game bug unrelated to DLSS.

Watch the fucking videos I posted. In game uses preset D that is prone to ghosting, yet nothing wrong with lights on crates. I'm also using lower base resolution that DF tried to achieve by forcing resolution parity with pro version.

Maybe this resolution tool caused bugs or maybe game caused it. No one else see the same problem...

 

[Karim]

You don't know the exact same parameters. All I know is that disabling DLSS, or using FSR fixed this issue back then. This issue was only present with DLSS enabled. End of fucking story. The only person with an agenda is you. I showed you clear evidence that at the time the issue was not present in the game with FSR. You tried to pass it off as a game bug unrelated to DLSS.

Arguing about PSSR flaws with pc gamers

Arguing about DLSS flaws with pc gamers

 

[Senua]

The next chapter in Three vs the world

 

[Zathalus]

The game released in 2023. It shipped with 3.1.13. DLSS 3.7 released in 2024. You may have updated to whatever version through the game or yourself. the version you used isn't important. It's that when that testing was done with 3.1 or 3.7 or whatever inbetween that issue was caused by DLSS 100%. Not sure why you're adamant to suggest it wasn't. I even ran into the issue myself and the fix was to not use DLSS. Boji himself was suggesting the fix is to use a different preset. Whatever version they were using in that comparison that accumulated frames for stability suffered from ghosting and that light issue occurred.

Alright, I downgraded my DLSS version to 3.1.13. No issues. Upgraded to 3.7 again, no issues. Used 3.7.20, the very latest version available when Alex made that video, again no problem. Using DLAA and DLSS Quality.

So what now? Three posters in this very thread that actually have the game have no issues, with video proof no less. So Alex has a magic DLSS version, or was it perhaps a bug of some kind?

 

[Three]

You used a single example. I not only used another example, but your own was also used and we couldn't reproduce it. So how the fuck do you conclude it was DLSS with that paltry amount of data? You need thorough testing, not one instance. Your conclusions is therefore hogwash.

A single example where ideally no settings are changed but the upscaler when making a comparison? Oh gee not good enough for you or my own experience of it but jumping to the conclusion that it is a "bug on the PC version" then being presented with clear evidence that with FSR instead of DLSS it worked fine on the exact same controlled test. Nope your own hogwash of "a PC game bug" is the only logical conclusion.

 

[Gaiff]

Arguing about PSSR flaws with pc gamers

Arguing about DLSS flaws with pc gamers

Oh, piss off. We went out of our way to reproduce that "flaw" and couldn't.

 

[Senua]

So what now? Three posters in this very thread that actually have the game have no issues, with video proof no less. So Alex has a magic DLSS version, or was it perhaps a bug of some kind?

 

[Gaiff]

A single example where ideally no settings are changed but the upscaler when making a comparison? Oh gee not good enough for you or my own experience of it but jumping to the conclusion that is is a "bug on the PC version" then being presented with clear evidence that with FSR instead of DLSS it worked fine on the exact same controlled test. Nope your own hogwash of "a PC game bug" is the only logical conclusion.

My conclusion is backed up by three posters testing a myriad of options trying to reproduce the issue. We all failed. Your conclusion is backed by your agenda. Now quit wasting everyone's time. Either you come up with something substantial or you get lost.

 

[Karim]

Oh, piss off. We went out of our way to reproduce that "flaw" and couldn't.

Quite easy when you have access to 50 different setting or module or whatever are named by nvidia in an open platform imo.