DF - Doom Dark Ages PS5/PRO & Series S/X showdown.

Its no surprise that the AI upscaling is behind DLSS its runnung on a 700 console vs 2000 GPUs its brand new tech on consoles and there hasnt been much experience with it.

It needs time to develop and as with all things AI it will develop fast. I saw some leak the other day of PSSR 2 and thats just a few months after launch. Then well get fram egen etc.

Shouting into the void that DLSS is better than PSSR is dumb and immature. pssr is a completely different level and target.
To be behind could be, it has some raw edges but it's not that far as some people claim and surely it's not broken and shitty. Also DLSS 4 has it's own issue and I doubt it's used to upscale from the same lower resolution on console.
 
Last edited:
Just in the last 3 weeks we've had on the Pro:

Expedition 33-very weak upgrade
Oblivion- one of the very worst Pro enhanced
Indiana Jones- low effort resolution upgrade
City of Wolves- no Pro update
Days Gone remaster- is this considered good? Yes and no. Graphics seem better but IQ is worse now.
Returnal- decent res boost but none of the settings were touched.
Doom The Dark Ages- in the running for worst Pro patch yet.

These games are not outliers either. Most upgrades have amounted to similarly tiny differences. I know as I have played just about all of them.
I mean what people expecting, 30-40% faster gpu rendering game with same settings but with around 40% more pixels usually. It is what it is, maybe little disappointing that apparenlty 2x-3x boost in rt performance is nowhere to seen.
 
Its no surprise that the AI upscaling is behind DLSS its runnung on a 700 console vs 2000 GPUs its brand new tech on consoles and there hasnt been much experience with it.

It needs time to develop and as with all things AI it will develop fast. I saw some leak the other day of PSSR 2 and thats just a few months after launch. Then well get fram egen etc.

Shouting into the void that DLSS is better than PSSR is dumb and immature. pssr is a completely different level and target.

This is idiotic. DLSS3 running on RTX 2060 from 2018 looks better than anything PSSR showed so far.
DLSS3 is objectively better than PSSR, there is no argue about it and even when comparing PSSR to 2020 DLSS2 - DLSS2 would win.

We will see next year if what Cerny promised is real because FSR4 is really good, quality like that on PS5 Pro would be a game changer.
 
Apparently HDR in this game is broken, and provides so much washed out picture, that its better to turn it off. Which is suprising, considering Eternal had amazing HDR implementation.
Also i find it weird that channel like DF, usually totally ignores HDR quality on their graphics reviews. I mean, id arque that proper HDR is just as impactful to image as resolution.
Yet they often praise games like Robocop or Expedition 33 without mentioning that overall picture quality is dull, because theres no HDR at all.
 
Apparently HDR in this game is broken, and provides so much washed out picture, that its better to turn it off. Which is suprising, considering Eternal had amazing HDR implementation.
Also i find it weird that channel like DF, usually totally ignores HDR quality on their graphics reviews. I mean, id arque that proper HDR is just as impactful to image as resolution.
Yet they often praise games like Robocop or Expedition 33 without mentioning that overall picture quality is dull, because theres no HDR at all.

Yeah it's broken, at least on Xbox and PC. Confirmed by GamingTech below:



Edit: he's posted a follow up video with a temporary fix or way to partially improve it

 
Last edited:
pcm.gif
 
This is idiotic. DLSS3 running on RTX 2060 from 2018 looks better than anything PSSR showed so far.
DLSS3 is objectively better than PSSR, there is no argue about it and even when comparing PSSR to 2020 DLSS2 - DLSS2 would win.

We will see next year if what Cerny promised is real because FSR4 is really good, quality like that on PS5 Pro would be a game changer.
Look better based on your feeling? I remind long conversation here amid MazingerDude (ex DF member) and you and how he tried to explain to you there are case by case which sometimes PSSR was even better and just to stop it with this annoying fanboy nonsense.. If for you DLSS3 it's better, good, but stop to spread your personal obstinacy as tech fact.
 
Last edited:
In your obsessed mind, sure.

Regarding jedi survivor:

Jedi Survivor Pro patch looks quite good overall, It's a major step up from base PS5 version.

But the biggest flaw of PSSR in this game (and in many others as well) is this "noise" in some places:

Pro:



PC with in game DLSS (3.1?) in performance mode:



I played few games with PSSR and DLSS and you have to be blinded by your love for Cerny (cool dude, I don't blame you) to say that PSSR looks better in any of them.
 
Last edited:
The noisy are caused to raytracing not PSSR. Just to show how deep are your knowledge about the matter

LOL. Game uses some standard denoiser on both PC and PS5 (no support for Ray reconstruction). PSSR IS THE PROBLEM here.
 
No your ignorance is the problem here. Because Jedi Survivor has notorious raytracing noisy problems.

On standard PS5 as well? PC also don't have issues with RT noise, I wonder why game with PSSR enabled is full of noise...

If PS5 Pro video capture was better I would show you the same stuff in Space Marine 2 that doesn't use RT.
 
Last edited:
Show me. I captured stock PS5 Pro and PC versions, I know the answer and videos show it as well.
Don't believe me. I don't care. It was reported on resetera by a developer who post regularly there and it explained most of the PSSR issues pointed out are not relative to the upscaler but to the lower buffer of the real time lighting and shadows effects as raytracing or screen space calculations.
 
Last edited:
The noisy are caused to raytracing not PSSR. Just to show how deep are your knowledge about the matter. Nvidia gpu have hardware raytracing denoiser. Ps5 unfortunately uses AMD gpu.
Jedi Survivor doesn't use ray reconstruction. It ships with a DLSS version that came out over 2 years ago, it was never updated with a newer DLSS version or RR. The PC and console versions all use the same software denoiser built into UE4.
 
Jedi Survivor doesn't use ray reconstruction. It ships with a DLSS version that came out over 2 years ago, it was never updated with a newer DLSS version or RR. The PC and console versions all use the same software denoiser built into UE4.
it uses raytracing for indirect lighting if I remember correctly and maybe for contact shadows. And surely for reflections.
 
Last edited:
Oh dear...

Nobody stood up there gaslighting and lying to people about the capabilities of the PS5 or even the Series X in comparison to hardware that is obviously more capable.

It is about setting realistic expectations, and they failed miserably with the Series S. So much so that they deleted the video that those screen grabs come from:

Yeah wait for it, but we thought the Xbone had the secret sauce, but it's actually gonna be the series S.
 
Don't believe me. I don't care. It was reported on resetera by a developer who explained most of the PSSR issues are not relative to the upscaler but to the lower buffer of real time lighting and shadows.

To be honest I don't care what the root of the issue is, they need to get rid of this.

PSSR is MOSTLY better than what standard PS5 versions are using, higher input res also helps but it's not on the level what people expected when PS5 Pro was announced. MAYBE it will change next year, I hope it will.
 
Last edited:
it uses raytracing for indirect lighting if I remember correctly and maybe for contact shadows. And reflections.
I'm aware, just pointing out there isn't any special denoiser that the PC version is using that the consoles are not. DLSS 3.1 (that the game ships with) is only upscaling and optional frame generation.
 
I'm aware, just pointing out there isn't any special denoiser that the PC version is using that the consoles are not. DLSS 3.1 (that the game ships with) is only upscaling and optional frame generation.
Nvidia gpu have hardware raytracing denoiser from awhile. Such gpu are more equipped than AMD gpu for such issues. If I'm not wrong you can't uses UE4/5 or any software denoiser because interfers with PSSR upscaler.
 
Last edited:
I'm aware, just pointing out there isn't any special denoiser that the PC version is using that the consoles are not. DLSS 3.1 (that the game ships with) is only upscaling and optional frame generation.

PSSR is causing this noise in more games, even ones without RT. There are also titles that don't have this noise at all like TLoU2.

Nvidia gpu have hardware raytracing denoiser from awhile.

All games with RT have platform agnostic denoisers that work on all RT capable hardware.
 
Last edited:
PSSR is causing this noise in more games, even ones without RT. There are also titles that don't have this noise at all like TLoU2.



All games with RT have platform agnostic denoisers that work on all RT capable hardware.
Not on AMD RT gpu. And it's not PSSR who cause noise but low buffer effect which DLSS also can handle separately at native resolution that's why are less noisy.
 
Last edited:
The noisy are caused to raytracing not PSSR. Just to show how deep are your knowledge about the matter. Nvidia gpu have hardware raytracing denoiser. Ps5 unfortunately uses AMD gpu.

nvidia does kinda have a "hardware raytracing denoiser", which is called Ray Reconstruction, but that has to be supported by the game in the game settings. Jedi Survivor doesn't support that as afar as I know
 
Nvidia gpu have hardware raytracing denoiser from awhile.
They do not. Or at least not in a general sense. Ray reconstruction is a ML powered denoiser that is accelerated by Tensor cores like DLSS is, but Nvidia doesn't have any hardware specific for denoising. AMD and Sony can do the same, just recently AND showed they were working on a ML powered denoiser for future release. PS5 Pro might support it in future, PS6 certainly will.

It doesn't really matter though, as Jedi Survivor does not even support ray reconstruction, the game has a version of DLSS that is 6 months older than the first version of ray reconstruction. It relies on the same software denoiser the consoles use.
 
Last edited:
nvidia does kinda have a "hardware raytracing denoiser", which is called Ray Reconstruction, but that has to be supported by the game in the game settings. Jedi Survivor doesn't support that as afar as I know
So it uses higher resolution for such effects. Though someone in resetera reported there were similar noise on pc also time ago in the foliage seen on ps5 pro. Not in same extent but there were.
 
Last edited:
They do not. Or at least not in a general sense. Ray reconstruction is a ML powered denoiser that is accelerated by Tensor cores like DLSS is, but Nvidia doesn't have any hardware specific for denoising. AMD and Sony can do the same, just recently AND showed they were working on a ML powered denoiser for future release. PS5 Pro might support it in future, PS6 certainly will.

It doesn't really matter though, as Jedi Survivor does not even support ray reconstruction, the game has a version of DLSS that is 6 months older than the first version of ray reconstruction. It relies on the same software denoiser the consoles use.
Well PSSR is also an older version on Jedi Survivor than the actual eh ...
 
Last edited:
Just in the last 3 weeks we've had on the Pro:

Expedition 33-very weak upgrade
Oblivion- one of the very worst Pro enhanced
Indiana Jones- low effort resolution upgrade
City of Wolves- no Pro update
Days Gone remaster- is this considered good? Yes and no. Graphics seem better but IQ is worse now.
Returnal- decent res boost but none of the settings were touched.
Doom The Dark Ages- in the running for worst Pro patch yet.

These games are not outliers either. Most upgrades have amounted to similarly tiny differences. I know as I have played just about all of them.
bro sell me your pro for peanuts, go hiking, or have a 1001 day sea navigation experience
both of us gonna be happy as shit that way
 
So it uses higher resolution for such effects. Though someone in resetera reported there were similar noise on pc also time ago in the foliage seen on ps5 pro. Not in same extent but there were.

rt related noise isn't uncommon. the issue is often a low internal resolution in combination with bad denoising.

some Antialiasing methods can expose bad denoising. that is very likely what happened in the case of Jedi Survivor and PSSR.
and since many modern engines use TAA smearing and reconstruction to make their graphics look cohesive, they are more dependent on how well any given TAA method works in conjunction with that than ever before.

what people often don't seem to realise is that any TAA is image reconstruction. TAA uses multiple frames, camera jittering and motion/depth data from the engine to create additional image information which it uses to smooth edges.
and engines like Unreal use this nowadays to render certain elements dithered at half resolution, and then relies on TAA to turn that into something that looks "completed".
PSSR being a TAA method not native to the engine means that it can clash with elements that rely on the TAA interacting correctly with them.

so issues like these can technically happen with any TAA method if the Devs aren't mindful of it. PSSR FSR2/3/4, DLSS etc.
 



Level 1 MinimumLevel 1 AverageLevel 1 MaximumLevel 2 MinimumLevel 2 AverageLevel 2 MaximumLevel 6 MinimumLevel 6 AverageLevel 6 Maximum3 Levels Average% Per Axis
Series S637p800p900p637p667p830p637p660p825p709p56%
Series X1019p1352p1440p1019p1226p1374p1019p1203p1440p1260p100%
PS5 1019p1147p1440p1019p1047p1170p1019p1038p1222p1077p85%
PS5 Pro1273p1520p1800p1273p1336p1549p1273p1314p1506p1390p110%
 
Last edited:
Well damn I was going to buy this but if they aren't putting any effort in fuck em. Didn't expect them to go crazy but damn.
 



Level 1 MinimumLevel 1 AverageLevel 1 MaximumLevel 2 MinimumLevel 2 AverageLevel 2 MaximumLevel 6 MinimumLevel 6 AverageLevel 6 Maximum3 Levels Average% Per Axis
Series S637p800p900p637p667p830p637p660p825p709p56%
Series X1019p1352p1440p1019p1226p1374p1019p1203p1440p1260p100%
PS5 1019p1147p1440p1019p1047p1170p1019p1038p1222p1077p85%
PS5 Pro1273p1520p1800p1273p1336p1549p1273p1314p1506p1390p110%


again, the power draw is just surprising me every time in these comparisons.

as for the Pro, I really wonder if they use all the RDNA2 features there that the base PS5 is missing, because I'm pretty sure they do use hardware VRS and sampler feedback streaming on Xbox, which both the base PS5 can't do but the Pro can as it's RDNA3+

if they used the base PS5 version as the base and then simply increased the resolution limit, it's no wonder the difference is the way it is here I think
 
Last edited:
again, the power draw is just surprising me every time in these comparisons.

as for the Pro, I really wonder if they use all the RDNA2 features there that the base PS5 is missing, because I'm pretty sure they do use hardware VRS and sampler feedback streaming on Xbox, which both the base PS5 can't do but the Pro can as it's RDNA3+

if they used the base PS5 version as the base and then simply increased the resolution limit, it's no wonder the difference is the way it is here I think
I don't think any of the additional features of the PS5 Pro were used.
The artifacts specific to HW VRS also do not appear in the Pro version.
I think the Pro version of this game just increases the resolution and slightly increases some settings.
7S1EHTnV_o.png

I only own the Xbox Series X version (Game Pass), so I can't compare it myself.
 
I don't think any of the additional features of the PS5 Pro were used.
The artifacts specific to HW VRS also do not appear in the Pro version.
I think the Pro version of this game just increases the resolution and slightly increases some settings.
7S1EHTnV_o.png

I only own the Xbox Series X version (Game Pass), so I can't compare it myself.

the base PS5 uses software VRS, with that being somewhat dynamic in general, it's hard to say I think how exactly it would manifest with different APIs n stuff... but the pro does indeed look much more like the software VRS of the base than the hardware VRS of the Xbox

but speaking of APIs, maybe Sony's API for the Pro doesn't support sampler feedback and VRS Tier 2 (yet)? is there any info on Alan Wake 2 and its use of Mesh Shaders? it uses Mesh Shaders on Series X, but on base PS5 they had to use Primitive shaders instead (again due to the missing RDNA2 features). I wonder if Sony didn't add this stuff to their APIs yet 🤔
 
Last edited:
the base PS5 uses software VRS, with that being somewhat dynamic in general, it's hard to say I think how exactly it would manifest with different APIs n stuff... but the pro does indeed look much more like the software VRS of the base than the hardware VRS of the Xbox

but speaking of APIs, maybe Sony's API for the Pro doesn't support sampler feedback and VRS Tier 2 (yet)? is there any info on Alan Wake 2 and its use of Mesh Shaders? it uses Mesh Shaders on Series X, but on base PS5 they had to use Primitive shaders instead (again due to the missing RDNA2 features). I wonder if Sony didn't add this stuff to their APIs yet 🤔
Primitive Shader still exists in hardware(PS5 & RDNA2 & 3), and Mesh Shader is supported with the idea that the way Mesh Shader is used is realized with Primitive Shader.
https://www-4gamer-net.translate.go...=ja&_x_tr_tl=en&_x_tr_hl=en-US&_x_tr_pto=wapp
 



Level 1 MinimumLevel 1 AverageLevel 1 MaximumLevel 2 MinimumLevel 2 AverageLevel 2 MaximumLevel 6 MinimumLevel 6 AverageLevel 6 Maximum3 Levels Average% Per Axis
Series S637p800p900p637p667p830p637p660p825p709p56%
Series X1019p1352p1440p1019p1226p1374p1019p1203p1440p1260p100%
PS5 1019p1147p1440p1019p1047p1170p1019p1038p1222p1077p85%
PS5 Pro1273p1520p1800p1273p1336p1549p1273p1314p1506p1390p110%



The difference in average resolution during gameplay is almost as wide between PS5 vs XSX as it is between XSX vs PS5 Pro. Some captures begins to make more sense.
TDZIa9c.png



On the other hand, again, the power consumption is revealing between the XSX and the rest. And seeing the XSS also gives me more hope of getting a Handheld with similar power in a not-too-long wait.
 
Apparently HDR in this game is broken, and provides so much washed out picture, that its better to turn it off. Which is suprising, considering Eternal had amazing HDR implementation.
Also i find it weird that channel like DF, usually totally ignores HDR quality on their graphics reviews. I mean, id arque that proper HDR is just as impactful to image as resolution.
Yet they often praise games like Robocop or Expedition 33 without mentioning that overall picture quality is dull, because theres no HDR at all.
Df always forgets to mention HDR even when RDR2 released with broken HDR it wasn't mentioned once in Johns initial review!

Dooms lack of good HDR is making the game look duller than it needs to!
 
Just in the last 3 weeks we've had on the Pro:

Expedition 33-very weak upgrade
Oblivion- one of the very worst Pro enhanced
Indiana Jones- low effort resolution upgrade
City of Wolves- no Pro update
Days Gone remaster- is this considered good? Yes and no. Graphics seem better but IQ is worse now.
Returnal- decent res boost but none of the settings were touched.
Doom The Dark Ages- in the running for worst Pro patch yet.

These games are not outliers either. Most upgrades have amounted to similarly tiny differences. I know as I have played just about all of them.
I'm not sure what people are expecting - res bumps and slightly improved frame rates are basically all the Pro can realistically offer. The generation's half over, and given the dev time for most AAA games, most devs are already done for this console cycle. Sony can mandate the inclusion of Pro enhancements, but devs are going to be extremely limited in what they can actually do with it at this stage. In the vast majority of cases, I imagine they'll swap out FSR2 for PSSR, let the increased clocks provide whatever scalability improvements their engine and assets already allow, and be done with it. I mean, what else are they honestly going to do here? Compared to the PS4, where the jump from base ~1080p to the ~4k resolves of the Pro was actually a noticeable step up, the jump here was always going to be smaller.
 
The difference in average resolution during gameplay is almost as wide between PS5 vs XSX as it is between XSX vs PS5 Pro. Some captures begins to make more sense.
TDZIa9c.png



On the other hand, again, the power consumption is revealing between the XSX and the rest. And seeing the XSS also gives me more hope of getting a Handheld with similar power in a not-too-long wait.
Jeez. That's not normal. Considered it runs also worse seems the playstation porting is quite rushed to me.
 
Last edited:
Top Bottom