Support NeoGAF

[WhartoX]

Testing it out with a 3,5gb/s samsung 970 ssd pci-e 3.0, 9900k 5ghz 8/16 core cpu, 32gb and windows 10, with 3080 tuf stock 10gb model.

3440x1440 ultrawide, everything maxed including textures to ultra high, and RT on, about ~50 fps with drops to high 30's.
With DLSS quality, ~50's, Dlss performance is about ~60's never saw it drop to 50's even in high demanding combat.

Screenshot taken at ultra settings + dlss quality max quality u can get.

Game looks like ass tho really.

Also v-ram consumption seems to halt at 8,8gb from what i saw on this resolution, memory is higher because i have a fuck ton of other programs running on the background. Didn't update nvidia drivers tho.

Load time is pretty much instant.

SSD usage highest i saw, was 360mb/s while roaming around.

How did you get the SSD Used metric for Riva Tuner?

 

[WhartoX]

Did you up core speed or memory speed (or both).

From the reading I've done around the subject it's the overclocking of memory speed that can yield better results, especially at higher resolutions.

I add +200 to the core and +1600 to memory. Core clock speed increases seems to increase FPS more than memory clock increases do but both a beneficial and it varies from game engine to game engine.

 

[DenchDeckard]

How is PC destroying ps5 so much with this game? Its a ps5 exclusive. Doesn't make sense to me.

Is there some strange bottleneck for this engine with ps5?

 

[Kenpachii]

How did you get the SSD Used metric for Riva Tuner?

No idea anymore, have that shit already for like years now.

 

[Kaachan]

3070 30fps

Lmao sure

 

[Gaiff]

How is PC destroying ps5 so much with this game? Its a ps5 exclusive. Doesn't make sense to me.

Is there some strange bottleneck for this engine with ps5?

I don't think PC destroys PS5 though. We haven't tested equivalent settings to see how it fares. Based on my 4090 dipping below 60fps at 4K/max settings, the PS5 dropping to 20 in Quality mode doesn't sound impossible or especially bad.

 

[DenchDeckard]

I don't think PC destroys PS5 though. We haven't tested equivalent settings to see how it fares. Based on my 4090 dipping below 60fps at 4K/max settings, the PS5 dropping to 20 in Quality mode doesn't sound impossible or especially bad.

Yeah but the ps5 is below 1440p and closer to 1080p in quality mode isn't it?

 

[octiny]

Yeah but the ps5 is below 1440p and closer to 1080p in quality mode isn't it?

RT mode drops it to 828p at times.

But yeah, as I noted in my previous post on last page, definitely seems to be more optimized towards AMD cards than Nvidia cards on PC.

 

[rodrigolfp]

How did you get the SSD Used metric for Riva Tuner?

Activate PerfCounter.dll on MSI Afterburner monitoring plugins.

 

[Gaiff]

Yeah but the ps5 is below 1440p and closer to 1080p in quality mode isn't it?

PS5 in the 30fps quality mode uses a dynamic resolution with the highest resolution found being 3840x2160 and the lowest resolution found being 1920x1080. The rendering resolution changes often in the 30fps quality mode and seems to rarely reach 3840x2160. PS5 in quality mode seems to be using FSR 2 to reconstruct a 3840x2160 resolution when rendering below this resolution.

So it seems that it's hard to pin down what the most common resolution is in Quality Mode.

The thing is, I was also using RT when running the game so I should have used RT mode as a basis.

PS5 in the 30fps ray tracing mode uses a dynamic resolution with the highest resolution found being 2880x1620 and the lowest resolution found being 1472x828. The rendering resolution changes often in the 30fps ray tracing mode and seems to rarely reach 2880x1620 and rarely drop to 1472x828. PS5 in the ray tracing mode seems to be using FSR 2 to reconstruct a 3840x2160 resolution.

So, 1620p is uncommon but running it at max settings on my 4090, I get 80-90fps with the lowest frame rate being 75. It's impossible to compare without equivalent settings and scenes but 828p is rough. By these metrics, it seems like the 4090 is like 4x the performance of the PS5 which would indeed mean that the PS5 is underperforming but once again, without equivalent settings, there's no way to tell. The PS5 could be performing like a 2080 Ti (probably not) for all we know but my PC with the 2080 Ti isn't with me right now.

 

[yamaci17]

so guys; what settingsp reset that "recommend" requirements plays at? should we accept it as the "standard" preset? does the PS5 use standard preset? I hope DF can point it out

i want to try EXACT settings they claim they need 24 GB RAM. so far, even at "high" preset, RAM usage is pretty tame on my 16 gigs. i know lots of you 32 gb folk hurdle in here saying game uses 20 GB on your rig, but that happened frequently with other games as well. the more you have, the more some engines will use it. but the game perfectly functions on my 16 gb budget so far.

they made it sound like 16 gb was at the tier where you would have to plat 720p 30 fps very low. i dont get that feeling right now. so that ram requirement is either bullshit or there's something I'm missing here. i will later provide a 1440p-standard video on the 3070

the video here:



also, how far demo goes? what is the most performance bound location in the demo, or is there any at all? does the demo performance resembles overall game performance, or is there more to it?

 

[Kataploom]

How is PC destroying ps5 so much with this game? Its a ps5 exclusive. Doesn't make sense to me.

Is there some strange bottleneck for this engine with ps5?

IDK I thought it was just me but my 6700 XT shits all over PS5, not even using dynamic res or FSR for achieving around 70 fps on High at 1080p... PS5 internal resolution spend most of the time below that on internal res with dips to mid 40s and with lower settings.

Maybe it's a CPU limitation, clearly bandwidth isn't.

 

[RoboFu]

Did they not put that bench mark in the demo?

 

[yamaci17]

So it seems that it's hard to pin down what the most common resolution is in Quality Mode.

The thing is, I was also using RT when running the game so I should have used RT mode as a basis.


So, 1620p is uncommon but running it at max settings on my 4090, I get 80-90fps with the lowest frame rate being 75. It's impossible to compare without equivalent settings and scenes but 828p is rough. By these metrics, it seems like the 4090 is like 4x the performance of the PS5 which would indeed mean that the PS5 is underperforming but once again, without equivalent settings, there's no way to tell. The PS5 could be performing like a 2080 Ti (probably not) for all we know but my PC with the 2080 Ti isn't with me right now.

if we can learn what settings ps5 uses, I can do pretty closely matched performance tests on my 3070. however I also find these kind of tests useless, since DLSS at 1440p produces almost 1200p (this is an arbitrary number) like image quality. even if rtx cards do end up underperforming, they will still get "better image quality per pixel" due to DLSS. DLSS in my opinion changes the entire comparison field. https://imgsli.com/MTQ4Mzcw look at this comparison shot of RDR2 at 1440p dlss performance and then native 1080p. can you argue that 1440p dlss performance, despite 720p internal resolution, resembles anything like 720p? if anything , it looks miles better than what native 1080p is supposed to look like.

problem with ps5 however, when it renders at 800 900p, it really do indeed look like that resolution, fuzziness simply shows it. dlss is the other way around.

i know this is dlss shilling but i have to do it. no one else does it, and i find it disturbing that dlss is being ignored for what it is. if a DLSS equivalent quality option was on consoles, I'm sure vocal minority console userbase would talk about it non stop.


"better image quality per pixel" i like this term lol. i will be using it from now on.

 

[SNG32]

I’m happy that my laptop with a 3070 runs this better than a PS5. 1440p over 30 Fps with ray tracing I really can’t complain.

 

[Reizo Ryuu]

Don't really understand the framerate in this game, I have no idea what random shit the renderer is doing, you can be standing still seeing 60 fps, spin your camera around, and suddenly you've lost 5 fps .
There's like zero consistency in what exactly is causing the fps to tank randomly, makes me almost want to go through each setting step by step and see what is causing the weird fluctuations; maybe tomorrow.

 

[yamaci17]

Don't really understand the framerate in this game, I have no idea what random shit the renderer is doing, you can be standing still seeing 60 fps, spin your camera around, and suddenly you've lost 5 fps .
There's like zero consistency in what exactly is causing the fps to tank randomly, makes me almost want to go through each setting step by step and see what is causing the weird fluctuations; maybe tomorrow.

i've found solace on gamepad
doing erratic movements with mouse seems to be the bane of the game; at least on my end

 

[SmokedMeat]

DLSS set to quality I’m getting a solid 60fps at 1440p, when setting my maximum frames to 60fps. Bumping maximum frames to 120fps I’m in the 80s. No stutters, and it loads pretty fast.

Got everything set to Standard with the exception of Model Detail and Texture filtering on High.

3070ti
Ryzen 7700X
32GB DDR5 6000

No desire to keep putzing around with this, I’d rather get back to Warhammer Inquisitor.

Edit: Just realized I haven’t installed the latest driver. Maybe performance is a little better with it.

 

[Arsic]

Looks like my 3080 is up to the task.

Sucks what I bought as a 4k card is really coming out to be the 1440p card for a lot of releases.

It is what it is. I plan to upgrade next year once the 5000 series is out this year, then 6000 is announced so I can get the 5000 at a good deal lol.

The reviews I looked at plus the performance impressions sounds like I got some fun combat and decent story for 15-20 hours. Going to turn cuff chatter down, set voices to Japanese, and enjoy the ride.

 

[Celcius]

I downloaded the pc demo but there’s no benchmark option on the main menu?

edit: I couldn't find the benchmark but the demo ran great on my 10700K + RTX 3090 at 4K max settings

 

[yamaci17]

Looks like my 3080 is up to the task.

Sucks what I bought as a 4k card is really coming out to be the 1440p card for a lot of releases.

It is what it is. I plan to upgrade next year once the 5000 series is out this year, then 6000 is announced so I can get the 5000 at a good deal lol.

The reviews I looked at plus the performance impressions sounds like I got some fun combat and decent story for 15-20 hours. Going to turn cuff chatter down, set voices to Japanese, and enjoy the ride.

can you find a compromise with 4k dlss performance? in my many tests, 4k/dlss perf was brutalizing (this is not a joke) native 1440p image quality. maybe you can relax texture settings a bit; i believe standard is fair game, they look okay. overall you will come across bad looking textures but it is just how the game is designed at times

 

[Kenpachii]

How is PC destroying ps5 so much with this game? Its a ps5 exclusive. Doesn't make sense to me.

Is there some strange bottleneck for this engine with ps5?

probably memory setup, both consoles have limited bandwidth to deal with.

 

[01011001]

How is PC destroying ps5 so much with this game? Its a ps5 exclusive. Doesn't make sense to me.

Is there some strange bottleneck for this engine with ps5?

I mean the game runs like shit for how it looks on PC as well. I'd say it runs basically as you'd expect from the hardware used. my 3060ti usually perfoms similarly to a PS5, slightly better in some games and worse in others (until RT is used then it kills RDNA2 of course)

and in this game it too perfoms similarly, slightly better I'd say, which is to be expected.

 

[DenchDeckard]

IDK I thought it was just me but my 6700 XT shits all over PS5, not even using dynamic res or FSR for achieving around 70 fps on High at 1080p... PS5 internal resolution spend most of the time below that on internal res with dips to mid 40s and with lower settings.

Maybe it's a CPU limitation, clearly bandwidth isn't.

Yeah, I feel it must be cpu bounded or something as bandwidth shouldn't be an issue.

I mean the game runs like shit for how it looks on PC as well. I'd say it runs basically as you'd expect from the hardware used. my 3060ti usually perfoms similarly to a PS5, slightly better in some games and worse in others (until RT is used then it kills RDNA2 of course)

and in this game it too perfoms similarly, slightly better I'd say, which is to be expected.

Are you running at 1080p and less to achieve 60fps on your setup? From what I've seen you could probably get closer to 60fps at 1440p with console settings which is over twice what the ps5 is performing.

Something is off and doesn't make sense for a studio that's been working on this since before the ps5 launched and its an exclusive. My theory is they came across dev issue and just stripped everything back including the budget and man power just to get it out of the door.

It looks like it was maybe targeting ps4 to begin with I dunno, it looks worse than FF XV.

 

[clampzyn]

ppl sh1tting on ps5



according to elanalista ps5 average resolution on quality mode (120hz) - 40fps is around 1728p. so don't say my gpu is way better than ps5 and achieves 1080p60fps on my 3070 LOL 1080p is nowhere near 1728p

it seems DRS is just pretty aggressive in some scenes/areas for the resolution to dip so low, even on pc, most scenes that have magic skills etc. cuts the fps to half. anyways this game runs sh1t anyways, doesn't justify and hardware performance for its visuals

 

[Thebonehead]

Have posts been merged into the weekend m wrong thread?

 

[rodrigolfp]

memory is higher because i have a fuck ton of other programs running on the background.

Why don't you measure just the RAM usage of the process?

 

[yamaci17]

ppl sh1tting on ps5



according to elanalista ps5 average resolution on quality mode (120hz) - 40fps is around 1728p. so don't say my gpu is way better than ps5 and achieves 1080p60fps on my 3070 LOL 1080p is nowhere near 1728p

it seems DRS is just pretty aggressive in some scenes/areas for the resolution to dip so low, even on pc, most scenes that have magic skills etc. cuts the fps to half. anyways this game runs sh1t anyways, doesn't justify and hardware performance for its visuals

"averages"? how does he know? how many samples he counted? there's no average to speak of unless you do a wide array of search. I simply can't take anyone seriously who says "averages" in cases like this. e, how many frames he personally counted and claims 1728p average? 5? 10? 5? outside of millions of frames?

historically elanalista's resolution claims were always off mark and overstated. I still remember how his misinformation of claims about horizon fw running at 1800p (no notion of checkerboarding) 60 FPS made people go crazy, whereas it was a checkerboarded 1800p instead.

as I said, per pixel quality is higher with DLSS so whole comparison could become moot from another perspective. I'd prefer believe VG tech who is more respectable than elanalista in all hardware scene

PS5 in performance mode uses a dynamic resolution with the highest resolution found being 2560x1440 and the lowest resolution found being 1280x720. In performance mode PS5 seems to usually render below 1920x1080. PS5 in performance mode seems to be using FSR 2 to reconstruct a 2560x1440 resolution when rendering below this resolution.

Despite the graphics settings being reduced; PS5 still is MOSTLY below 1080p on a 60 FPS mode, all the while still dropping frames. How the hell do you expect me to believe that it averages 1512p with ray tracing or 172p with 40 FPS mode?

 

[Kenpachii]

Why don't you measure just the RAM usage of the process?

Didn't interest me, SSD performance, load speeds, 3080 performance on ultra wide, v-ram usage and how the game looks like maxed out is what interested me.

 

[rodrigolfp]

Didn't interest me, SSD performance, load speeds, 3080 performance on ultra wide, v-ram usage and how the game looks like maxed out is what interested me.

What I mean is that you are already measuring all RAM usage at that 21909GB. Change to just measure process RAM usage.

 

[Kenpachii]

What I mean is that you are already measuring all RAM usage at that 21909GB. Change to just measure process RAM usage.

Ah yea could have done that, memory was never really much of a metric to look at however as its like a years old profile i have no clue how to change it anymore to be honest.

 

[yamaci17]

What I mean is that you are already measuring all RAM usage at that 21909GB. Change to just measure process RAM usage.

memory usage analysis on higher amounts of RAm is super misleading. most games will offload more ram then they require for the sake of doing it. I see this game as such an example. every one and their mothers who upgraded 32 gigs will easily total ram usage upwards of 20 gb while game usage might also be around 11 12 mg which might lead them to think 16 gb is ded

in my case, my total ram usage caps around at 12 13 gb and game usage hurdles around 9-10 gb ram. it is totally playable and enjoyable on 16 gigs despite 24 gb being recommended

 

[yamaci17]

so i've made a comparison between native 1620p, 1620p dlss quality and PS5's 40 fps mode

Imgsli

imgsli.com

it really does not look like ps5 in this scene and shot renders at a 1728p average. if you want my honest opinion, I'd say it is around 1080p or 1200p or something. it looks too fuzzy to be anything above 1440p. even 1620p dlss quality looks much better than it while averaging 60 FPS

good luck

shots taken from vg tech: https://drive.google.com/drive/folders/1MLRQkOYJF2JrrQU4FQAeYXOj3PwGGA1G

 

[Gaiff]

ppl sh1tting on ps5



according to elanalista ps5 average resolution on quality mode (120hz) - 40fps is around 1728p. so don't say my gpu is way better than ps5 and achieves 1080p60fps on my 3070 LOL 1080p is nowhere near 1728p

it seems DRS is just pretty aggressive in some scenes/areas for the resolution to dip so low, even on pc, most scenes that have magic skills etc. cuts the fps to half. anyways this game runs sh1t anyways, doesn't justify and hardware performance for its visuals

Eh, this guy is usually pretty bad with his pixel count. VGTech states the following:

PS5 in the 40fps quality mode seems to have the same resolution setup as the 30fps quality mode but the average rendering resolution is lower as the DRS system is targeting 40fps. The 40fps quality mode seems to render at 1920x1080 much more often than the 30fps quality mode does.

And by all accounts, including DF and VGTech, it's pretty much impossible to get an "average" because of the aggressive DRS that causes the resolution to change often. I don't think it'll be possible to do an apples to apples comparison with an equivalent PC with such an intrusive level of resolution scaling. That said, I don't think PS5 performs particularly poorly there. Seems equivalent PC parts would perform in the same ballpark but it's hard to tell for sure.

 

[Gaiff]

so i've made a comparison between native 1620p, 1620p dlss quality and PS5's 40 fps mode

http://[URL][URL]https://imgsli.com/MTUwMTk0/0/1

[/URL][/URL]

it really does not look like ps5 in this scene and shot renders at a 1728p average. if you want my honest opinion, I'd say it is around 1080p or 1200p or something. it looks too fuzzy to be anything above 1440p. even 1620p dlss quality looks much better than it while averaging 60 FPS

good luck

shots taken from vg tech: https://drive.google.com/drive/folders/1MLRQkOYJF2JrrQU4FQAeYXOj3PwGGA1G

Yeah, no way that's 1728p on the PS5 side. Much closer to 1080p. If it's really 1728p, it would be the target resolution for FSR, not the base resolution it upscales from. A base resolution of 1728p with FSR 2.0 looks rather clean. This looks very soft.

 

[yamaci17]

Yeah, no way that's 1728p on the PS5 side. Much closer to 1080p. 1728p would be the target resolution for FSR, not the base resolution it upscales from. A base resolution of 1728p with FSR 2.0 looks rather clean. This looks very soft.

not to mention 1620p dlss quality actually have elements aspects that looks better than native 1620p itself (look at frey's hair and intricate details on her cloths. they're all reconstructed better at 1080p with DLSS. like; native TAA at 1620p cannot even construct the chains between her necklace whereas DLSS can at a much lower resolution. however I do accept that 1620p dlss quality is also bit softer, but nowhere near whatever PS5 is outputting there

 

[Gaiff]

not to mention 1620p dlss quality actually have elements aspects that looks better than native 1620p itself (look at frey's hair and intricate details on her cloths. they're all reconstructed better at 1080p with DLSS. like; native TAA at 1620p cannot even construct the chains between her necklace whereas DLSS can at a much lower resolution. however I do accept that 1620p dlss quality is also bit softer, but nowhere near whatever PS5 is outputting there

It's probably significantly lower than 1728p. It bottoms out to 1080p there so it could easily be 1080 in that scene. Alex said it would take at least 5-6 days yesterday to get his analysis out. I also imagine he might not compare it to the PS5's performance due to the fickle nature of the DSR in that game.

Anyway, the game runs well for me. Stable, no major hiccups, bad frame pacing, or intrusive stutters. Load times are short as well. The visuals don't justify the performance though. 55fps minimum on a 4090 at 4K max settings. Generally 60-70fps. I get way better performance in open-world games that look better than this. I'm guessing the particle effects, draw distances, and LOD have a lot to do with this.

 

[rofif]

so i've made a comparison between native 1620p, 1620p dlss quality and PS5's 40 fps mode

Imgsli

imgsli.com

it really does not look like ps5 in this scene and shot renders at a 1728p average. if you want my honest opinion, I'd say it is around 1080p or 1200p or something. it looks too fuzzy to be anything above 1440p. even 1620p dlss quality looks much better than it while averaging 60 FPS

good luck

shots taken from vg tech: https://drive.google.com/drive/folders/1MLRQkOYJF2JrrQU4FQAeYXOj3PwGGA1G

This is a weird shot. I've got a screenshot from almost that location from the ps5 demo.
I already deleted the demo and I am playing full game. Yes, the FSR break up is kinda bad but it doesn't look like this.

This is how ps5 demo looked in quality here... looks nothing like their shot tbh

I am not there in the full version. The only front facing pic I have from full version on ps5 is below. So the FSR break up is not always bad but it's an exception where it looks stable. They really should dich FSR.

 

[rofif]

It's probably significantly lower than 1728p. It bottoms out to 1080p there so it could easily be 1080 in that scene. Alex said it would take at least 5-6 days yesterday to get his analysis out. I also imagine he might not compare it to the PS5's performance due to the fickle nature of the DSR in that game.

Anyway, the game runs well for me. Stable, no major hiccups, bad frame pacing, or intrusive stutters. Load times are short as well. The visuals don't justify the performance though. 55fps minimum on a 4090 at 4K max settings. Generally 60-70fps. I get way better performance in open-world games that look better than this. I'm guessing the particle effects, draw distances, and LOD have a lot to do with this.

If you put PC version to performance FSR at 4k (so 1080po internal) it still doesn't look this pixelated. The screenshot has some bug or it's manipulated.
But the break up is still a bit less on pc

 

[yamaci17]

This is a weird shot. I've got a screenshot from almost that location from the ps5 demo.
I already deleted the demo and I am playing full game. Yes, the FSR break up is kinda bad but it doesn't look like this.

This is how ps5 demo looked in quality here... looks nothing like their shot tbh

I am not there in the full version. The only front facing pic I have from full version on ps5 is below. So the FSR break up is not always bad but it's an exception where it looks stable. They really should dich FSR.

that first shot still does not look like high resolution such as 1440p and upwards. a lot of pixel fuzzines and breakup within the foliage can be observed

btw trees look horrible (on ps5 too). they only load proper lods once you get close to them. i guess it is the same on PS from the looks of it

this game is a disaster lol

 

[rofif]

that first shot still does not look like high resolution such as 1440p and upwards. a lot of pixel fuzzines and breakup within the foliage can be observed

btw trees look horrible (on ps5 too). they only load proper lods once you get close to them. i guess it is the same on PS from the looks of it

this game is a disaster lol

haha yeah there are some issues. The screenshot is 1mb compressed by ps5 but it's not 1800p for sure anyway.
The lods in pc demo didn't load for me on the trees too. I've not got there in ps5 versions yet.

Thankfully ignorance is bliss and I m just playing the game on ps5. No settings or stuff to check, experiment and change aside from mode

 

[yamaci17]

haha yeah there are some issues. The screenshot is 1mb compressed by ps5 but it's not 1800p for sure anyway.
The lods in pc demo didn't load for me on the trees too. I've not got there in ps5 versions yet.

Thankfully ignorance is bliss and I m just playing the game on ps5. No settings or stuff to check, experiment and change aside from mode

no need, your screenshot gives it away (i mean look at those trees, they look like blobs. they get actual quality once you get near them)

honestly I wont be playing this game. maybe if it gets a %70 discount (which it should, considering the %40 review score on Steam. they will have to recoup the investment by doing a huge flash sale. it is impossible to recoup investment mooney and more by making your game only run on niche 4070ti+ hardware comfortably. most PC folks won't simply accept 30 FPS as a compromise as opposed to console folks

i have no problems with 30 FPS but at least with their attitude, we can force devs to target 60 FPS on their ports. so this game gave me hop on that regard. devs will know their place next time they target 30 FPS on medium hardware

then; i would've considered it buying for 40 bucks if itwas a decent port. i wouldnt get this on ps5 either. it disrespects the hardware huge time. we shouldn't promote this man. 70 bucks for this much unpolished performance is a disgrace to all other studios who give all their efforts to make people play games comfortably and high quality. think of horizon fw and ragnarok. they look better than this game; miles better no? and they run nearly native 4k 60 fps both. it is an enormous gulf between them. this cannot be explained by studio quality alone. if they are not big as them then that means their game does not deserve the 70 bucks premium price tag too.

 

[rodrigolfp]

Kb+m controls are glorious. No retard aiming/camera turning, no claw grip for magic parkour, instante magic swaps...

 

[SNG32]

"averages"? how does he know? how many samples he counted? there's no average to speak of unless you do a wide array of search. I simply can't take anyone seriously who says "averages" in cases like this. e, how many frames he personally counted and claims 1728p average? 5? 10? 5? outside of millions of frames?

historically elanalista's resolution claims were always off mark and overstated. I still remember how his misinformation of claims about horizon fw running at 1800p (no notion of checkerboarding) 60 FPS made people go crazy, whereas it was a checkerboarded 1800p instead.

as I said, per pixel quality is higher with DLSS so whole comparison could become moot from another perspective. I'd prefer believe VG tech who is more respectable than elanalista in all hardware scene

PS5 in performance mode uses a dynamic resolution with the highest resolution found being 2560x1440 and the lowest resolution found being 1280x720. In performance mode PS5 seems to usually render below 1920x1080. PS5 in performance mode seems to be using FSR 2 to reconstruct a 2560x1440 resolution when rendering below this resolution.

Despite the graphics settings being reduced; PS5 still is MOSTLY below 1080p on a 60 FPS mode, all the while still dropping frames. How the hell do you expect me to believe that it averages 1512p with ray tracing or 172p with 40 FPS mode?

Also is the ps5 running this at high texture quality or standard?

 

[GHG]

Kb+m controls are glorious. No retard aiming/camera turning, no claw grip for magic parkour, instante magic swaps...

I played the demo on PC with a controller, maybe I made a mistake.

Especially considering the fact that I think the combat is the only redeeming feature of this game and think it plays like a power fantasy ARPG but in 3rd person instead of being top-down.

 

[yamaci17]

Also is the ps5 running this at high texture quality or standard?

i don't know, from the looks of if, there is no texture quality setting, just the budget of it

once you run out of budget, game starts loading super low res textures

 

[Bojji]

Also is the ps5 running this at high texture quality or standard?

Probably the highest (I'm running demo now on PS5 and comparing it to my 3060ti).

It's "funny" that console that have ~13GB of TOTAL memory can fit this game with the highest textures but cards without 12GB of Vram can't. And what system ram is used for?

 

[yamaci17]

Probably the highest (I'm running demo now on PS5 and comparing it to my 3060ti).

It's "funny" that console that have ~13GB of TOTAL memory can fit this game with the highest textures but cards without 12GB of Vram can't. And what system ram is used for?

you simply run out of budget. chances are, you will get similar texture quality if you rather set your texture quality to standard or maybe even low (if you have a lot of background tasks, that is)

also its not that funny, but I project consoles to have around 9-10 GB usable video memory for games.

system ram is mostly used for duplicate data on PCs, and then some actual RAM data that is also used on consoles (could be around 2-4 GB). so 9 gb vram load + 4-5 gb duplicate texture data + 3-5 GB raw CPU bound memory data would mean around 7-10 GB of system ram usage which is in line with what happens

 

[Bojji]

you simply run out of budget. chances are, you will get similar texture quality if you rather set your texture quality to standard or maybe even low (if you have a lot of background tasks, that is)

also its not that funny, but I project consoles to have around 9-10 GB usable video memory for games.

system ram is mostly used for duplicate data on PCs, and then some actual RAM data that is also used on consoles (could be around 2-4 GB). so 9 gb vram load + 4-5 gb duplicate texture data + 3-5 GB raw CPU bound memory data would mean around 7-10 GB of system ram usage which is in line with what happens

10GB of video memory on consoles is understandable but why even 10gb rtx 3080 can't use PS5 textures? Something is very wrong with the software and that's it. Memory jump was much smaller this gen compared to PS3->PS4.

This is like far cry 6 where even 3090 was showing low res textures at launch.

 

[KXVXII9X]

haha yeah there are some issues. The screenshot is 1mb compressed by ps5 but it's not 1800p for sure anyway.
The lods in pc demo didn't load for me on the trees too. I've not got there in ps5 versions yet.

Thankfully ignorance is bliss and I m just playing the game on ps5. No settings or stuff to check, experiment and change aside from mode

You just enjoy the game! Don't mind us.