• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Digital Foundry: Horizon Forbidden West PC - Optimised Settings vs PS5

Gaiff

SBI’s Resident Gaslighter


Part III of the series.

k5ez9NO.png
 
Last edited:

Gaiff

SBI’s Resident Gaslighter
yeah no ps5 equivalent settings
none of his famous ps5 vs 3070 matchups :)

of course he can't do that since even high textures brutalize 8 GB VRAM at 4K upscaling
He does mention the 3070 can massively drop in performance and recommends lowering the textures in such case. He also compares the drop in texture quality when doing so.

Can’t really do a match up because of DRS and CBR vs DLSS/FSR.
 

yamaci17

Member
He does mention the 3070 can massively drop in performance and recommends lowering the textures in such case. He also compares the drop in texture quality when doing so.

Can’t really do a match up because of DRS and CBR vs DLSS/FSR.
he can find a specific scene that reports similar resolutions on PS5. dynamic resolution changes betwen scenes and camera motions. if you stand in a place with no NPCs, robots etc. and just stand still, resolution should stabilize. and you can punch in that exact same internal resolution on PC with tools like DLSS tweaks

they're DF, they can find a way to do it. he's just avoiding doing it because it would make 8 GB cards look weak against PS5. and despite being a huge defender of DLSS and ray tracing and stuff and even accepting the compromises, I have a problem with his approach to this

either do whatever you're doing for all games as much as possible or don't do it at all. the second he finds a game where 3070 can have the edge without running into VRAM problems, he will go extremes to make the comparison happen. when reverse happens, he will resort to endless amount of excuses

not to mention DSR being a thing did not stop Rich from comparing 4070 to the PS5 with DSR driven games.
 
Last edited:

Gaiff

SBI’s Resident Gaslighter
he can find a specific scene that reports similar resolutions on PS5. dynamic resolution changes betwen scenes and camera motions. if you stand in a place with no NPCs, robots etc. and just stand still, resolution should stabilize. and you can punch in that exact same internal resolution on PC with tools like DLSS tweaks
Or you can run into a CPU bottleneck on PS5. Furthermore, there's really no point in comparing performance when the reconstruction method isn't available on PC. We're not even sure of the impact of CBR.
they're DF, they can find a way to do it. he's just avoiding doing it because it would make 8 GB cards look weak against PS5. and despite being a huge defender of DLSS and ray tracing and stuff and even accepting the compromises, I have a problem with his approach to this
The 8GB card already looks bad as said in this video. He says that at above 1440p or something, you need to drop the textures down to Medium when the PS5 uses Very High. He also demonstrates how much the performance drops when the 3070 exceeds is VRAM capacity. What more do you want exactly? To know precisely how much it drops compared to the PS5?
not to mention DSR being a thing did not stop Rich from comparing 4070 to the PS5 with DSR driven games.
Because most of these games are 3rd-party, and easy to find a spot where the DRS bottoms out. In HFW, good luck with that. The only drops below 60 are mostly CPU-related or during intensive combat which are impossible to reproduce. Hell, even in Frontiers of Pandora, they initially did not have a spot to compare until Tom Morgan found one for Rich and these take a while to get a hold of. These games also use FSR which we have on PC, or no upscaler. If you introduce a third variable that can skew the results, it's a bad methodology that will give bad results.

I really don't have much of a problem with this video. We got optimized settings, we know 8GB cards die with Very High textures, and we know how the PS5 settings stack up more or less and we also know how to get a decent PC experience which is the entire point of this.
 
Last edited:

Skifi28

Member
Interesting that a 3600 has issues even though it's supposedly the console equivalent CPU in performance.
 

Gaiff

SBI’s Resident Gaslighter
Can't say I noticed consistent traversal stutter. I'm not talking about a dropped frame here or there, that's pretty minor.
It’s minor but during intense sequences, it can drop to below 60 and at times it struggles to go above 60fps even when the frame rate is unlocked.
 

shamoomoo

Member
Or you can run into a CPU bottleneck on PS5. Furthermore, there's really no point in comparing performance when the reconstruction method isn't available on PC. We're not even sure of the impact of CBR.

The 8GB card already looks bad as said in this video. He says that at above 1440p or something, you need to drop the textures down to Medium when the PS5 uses Very High. He also demonstrates how much the performance drops when the 3070 exceeds is VRAM capacity. What more do you want exactly? To know precisely how much it drops compared to the PS5?

Because most of these games are 3rd-party, and easy to find a spot where the DRS bottoms out. In HFW, good luck with that. The only drops below 60 are mostly CPU-related or during intensive combat which are impossible to reproduce. Hell, even in Frontiers of Pandora, they initially did not have a spot to compare until Tom Morgan found one for Rich and these take a while to get a hold of. These games also use FSR which we have on PC, or no upscaler. If you introduce a third variable that can skew the results, it's a bad methodology that will give bad results.

I really don't have much of a problem with this video. We got optimized settings, we know 8GB cards die with Very High textures, and we know how the PS5 settings stack up more or less and we also know how to get a decent PC experience which is the entire point of this.
Yeah. There aren't any exact performance cost for CBR but it isn't free on the PS4-5.
 

Gaiff

SBI’s Resident Gaslighter
Yeah. There aren't any exact performance cost for CBR but it isn't free on the PS4-5.
I know it isn't, which furthers my point. If you can't do an exact match or at least a very close one, then don't bother comparing metrics.
 
Last edited:

shamoomoo

Member
I know it isn't, which furthers my point. If you can't do an exact match or at least a very close one, then don't bother comparing metrics.
There isn't any definitive metrics given about CBR but here's a take on CBR from Johnathan Blow via an old WCCFtech Q&A/interview:

I think it depends on the particular game and engine. Different rendering pipelines are structured differently; for some pipelines, the cost of adding checkerboard rendering would be very low, because they are already computing a lot of the information that checkerboard rendering needs. For other pipelines the cost might be higher. In our case we're just not sure of the total cost yet, but we think it is probably high enough that we may prefer to do a straight upscale. But we're not completely sure.

(It is true that, as Sony has announced, the PS4 Pro provides hardware support for checkerboard rendering that makes it faster than it would otherwise be. I think in some places I have seen the rumor that checkerboard is completely free, but I would consider that an exaggeration: the cost is going to vary per game. Unfortunately due to NDAs I can't provide details; I can't say anything more about Sony technologies than what they have announced. It is definitely true that if you had a game running on the original PS4, and the developer wants to do the most straightforward thing to make the game look better on the Pro, that developer could enable checkerboard rendering and the game will look better and run faster; so it's "free" in that sense. But if you are going to get picky about how you are spending the GPU memory and bandwidth of the new machine, then there are tradeoffs here, like with anything.)


 
Last edited:

ChiefDada

Gold Member
PS5 can also become CPU-limited.

It’s minor but during intense sequences, it can drop to below 60 and at times it struggles to go above 60fps even when the frame rate is unlocked.

Pretty sure it's gpu limited and they tightened the drs range once the vrr patch went live, which is why they're comfortable letting it run below 60 when it needs to.
 

Gaiff

SBI’s Resident Gaslighter
Pretty sure it's gpu limited and they tightened the drs range once the vrr patch went live, which is why they're comfortable letting it run below 60 when it needs to.
Possibly but on a 4070 at 1080p with DLSS Quality and a 3600, it can dip to the high 50s during intense action sequences. I watched a video of the PS5 in its unlocked mode and it did drop to 59 or 58fps, similarly to the 3600 but it could be a GPU limitation. I'm more inclined to say CPU because the behavior is suspiciously similar and DRS is purely for GPU scaling so when it cannot maintain it, I tend to think it's more due to the CPU than the GPU.
 
Last edited:

shamoomoo

Member
I'm not sure if it still applies,but DF did a comparison of CBR vs DLSS for Death stranding on PC vs PS4 Pro.




Alex Battaglia avatar
Feature by Alex Battaglia Video Producer, Digital Foundry
Updated on 28 Jul 2020
80 comments
Follow Death Stranding
The concept of native resolution is becoming less and less relevant in the modern era of games and instead, image reconstruction techniques are coming to the fore. The idea here is remarkably simple: in the age of the 4K display, why expend so much GPU power in painting 8.3m pixels per frame when instead, processing power can be directed at higher quality pixels instead, interpolated up to an ultra HD output? Numerous techniques have been trialled, but Kojima Productions' Death Stranding is an interesting example. On PS4 Pro, it features one of the best checkerboarding implementations we've seen. Meanwhile, on PC, we see a 'next-gen' image reconstruction technique - Nvidia's DLSS - which delivers image quality better than native resolution rendering.



Checkerboard rendering as found in Death Stranding is non-standard and it's the result of months of intensive work by Guerrilla Games during the production of Horizon Zero Dawn. Curiously, it does not use PS4 Pro's bespoke checkerboarding hardware. Base resolution is 1920x2160 in a checkerboard configuration, with 'missing pixels' interpolated from the previous frame. Importantly, Decima does not sample a pixel from its centre, but from its corners over two frames. By combining these results over time in a specialised way similar to the game's TAA and a very unique pass of FXAA, a 4K pixel grid is resolved and the perception of much higher resolution is achieved. According to presentations from Guerrilla, of the engine's 33.3ms per-frame render budget, 1.8ms is spent on the checkerboard resolve.

Although it upsamples from a much lower resolution, DLSS works differently. These is no checkerboard, no pixel-sized holes to fill. Rather, it works more like accumulation temporal anti-aliasing, where multiple frames from the past are queued up and information from these frames is used to smooth lines and add detail into an image - but instead of adding detail to an image of the same resolution as TAA does, it generates a much higher output resolution instead. As a part of these frames from the past, motion vectors from those frames for every object and pixel on screen are integral for DLSS working properly. How all of this information is used to create the upscaled image is decided upon by an AI program running on the GPU, accelerated by the tensor cores in an RTX GPU. So while DLSS has fewer base pixels to work from, it has access a vast amount of compute power to help the reconstruction process.

Cover image for YouTube videoDeath Stranding PC DLSS 2.0 vs PS4 Pro Checkerboarding: Image Reconstruction Analysis
An in-depth look at how Nvidia's DLSS 2.0 improves in most respects over PS4 Pro's checkerboarding.





On an RTX 2080 Ti at 4K, DLSS completes in around 1.5ms - meaning it's faster than checkerboarding on PS4 Pro. However, the hardware comparison is obviously lopsided. The weakest capable GPU is the RTX 2060 (still significantly more powerful than the Pro) and DLSS has an overhead in excess of 2.5ms on this card. That's heavy, especially if you're targeting 60fps where the entire frame rendering budget is just 16.7ms. However, the core advantage is that the base resolution is so much lower. DLSS in Death Stranding comes in two flavours: the performance mode achieves 4K quality from just a 1080p internal resolution. Meanwhile, the quality mode delivers better-than-native results from a 1440p base image. In both cases, that's much lower than PS4 Pro's 1920x2160 core resolution. By running everything else in the GPU pipeline at much lower resolutions, the cost of processing DLSS is more than offset - to the point where mildly overclocking the RTX 2060 allows Death Stranding to deliver 4K gaming at 60fps.

In the video on this page, you'll see detailed comparisons of how Death Stranding's checkerboarding on PS4 Pro stands up against DLSS on PC and it's fascinating to see what is effectively a state of the art current generation reconstruction technique and how it compares to a next-gen equivalent. Despite running with much lower pixel count, DLSS is undoubtedly sharper and delivers more detail and clarity than checkerboard rendering. Transparent elements like hair see checkerboard artefacts on PS4 Pro that are totally gone with DLSS. In motion, this translates to more temporal stability with DLSS, with the subtle flicker seen on the PS4 Pro version completely gone. In general, pixel crawl and popping is also much reduced. DLSS does have a weakness though: certain objects at a distance exhibit particle trails that are not visible on PS4 Pro, nor in native rendering. It's a small blemish and the only negative point in the presentation.

Advertisement

Comparisons between the two techniques are fascinating but the big takeaway is that DLSS image reconstruction from 1440p looks cleaner overall than native resolution rendering. We've reached the point where upscaling is quantifiably cleaner and more detailed - which sounds absurd, but there is an explanation. DLSS replaces temporal anti-aliasing, where all flavours of TAA exhibit softening or ghosting artefacts that Nvidia's AI upscaling has somehow managed to mostly eliminate. And this poses an interesting question: why render at native resolution at all, if image reconstruction is better and cheaper? And what are the applications for next-gen consoles?

Cover image for YouTube videoDeath Stranding PC Tech Review: The Upgrade We've Been Waiting For
Our Death Stranding PC tech review includes analysis of DLSS upscaling up against native 4K rendering.
There's an important point of differentiation between Nvidia's hardware and AMD's, however. The green team is deeply invested in AI acceleration across its entire business and it's investing significantly in die-space on the processor for dedicated AI tasks. AMD has not shared its plans for machine learning support with RDNA 2, and there is some confusion about its implementation in the next-gen consoles. Microsoft has confirmed support for accelerated INT4/INT8 processing for Xbox Series X (for the record, DLSS uses INT8) but Sony has not confirmed ML support for PlayStation 5 nor a clutch of other RDNA 2 features that are present for the next generation Xbox and in PC via DirectX 12 Ultimate support on upcoming AMD products.

Advertisement
Broadly speaking then, the Xbox Series X GPU has around 50 per cent of the RTX 2060's machine learning processing power. A notional DLSS port would see AI upscaling take 5ms to complete, rather than a 2060's circa 2.5ms. That's heavy, but still nowhere near as expensive as generating a full 4K image - and that's assuming that Microsoft isn't working on its own machine learning upscaling solution better suited to console development (spoilers: it is - or at least it was a few years back). In the meantime though, DLSS is the most exciting tech of its type - we're sure to see the technology evolve and for Nvidia to leverage a key hardware/software advantage. The only barrier I can see is its status as a proprietary technology requiring bespoke integration. DLSS only works as long as developers add it to their games, after all.

As exciting as the prospects for machine learning upscaling are, I also expect to see continued development of existing non-ML reconstruction techniques for the next-gen machines - Insomniac's temporal injection technique (as seen in Ratchet and Clank and Marvel's Spider-Man) is tremendous and I'm fascinated to see how this could evolve given access to the PS5's additional horsepower. Maybe the developer is leaning into this to achieve its 60fps mode for Marvel's Spider-Man: Miles Morales? Even with the arrival of Xbox One X and its 'true 4K' marketing - and even the focus on native 4K at the PS5 games reveal - the truth is that the concepts of dynamic resolution scaling, temporal supersampling and perhaps even checkerboarding may well persist into the next generation. Despite the move to next-gen hardware, GPU resources will still be finite - and it's likely that ultra HD will still be more of a 'destination' and the route taken to get there will vary very much on a game by game basis.




Hideo Kojima taking an "arthouse approach" with Death Stranding movie


Death Stranding is getting a movie adaptation
 
Last edited:
There isn't any definitive metrics given about CBR but here's a take on CBR from Johnathan Blow via an old WCCFtech Q&A/interview:

I think it depends on the particular game and engine. Different rendering pipelines are structured differently; for some pipelines, the cost of adding checkerboard rendering would be very low, because they are already computing a lot of the information that checkerboard rendering needs. For other pipelines the cost might be higher. In our case we're just not sure of the total cost yet, but we think it is probably high enough that we may prefer to do a straight upscale. But we're not completely sure.

(It is true that, as Sony has announced, the PS4 Pro provides hardware support for checkerboard rendering that makes it faster than it would otherwise be. I think in some places I have seen the rumor that checkerboard is completely free, but I would consider that an exaggeration: the cost is going to vary per game. Unfortunately due to NDAs I can't provide details; I can't say anything more about Sony technologies than what they have announced. It is definitely true that if you had a game running on the original PS4, and the developer wants to do the most straightforward thing to make the game look better on the Pro, that developer could enable checkerboard rendering and the game will look better and run faster; so it's "free" in that sense. But if you are going to get picky about how you are spending the GPU memory and bandwidth of the new machine, then there are tradeoffs here, like with anything.)


I’m always impressed you can quickly find articles several years old how do you do it
 
These techpowerup benchmarks are screwy sometimes

mgSJZTz.jpg




I think I'm going to try this one on my 3060 PC with 1440p DLSS. Obviously no RT.

Unlike most people it seems I was intrigued by the science fiction story of the first game and didn't mind Aloy being sassy.

Imagine knowing about technology and starships and you're surrounded by fucking pagans worshipping trees. Ofcourse you're angry.
 

Gaiff

SBI’s Resident Gaslighter
So, yeah, 8GB isn't enough for 1440p/Very High. Steve from HU said he tried the game both on a 16GB and 8GB 4060 Ti and the 8GB experienced sucked at 1440p.

vram.png


 

Fredrik

Member
Did he address the hair and fur quality seeming lower than ps5 I assume it’s a bug
It’s a zoom in photo mode difference. I can do A-B switching now and was just now standing 30 cm from the 65 inch TV staring at the screen lol. There is a clear difference when maxing out focal length to get super close, PS5 version looks sharper up close in photo mode, more texture details, easiest seen on the wrinkles on Aloy’s lips, they’re blurred on PC. These differences are in both performance and resolution mode so it’s not about the resolution. I think they use different assets in photo mode and higher fidelity on PS5. I can take photos for proof but I advice people to go play the game instead.

In gameplay the PC version looks better, crisper. PS5 has a smooth enough performance mode but the image quality takes a hit, looks fuzzy, easiest seen on grass and environment textures.
Colors are punchier too on PC but I don’t know what that is about, could be a color settings difference.

It really is a superb port anyhow. Just play it and enjoy is my advice!
 
he can find a specific scene that reports similar resolutions on PS5. dynamic resolution changes betwen scenes and camera motions. if you stand in a place with no NPCs, robots etc. and just stand still, resolution should stabilize. and you can punch in that exact same internal resolution on PC with tools like DLSS tweaks

they're DF, they can find a way to do it. he's just avoiding doing it because it would make 8 GB cards look weak against PS5. and despite being a huge defender of DLSS and ray tracing and stuff and even accepting the compromises, I have a problem with his approach to this

either do whatever you're doing for all games as much as possible or don't do it at all. the second he finds a game where 3070 can have the edge without running into VRAM problems, he will go extremes to make the comparison happen. when reverse happens, he will resort to endless amount of excuses

not to mention DSR being a thing did not stop Rich from comparing 4070 to the PS5 with DSR driven games.
The PS5 runs at 1800p internally in performance mode? The 3070 is fine as long as your output resolution is 1440p in this game which makes sense because the 3070 was never a 4k card.
 

Senua

Member
Its so annoying how they refuse to do performance comparissons vs consoles half the time.
I don't see how, just ignore them if they upset you.

Some nonsense about the PC port of HFW vs the PS5 version. Senua said arguing with Rofif was like arguing with a potato and Rofif answered that the only potato was the one in Senua’s ass.
I'm still scarred, mentally and anally.
 

Gaiff

SBI’s Resident Gaslighter
So, I realized that by far the biggest improvement (teaching no one anything new here) is the texture work in Forbidden West. It's phenomenal and arguably the best in the business. However, things like lighting and object quality don't seem to have been massively improved? Obviously, character facial animations have received a dramatic enhancement, especially secondary or tertiary NPCs. Oh, and can't forget about water physics and reflections.

This is Zero Dawn. Still quite a looker. Just don't talk to any NPCs. Also, HDR is off because Photo Mode for some reason doesn't play well with it and washes out the colors.

QEutexn.jpeg

7W2A83Q.jpeg

pcDzNuO.jpeg
 
Top Bottom