Current Generation Games Analysis Technical Discussion [2023] [XBSX|S, PS5, PC]
- Thread starter BRiT
- Start date
- Status
I'm sorry, but the path tracing implementation in Alan Wake 2 is one of the worst implementations I've seen of Raytracing, period (yes even worse than those low effort AMD ones, atleast they don't hammer performance). Especially after watching Alex' video.
First, path traced lighting does not make sense in a static title like this where every corridor has pre-baked lighting made by the developers. Thus, the differences are minimal most of the time and the performance hit is abysmal (110 fps to 31 fps or something on a 3080? Oof). Even in side by sides the differences are hard to tell. My buddy has an RTX 3080 btw and even with DLSS at 1440p and optimized settings, he was not able to hold 60 FPS in the later sections of the game with Raytracing on low.
Reflections do make quite a difference sometimes, however in Alan Wake 2 they are locked behind expensive path tracing, you can only enable transparencies seperated from PT lighting. This basically makes RT unplayable for Turing and Ampere users. Why didn't they include an efficient Raytraced reflection system like the one they had in Control? Makes zero sense to me.
First, path traced lighting does not make sense in a static title like this where every corridor has pre-baked lighting made by the developers. Thus, the differences are minimal most of the time and the performance hit is abysmal (110 fps to 31 fps or something on a 3080? Oof). Even in side by sides the differences are hard to tell. My buddy has an RTX 3080 btw and even with DLSS at 1440p and optimized settings, he was not able to hold 60 FPS in the later sections of the game with Raytracing on low.
Reflections do make quite a difference sometimes, however in Alan Wake 2 they are locked behind expensive path tracing, you can only enable transparencies seperated from PT lighting. This basically makes RT unplayable for Turing and Ampere users. Why didn't they include an efficient Raytraced reflection system like the one they had in Control? Makes zero sense to me.
Last edited:
I play everything in the maximum settings. For me, the visual advantages are clearly visible. They're not as big as in Cyberpunk 2077 but everything still looks better embedded in the world. There are non artifacts due to screenspace effects. At maximum settings Alan Wake 2 looks very polished. If you want almost flawless and polished visuals in a videogame then you have to play Alan Wake 2 on the maximum settings.
I need to have another look at wooden chairs etc.
Good observation and thank you for emphasizing it again here. Reading this point after watching your video makes it easier to internalize. In the video it was shown on the wooden door which was matt without pathtracing and glossy with pathtracing. Enabling pathtracing makes materials more realistic.
I need to have another look at wooden chairs etc.
Last edited:
Don't think it would look worse. However, it would appear different in certain areas - it could be darker or lighter. Since the game is already darkly lit in many places, such changes can affect gameplay.
I am sure they wanted to keep the lighting from becoming darker or brighter to prevent any major reworking of the game maps' lighting.
Regarding the use of baked GI for additional bounces or improved stability, one would typically expect the low-resolution baked GI to act as a cache at the hit points of the per-pixel RT GI, rather than being directly visible.
However, this is not the case in the game, likely for the reasons I highlighted above.
Yep, you can't bake glossy GI as it's view dependent, so it doesn't contribute to lighting in games like TLOU.
Last edited:
If lights are static, you can bake specular information just as you can diffuse such as in spherical light probes. TLoU was matte from an art choice I think, avoiding photorealism. IIRC ND said they went with a graphic novel look for Uncharted and TLoU is an extension of that.
This is one of the strangest takes I've seen on the game. I'm not sure what side by sides you're comparing but there are definitely huge differences between RT on and RT off which I'm not sure how anyone could fail to notice. For example the terrible flickering shadows that are completely resolved with RT, or the vast reduction in image noise, or as you state below, the much improved reflections which can completely transform the look of materials.
Alex's optimised settings address this. You can get around 60fps with RT Direct lighting on, but PT Indirect lighting off at DLSS Balanced. Dropping that to DLSS performance mode should give a pretty solid 60fps.
Turing and Ampere users can either use the DLSS Ultra Performance mode (which works pretty well in this game) or just stick with 30fps. I know it's frowned upon in PCMR circles to accept anything less than 60fps, but if it's good enough for the millions of console gamers that will be picking this title up then it should be good enough for owners of half decade old Turing GPU's or lower end 3 year old Ampere GPU's. Personally I celebrate the option to ramp settings up beyond the limits of 2 generation old GPU's.
The Ps3/360 had a lot of sub 720p games yet that generation pushed things that the current generation can’t touch. Physics, environmental interactivity, destruction, massive online multiplayer, etc. People were raving about 100 player battle royals meanwhile the ps3 had Mag with 256 players online. Many of the best games with those features come from that era.
I don’t have high hopes for UE5 at all. For me, unreal engine has gotten worse for gaming not better over time. UE3>UE4>UE5. While the newer engines have more features, they have more problems ranging all performances centric. It’s become more bloated and less performant. It’s an engine with a lot of tech debt.
If devs look to these technologies as the first step in their optimization process, they’ve completely lost the plot. As we can see in the more recent releases, many devs have completely lost the plot. Releasing games at ps360 resolutions with very little world interactivity and poor physics is beyond unacceptable. Especially when those compromises are made for some bad checkerboard/low quality rt “effects”.
Last edited:
Those flickering shadows are minor differences, not major. Most of the times, the differences are just visible in cherry picked scenes. And they are certainly not worth the heavy performance loss.
This is the difference you'll see most of the time, and personally I don't care about flickering shadows. If the overall visuals are not improved signifcantly in every scene for THAT performance loss, then it's not worth it. There's an entire video by DF where you can see the differences between maxed out PC and console versions are minor.
Just an absolutely horrible visuals to performance ratio that is completely inacceptable. Same as Ultra Performance mode or 30 FPS on a high end Ampere GPU.
And now compare that to the RTGI found in Metro Exodus Enhanced Edition which transforms EVERY scene to a point where everything looks completely different and runs at 1080p 60 FPS on an RTX 2060. This is what I call efficient. (Before you comment, I'm aware Alan Wake 2 is doing a lot more than Metro and is just performance intense in general, I'm speaking about the raytracing solution here with a comparison point of the older game without RT and the Enhanced Edition)
Why do we just accept how Alan Wake 2 performs with Raytracing? They could easily have built in a hybrid Raytracing mode with reflections as well as some hybrid RT shadows and I guarantee you it would look almost the same as the maxed out path tracing version for a fraction of the performance cost.
And yes, my friend was already using optimized settings and aggressive DLSS presets. 60 FPS was not possible near the end of the game with any sort of Raytracing enabled. The forest from the beginning is not the heaviest scene at all and he played through the entire game.
The game was absolutely not made with RT in mind and you can clearly see that. It was made for the consoles and Nvidia later tacked on a path tracing solution, that only runs well enough on Lovelace (and even that struggles a lot).
Normally I'm a huge fan of any sort of Raytracing, but in this game my clear recommendation is to turn everything off. It's clearly not worth it.
Last edited:
For you the shadows may not make much of a difference but for me the path tracing shadows contribute enormously to image stability. I was already playing The Division with HFTS hybdrid raytracing shadows back then.
The loss of performance may not be worth it for you but it is for others. In the future and with the release of even faster GPUs many more players will enjoy this game with path tracing. These high end settings helping to make it well equipped for the future. In addition. If the game hardly looks any better with path tracing why does it bother you if it runs so badly with path tracing? Then you can just play without it and as you say you have the same visuals.
Why should the performance to visuals ratio be terrible when it runs slow but almost no game in the industry even comes close to the game graphics? By the way the performance in the Alan Wake NYC sections is not that bad either.
I would be careful with such claims. They were one of the first studios to show ray tracing in tech demos and Control already used raytracing heavily. Remedy praised raytracing back then so it's highly likely that they had raytracing in mind.
Frenetic Pony
Veteran
Remedy loves pushing things as much as possible, but the difference between Control and Alan Wake II is primarily in other parts of the rendering pipeline. They cleaned up their horrid image quality issues, moved from forward shading to deferred/mesh shaders, stuff like that. The lighting is otherwise the same as Control's SDF/screenspace reflections/probe grid GI. The RT/Pathtracing was more for fun to see what they could do.
For a future title we'll likely see post processing cleaned up, it costs way too much per pixel for what it does, and maybe then we'll see an overhaul of lighting moving to all realtime as well. Considering Remedy have been at the forefront of lighting for a good while it'll be interesting to see, I'd love to see more than 3 bounces, but the pathtracing here is as much a tech demo as anything.
The path tracing looks so good, and all of the content is tuned for it, I think it was a major priority. If you have a fully featured PT renderer and a fully featured "traditional" renderer (with a similar featureset) and you target both for launch, the image just isn't going to look as dramatically different as if you go back through the game a year after release and change all of the lighting.
There are a couple of posters looking for absolutely any excuse to hate on devs. If your game scales badly too low end, it's unoptomized -- if your game scales well to low-end, the high endisn't worth the cost. If you include baked lighting, it's fake PT -- if you PT everything, it's a diaster, poorly optimized, doesn't run at 180fps, etc. Some of the very same posters flipping their lids about software nanite seem to think that only an idiot would switch from remedy's voxel/sdf based gi to the RT gi, or that anything more expensive than a coarse probe based solution is terrible engineering...
Last edited:
DavidGraham
Veteran
A detailed analysis of why Cities Skyline 2 performs so bad, in short it's due to Unity, poorly coded geometry culling schemes, and shadows that are horribly optimized.
40ms for poor looking cascaded shadow maps, yikes.
As a fun aside, from reading the article, I spent some time reading through MS's documentation on CSMs and found it neat that their official docs point to a Beyond3d thread! Alas, the link is from before the redesign and is now broken. I sent them an email; hopefully they fix it.
I'm assuming this is the new version of the thread they were trying to link to: https://forum.beyond3d.com/threads/layered-variance-shadow-maps.42174/
Yah, I don’t really understand about the recent uproar against upscaling. I guess it’s mostly pc players that can no longer buy their way out of upscaling.
I do think FSR looks bad and would like to see more alternatives. When I’ve tried UE5 TSR I’ve thought it looked okay but for whatever reason it’s not being adopted. Unless we all buy 1500 Watt pcs it seems like hardware scaling won’t be a viable solution and we’ll need continuous improvements to upscaling solutions.
The comparisions between "PS5 optimized" and full Raytracing makes no sense. Raytracing replaces most part of the lighting systems. So going from reduced settings to Raytracing will always increase the processing...
Did a comparision in Alan Wake 2 with everything on and without/with Pathtracing@low + denoiser@low + transparency@high:
At least on Lovelace the indirect Pathtracing is very fast. ~20% performance lost for GI, reflections and AO is really good.
Hardware Raytracing has worked from day 1. Within one year games like BF5, Metro Exodus and Control had been relased, within 2 1/2 years a fully real time multi bounce GI solution had been relased by 4A Games with Metro Exodus EE and within 4 1/2 years we got real time Pathtracing.
The pace of hardware Raytracing is incredible. "Deliver us the Moon" was the first UE4 game released with hardware Raytracing support for reflections earlier 2020. You got something like this:
This here is state of the art in UE5 games (Robocop) released nearly 4 years later:
So, i'm open for your arguments why nVidia is responsible for this regression in image quality.
Did a comparision in Alan Wake 2 with everything on and without/with Pathtracing@low + denoiser@low + transparency@high:
At least on Lovelace the indirect Pathtracing is very fast. ~20% performance lost for GI, reflections and AO is really good.
Blaming 3rd parties for the situation with UE5 is a weird way to defend Epic.
Hardware Raytracing has worked from day 1. Within one year games like BF5, Metro Exodus and Control had been relased, within 2 1/2 years a fully real time multi bounce GI solution had been relased by 4A Games with Metro Exodus EE and within 4 1/2 years we got real time Pathtracing.
The pace of hardware Raytracing is incredible. "Deliver us the Moon" was the first UE4 game released with hardware Raytracing support for reflections earlier 2020. You got something like this:
This here is state of the art in UE5 games (Robocop) released nearly 4 years later:
So, i'm open for your arguments why nVidia is responsible for this regression in image quality.
Last edited:
Lightmap has a resolution limit. Especially on complex geometries like static vegetations which have massive surface areas. On lower end platform titles like Metroid Prime remastered on Switch, you can clearly spot blocky light maps on larger scenes.
Plus the development efficiency is extremely low due to the low baking process. Probe baking is a lot faster, but the resolution is heavily reduced.
And with all these cons, you can get the diffuse lighting kinda correct, any view dependent bxdf cannot be baked (specular lighting, more complex diffuse materials)
I see direct lighting and specular occlusion, but not large-scale GI shadows. You need to toggle direct lighting on/off to be sure that it's indirect.
In the same hotel there's a wooden beam which casts a jagged shadow which doesn't change when you toggle PT on/off, as if it's baked in a lightmap. It doesn't change when you toggle direct lighting. Notice also that character doesn't cast any indirect shadow in the same direction as static objects, even though that beam shadow has lots of definition.
When you look from another angle at it you can see character shadow on the wooden beam, but in a completely different direction, suggesting that it's done through some different technique.
When you toggle all GI, Reflection and PT setting off there's still the same character shadow, as if it would be some capsule traced shadow using direction from a lightmap:
You can see it in the first pair of screenshots I posted earlier (location: staircase to the morgue). You will also find out that, in the same location, the sun position for in-game lighting does not match the prebaked lighting precisely.
Also, you're confusing direct and indirect lighting. Direct lighting would be under the direct sun rays, outside of the windows in the screenshots, while the indirect lighting is the light that has bounced off other surfaces outside and then enters the room through the windows. There are no other light sources besides the windows in the spots where I took these screenshots.
I checked this in the diner area. It was the GI option that produced shadows for all the items on the walls. Can't say anything about the other scenes, as I checked them with RT turned on/off only, and I am currently playing the Alan part of the game. However, it would be clever to approximate GI by using area lights in place of windows (considering the window in GI essentially functions as the same area light). This would result in similar visuals, yet it would be of higher quality with fewer samples and less noise.
Last edited:
Showcasing new Northlight technology in Alan Wake 2
www.remedygames.com
How Northlight makes Alan Wake 2 shine
Showcasing new Northlight technology in Alan Wake 2
www.remedygames.com
- Status
Similar threads
