Digital Foundry Article Technical Discussion [2025]

[pcchen]

I'm not sure I fully agree with John's argument about UC4 and static environments, especially as they are inserting clips of a jeep plowing through dozens of movable physics objects while he's explaining it. Dynamic environments and objects don't need RT. RT just allows you to have movable objects with closer to baked lighting quality.

There are games today and in the past that have quite a bit of interactivity without RT. Astro Bot and TOTK are two modern examples. TOTK even has a crude system for dynamic diffuse GI and reflections on very low-end hardware. Fortnite is maybe the best example of what he's saying, and it's dynamicism is likely why Epic moved in the direction of Luman. But games like Fortnite are the exception.

Imo, I think the reason people are unimpressed is because realtime RT lighting brings with it heavy processing requirements, noise, and low resolutions on common hardware, all for marginal visual gains if you already had good baked lighting. Plus, many games have no real increase in interactivity. (Silent Hill 2 Remake comes to mind) Basically, many games are filled with new artifacts that weren't there before, and many people often couldn't see the artifacts in the old rasterized way in the first place; especially if they are comparing the best rasterized examples from the past.

Of course RT is the future, and has been the future since programmable shading came out in 2000, but perhaps it's been pushed too hard when most people really don't have the raw GPU power to make it look good yet. Moore's Law slowing to a crawl just makes that worse.

Isn't that his point entirely? When you have some restrictions, it's possible to make some impressive visual effects. The point of RT is to remove more of these restrictions, however it'd cost more.
For example, I still remember seeing the first Little Big Planet game. It's lighting is so good it almost feels pre-render quality. However, the restriction is that it's a semi-2D game (the camera can't move much more the fixed angle). Many tricks were applied to make it looks very good, with the condition that the camera can't move much.

Saying that RT has been "pushed too hard" is a bit like saying "3D gaming was pushed too hard" in the era of Virtual Fighter. RT hardware was out for 6 years. It's about time people should do something with it.

 

[trinibwoy]

Imo, I think the reason people are unimpressed is because realtime RT lighting brings with it heavy processing requirements, noise, and low resolutions on common hardware, all for marginal visual gains if you already had good baked lighting. Plus, many games have no real increase in interactivity.

Good baked lighting requires you to design your game around static environments so it’s a self fulfilling prophecy. Real time lighting opens up more possibilities in game design. Sure games that are inherently static don’t benefit from RT but that’s not really the point.

 

[GhostofWar]

Monster Hunter Wilds might be the worst reviewed game on steam with the biggest ccu. ~53% positive currently with ~1.3 million ccu. This isn't how we get pc game performance issues taken seriously, I can't see that success making Capcom try harder next time or even try fix MHW promptly if at all.

 

[homerdog]

Isn't that his point entirely? When you have some restrictions, it's possible to make some impressive visual effects. The point of RT is to remove more of these restrictions, however it'd cost more.
For example, I still remember seeing the first Little Big Planet game. It's lighting is so good it almost feels pre-render quality. However, the restriction is that it's a semi-2D game (the camera can't move much more the fixed angle). Many tricks were applied to make it looks very good, with the condition that the camera can't move much.

Saying that RT has been "pushed too hard" is a bit like saying "3D gaming was pushed too hard" in the era of Virtual Fighter. RT hardware was out for 6 years. It's about time people should do something with it.

The slowing of progress has made RT feel forever out of reach for many gamers. Compare 8800GTX to GTX560. The 560 is way faster. Now compare RTX2080Ti to RTX4060. The 4060 is still slower than 2080Ti. Half a decade of progress doesn't look like it used to.

 

[pcchen]

The slowing of progress has made RT feel forever out of reach for many gamers. Compare 8800GTX to GTX560. The 560 is way faster. Now compare RTX2080Ti to RTX4060. The 4060 is still slower than 2080Ti. Half a decade of progress doesn't look like it used to.

But this has nothing to do with RT. The slowing down of progress is universal. It's not because NVIDIA put more efforts on RT while neglected on traditional raster performance. The evidence is that AMD was having the same problem.
I think blaming poor performance on RT is IMHO misguided. Today many games perform pretty poor even with RT disabled, or they have RT as an afterthought. The problem really lies elsewhere.

 

[homerdog]

But this has nothing to do with RT. The slowing down of progress is universal. It's not because NVIDIA put more efforts on RT while neglected on traditional raster performance. The evidence is that AMD was having the same problem.
I think blaming poor performance on RT is IMHO misguided. Today many games perform pretty poor even with RT disabled, or they have RT as an afterthought. The problem really lies elsewhere.

I know but it gives the impression to the average user that RT is pointless without a quite expensive GPU. This reality hasn't changed much in the last 5 years. This incredible new tech (that is very performance intensive) comes along right when gen-on-gen improvements slow way down and game optimization takes a dive. That's not the fault of RT but it's the reality we have to deal with. According to the Steam survey, the majority of PC gamers won't be able to use much RT in modern games.

I bet RT would be viewed very differently if general performance (and more importantly perf/$) improvements had maintained their pace from 10-15 years ago.

 

[Remij]

Did Digital Foundry, or more specifically Alex ever take a look at Spider-Man 2 on PC? My god this port is trash. Nixxes should be ashamed of releasing a game like this.

I simply can't believe how much this game stutters on camera cuts. Is this what we have to look forward to with DirectStorage?

This has to be the worst Nixxes port yet.

 

[Flappy Pannus]

Did Digital Foundry, or more specifically Alex ever take a look at Spider-Man 2 on PC? My god this port is trash. Nixxes should be ashamed of releasing a game like this.

I simply can't believe how much this game stutters on camera cuts. Is this what we have to look forward to with DirectStorage?

This has to be the worst Nixxes port yet.

Nah not yet, huge backlog but eventually. He just looked at it on a DF Direct briefly so far.

 

[Remij]

Nah not yet, huge backlog but eventually. He just looked at it on a DF Direct briefly so far.

Oh ok.

Hopefully they cover it and hold them to task.

 

[Lurkmass]

I think there's enough sources out there now that Digital Foundry should look specifically at DirectStorage with GPU decompression and call out how sporadic frametimes are when simply turning the camera. This is greatly exacerbated with kb/m controls. You can tank the framerates by just looking around quickly in games like Ratchet and Clank, Spider-Man 2, and a couple others I know of but am forgetting offhand. It's terrible.

Nvidia specifically seems to be affected more disproportionately compared to AMD, for whichever reasons.. but its common enough now with these games that use GPU decompression that it doesn't seem like a developer specific implementation issue.

I think devs should stick to CPU DirectStorage support until this is either solved or there's some new solution (neural decomp I guess)

Did Digital Foundry, or more specifically Alex ever take a look at Spider-Man 2 on PC? My god this port is trash. Nixxes should be ashamed of releasing a game like this.

I simply can't believe how much this game stutters on camera cuts. Is this what we have to look forward to with DirectStorage?

This has to be the worst Nixxes port yet.

If a particular IHV has a problem with a said specific desirable feature then it's more constructive for the hardware vendor to improve their future hardware design around it!

 

[Remij]

If a particular IHV has a problem with a said specific desirable feature then it's more constructive for the hardware vendor to improve their future hardware design around it!

That doesn't help anyone who already owns cards... and this was never a stated drawback of DirectStorage when they marketed it to us. So whoever needs to fix it needs to fix it.

 

[Lurkmass]

That doesn't help anyone who already owns cards... and this was never a stated drawback of DirectStorage when they marketed it to us. So whoever needs to fix it needs to fix it.

After recent developments, it's obvious to anyone else that it's the developer's job to bully and threaten hardware vendors out of improving their implementation some more for features that they do want to use. Ironic that the vendor who originally proposed the GDeflate algorithm (backbone of DirectStorage) are having the most visible purported issues with it ...

 

[Remij]

After recent developments, it's obvious to anyone else that it's the developer's job to bully and threaten hardware vendors out of improving their implementation some more for features that they do want to use. Ironic that the vendor who originally proposed the GDeflate algorithm (backbone of DirectStorage) are having the most visible purported issues with it ...

I don't care who needs to fix it.. or who's fault it is... it needs to be fixed... and not just for future hardware.. for current hardware that's supported.

 

[Lurkmass]

I don't care who needs to fix it.. or who's fault it is... it needs to be fixed... and not just for future hardware.. for current hardware that's supported.

Well the solution is to just get better hardware because what AAA game developer even in their own world wants their final release of ever becoming more complex projects to take up more storage space and higher memory consumption ?

The developer's stance is that unsuitable hardware should just rot in eternity ...

 

[Remij]

Well the solution is to just get better hardware because what AAA game developer even in their own world wants their final release of ever becoming more complex projects to take up more storage space and higher memory consumption ?

The developer's stance is that unsuitable hardware should just rot in eternity ...

No, that's not the solution. You can't get better hardware than what I have.

 

[Lurkmass]

No, that's not the solution. You can't get better hardware than what I have.

Really ? If your "better hardware" isn't delivering a superior 'experience' then is it really better hardware ?

If there's a strong relationship in an issue affecting customers then developers are just going to devolve into pointing fingers at the source of that problem rather than do any self reflection ...

 

[DavidGraham]

If there's a strong relationship in an issue affecting customers then developers are just going to devolve into pointing fingers at the source of that problem rather than do any self reflection ...

Whose developers? This is an unstable buggy game that asks the player on launch to enable frame generation to compensate for it's horrendous optimization issues and glaring visual bugs. When a game that looks like ass performs like a path traced game and asks the player to enable frame generation then a proper self reflection is 100% warranted.

 

[MystWalker]

Isn't that his point entirely? When you have some restrictions, it's possible to make some impressive visual effects. The point of RT is to remove more of these restrictions, however it'd cost more.
For example, I still remember seeing the first Little Big Planet game. It's lighting is so good it almost feels pre-render quality. However, the restriction is that it's a semi-2D game (the camera can't move much more the fixed angle). Many tricks were applied to make it looks very good, with the condition that the camera can't move much.

Saying that RT has been "pushed too hard" is a bit like saying "3D gaming was pushed too hard" in the era of Virtual Fighter. RT hardware was out for 6 years. It's about time people should do something with it.

Good baked lighting requires you to design your game around static environments so it’s a self fulfilling prophecy. Real time lighting opens up more possibilities in game design. Sure games that are inherently static don’t benefit from RT but that’s not really the point.

He's framing it as if RT is a necessary step to get us back to more games with greater interactivity, when interactivity is a dev choice, separate from RT. Will a game's lighting be as high quality if it has a very dynamic world with no RT? No, of course not. But there are different types of precomputed lighting that give varying degrees of interactivity, from lightmaps to irradiance volumes to PRT. There's also methods of realtime lighting that are not hardware RT like SVOGI, real time cubemaps, etc. These all have their own drawbacks, but it's not like updating a BVH with lots of dynamic objects is particularly viable right now either. The point is, there's lighter weight options that still can look "good" and not require a game to be noisy and run at 720p internal.

I think devs need to back off the throttle on RT a bit, and look into other methods of achieving their goals that are better fitted to the hardware we have now in mainsteam PCs and consoles. Maybe that's "smart" pre-computation, rasterization based techniques, or just lighter weight RT. (For example The Finals and Indy's use of RT probe GI. Both games run well at acceptable resolutions on common hardware)

Unlike past transitionary times in real time 3d graphics, we can't just expect more hardware to save us. The progress of performance is too slow now, so devs need to be more intentional in the choices they make and how they use the hardware we have.

But this has nothing to do with RT. The slowing down of progress is universal. It's not because NVIDIA put more efforts on RT while neglected on traditional raster performance. The evidence is that AMD was having the same problem.
I think blaming poor performance on RT is IMHO misguided. Today many games perform pretty poor even with RT disabled, or they have RT as an afterthought. The problem really lies elsewhere.

This is true as well. UE5 games tend to be heavy regardless, but we know Nanite and Luman are expensive techniques. Alan Wake II is also very heavy even when not using RT, and it uses mostly baked lighting. (Maybe it's geometry overdraw?) I don't have the answers, it's getting into a different topic, but I think what I said above applies here too.

 

[DavidGraham]

The point is, there's lighter weight options that still can look "good" and not require a game to be noisy and run at 720p internal.

720p internal is not usually caused by ray tracing, 90% of the time it's caused by raster games. Like the recently released Monster Hunter Worlds. There is also A Plague Tale: Requiem, Gotham Knights, Star Wars Jedi Survivor and Forspoken (Performance mode doesn't support ray tracing yet runs at 720p internal), or the usual UE5 suspects, like Immortals of Aveum, Avowed, Black Myth Wukong .. etc.

 

[Lurkmass]

Whose developers? This is an unstable buggy game that asks the player on launch to enable frame generation to compensate for it's horrendous optimization issues and glaring visual bugs. When a game that looks like ass performs like a path traced game and asks the player to enable frame generation then a proper self reflection is 100% warranted.

Weak hardware or subpar implementations aren't the developer's problems anymore ever since PC consumers have continuously normalized the trend of performance destructive technologies outpacing the growth of hardware capabilities than what would've been naturally possible. If every game in the future is going to come with the expectation that they'll be released with hacks like both upscaling and frame generation then it'll happen and developers will have more important things to attend to rather than catering to a lower common denominator ...

If you you think some developer making use of a said technology are the problem then it's better if someone else breaks the bad news to you since every big publisher has used it before and it doesn't look like interest will wane anytime soon because they have genuine admiration for the concept to keep using it ...