BenchmarKing's banner

BenchmarKing

@BMKing23 • 6,269 subscribers

▶️YouTube Content Creator | +150K Subs. Technical Reviews, Optimization Guides & Graphics Settings Comparisons to help you get the best performance and visuals.

Shorts

Looks like #ForzaHorizon6 changes the DLSS preset depending on the mode: • Ultra Performance → Preset L • Performance → Preset M • Balanced & Quality → Preset K If you’re on an RTX 20 or 30 series card, you might notice that DLSS Performance runs almost the same as or even worse than DLSS Quality. I think it would’ve been better if the game offers an option to manually switch between DLSS 4 and 4.5. It’s also a bit disappointing that they dropped MSAA support in this one.

Looks like #ForzaHorizon6 changes the DLSS preset depending on the mode: • Ultra Performance → Preset L • Performance → Preset M • Balanced & Quality → Preset K If you’re on an RTX 20 or 30 series card, you might notice that DLSS Performance runs almost the same as or even worse than DLSS Quality. I think it would’ve been better if the game offers an option to manually switch between DLSS 4 and 4.5. It’s also a bit disappointing that they dropped MSAA support in this one.

75,149 görüntüleme

If the shadow noise is bothering you, try using PCF instead of PCSS that should get rid of it. but you’ll lose the soft shadows and start noticing more pixelated shadows particularly on multiplayer maps from what I’ve seen, this setting doesn’t work in the campaign. #Battlefield6

If the shadow noise is bothering you, try using PCF instead of PCSS that should get rid of it. but you’ll lose the soft shadows and start noticing more pixelated shadows particularly on multiplayer maps from what I’ve seen, this setting doesn’t work in the campaign. #Battlefield6

137,155 görüntüleme

It looks like the game only renders correctly when DLSS is enabled and the internal resolution is below 1080p. Above that threshold, a lighting pass is missing. This explains why DLSS doesn’t perform as expected at 1440p or lower, since performance is impacted by the extra lighting pass. It also matches why Digital Foundry saw a bigger boost, since they tested DLSS at 4K. and It explains why DLAA looks worse than DLSS Quality at 1440p or below. Big thanks to ハイ／🌼 and yeradd for the info.

It looks like the game only renders correctly when DLSS is enabled and the internal resolution is below 1080p. Above that threshold, a lighting pass is missing. This explains why DLSS doesn’t perform as expected at 1440p or lower, since performance is impacted by the extra lighting pass. It also matches why Digital Foundry saw a bigger boost, since they tested DLSS at 4K. and It explains why DLAA looks worse than DLSS Quality at 1440p or below. Big thanks to ハイ／🌼 and yeradd for the info.

14,420 görüntüleme

#WhereWindsMeet DirectX 11 vs DirectX 12. DX11 is a bit faster and delivers more consistent frame times, at least on the 4070 Super and 3060 Ti. However, The main drawback of using DX11 is the lack of DLSS and XeSS support, both upscalers are only available when using DX12.

#WhereWindsMeet DirectX 11 vs DirectX 12. DX11 is a bit faster and delivers more consistent frame times, at least on the 4070 Super and 3060 Ti. However, The main drawback of using DX11 is the lack of DLSS and XeSS support, both upscalers are only available when using DX12.

42,577 görüntüleme

Some replies on my previous post mentioned it’s normal for performance to drop when there are lots of alpha effects/particles on screen. But in this broken game, you don’t need a lot, just a single roll can tank the FPS even on a 4070 super. 1/2 #MGSDelta

Some replies on my previous post mentioned it’s normal for performance to drop when there are lots of alpha effects/particles on screen. But in this broken game, you don’t need a lot, just a single roll can tank the FPS even on a 4070 super. 1/2 #MGSDelta

52,605 görüntüleme

Looks like Ray Tracing in #DyingLightTheBeast occupies the Ultra slot for each setting (Global Illumination, Shadows and Reflections Quality) and it’s basically just an on/off toggle. Pretty disappointing, I was hoping for more options and better control over RT quality.

Looks like Ray Tracing in #DyingLightTheBeast occupies the Ultra slot for each setting (Global Illumination, Shadows and Reflections Quality) and it’s basically just an on/off toggle. Pretty disappointing, I was hoping for more options and better control over RT quality.

11,128 görüntüleme

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Ray Tracing modes in #AssassinsCreed Black Flag Resynced: Off: Precomputed Lighting. Standard: Ray Traced Global Illumination (RTGI). Extended: RTGI + Ray Traced Reflections.

Ray Tracing modes in #AssassinsCreed Black Flag Resynced: Off: Precomputed Lighting. Standard: Ray Traced Global Illumination (RTGI). Extended: RTGI + Ray Traced Reflections.

107,818 görüntüleme • 8 gün önce

So far, my impressions of #AssassinsCreed Black Flag Resynced are positive, but there are a few annoyances that I hope Ubisoft addresses in the future. ✅The game takes full advantage of the Anvil engine's latest features (RTGI, RT Reflections, Micropolygon Geometry, etc.), and it looks great. ✅Performance is stable, and the frame time graph stays mostly flat, even in dense, crowded areas like Havana. ⚠️Cutscenes are locked to 30 FPS with no in-game option to remove the cap. This makes the transition between cutscenes and gameplay feel jarring. ⚠️The user interface experience is rough. The game offers a lot of graphics settings and options, but most of them require either a full restart or a save reload to apply, even post processing effects.

So far, my impressions of #AssassinsCreed Black Flag Resynced are positive, but there are a few annoyances that I hope Ubisoft addresses in the future. ✅The game takes full advantage of the Anvil engine's latest features (RTGI, RT Reflections, Micropolygon Geometry, etc.), and it looks great. ✅Performance is stable, and the frame time graph stays mostly flat, even in dense, crowded areas like Havana. ⚠️Cutscenes are locked to 30 FPS with no in-game option to remove the cap. This makes the transition between cutscenes and gameplay feel jarring. ⚠️The user interface experience is rough. The game offers a lot of graphics settings and options, but most of them require either a full restart or a save reload to apply, even post processing effects.

72,159 görüntüleme • 11 gün önce

There are two things worth pointing out about upscaling in #AssassinCreed Black Flag Resynced. 1. Incorrect resolution scaling with Balanced and Performance modes: Using Balanced or Performance with any upscaler causes the game's Resolution Scale (output resolution) to drop, which also lowers the internal render resolution. As a result, these modes apply incorrect scaling and the image quality is noticeably worse than it should be. For now, the workaround is to set the Resolution Scale to 100% and manually adjust the Max Dynamic Resolution (internal resolution) to match the expected scaling for Balanced or Performance mode. 2. The game defaults to the older DLSS 3 CNN model: By default, the game uses the older DLSS 3 Preset E. While it's still a fine option, it doesn't match the image quality of the DLSS 4 Transformer model (Preset K) or the newer DLSS 4.5 (Preset L and M). So I highly recommend forcing at least Preset K through the NVIDIA App (or another tool). Even on RTX 20/30 series GPUs, with Preset K, you can maintain the same performance as before while getting better image quality by using a lower upscaling mode, like Balanced instead of Quality. Hopefully Ubisoft Support fixes the scaling issue and adds an in-game option to choose between DLSS 3, DLSS 4, and DLSS 4.5.

There are two things worth pointing out about upscaling in #AssassinCreed Black Flag Resynced. 1. Incorrect resolution scaling with Balanced and Performance modes: Using Balanced or Performance with any upscaler causes the game's Resolution Scale (output resolution) to drop, which also lowers the internal render resolution. As a result, these modes apply incorrect scaling and the image quality is noticeably worse than it should be. For now, the workaround is to set the Resolution Scale to 100% and manually adjust the Max Dynamic Resolution (internal resolution) to match the expected scaling for Balanced or Performance mode. 2. The game defaults to the older DLSS 3 CNN model: By default, the game uses the older DLSS 3 Preset E. While it's still a fine option, it doesn't match the image quality of the DLSS 4 Transformer model (Preset K) or the newer DLSS 4.5 (Preset L and M). So I highly recommend forcing at least Preset K through the NVIDIA App (or another tool). Even on RTX 20/30 series GPUs, with Preset K, you can maintain the same performance as before while getting better image quality by using a lower upscaling mode, like Balanced instead of Quality. Hopefully Ubisoft Support fixes the scaling issue and adds an in-game option to choose between DLSS 3, DLSS 4, and DLSS 4.5.

38,242 görüntüleme • 10 gün önce

Ray Tracing vs Path Tracing in #PRAGMATA . Path Tracing looks stunning, no doubt, but do we really need it for basic effects like casting a shadow or adding shading beneath the car? Or to properly shade characters’ faces? I think the answer is a clear NO.

Ray Tracing vs Path Tracing in #PRAGMATA . Path Tracing looks stunning, no doubt, but do we really need it for basic effects like casting a shadow or adding shading beneath the car? Or to properly shade characters’ faces? I think the answer is a clear NO.

175,695 görüntüleme • 3 ay önce

I've tested several scenes on the 3060 Ti, and DLSS is still nowhere near FSR 3 in terms of performance. What's even weirder is that DLSS Quality seems to handle certain elements better than DLAA. that's a pretty clear sign that something is wrong. #007FirstLight

I've tested several scenes on the 3060 Ti, and DLSS is still nowhere near FSR 3 in terms of performance. What's even weirder is that DLSS Quality seems to handle certain elements better than DLAA. that's a pretty clear sign that something is wrong. #007FirstLight

60,380 görüntüleme • 1 ay önce

One of the most impressive and often overlooked visual highlights of #PRAGMATA is its hair rendering. While the RE4 Remake featured strand based hair, PRAGMATA's implementation looks significantly better and doesn’t come with a large performance hit, around 10–12% cost.

One of the most impressive and often overlooked visual highlights of #PRAGMATA is its hair rendering. While the RE4 Remake featured strand based hair, PRAGMATA's implementation looks significantly better and doesn’t come with a large performance hit, around 10–12% cost.

318,156 görüntüleme • 7 ay önce

#Borderlands4 on the Badass preset vs Cyberpunk 2077 with full Path Tracing. Both at 1440p DLSS 4 Quality on the RTX 4070 Super. And guess what? Performance is almost the same between them. And I suspect this applies to most recent UE5 games too. Absolutely ridiculous.

#Borderlands4 on the Badass preset vs Cyberpunk 2077 with full Path Tracing. Both at 1440p DLSS 4 Quality on the RTX 4070 Super. And guess what? Performance is almost the same between them. And I suspect this applies to most recent UE5 games too. Absolutely ridiculous.

362,799 görüntüleme • 10 ay önce

#PRAGMATA Demo. Ray Tracing Off vs On. Ray Tracing in this game is quite efficient and doesn’t cost much but at the same time its quality is not particularly high. At native 1440p, there is noticeable noise in RT reflections. The game could definitely benefit from additional RT quality settings, or support for DLSS Ray Reconstruction and FSR Ray Regeneration.

#PRAGMATA Demo. Ray Tracing Off vs On. Ray Tracing in this game is quite efficient and doesn’t cost much but at the same time its quality is not particularly high. At native 1440p, there is noticeable noise in RT reflections. The game could definitely benefit from additional RT quality settings, or support for DLSS Ray Reconstruction and FSR Ray Regeneration.

194,710 görüntüleme • 7 ay önce

I’m working on a video for #ArcRaiders Update 1.7.0, and I noticed that the frame times are less consistent with this update compared to the launch version. After some testing, it turns out NVIDIA Reflex is the main culprit. If you’re noticing micro-stuttering or uneven frame pacing, disable Reflex.

I’m working on a video for #ArcRaiders Update 1.7.0, and I noticed that the frame times are less consistent with this update compared to the launch version. After some testing, it turns out NVIDIA Reflex is the main culprit. If you’re noticing micro-stuttering or uneven frame pacing, disable Reflex.

179,723 görüntüleme • 6 ay önce

#PRAGMATA really shows how far hair rendering has come on the RE Engine since Resident Evil 4 Remake. Even with the performance and VRAM hit, it’s hard to give up on the strand-based hair, as it plays a huge role in the game’s visual presentation.

#PRAGMATA really shows how far hair rendering has come on the RE Engine since Resident Evil 4 Remake. Even with the performance and VRAM hit, it’s hard to give up on the strand-based hair, as it plays a huge role in the game’s visual presentation.

80,811 görüntüleme • 3 ay önce

The latest Patch 1.08.00 for #CrimsonDesert added a new option to enable Ray Traced Shadows for sunlight and moonlight.

The latest Patch 1.08.00 for #CrimsonDesert added a new option to enable Ray Traced Shadows for sunlight and moonlight.

52,925 görüntüleme • 1 ay önce

If you're using Path Tracing in #PRAGMATA, set Shadow Quality to Lowest. This saves a significant amount of VRAM with no noticeable visual impact.

If you're using Path Tracing in #PRAGMATA, set Shadow Quality to Lowest. This saves a significant amount of VRAM with no noticeable visual impact.

75,226 görüntüleme • 3 ay önce

DLSS 4.5 performance cost on the RTX 3060 Ti, tested in #TheLastOfUs and #ArcRaiders at 1440p. The Last of Us Part I Performance mode – CNN (Preset A) → Transformer Preset K: ~9% cost – Preset K → Preset L: ~16% hit – Preset K → Preset M: ~10% hit Balanced mode – CNN (Preset A) → Transformer Preset K: ~9% cost – Preset K → Preset L: ~23% hit – Preset K → Preset M: ~14% hit Quality mode – CNN (Preset A) → Transformer Preset K: ~9% cost – Preset K → Preset L: ~23% hit – Preset K → Preset M: ~20% hit Arc Raiders Performance mode – CNN (Preset E) → Transformer Preset K: ~9% cost – Preset K → Preset L: ~20% hit – Preset K → Preset M: ~15% hit Balanced mode – CNN (Preset E) → Transformer Preset K: ~8% cost – Preset K → Preset L: ~25% hit – Preset K → Preset M: ~21% hit Quality mode – CNN (Preset E) → Transformer Preset K: ~7% cost – Preset K → Preset L: ~28% hit – Preset K → Preset M: ~25% hit Preset L and M are brutal on the 3060 Ti , which is expected since the 20/30 series GPUs lack native FP8 Tensor Core support, and the performance cost increases even further in Balanced and Quality modes. Visually, unlike the jump from the CNN model to the Transformer, I’m not seeing enough improvement to justify this massive performance hit. Most of the time it’s just a small sharpness boost (which can sometimes look overly sharp) and slightly reduced ghosting. On the flip side, the newer presets also introduce issues that Preset K doesn’t have (check the distant Salt Lake City sign in The last of Us comparison).

DLSS 4.5 performance cost on the RTX 3060 Ti, tested in #TheLastOfUs and #ArcRaiders at 1440p. The Last of Us Part I Performance mode – CNN (Preset A) → Transformer Preset K: ~9% cost – Preset K → Preset L: ~16% hit – Preset K → Preset M: ~10% hit Balanced mode – CNN (Preset A) → Transformer Preset K: ~9% cost – Preset K → Preset L: ~23% hit – Preset K → Preset M: ~14% hit Quality mode – CNN (Preset A) → Transformer Preset K: ~9% cost – Preset K → Preset L: ~23% hit – Preset K → Preset M: ~20% hit Arc Raiders Performance mode – CNN (Preset E) → Transformer Preset K: ~9% cost – Preset K → Preset L: ~20% hit – Preset K → Preset M: ~15% hit Balanced mode – CNN (Preset E) → Transformer Preset K: ~8% cost – Preset K → Preset L: ~25% hit – Preset K → Preset M: ~21% hit Quality mode – CNN (Preset E) → Transformer Preset K: ~7% cost – Preset K → Preset L: ~28% hit – Preset K → Preset M: ~25% hit Preset L and M are brutal on the 3060 Ti , which is expected since the 20/30 series GPUs lack native FP8 Tensor Core support, and the performance cost increases even further in Balanced and Quality modes. Visually, unlike the jump from the CNN model to the Transformer, I’m not seeing enough improvement to justify this massive performance hit. Most of the time it’s just a small sharpness boost (which can sometimes look overly sharp) and slightly reduced ghosting. On the flip side, the newer presets also introduce issues that Preset K doesn’t have (check the distant Salt Lake City sign in The last of Us comparison).

138,748 görüntüleme • 6 ay önce

Lego Batman: LOTDK vs Batman Arkham Knight.

Lego Batman: LOTDK vs Batman Arkham Knight.

22,819 görüntüleme • 25 gün önce

#ForzaHorizon6 RT Global Illumination on the 3060 Ti. RTGI can look excellent in the right scenes, but it also introduces quite a bit of noise. The only way to reduce it is by increasing the resolution or using High, and both options come with a significant performance cost.

#ForzaHorizon6 RT Global Illumination on the 3060 Ti. RTGI can look excellent in the right scenes, but it also introduces quite a bit of noise. The only way to reduce it is by increasing the resolution or using High, and both options come with a significant performance cost.

52,409 görüntüleme • 2 ay önce

#ResidentEvilReqiuem Raster Lighting vs Ray Tracing vs Path Tracing on the RTX 4070 Super. Path Tracing significantly enhances the game’s visuals. Regular RT still looks fine, but its denoiser is underwhelming and produces noticeable boiling and noise artifacts.

#ResidentEvilReqiuem Raster Lighting vs Ray Tracing vs Path Tracing on the RTX 4070 Super. Path Tracing significantly enhances the game’s visuals. Regular RT still looks fine, but its denoiser is underwhelming and produces noticeable boiling and noise artifacts.

103,593 görüntüleme • 4 ay önce

Some examples showing what the missing lighting pass bug (when rendering above 1080p) affects visually in #007FirstLight. It primarily impacts Ambient Occlusion coverage, the softness of some shadows, as well as the definition and stability of fine geometry especially foliage.

Some examples showing what the missing lighting pass bug (when rendering above 1080p) affects visually in #007FirstLight. It primarily impacts Ambient Occlusion coverage, the softness of some shadows, as well as the definition and stability of fine geometry especially foliage.

26,091 görüntüleme • 1 ay önce

Some DLSS 4 and 4.5 issues: DLSS 4 → worse particle handling, with more trailing and reduced rain visibility. DLSS 4.5 → More RT "noise" and more shadow flickering, especially in distant/mid-range cascades. Hope this video doesn’t get compressed to death. #CrimsonDesert

Some DLSS 4 and 4.5 issues: DLSS 4 → worse particle handling, with more trailing and reduced rain visibility. DLSS 4.5 → More RT "noise" and more shadow flickering, especially in distant/mid-range cascades. Hope this video doesn’t get compressed to death. #CrimsonDesert

79,183 görüntüleme • 3 ay önce

Sun Shadow Quality. #Battlefield6

Sun Shadow Quality. #Battlefield6

164,323 görüntüleme • 9 ay önce

Ryzen 3600 vs Ryzen 5700x3D in "Bug-Hill" This spot is extremely CPU-intensive and the performance gap between the two chips is on the higher side. #CrimsonDesert

Ryzen 3600 vs Ryzen 5700x3D in "Bug-Hill" This spot is extremely CPU-intensive and the performance gap between the two chips is on the higher side. #CrimsonDesert

77,671 görüntüleme • 3 ay önce