• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Fidelity Super Resolution 2.0

You said game runs very well, but then the frametime did not look smooth which was intriguing ;)

Why do you think I've included 1% low and minimum FPS data in my videos? It'll be included in the 4K native video too with RT enabled. :)
Here's a completely flat frame time graph from a great video with a 12900K and a 6950 XT overclocked, getting worse 0.1%, minimum and average FPS than my 5800X3D and my 6950 XT setup, which had a frame time graph that is not always flat, especially in cut scenes.

What's the moral of the story I hear you ask? Always include the 0.1% and minimum FPS on your overlay because they will show any issues with frame pacing and VRAM saturation. A frame time graph is clearly not enough on its own. :)

4K Very High Settings + FSR2.

4K Very High Settings No FSR2.

Game is comfortably using 12GB at 4K too.
 
Here's a completely flat frame time graph from a great video with a 12900K and a 6950 XT overclocked, getting worse 0.1%, minimum and average FPS than my 5800X3D and my 6950 XT setup, which had a frame time graph that is not always flat, especially in cut scenes.

What's the moral of the story I hear you ask? Always include the 0.1% and minimum FPS on your overlay because they will show any issues with frame pacing and VRAM saturation. A frame time graph is clearly not enough on its own. :)

4K Very High Settings + FSR2.

4K Very High Settings No FSR2.

Game is comfortably using 12GB at 4K too.
@TNA I figured out what was causing the (false) spikes in the frame time overlay, my custom overlay. Oddly, it only affected a few games, Spiderman, Far Cry 6, Days gone and a few others. Other games like COD, Dying Light 2, Red Dead Redemption 2 were unaffected. I rebuilt it from scratch and now its back to a flat line in all games. :D

Got my hands on a 3090 TI recently and was able to compare it to the RX 6950 XT at 4K in Spiderman, very high settings with high RT enabled.

The 3090 TI averages 86 FPS with 1% lows of 58.

The RX 6950 XT averages 81 FPS with 1% lows of 57.

The 3090 TI video was recorded with the overlay glitch still present (same as my 6950 XT video), however all videos going forward will now be back to the usual flat graph.
 
Suspect the modded version using dlss files will look/work better overall than the official integration. Reminds me, I need to do another play through and choose different scenarios to my first playthrough.
 
Suspect the modded version using dlss files will look/work better overall than the official integration. Reminds me, I need to do another play through and choose different scenarios to my first playthrough.

Hope to see some comparisons, I'm not flush enough to get Dying Light 2 yet otherwise I'd check for myself.
Be handy as I loaded the game up with mods including the fsr mod but haven't played it yet.
 
This will be a good test for fsr 2:


Apparently taa is also improved so be curious to see how it looks now as it was absolutely awful.
 
FSR 2.0 looks fine in RDR2. Initially it started quite well as I booted into my save in St Dennis, looking quite sharp overall and very 4K-like even on performance mode & could only really tell it wasn't when I saw the pole wires because it's an obvious example given how these methods handle them. Unfortunately once I made my way to the train station and got on-board I started seeing the dreaded ghost streaks. First it was around the trees with water behind them, but then all the insects buzzing about also had them (like how DLSS was initially in Death Stranding if anyone remembers). Overall probably still viable over simply reducing render resolution with the native TAA but hardly worth much - but can't say how it differs at lower resolution as I already deleted the game after testing.

As for TAA itself, it looks good to me at 4K, can't remember if better than before because I haven't logged on in over a year, but definitely no complaints with regards to it being blurry or any of that (but also remember to use medium and not high TAA!). Clearly sharper and more detailed than FSR 2.0 at Q tho - which it should be. If I were to play it again I'd definitely stick to TAA, it's the best quality out of all options atm.
 
Here's a completely flat frame time graph from a great video with a 12900K and a 6950 XT overclocked, getting worse 0.1%, minimum and average FPS than my 5800X3D and my 6950 XT setup, which had a frame time graph that is not always flat, especially in cut scenes.

What's the moral of the story I hear you ask? Always include the 0.1% and minimum FPS on your overlay because they will show any issues with frame pacing and VRAM saturation. A frame time graph is clearly not enough on its own. :)

4K Very High Settings + FSR2.

4K Very High Settings No FSR2.

Game is comfortably using 12GB at 4K too.
Uhmm naaa.

Vram saturation looks like this:

The opposite of stutter.
Observe maxed GPU usage, low power draw and low FPS, that's the bottleneck on modern hardware with SSD's.

 
Yup, my post was completely unrelated to stuttering caused by VRAM usage. It turned out to be an incorrect setting in my overlay which was causing false spikes. I redid it from scratch and the problem was solved. :confused:
Did not see your later post, my bad, helps to distinguish what Vram capping out looks like though.
 
Back
Top Bottom