• 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.

Resident Evil 8 performance

Soldato
Joined
4 Feb 2006
Posts
3,204
So as many people predicted AMD cards this generation will age better (with exception of the 3090 for obvious reasons), generation literally just started and Nvidia cards already struggling, even worse considering how much 3070s are selling for.

My old R9 390 has 8Gb and I felt that hit the VRAM limit in some games if I used Ultra textures. There is no excuse having only 8GB for the latest gen gpu's. It was planned obsolescence all along.
 
Permabanned
Joined
21 Feb 2021
Posts
474
And a good SSD on top of it, wasnt Nvidia claiming that a 3070 would be as fast as a 2080TI months ago?
No, as you can see RTX 2080ti is arealdy out of juice, rendering 70 FPS at 4K. It should render 120 fps or bust!! if it can't render 120 fps then it's not fit for 4k anyways /s
 
Associate
Joined
15 Feb 2021
Posts
51
Nah, I'm talking about this kind of stuff:

https://www.reddit.com/r/nvidia/comments/n5l5hk/resident_evil_viii_benchmark_guru3d_8gb_not/gx1v05m/

Their tune will also change. When? Obvious, once 16 GB Nvidia cards flood the market!

"AMD led development in RT will over burden the vram as it is one of the advantages the radeon cards have at the moment"

Conspiracy at its best. Exact same arguments I've faced when I presented the Godfall FPS drop video. Inevitable "crap game anyways!1" followed. :p At least RE:Village seems like a decent game. We shall see.

Lol, thread is locked. Funny

I found the TechpowerUp link quite interesting in that thread regarding their own performance analysis, that found that the 3070/3060Ti didn't show the same results despite running the same settings. The only changes being the setups used. TPU used a 5800x with 4000Mhz RAM and Guru3D used a 5950x with 3600Mhz RAM.

To be fair, any YT videos that I've seen with a 3070 being used on RE8 hasn't shown performance results in line with the Guru3D. It seems to be a outlier for whatever reason.
 
Permabanned
Joined
21 Feb 2021
Posts
474
I found the TechpowerUp link quite interesting in that thread regarding their own performance analysis, that found that the 3070/3060Ti didn't show the same results despite running the same settings. The only changes being the setups used. TPU used a 5800x with 4000Mhz RAM and Guru3D used a 5950x with 3600Mhz RAM.

To be fair, any YT videos that I've seen with a 3070 being used on RE8 hasn't shown performance results in line with the Guru3D. It seems to be a outlier for whatever reason.
It's not outlier, it can happen for random reasons based on configurations or gameplay, even.

Maybe Guru3d did a more comprehensive test, maybe they've waited until game firmly utilizes more VRAM. That is a possibility.

I can testify having problems on my own 3070. People here are aware that I've been constantly hitting weird VRAM constraints on all kind of games, ranging from Cyberpunk to Godfall and now RE:Village.

TBH, even at 1080p, this card struggles with RT enabled. It's clear that it has the power pushing high settings and RT at 1080p/1440p, but pushing high settings stresses VRAM even more, which defeats the whole purpose of being as "powerful" as a 2080Ti
 
Associate
Joined
15 Feb 2021
Posts
51
It's not outlier, it can happen for random reasons based on configurations or gameplay, even.

Maybe Guru3d did a more comprehensive test, maybe they've waited until game firmly utilizes more VRAM. That is a possibility.

I can testify having problems on my own 3070. People here are aware that I've been constantly hitting weird VRAM constraints on all kind of games, ranging from Cyberpunk to Godfall and now RE:Village.

TBH, even at 1080p, this card struggles with RT enabled. It's clear that it has the power pushing high settings and RT at 1080p/1440p, but pushing high settings stresses VRAM even more, which defeats the whole purpose of being as "powerful" as a 2080Ti

Perhaps in the benchmarking sessions, they are playing the game for longer. KitGuru are also reporting the same as TechPowerUp as well. Would be interesting to hear from more 3070/3060Ti owners once the game has been released.
 
Caporegime
Joined
12 Jul 2007
Posts
40,567
Location
United Kingdom
TPU are always very vague about what sections of games they use to benchmark.

Kitguru said this.
"Our 1-minute benchmark pass came from quite early on in the game, as the player descends down into the village for the first time".

Guru3D have a video showing the area they tested.

So one thing is clear is that they are testing different areas and this likely explains the performance difference.

I can't help but think a 1 minute benchmark pass is not sufficient though regardless of what area you test.
 
Permabanned
Joined
21 Feb 2021
Posts
474
TPU are always very vague about what sections of games they use to benchmark.

Kitguru said this.
"Our 1-minute benchmark pass came from quite early on in the game, as the player descends down into the village for the first time".

Guru3D have a video showing the area they tested.

So one thing is clear is that they are testing different areas and this likely explains the performance difference.

I can't help but think a 1 minute benchmark pass is not sufficient though regardless of what area you test.
Same story on Godfall. If you only do benchmark and done with it, it will seem as you're fine at 1440p:

VdH3uGC.png

But I'm suffering from severe drops at 1440p in "actual" gameplay:

https://www.youtube.com/watch?v=sJ_3cqNh-Ag

If this game was any decent, I bet this would have more exposure. Game is such a joke that only RT benchmark I could find belongs to this channel, and they just tested with in built benchmark.

I tried the same benchmark, and of course not surprisingly, it runs good.

The problems occur when you actually play the game, change maps and come across more enemies.
 
Associate
Joined
20 Nov 2020
Posts
1,120
It looks like the Metro Exodus EE will also run ok on AMD which means the Big Navi is much closer to Ampere in RT performance than many people think, and the games where its running bad are not optimized, i will not say they are not optimized by design :)

https://www.hardwareluxx.de/index.p...tion-ein-blick-in-die-raytracing-zukunft.html

aa.jpg


That is ok'ish for a full RT game, remember that some people claimed RDNA2 can only run ray traced shadows. :)
 
Caporegime
Joined
12 Jul 2007
Posts
40,567
Location
United Kingdom
It looks like the Metro Exodus EE will also run ok on AMD which means the Big Navi is much closer to Ampere in RT performance than many people think, and the games where its running bad are not optimized, i will not say they are not optimized by design :)

https://www.hardwareluxx.de/index.p...tion-ein-blick-in-die-raytracing-zukunft.html

aa.jpg


That is ok'ish for a full RT game, remember that some people claimed RDNA2 can only run ray traced shadows. :)
Not bad, minimums are disappointing all round though, and this is only 1440P.
 
Associate
Joined
20 Nov 2020
Posts
1,120
Yeah well the Nvidia owners will be saved by the DLSS anyway but that's because your lazy software developers are not yet replaced and you are wasting your time on forums instead of working on FSR. :)
It doesn't matter if these are the results, it means that the gap between the cards is much closer that many people think and it is good news for RDNA3.
 
Caporegime
Joined
12 Jul 2007
Posts
40,567
Location
United Kingdom
Same story on Godfall. If you only do benchmark and done with it, it will seem as you're fine at 1440p:

VdH3uGC.png

But I'm suffering from severe drops at 1440p in "actual" gameplay:

https://www.youtube.com/watch?v=sJ_3cqNh-Ag

If this game was any decent, I bet this would have more exposure. Game is such a joke that only RT benchmark I could find belongs to this channel, and they just tested with in built benchmark.

I tried the same benchmark, and of course not surprisingly, it runs good.

The problems occur when you actually play the game, change maps and come across more enemies.
That's why short benchmark runs are not always representative, that VRAM buffer will fill up over time. These limitations won't usually be exposed in 1 minute of gameplay.
 
Associate
Joined
15 Feb 2021
Posts
51
It looks like the Metro Exodus EE will also run ok on AMD which means the Big Navi is much closer to Ampere in RT performance than many people think, and the games where its running bad are not optimized, i will not say they are not optimized by design :)

https://www.hardwareluxx.de/index.p...tion-ein-blick-in-die-raytracing-zukunft.html

aa.jpg


That is ok'ish for a full RT game, remember that some people claimed RDNA2 can only run ray traced shadows. :)

Was much better than I was expecting for a 6800XT. I play at 1440p anyway, so with a nice overclock, I should be hitting around 67-70FPS, which is very acceptable.
 
Associate
Joined
20 Nov 2020
Posts
1,120
Was much better than I was expecting for a 6800XT. I play at 1440p anyway, so with a nice overclock, I should be hitting around 67-70FPS, which is very acceptable.
Probably they put a lot of work to launch the game on consoles and some of that optimization went into pc version too. But if these are the results, it is very good news for next gen card from AMD.
 
Soldato
Joined
18 Oct 2002
Posts
7,028
Location
Melksham
Comparing RT performance from two very different hardware approaches in software like Control that was very much targeted at one of those hardware designs was never going to give the full picture...

And this was mentioned multiple times, but didn't stop the fanboys. Of course this may be an outlier, it's kinda hard to tell at this point but I suspect this is much more like what we'll see going forwards, especially with the new consoles.
 
Caporegime
Joined
4 Jun 2009
Posts
31,045
Hard to tell from review as only skimmed it really (not got time to read properly till later) but aren't those benchmark presets with "low" ray tracing effects????

From the article:

A GeForce RTX 3080 delivers even more FPS than a Radeon RX 6900 XT, even in the "high" preset. As I said, the ray tracing effects are always active at their lowest level, regardless of which preset is selected.

EDIT:

Just had another look, looks like they show ray tracing specific benchmarks too, paints a slightly different picture to just the "preset" bar chart but still impressive enough showing from amd! Hopefully this will get more games onboard with ray tracing now if amds fans herald ray tracing as the next best thing going by this ;)
 
Soldato
Joined
2 Nov 2003
Posts
5,532
I won't be buying this on release, the demo did nothing for me personally and it did not feel resident evil enough for me.

May pick it up when it's on sale.
 
Back
Top Bottom