• 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.

Can someone explain why the 2080ti is beating the 3080ti in..?

Grim5

Grim5

Grim5

Soldato
Joined
6 Feb 2019
Posts
20,349
Alan Wake Remastered using DLSS

these are official benchmarks from Nvidia and they don't make sense.

Maybe this is what they call Turing FineWine :p

 
It's not odd really, because it's solely DLSS based. The odd thing is that There is no AMD representation in the slide, you would have thought there would have been.
 
Seems odd but guess there is a CPU bottleneck or a game engine bottleneck.

The fact that 4 GPU's are all getting around 120FPS

It's an old game even though it's been "remastered" wouldn't be surprised if it's still running on a single core or 2.
 
Well if anybody wants to swap their 3080ti card for my clearly faster 2080ti then I'm open to offers.
The non DLSS numbers are in line with expectations.
The 120 cap may be to stop the game engine breaking, can get weirdness with high FPS and old games.
 
It could be related to the changes in tensor cores, RTX30x0 series has fewer tensor cores but ones that are much more advanced. This might just be a situation where having lots of the older more basic ones are slightly better.
 
newuser said:
Well if anybody wants to swap their 3080ti card for my clearly faster 2080ti then I'm open to offers.
The non DLSS numbers are in line with expectations.
The 120 cap may be to stop the game engine breaking, can get weirdness with high FPS and old games.
Click to expand...

They will regret that swap :D:D:D
 
Grim5 said:
Alan Wake Remastered using DLSS

these are official benchmarks from Nvidia and they don't make sense.

Maybe this is what they call Turing FineWine :p
Click to expand...

1 FPS is margin of error as we know the 3080 Ti is a hell of a lot faster than the 2080 Ti, Plus the fact the 2080 Ti and 3080 Ti are both locked around 120FPS means there's either a frame cap or CPU bottleneck going on.
 
I'm guessing there are 2 aspects here - the 2080ti has more Tensor core performance than anything but the 3080 and 3080ti in that list and there is probably a fixed minimum time for stuff like the Tensor cores take to operate, at the end of the frame, which is producing a bottleneck at around 8ms frame times [in this particular case].
 
Last edited:
Rroff said:
I'm guessing there are 2 aspects here - the 2080ti has more Tensor core performance than anything but the 3080 and 3080ti in that list and there is probably a fixed minimum time for stuff like the Tensor cores take to operate, at the end of the frame, which is producing a bottleneck at around 8ms frame times.
Click to expand...

I'd put 1 frame per second down to margin of error rather than anything else.
 
This is simple, clearly a bottleneck there in my opinion and 2080ti clearly in margin of error imo
 
I find the the large gap between the 3080 and Ti without DLSS more noteworthy, I doubt it's a VRAM limitation as the 3070 beats the 2080 Ti. 20% is more than the usual difference I saw in reviews.
 
Performance mode is likely a CPU bottleneck, would prefer to see the benchies in quality mode. There is more to how a game looks and plays than resolution and FPS.
 
You must log in or register to reply here.
Back
Top Bottom