• 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.

Neon Noir benchmarks

Kaapstad said:
2160p Ultra

vHo02w0.jpg
Click to expand...
Nice one Kaaps. I got 1525 with my Vega 56 on the same settings :p

I love the scene with the bullets. This benchmark kind of reminds me of the first Cyberpunk 2077 teaser trailer from 6+ years ago.

Jensen needs to pull his finger out and get the RTX 3000 series out already!
 
Image is for Ultra setting 1080p, I think it was about 30fps average I got, not very smooth.
Just adjusted RAM timings and got 3181, a 79 point increase. CPU and GPU not overclocked.
A few years and enthuisasts will have Real Time Ray Tracing without the need for nvidias dedicated Tensor cores, bring it on.

MwNQL6Y.jpg
Sorry for huge image, don't know how to make it smaller.
 
2160p ultra

2080 Ti - 3980

GPU core averaged about 1850Mhz.

lsdxvQ7.png

In SLI mode on 2x 2080 Ti's - 6170

xopTRIz.png

GPU usage with two cards seemed to average about 85% load on both, but just wanted to see if I could at least get SLI working.

Used NV inspector and just added the benchmark (GameLauncher.exe) to the Crysis 3 profile, ran ok though.
 
Comparing AMD cross generation performance


Doesn't look like there is much difference other than the existing performance variance between these cards.

Interesting details I found about this benchmark. Even with the graphics set to Ultra, the amount of rays and bounces generated is only equal to Low or Medium settings in a game like Battlefield V that uses RTX. I don't see anything wrong with this and it's totally expected even for next year's AMD ray tracing implementations - without the raw power from fixed function hardware, I expect developers and AMD to need to be very specific about the quality of the ray tracing implementation in order to maintain some performance standards. The good news, is that even for me, I can't really tell much different between Medium and Ultra ray tracing in BFV, so I think lowering the amount of rays and bounces is good.

However, what it does do is place more work into the hands of the developers. Not only do they need to keep using SSAO and Spot Lights for cards that don't do Ray Tracing. They need to put a fair bit of effort into Ray Tracing on AMD hardware to keep the performance high by choosing only to apply it to specific light sources and reflections and carefully adjusting the amount of light bounces etc. Compared to RTX, the developers can be a bit lazier where Ray Tracing is closer to just hitting an on switch.
 
Grim5 said:
Comparing AMD cross generation performance


Doesn't look like there is much difference other than the existing performance variance between these cards.

Interesting details I found about this benchmark. Even with the graphics set to Ultra, the amount of rays and bounces generated is only equal to Low or Medium settings in a game like Battlefield V that uses RTX. I don't see anything wrong with this and it's totally expected even for next year's AMD ray tracing implementations - without the raw power from fixed function hardware, I expect developers and AMD to need to be very specific about the quality of the ray tracing implementation in order to maintain some performance standards. The good news, is that even for me, I can't really tell much different between Medium and Ultra ray tracing in BFV, so I think lowering the amount of rays and bounces is good.

However, what it does do is place more work into the hands of the developers. Not only do they need to keep using SSAO and Spot Lights for cards that don't do Ray Tracing. They need to put a fair bit of effort into Ray Tracing on AMD hardware to keep the performance high by choosing only to apply it to specific light sources and reflections and carefully adjusting the amount of light bounces etc. Compared to RTX, the developers can be a bit lazier where Ray Tracing is closer to just hitting an on switch.
Click to expand...

So a 30-40% difference between vega 56 and the 5700 is usual. Plenty of times i can see the 5700 pulling this gap. That's not the usual difference between them. RDNA certainly seems to be better when it loads up. Probably need to watch the vids before commentating never looked to hard at the others but the 5700 does seem a good bit better at times comparing to Vega.

low or medium on this seems to be better than ultra in battlefield. What we care about is the overall effect and this demo looks as good as any lately. Overall impressed with what they done and can do with cards that don't portray themselves as rt card. Even turing using no tensor cores runs this better than the amd counterparts. Great demo from Crytek showing what could be achieved without a 1000 squididly's
 
Grim5 said:
Doesn't look like there is much difference other than the existing performance variance between these cards.

Interesting details I found about this benchmark. Even with the graphics set to Ultra, the amount of rays and bounces generated is only equal to Low or Medium settings in a game like Battlefield V that uses RTX. I don't see anything wrong with this and it's totally expected even for next year's AMD ray tracing implementations - without the raw power from fixed function hardware, I expect developers and AMD to need to be very specific about the quality of the ray tracing implementation in order to maintain some performance standards. The good news, is that even for me, I can't really tell much different between Medium and Ultra ray tracing in BFV, so I think lowering the amount of rays and bounces is good.

However, what it does do is place more work into the hands of the developers. Not only do they need to keep using SSAO and Spot Lights for cards that don't do Ray Tracing. They need to put a fair bit of effort into Ray Tracing on AMD hardware to keep the performance high by choosing only to apply it to specific light sources and reflections and carefully adjusting the amount of light bounces etc. Compared to RTX, the developers can be a bit lazier where Ray Tracing is closer to just hitting an on switch.
Click to expand...

Thing is though as impressive as this is - and even having this level running in real time in software is a feat - what really makes ray tracing is the subtle bounced and scattered light, etc. the distorted light reflected on a variety of surfaces, etc. and once you add that type of stuff in at a quality level comparable to what is possible with RTX this kind of approach will not be able to provide real time viable performance.
 
When everyone is saying 1440, is this 3440 x 1440, 2560 x 1440 or 1920 x 1440?

Anyway -
1920 x 1080 - 9001
3440 x 1440 - 4328

EVGA 1080 Ti@ stock
16GB Ram
i7 4790k @4.5ghz
 
You must log in or register to reply here.
Back
Top Bottom