Nvidia Turing Product Reviews and Previews: (Super, TI, 2080, 2070, 2060, 1660, etc)

That's nVidia power and influence here...

I know it's not the same thing but I'll love to see some studios saying "eh, we'll add fp16 support in our games" when Vega cames out...
 
Just not power and influence ... all it takes is an in-house demo to see that the tech works and is easy to implement. The performance gained from having DLSS likely out weights any reason not to implement it.
 
Rootax said:
That's nVidia power and influence here...

I know it's not the same thing but I'll love to see some studios saying "eh, we'll add fp16 support in our games" when Vega cames out...
Click to expand...
Not very many developers did that last time, Nvidia was offering increased FP16 throughput. So they went ahead and did it themselves. :D
 
NVIDIA GeForce RTX 2080 3DMark Graphics Performance Benchmarks Leaked – Faster Than Titan XP, Running on Latest 411.51 Drivers
In the benchmark, the graphics card scored 10,147 points and 10,659 graphics points. The memory was clocked at reference speeds of 7000 MHz (that’s 14,000 MHz effective) and the chip clocked in at 1875 MHz which should be the standard boost clock for Turing TU104 based GPUs. Compared to the GeForce GTX 1080 Ti and Titan XP, we see faster performance when comparing at stock and we can definitely expect higher performance with overclocking.

In addition to that, the 3DMark Firestrike graphics score is said to be in the range of 27,000 points which is a good bit higher than the GeForce GTX 1080 Ti even when its overclocked.
Click to expand...
https://www.3dmark.com/compare/spy/4404946

https://wccftech.com/nvidia-geforce-rtx-2080-3dmark-performance-benchmark-leak/
 
Last edited by a moderator:
pharma said:
NVIDIA GeForce RTX 2080 3DMark Graphics Performance Benchmarks Leaked – Faster Than Titan XP, Running on Latest 411.51 Drivers

https://www.3dmark.com/compare/spy/4404946

https://wccftech.com/nvidia-geforce-rtx-2080-3dmark-performance-benchmark-leak/
Click to expand...

The graphics score of this card is way too low for Firestrike. 27k is only 2k higher than the fastest submitted score for a GTX 1080 (non-Ti) on air/water with an 8700k. Either this isn't a 2080, or it's not regular Firestrike. Too bad the submission contains no content now so we can't know for sure which is the case.
 
It's interesting to note that some of the drops from HDR to SDR on Pascal cards can reach up to 12% in some titles. And these are official numbers from NVIDIA, why is HDR slower than SDR in the first place? Also HDR YUV422 drops more than HDR RGB444.
 
DavidGraham said:
It's interesting to note that some of the drops from HDR to SDR on Pascal cards can reach up to 12% in some titles. And these are official numbers from NVIDIA, why is HDR slower than SDR in the first place? Also HDR YUV422 drops more than HDR RGB444.
Click to expand...
I noticed that too. Probably some bugs still to be worked out in the drivers.
 
DavidGraham said:
NVIDIA's reviewer guide:
2080Ti is 40% faster than 1080Ti in 12 games @4K SDR. Slightly more than that @4K HDR.
2080 is 10% faster than 1080Ti @4K SDR and slightly more than that @4K HDR.
https://videocardz.com/77983/nvidia-geforce-rtx-2080-ti-and-rtx-2080-official-performance-unveiled
Click to expand...

4k performance even better than expected.

The price sucks but if that performance is upheld in independent reviews, kinda wished I had preordered the 2080Ti. That way I can game at 4k while everyone else bickers about the price/value.
 
NVIDIA GeForce RTX 2060
The GeForce RTX 2060 will be a graphics card by NVIDIA. Built on the 12 nm process, and based on the TU116 graphics processor, the card supports DirectX 12.0. It features 1536 shading units, 128 texture mapping units and 32 ROPs. Also included are 192 tensor cores which help improve the speed of machine learning applications. Also included are 24 raytracing acceleration cores. NVIDIA has placed 8,192 MB GDDR6 memory on the card, which are connected using a 128-bit memory interface. The GPU is operating at a frequency of 1410 MHz, which can be boosted up to 1560 MHz, memory is running at 1750 MHz.

Being a dual-slot card, the NVIDIA GeForce RTX 2060 draws power from 1x 6-pin power connectors, with power draw rated at 120 W maximum. Display outputs include: 1x HDMI, 3x DisplayPort, 1x USB Type-C. GeForce RTX 2060 is connected to the rest of the system using a PCI-Express 3.0 x16 interface. The card measures 241 mm in length, and features a dual-slot cooling solution.
Click to expand...
https://www.techpowerup.com/gpudb/3310/geforce-rtx-2060
 
NVIDIA RTX 2080 & 2080Ti spotted in FINAL FANTASY XV Benchmark

kiGtkX5.png

5LJ9e0f.png
Click to expand...


http://benchmark.finalfantasyxv.com/result/
 
Last edited by a moderator:
Malo said:
TU106 is 2070.
Click to expand...

Yeah but what’s the point? It has the same bus width as TU104 and 75% of the SMs. Are yields so wacky on 12nm that it made sense to spin a separate chip instead of salvaging TU104 dies? Maybe yields are pretty good and they didn’t want to have to sell an expensive-to-manufacture TU104 chip with 1/4 of the units disabled.

I could understand if RTX 2060 used a salvaged TU106 but if there’s really a TU116 the whole setup seems strange.
 
You must log in or register to reply here.
Back
Top