Just stumbled upon some of the most interesting benchmark results I've seen in a long time - some reviewers actually tested real-world performance with RTX 2070, GTX 1080 and Vega 64 in many classic VR games using Rift CV1.
One problem when let's say evaluating RTX 3080 performance is that we assume that results from pancake gaming will translate into VR performance. But that might not be true - as we'll check later. First let's see some VR results.
System setup:
One problem when let's say evaluating RTX 3080 performance is that we assume that results from pancake gaming will translate into VR performance. But that might not be true - as we'll check later. First let's see some VR results.
System setup:
- Intel Core i7-8700K (HyperThreading and Turbo boost is on to 4.8 GHz for all cores; Coffee Lake DX11 CPU graphics).
- EVGA Z370 FTW motherboard (Intel Z370 chipset, latest BIOS, PCIe 3.0/3.1 specification, CrossFire/SLI 8x+8x), supplied by EVGA
- T-Force XTREEM 16GB DDR4 (2×8 GB, dual channel at 3866MHz), supplied by Team Group
- RTX 2070 8 GB Founders Edition, stock FE clocks, on loan from NVIDIA
- GTX 1080 8GB Founders Edition, stock FE clocks, on loan from NVIDIA
- Liquid-Cooled RX Vega 64, 8GB at liquid-cooled clocks
- 2 x 480GB Team Group SSDs – 1 for Radeon and 1 for GeForce
- 1.92 TB San Disk enterprise class SSD
- 2 TB Micron 1100 enterprise class SSD
- Seasonic 850W Gold Focus power supply unit
- EVGA CLC 280mm CPU water cooler, supplied by EVGA
- EVGA Nu Audio stereo PCIe sound card, supplied by EVGA
- Edifier R1280T active desktop speakers
- EVGA DG-77, mid-tower case supplied by EVGA
- Monoprice Crystal Pro 4K
- Oculus Rift
More info here about the software used etc.:
https://babeltechreviews.com/vr-wars-the-rtx-2070-vs-the-gtx-1080-vs-the-vega-64/
- and the results in short:
In average RTX 2070 is only 7 % faster than GTX 1080. And 7 % - how many would notice that in a blind test!? (Especially with vsync on) Sure some games are up to nearly 20 % faster using RTX 2070 than GTX 1080, but one game was even 5 % slower! (Ethan Carter).
Has the old Series 10 been more optimized for VR than Series 20? And how about Series 30? (Huang did not mention VR with one word when introducing 3080, clearly VR has low priority)
Now, what if we compare 2070 FE vs. 1080 FE in 23 pancake games - then we can see - average results:
Source: https://www.techpowerup.com/review/nvidia-geforce-rtx-2070-founders-edition/
Using 1920x1080 res, 2070 FE is 12 % faster than 1080 FE. That's nearly twice the difference we saw when combining 11 VR games. Thus using pancake games to assume VR performance difference may lead to overestimating performance differences - and it's much worse for 4K pancakes:
Source: https://www.techpowerup.com/review/nvidia-geforce-rtx-2070-founders-edition/
In 4K res, the 2070 is average 19 % faster than GTX 1080.
Now, the VR results were obtained using CV1 - might be different with Index or G2, we don't know that. What the above results do suggest is that VR performance differences may be much more subtle than performance differences in pancake games. For now using pancake games to estimate performance gains in VR games and apps when upgrading a video card may not be easy and may lead to wrong estimation, so beware.
The reviewers behind the testing of 11 VR games also have a RTX 3080 - I'll contact them and hear if they would consider some RTX 3080 VR testing - we need to know the real numbers!
