VR has amazing potential, but will only succeed when there are enough users who are capable of enjoying the experience. NVIDIA examines the current state of the PC market for VR, and discusses its forecast for the coming years and technologies that will propel the industry forward.
7. 7
NVIDIA VR SDK
WARP AND BLEND
Increase
performance via an
innovative new way
to render for VR
Scale performance
with multiple GPUs
Minimize head
tracking latency
with asynchronous
time warp
Plug and play
compatibility
from GPU to HMD
Reduce latency by
rendering directly
to the front buffer
API for geometry and
intensity adjustments
for seamless VR
Provides tear free VR
environments by
synchronizing multi GPU’s
Fine grain control
to pin OGL contexts
to specific GPU’s
Reduces latency for
video transfer to
and from the GPU
SYNCHRONIZATION GPU
AFFINITY
GPUDIRECT
FOR VIDEO
10. 10
MULTI-RESOLUTION SHADING
Multi-Resolution Rendering
Splits 30% in from each edge,
Periphery rendered at 60%
resolution
58% of original fragment count
Theoretical (perfectly fragment-
bound) speedup of 1.7x
Image: UE4 Reflections Subway, courtesy of Epic Games
16. 16
100M GEFORCE VR READY PCS BY 2020
0M
20M
40M
60M
80M
100M
120M
140M
2014 2015 2016 2017 2018 2019 2020
100M
13M
17. 17
MULTI-RES EXPANDS INSTALLED BASE
OF GEFORCE VR READY PCS
0M
20M
40M
60M
80M
100M
120M
140M
2014 2015 2016 2017 2018 2019 2020
130M
25M
Baseline
Multi-Res
18. 18
CLIMBING EVEREST
BASE CAMP
Create technology to enable the ecosystem
CROSS THE CREVASSE
Build large installed base of VR Ready PCs
SUMMITING
Make VR a pervasive part of our lives
We talk a lot about VR in the context of gaming and traditional pro/design apps.
VR is the next app and computing platform. The next smartphone.
Opportunity well beyond gaming and traditional pro markets
Virtual reality is extremely demanding with respect to rendering performance. Both the Oculus Rift and HTC Vive headsets require 90 Hz, a very high framerate compared to gaming on a screen. We also need to hit this framerate while maintaining low latency between head motion and display updates.
Mike Abrash at Oculus has done some research on this, and his results show that the motion-to-photons latency should be at most 20 milliseconds to ensure that the experience is comfortable for players.
The problem is that we have a long pipeline, where input has to first be processed and a new frame submitted from the CPU, then the image has to be rendered by the GPU, and finally scanned out to the display. Each of these steps adds latency.
Traditional real-time rendering pipelines have not been optimized to minimize latency, so this goal requires us to change how we think about rendering to some extent.
Headsets Low Latency, High resolution, Fast FPS; from GeForce driving Oculus and HTC to GameFace Labs – building mobile headset using Tegra K1
Platforms / Stores Plug and play compatibility
App Developers Fast FPS / Rich Graphics
Engines Fast FPS / Rich Graphics – working with UE4 to integrate GameWorks VR
Input & Tracking Low Latency, Computer Vision, Deep Learning – using NVIDIA GPUs to do Deep Learning to train
Content Capture & Streaming Encoding, Stitching, Streaming – using NVIDIA CUDA to accelerate video stitching, and Direct to Video for fast transcode
NVIDIA has created a VR SDKs for headset and HMD developers that improves performance, reduces latency, improves compatibility, enables immersive environments, and accelerates 360 video broadcast.
Multi-res doubles TAM in 2016
Pulls in the installed base by 1 year
And speaking of Summiting Everest – let me leave you with a small teaser. This isn’t a photograph, but rather a first look screenshot from a VR experience from Icelandic VR developer, Solfar, which we will be integrating GameWorks VR with. Stay tuned for an announcement and more details from Solfar tomorrow!
We are excited to be working with Icelandic based Solfar to bring -- stay tuned for an exciting announcement from Solfar on Tuesday.