2 HN users recommended this talk2 months2014–2021score 2.5
Years with an HN recommendation, 2007–2026
HN recommends it for Carmack's detailed engineering insight into VR rendering and latency challenges.
John Carmack's keynote at Oculus Connect 2014 about the Gear VR and development stories from building it. He discusses mobile GPU optimization, latency, and display ideas like programmable interlacing.
“Carmack's 2014 Oculus Connect keynote is well worth watching” (HN, 2015)
While this may be true for high fidelity gaming, for moving some textured triangles around, mobile GPUs are fine. Also, they can (and do) perform much better clock-for-clock vs a PC (and especially a Mac) thanks to the ability to optimize drivers. For a fascinating discussion of the optimizations possible, Carmack's 2014 Oculus Connect keynote is well worth watching: [this video] This talk was a slightly long but very informative discussion of the various challenges in reducing latency and dealing w/ PC rendering: [link]
Carmack explains mobile GPU optimizations and latency challenges in VR rendering.
> What was Carmack's development focus during Oculus? I'm not sure exactly but there's a bunch of keynotes from Carmack around Oculus Connect on YouTube. Here's one from 2014: [this video] , and another one from a year later where he's live coding in a VR environment [link] . You'll find a bunch more if you search too. I remember skimming through them years ago not because I'm into VR development but I've been a life long fan of Carmack's work and like listening to him talk about things.
Speaking of "frameless" rendering, I noticed during Carmack's Oculus keynote ( [this video] ), he talks about trying to persuade Samsung to integrate programmable interlacing into their displays in order to give dynamic per-frame control over which lines are being scanned. This would give you the same "adaptive per-pixel updating" seen in your link, though primarily to tackle the problems with HMDs (low-persistence at high frame-rates).
Carmack on programmable interlacing for HMD displays
Found where Carmack talked about it[1]. For VR, the consensus I've seen seem to push for multi-sample AA > full-screen AA > temporal AA (I feel like I've even seen no AA is better than temporal AA). I'm not quite sure if it's performance, architecture (they really prefer forward renderers instead of deferred renderers for low latency), or aesthetics. In the little bit I've done, when just playing with knobs temporal AA looks better to me. Without AA specular highlights are way too distracting and pop too much. This reference[2] talks about the extra velocity buffer needed for temporal AA and the fact it tends to over-blur which can fuzz out fine details (low resolution is a touchy subject in VR). There's a lot of moving pieces when making decisions. In talking about interlacing, he was ideally talking about not having to create a whole image buffer before sending data to the hardware--just rendering parts (well, scanline) of the image that moved. [1] [this video] [2] [link]
Carmack discusses rendering approaches like interlacing and partial scanline rendering for VR.
Excellent list of things to keep in mind. Of course, being The Statue is allowed, if you happen to be John Carmack, legs planted firmly on the stage like a colossus: [this video]
Video list and Hacker News comment labels from HackerTube (prompted by @karpathy), released under ODC-By 1.0, built on the open-index/hacker-news archive. Hacker News data is provided by Y Combinator; comments remain the work of their authors. Video titles, channel names and thumbnails belong to their creators and are shown via YouTube. HackerTube Watch is an unofficial edition. Data export of 2026-10-31.
Plays via YouTube's privacy-enhanced embed
5 Hacker News comments
From the HN archive; up to 20 kept per video, favoring different commenters. Recommendations first.
Citations, criticism and jokes
Read more