Kyouki wrote: ↑15 Aug 2022, 02:11
Chief Blur Buster wrote: ↑15 Aug 2022, 00:08
Also, this forum, is a very weird bridge between the esports community (e.g. General / Latency) and the researcher/science community (e.g. Area 51).
Can confirm that I am indeed a weird mix of the in-between both. Having started my interest from esport and growing more into the researcher or science portion (to also aid my esport/help others understand)
I love your posts Chief. Thanks.
On-topic-ish:
Can indeed confirm that playing myself and going from the 165hz casual gamer, while trying esport games at the time as well but always found myself playing terrible/miss information in my aim to compete. At some point decided to try using ULMB on 120hz on that 165hz monitor (PG279Q) and started to notice I was able to track much much better overtime of using it while I thought of it as a gimmick at first.
That interest led me to investigate the modes or possible options as well as learning more and more about displays.
Figured out why I always had the craves for tweaking systems, getting the most optimal performance. This is indeed like Chief once posted that we do get into territories where we start notice the sub 1ms lag/latency differences. Response times on monitors, strobe tunes, system latency, network latency and all the variances of those. Even EMI/Power appears to be impactful which was eye-opening to me and a new topic I would love to dive in to.
After purchasing the 360hz monitor (with 240hz ULMB) it started to creep even more onto me and I feel today that with all the knowledge and understanding in which scenarios I would use which display tech and I like to educate other people about them as well as I find that Gsync (freesync)/ulmb (any form of blur reduction) is severely underrated.
Fantastic! I love meeting researchers, even if they're anonymous.
To every new researcher, I love reminding them to avoid making assumptions based on today's unfortunate refresh rate incrementalism - You should know about the geometric-upgrade problem of refresh rates and frame rates -- where human-visibility needs something like 2x-4x upgrades such as
60 -> 120 -> 240 -> 480 -> 1000 -> 2000 -> 4000 (until the vanishing point of diminishing curve of returns)
or
60 -> 240 -> 1000 -> 4000 (fps=Hz)
esports players will see differences as small as 1.1-1.5x while average users may need 4x (e.g. 240Hz vs 1000Hz is easily visible to my grandma, like the temporal equivalent of the spatial geometric curve VHS-vs-4K, unlike VHS-vs-DVD). We are (~2024) gradually building a 4K or 8K 1000fps 1000Hz demonstration that is possible with today's technology, perhaps to "Douglas Engelbart The Demo"-ize the knowledge of the refresh rate curve. We found creative unorthodox solutions for all the technology bottlenecks -- from production to display -- and are weaving them all together like 100 different brilliant TestUFO/BlurBusters inventions into one.
Also, 240Hz-vs-360Hz is like a 1.1x difference not 1.5x difference, because high-frequency jitter (VSYNC OFF, game engine flaw, microjitter, fps-mismatch-Hz, etc) blends into blur like fast-vibrating guitar string, as well as the LCD GtG throttling differences between Hz. So 240Hz-vs-360Hz conclusions should not be misinterpreted with this knowledge of error margins. 240Hz-vs-360hz is only an exact 1.5x blur difference at perfect GtG=0ms and stutter error=0ms.
Also, usefully need to be familiar with the
Vicious Cycle Effect where higher resolutions and FOV amplify refresh rate limitations. I am slow at creating new explainer articles, so some of these are piecemeal all over the place.
Certain stratospheric refresh rates are crapshoot unimportant on a VHS-resolution display, but supremely important on 8K 0ms-GtG displays (e.g. laser, DLP, OLED, MicroLED) -- and it is unfortunate how LCD GtG throttles differences between Hz unlike the prototype displays I am seeing.
If you ever use Blur Busters inspirations, please find a way to cite us! Over the longer term, I'm currently looking to try to create TestUFO DOI's because many papers are starting to cite TestUFO links (and certain Blur Busters articles). Over 100 research papers have obvious Blur Busters inspirations, with only approx 30-35 of them citing my name or TestUFO or Blur Busters (about 20% academia papers, 80% corporate papers) at
www.blurbusters.com/research-papers
By now, you've probably seen my simplified "Cole Notes" explainer articles at
www.blurbusters.com/area51
I am totally happy to vet flaws in research papers, so you are welcome to back-channel contact me at mark [at] blurbusters.com ... I review research for free and commentary on potential error margins many researchers overlooked.
I only mention all this, only simply because many researchers still accidentally make Hz-limit assumptions (the oft-quoted-by-amateurs 255Hz fighter pilot paper only tests one attribute, and the 500Hz mentioned by Oxford university researcher, etc) forgetting some scientific variables I am familiar with that they have overlooked (e.g. making MPRT in milliseconds less than jitter error in milliseconds, for a specific kind of test). And certain tests don't force eye-tracking (e.g. trying to read tiny moving text) -- so people don't notice motion blur while others do, because of differences between eye-tracking behaviours in different humans in different content, but more than >90% people suddenly see 240Hz-vs-1000Hz when I force eye-tracking scrolling-text-identification tests, etc (it's a whopping 4x motion blur difference, after all, 4 pixels of motion blur per frame at 240Hz is a slowly scrolling 1000 pixels/sec on an 8K 240Hz display, unlike yesteryear VGA resolution displays where it was too fast to eye-track) -- so content and job can create major differences in Hz-difference tests.
Even jitter is a major error margin now in these refresh rate stratospheres, since 70 single-pixel microjitters per second at 360Hz blends into +1 extra pixel motion blur (on top of GtG blur and MPRT blur), etc. -- where stutter during stratospheric refresh rates is so fast (beyond flicker fusion threshold) that even wider-amplitude stutter is now generating extra display persistence motion blur blur on top of existing GtG & MPRT blur (as MPRT tests only measures regular pixel steps, without factoring in the fact that high-frequency stutters is effectively worse MPRT -- the regular staircase of the sample and hold effect becoming an erratic staircases due to non-consistent pixel-steps per frame, with the bigger steps becoming the new motion blur weak link). Not to mention that legacy measurement standards have very weak cutoff thresholds. Motion blur below GtG10% / MPRT10% and beyond GtG90% / MPRT90% is still human visible (so VESA misses the 20% slowest portion of human visible blur that can be more millimeters long than the 10%-90% obvious blur on the same said display) -- so those 1ms GtG/MPRT advertisements are not always trusted anymore by today's tech-savvy audience. So many error margins are overlooked by today's researchers!
Also, while many tolerate flicker, Blur Busters is a big fan of eliminating motion blur without strobing. But it would take 3333fps 3333Hz for a fully flickerless display of any kind to match the motion clarity of an Oculus Quest 2 (0.3ms MPRT100% strobe pulses). So we have a long technological engineering curve when Quest 2 VR headsets currently have less than 1/10th the motion blur of a non-strobed 360Hz monitor...
I'll be resuming writing content for Area 51, including converting some of the best forum posts I've made in the last 12 months into Area51-worthy articles -- like a sequel to the 1000 Hz Journey article that has relevant content for researchers of various elements of this refresh rate race -- to help vet research for overlooked / missed variables.