QUICKSTART: Begin UltraHFR video making!

Discussion about 120fps HFR as well as future Ultra HFR (240fps, 480fps and 1000fps) playing back in real time on high refresh rate displays. See Ultra HFR HOWTO for bleeding edge experimentation.
Post Reply
User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

QUICKSTART: Begin UltraHFR video making!

Post by Chief Blur Buster » 29 Apr 2018, 17:09

QuickStart: What Do I Need For UltraHFR Video Making?
Realtime UltraHFR (not slow-motion) playing full framerate on full high-Hz displays.

Image

More recent amateur-league UltraHFR experiments are occuring between a few people around here, online and offline.

Recommended Computer Specs
- Any recent CPU made in the last few years (i5, i7, Ryzen, etc).
- A recent NVIDIA or AMD GPU, the fastest you can afford.
- A recent SSD at >500 MBytes/sec. For faster, consider M.2 SSD slots on motherboard or PCI Express cards.
- Computers capable of 4K60 can generally do up to 1080p240.
- Computers capable of 8K60 can generally do up to 1080p960.
- For displaying the files in a compatible workflow, you are currently limited to 4K120Hz, or 1080p240Hz, or 540p480Hz.
- Cheap 240fps UltraHFR: A basic midrange NVIDIA/Radeon GPU.
- Cream-of-crop UltraHFR: GTX 1080Ti or GTX Titan X Black+ on recent i7/Ryzen and Samsung 960Pro M.2 SSD or 970Pro M.2 SSD.
- Basic UltraHFR displays are surprisingly not too expensive, 240Hz costs under $500 and 480Hz costs under $1000.

Test System Load for 720p240fps H.264 .MP4 video files at 100 megabits/sec on i7-7740X with GTX 1080Ti
- CPU usage 18% (via Task Manager)
- GPU usage 4% (via GPU-Z from TechPowerUp)
- Predicted capability: 1080p960fps or 1440p960fps UltraHFR (to be tested this year)

240fps Players with perfect framepacing with UltraHFR files on 240Hz displays (H.264)
- Windows Media Player
- MPV player
- Chrome browser embedded player

480fps Players with perfect framepacing with UltraHFR files on 480Hz displays (H.264)
- MPV player

Failures
- Fail 480Hz+ Windows Media Player refuses to play 480fps files completely (thinks they're invalid files)
- Fail 480Hz+ Chrome embedded player microstutters too much (not 1:1 puldown) at 480Hz and up UltraHFR
- Fail 240Hz+ FireFox embedded player microstutters too much (not 1:1 pulldown) at any UltraHFR framerate or Hz
- Fail 240Hz+ VLC microstutters too much (not 1:1 pulldown) at any UltraHFR framerate or Hz

NOTE: We have found frame pacing precision is usually the video player's author fault. Sub-millisecond-accurate framepacing is absolutely essential to round-off played frames to the refresh cycles correctly. Millisecond-accurate framepacing is too low-resolution for 480Hz (frames are only 2ms, so 1ms rounding errors will often miss refresh cycles and create random-pulldown like 0:1:2:0:1:2:0:1:2 instead of correct 1:1:1:1:1:1:1:1 perfect frame:Hz sync). The conclusion is PLEASE use things like RTDSC or QueryPerformanceCounter() or std::chrono::high_resolution_clock::now() instead of old fashioned millisecond clocks, for software developers writing open source video players. Submit appropriate BugZilla ticket reports where applicable.

240Hz displays: Buy one from our monitor lists
- Acer Predator XB251HQT
- Acer Predator XB252Q
- Acer Predator XB272
- Acer Predator XF270HA
- Acer Predator XF250Q
- AOC AGON AG251FG
- AOC AGON AG251FZ
- ASUS ROG PG258Q
- ASUS ROG Strix XG258Q
- BenQ ZOWIE XL2540
- BenQ ZOWIE XL2546
- BenQ ZOWIE XL2740
- Dell Alienware AW2518H
- Dell Alienware AW2518Hf
- LG 27GK750F
- Viewsonic XG2530
- Viewsonic XG2560

480Hz displays: Current true-480Hz monitors:
- Zisworks 4K120 http://www.zisworks.com/shop -- Supports 4K 120Hz, 1080p 240Hz, 720p 360Hz and 540p 480Hz
- A little under $1000 for the preassembled engineering sample, must supply your own VESA mount (~$20 from Staples or Monoprice)

Video workflows used
- Blender works with UltraHFR!! It can speed slo-mo camera videos to 240fps and 480fps. Audio is preservable on some cameras (e.g. Hallelujah!!!
- Slomo cameras sped-up to realtime using instructions here
- Very few compatible workflows yet. For source engine videos, you can use "makemovie" -- that can "screenshot-every-frame" (using SSD or SSD RAID) -- it is possible to create true 480fps gameplay video files using Half Life engined games. Many SSDs including Samsung 960 Pro M.2 SSD seems to have no problems keeping up with 480 screenshots per second (960 files written to disk per second).

Cameras successfully used
- GOPRO Hero 4
- GOPRO Hero 5
- GOPRO Hero 6
- Sony FDR-X1000V action cam
- If you have lots of money, Phantom Flex cameras also work. However, most amateurs cannot afford them.
If you tested 240fps on any action cam and it works well enough for UltraHFR, please follow up and let me know
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

QUICKSTART: Begin UltraHFR video making!

Post by Chief Blur Buster » 29 Apr 2018, 17:13

You can use multiple separate GoPro cameras to record higher framerates -- e.g. two 240fps GoPros recording simultaneously to generate 480fps video files, see How to use two 240fps GoPros simultaneously to create true-480fps video.

phpBB [video]


If you need to purchase UltraHFR compatible cameras, please support Blur Busters via using these links.
- GOPRO Hero 4
- GOPRO Hero 5
- GOPRO Hero 6
- Sony FDR-X1000V action cam
If you tested 240fps on any action cam and it works well enough for UltraHFR, please follow up and let me know

And please, participate in these forums to help us improve the art of UltraHFR video making and discovering workflows that successfully work in this bleeding edge pioneer territory. An example thread includes this thread illustrating our progress (previous pages includes a links to 480fps gameplay video tests)

To share to others, you must share video files natively (e.g. links to .MP4 files). Cloud services work, such as 1TB Dropbox or GoogleDrive accounts. Doing YouTube/Vimeo won't work. Must be an .MP4 file download. Embedded .MP4 playback will however work in Google Chrome (up to 240fps), but .MP4 downloads are recommended, and generally recommending users to use MPV player because it is the most reliable UltraHFR capable video player at the moment.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
POVHFR Videos
Posts: 9
Joined: 27 Apr 2018, 00:44
Location: Atlanta, GA
Contact:

Re: QUICKSTART: Begin UltraHFR video making!

Post by POVHFR Videos » 30 Apr 2018, 00:07

NCH Video Pad Video Editor is also another program I used and use for Ultra HFR video. Video Pad can export videos at 1000fps. The 'Detect..' tab can detect any frame rate but can save videos up to 1000fps. Blender 2.79 can now save any frame rate you throw at it*! I'm aiming for 1920fps now, but blender can save as high as 12000fps or higher!! :shock: :shock: :shock:

User avatar
YeahWhiplash
Posts: 1
Joined: 27 May 2022, 01:13
Location: Los Angeles
Contact:

Re: QUICKSTART: Begin UltraHFR video making!

Post by YeahWhiplash » 27 May 2022, 01:34

Hi, apologies if this isn't the right place to post this but this does seem to be quite a niece thing. I am a cinematographer based out of LA that is very interested in experimenting with ultra high frame rate. I have talked to various cinematographers and even some ASC members, but nobody seems to have an interest in UHFR cinematography.

With the release of the V-Raptor, I believe there is a camera with suitable resolution, color science, raw codec, and framerates to do some iterative experimentation with. I'm interested in doing tests at various shutter speeds/lighting conditions/movement/exposure around the 120 and 240 fps framerates (I would do 500 or 1000, but one step at a time, the cameras/displays will catch up and I do have a 244hz monitor for viewing). I would like to do these experiments so I can get a better understanding of how shooting conditions differ while shooting at higher framerates and to better inform myself when it may be appropriate to use UHFR as an artistic consideration. I would even be open to doing some narrative work with UHFR under the right conditions. I feel like this is one of the only places in the world where I might find another person interested in this sort of experimentation, so if you're local to LA and would like to do some tests with me sometime please reach out!

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: QUICKSTART: Begin UltraHFR video making!

Post by Chief Blur Buster » 27 May 2022, 17:07

YeahWhiplash wrote:
27 May 2022, 01:34
Hi, apologies if this isn't the right place to post this but this does seem to be quite a niece thing. I am a cinematographer based out of LA that is very interested in experimenting with ultra high frame rate. I have talked to various cinematographers and even some ASC members, but nobody seems to have an interest in UHFR cinematography.

With the release of the V-Raptor, I believe there is a camera with suitable resolution, color science, raw codec, and framerates to do some iterative experimentation with. I'm interested in doing tests at various shutter speeds/lighting conditions/movement/exposure around the 120 and 240 fps framerates (I would do 500 or 1000, but one step at a time, the cameras/displays will catch up and I do have a 244hz monitor for viewing). I would like to do these experiments so I can get a better understanding of how shooting conditions differ while shooting at higher framerates and to better inform myself when it may be appropriate to use UHFR as an artistic consideration. I would even be open to doing some narrative work with UHFR under the right conditions. I feel like this is one of the only places in the world where I might find another person interested in this sort of experimentation, so if you're local to LA and would like to do some tests with me sometime please reach out!
Good points, but we discovered something new recently in UltraHFR experiments.

The big problem is refresh rate incrementalism.

60fps-vs-120fps is an 8.3ms difference in display motion blur (on sample-and-hold).

120fps-vs-1000fps is a 7.3ms difference in display motion blur (on sample-and-hold).

Display motion blur physics is actually really simple for a GtG=0ms display (where all display blur is MPRT-based)
Remember, 7.3ms better = 7.3 fewer pixels of motion blur at 1000 pixels/sec panning motion
Remember, 7.3ms better = 73 fewer pixels of motion blur at 10,000 pixels/sec panning motion
Remember, 7.3ms better = 7.3 fewer millimeters of motion blur at 1000 millimeters/sec panning motion.
Remember, 7.3ms better = 73 fewer arcminute of motion blur at 10,000 arcminute/sec panning motion.
You get the idea.

Remember higher resolutions amplify refresh rate limitations. At 1080p, motion at 10,000 pixels/sec would be too fast to eye-track. But at 8K, the motion of 10,000 pixels/sec is barely more than one screenwidth per second. That's enough time for human eyes to track along an 8K pan moving at 10,000 pixels/sec. And then notice there's 73 more pixels of forced display motion blurring (above and beyond natural human vision) at 120fps 120Hz sample-and-hold, versus 1000fps 1000Hz sample-and-hold.

You need to go ultra-dramatically up the diminishing curve of returns when increasing HFR frame rates. Although there were benefits visible to many, 120Hz-vs-240Hz is almost population-worthless for UltraHFR movies. Go big or go home, if you're trying to wow a bigger % of human population.

240fps-vs-480fps is just a temporal equivalent of VHS-vs-DVD (2ms blur difference). Many grandmas couldn't see that.
120fps-vs-1000fps is more a temporal equivalent of VHS-vs-4K (7.3ms blur difference). Most of the population could tell the difference.

Now if you're avoiding strobing as a method of eliminating motion blur (The Stroboscopic Effect of Finite Frame Rates).

We even do multiple Hz-vs-Hz presentations and show the need for this "dramatic jump up diminishing curve of returns" off in multiple presentations.

Image

Because we work with prototype displays, our knowledge is slightly ahead of cinematographers when it comes to "humankind population benefit".

Also, things like refresh rate combining (e.g. 8 different 120Hz projectors with separate 120fps video files each, stacking onto a screen strobing round-robin, to create 960fps 960Hz using today's projector technology). 8 different systems on a genlock bus can display one frame each, etc.

4K 1000fps 1000Hz is possible with today's technology using some parallelism tricks. Sped-up Phantom Flex video files, or round-robin capture (8 concurrent 120fps cameras taking frames 1/960sec out of phase of each other). Parallelism can be applied to both camera-side and display-side to enable 8K 10,000fps 10,000Hz in less than one human generation (20-years), for retina-refresh-rate tests.

The high budgets available for things like simulators, can make this happen, as infinite refresh rate scaling is possible by adding together projector refresh rates, with certain types of projector technologies (especially LCD and LCoS that has fast enough GtG to be hidden in the strobe) -- just refresh low enough to hide LCD GtG, and create a single sample-and-hold projector screen out of multiple strobed projectors. I'm currently in the process of creating and releasing a white paper about this as we speak.

I'm currently working with multiple parties to help them to deploy enabling technologies necessary to jump more dramatically up the diminishing curve of returns faster.

As refresh rate snobs ourselves, we frown on refresh rate incrementalism, since human visible benefits in the refresh rate stratospheres requires bigger jumps up the diminishing curve of returns. Otherwise mainstream media squeals "Not worth going above 120Hz".

If you must do just 240fps HFR, a good stepping stone is a strobed 240Hz display. 240fps 240Hz strobed will break the motion-blur barrier (1/240sec of guaranteed display motion blur on current E-Cinema projectors). Now that being said, even Christie can't do 30-bits of color in 1/240sec, so you get some temporal loss per refresh cycle with DLP technologies, unless you add more projectors to cram more bits of temporal dithering per Hz. The fastest DLP chips are 2880Hz pixels that are binary 1-bit mirrors (fullwhite and off) that are toggled rapidly to generate color. This enable 48-bit color at 60Hz, but only 6-bits of color per channel per 480Hz refresh cycle on a 3-chip DLP, so infinite refresh rate scaling (via projector stacking to add more Hz) is easier on other projector technologies that don't generate subrefresh temporals.

Remember source blur and destination blur is additive. Sample-and-hold 240Hz means 1/240sec display persistence + 1/240sec shutter speed = 2/240sec = 1/120sec of human perceived display motion blur during fast panning. That's the same blur as a 1/120sec SLR photograph of a fast-moving camera lens. Definitely far from retina refresh rate for fast-motion material.

Relevant thread: Properly Designing A Blind Test That >90% Of Humans Can See 240Hz-vs-1000Hz (non-Game Use Cases Too!)

We are coaxing multiple parties towards a "1000fps 1000Hz Mother of All Demos" which should silence a lot of old assumptions of Hz/fps benefits.

While 240fps HFR is vastly superior to 48fps HFR, I have to admit that 120fps HFR vs 240fps HFR is ho-hum. Think higher. Many people don't realize 8K 1000fps 1000Hz is achievable sooner than many think, through parallelism tricks.

Currently, 240fps 240Hz is doable with a one-camera one-projector workflow, but if you don't mind parallelism techniques, you can go well beyond -- which is super important in showing off to other cinematographers the benefits of jumping more dramatically up the diminishing curve of returns.
YeahWhiplash wrote:
27 May 2022, 01:34
I would do 500 or 1000, but one step at a time, the cameras/displays will catch up and I do have a 244hz monitor for viewing
Make sure you upgrade quickly to 240Hz OLED as soon as such options become available. LCD GtG limitations reduce the difference between 120Hz and 240Hz less than the blur difference of 120Hz vs 240Hz on the new Christie DLP projector.

GtG is an error margin in Hz-vs-Hz tests, and is a big reason why 240Hz-vs-360Hz is only a 1.1x blur difference rather than the proper 1.5x blur difference.

Don't forget that there are two pixel response benchmarks, GtG versus MPRT, so we want GtG near 0 as possible, so that all remaining blur is display persistence related (sample and hold effect) for much simpler blur mathematics of Blur Busters Law.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

stl8k
Posts: 23
Joined: 15 May 2019, 07:59

Re: QUICKSTART: Begin UltraHFR video making!

Post by stl8k » 14 Jun 2022, 10:15

YeahWhiplash wrote:
27 May 2022, 01:34
Hi, apologies if this isn't the right place to post this but this does seem to be quite a niece thing. I am a cinematographer based out of LA that is very interested in experimenting with ultra high frame rate. I have talked to various cinematographers and even some ASC members, but nobody seems to have an interest in UHFR cinematography.

With the release of the V-Raptor, I believe there is a camera with suitable resolution, color science, raw codec, and framerates to do some iterative experimentation with. I'm interested in doing tests at various shutter speeds/lighting conditions/movement/exposure around the 120 and 240 fps framerates (I would do 500 or 1000, but one step at a time, the cameras/displays will catch up and I do have a 244hz monitor for viewing). I would like to do these experiments so I can get a better understanding of how shooting conditions differ while shooting at higher framerates and to better inform myself when it may be appropriate to use UHFR as an artistic consideration. I would even be open to doing some narrative work with UHFR under the right conditions. I feel like this is one of the only places in the world where I might find another person interested in this sort of experimentation, so if you're local to LA and would like to do some tests with me sometime please reach out!
Great to see you here! I'm STL8K an innovation pro with a passion for imaging (capture) and display.

In addition to Mark's thoughts, let me suggest a few ideas for finding collaborators and immersing generally:

- https://www.linkedin.com/in/demetri-portelli-4a31a44/ is a Canadian who worked with Ang Lee.
- For HFR imaging, Japan is where much of the innovation is happening. Perhaps look for the folks who worked on these projects: https://www.nhk.or.jp/bs4k8k/eng/projects/ especially the newer ones. For example, https://www.linkedin.com/company/eleven-arts-inc/ is in LA. I presume some of those projects used HFR. All were 8K.
- If you don't have SMPTE Digital Library access, get it and get your HFR nerd on. https://innovate.ieee.org/smpte-digital-library/.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: QUICKSTART: Begin UltraHFR video making!

Post by Chief Blur Buster » 14 Jun 2022, 15:25

Thanks for the follow up!

The bottom line is that the next HFR step after 120fps HFR should be 1000fps UltraHFR, unless you're stuck with a live-film single-camera workflows, or have to distribute films to existing HFR theatres of a target frame rate, have to work with existing equipment, and can't afford the framerate-multiplying/parallelism workarounds.

The technology exists with creativity, if we're bothering to demonstrate HFR and spending lots on equipping a ride-simulator theater, a figher jet simulator, or a HFR cinema -- then I say Go Big Or Go Home.

To sufficiently properly wow the population, the 120-vs-1000 wow effect is as big (or bigger) than the 60-vs-120 wow effect. The 120-vs-240 is pretty minor. That being said, you can still distribute 960fps Ultra HFR to 120fps and 240fps HFR theatres.

The problem is that mere petty refresh rate incrementalism does not acknowledge the fact we need to jump geometrically up the diminishing curve of returns.

Understandably (initially) computer generated (CGI) material might be more appropriate demo reel for Ultra HFR, since it eliminates the need to purchase 8 separate 8K 120Hz cameras for parallelization for 960fps Ultra HFR tests. Bonus is that open source apps (e.g. Blender) are more easily modified with minor changes to support Ultra HFR frame rates.

We want to see a The Great Demo (8K 1000fps 1000Hz) Ultra HFR demonstration as soon as possible, because it is possible with today's technology (refresh rate parallelization and combining via camera array & projector array) -- it just expensive to parallelize. But as tech improves, fewer cameras/projectors need to be parallelized, until it's achieved by a single camera.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

Post Reply