jcs

Why not shoot something around 1 minute first? Then show it for feedback...

The D850 has better color/skintones vs. Sony or Panasonic, however AF isn't really usable and rolling shutter is very high. Do you still have a 1DX II? Curious why you didn't compare the 1DX II, which currently has the best balance of usable features for high-end stills and 4K DSLR video. 4K MJPEG is indeed heavy for storage, though the latest version of Resolve can play the footage smoothly on somewhat old hardware (e.g. 2010 MacPro and GTX 980-level GPU). A modern PC can play the 4K MJPEG (including 60p) in real-time in Premiere Pro CC. When aliasing isn't an issue and doing mostly medium and closeups, 1080p full frame looks excellent / filmic at 24, 60, and even 120fps (DPAF works in all modes), along with fairly small H.264 files.

@fuzzynormal Isaac Asimov wrote a short story called "The Last Question" which is relevant to your comments and this discussion: http://multivax.com/last_question.html (full text here, just a few minutes to read). @Kisaha it's super cool that Nintendo put shrooms (Amanita muscaria) in Super Mario Bros. The 'enlargening" isn't physical in real life with shrooms, it's mental/spiritual These kinds of conversations ultimately affect how we treat each other. It doesn't matter what each of us thinks individually. It does matter when we physically interact with the world and each other. Omniscient, omnipresent, and omnipotent: that's what all humans connected via the internet compose. Quantum physics experiments show that consciousness effects experiments, and Princeton's experiments show further intriguing interactions with thought and physical reality. Thus there may be some truth to visions on DMT/shrooms that we are all connected in a metaphysical way* and collectively co-create reality together (where consciousness is primary). If that's all "true", what does that make us? *related to entanglement in quantum physics and the Big Bang (everything emerging from a single point of unity)

You can get to 24p from 30p by retiming to 80% (24/30 = .8) in Premiere etc. for 120/24 = 5x slomo. We recently shot 1DX II 120fps (PDAF worked well). Without pixel peeping didn't see any resolution reduction vs. 60p (full frame, no crop for both cases). Provides enough resolution/detail for close up and medium shots. For wides best to use 4K60p if detail is important for the shot.

Received a PM asking about this: They cover a few topics, namely science & spirituality, oneness & psychedelics, and is consciousness primary. When we absolutely believe in something spiritual, e.g. something we can't test with the scientific method, that becomes religion. If we absolutely believe in science in the same way, even though scientific beliefs are based on things we can test, that too becomes religion, because we totally believe the results and stop questioning. If someone is certain that there is no God / nothing 'spiritual' happening or anything beyond what science can test (atheist), that's also religion, because they've stopped questioning. This ties in with the latest studies on quantum physics, which has many scientists wondering if perhaps some philosophers were right in asking, "Is consciousness primary?". Which leads some people to take a deeper look at the theory that the universe behaves like a virtual reality simulation, where nothing is rendered/"created" unless there is an observer present to require the energy / CPU expense to "render" that snippet of reality. In "normal consciousness", in human form, we feel we exist as separate entities. When we meditate deeply and/or take psychedelics (DMT, LSD etc.), we perceive reality totally differently- that we are all one, part of a collective living being, all connected, all one with each other. All of us making up "God". This awareness / belief can help us to work together and be kinder to each other, animals, and the environment. And also less likely to follow authoritarian rules (unlike say alcohol which reduces perception/awareness). Fortunately psychedelics are now being used in research and select clinics in helping cure people of PTSD, anxiety, and addiction. While some folks are micro-dosing psychedelics in silicon valley for "competitive creative benefits", most wouldn't want the increased awareness provided by psychedelics all the time, as it would be very overwhelming and hard to deal with everyday tasks for survival. Which makes sense that we have this "human filter" running most of the time for daily living. Humans evolved with psychedelics and they are useful tools for living in the human condition. By themselves, they can't do anything. People must choose to change their beliefs to change their lives. Psychedelics simply provide a brief view of alternate ways to think about literally anything and everything. This "journey" typically has people believing/not believing in many things, and ultimately when one realizes how easy it is to deceive ourselves, as we pull away layers of illusions, we realize that the layers of illusion may never stop. That we can't know anything "for sure". So how do we live, what should we believe in? The scientific method is a useful tool, with limitations. Spiritual/philosophical ideas can lead to new scientific discoveries (such as the Big Bang predicted at least thousands of years ago). What today seems like meta-physics, someday may become regular physics. We can live by keeping an open mind, not "believing" in anything without question, always understanding that our personal perceptions of reality can be illusions. A classic example is a crowd at a crime scene. When interviewed, everyone has a different report, and figuring out what "really" happened takes time. Eye-witness reports aren't always accurate. So the philosophy student asks, "if we can't know anything for sure, how are we supposed to live our lives"? Zen makes sense to me: When hungry, eat. When cold seek warmth. When in rain seek shelter. Understand attachment causes suffering. Be kind to others and forgive quickly- as much for the relief of your own suffering as well as for the relief of a remorseful person. Even if the other person isn't remorseful, it's helpful for the self to forgive, to reduce self suffering. In human form, what else do you need?

Did a quick test- it appears to stop / not-export on sections where the audio has been swapped. I would expect it to render each video section as a separate clip, no matter what audio is included (even if the audio bridges multiple video clips). That would be the 'obvious' behavior, and thus the current behavior looks like a bug. From a code/design point of view, running the renderer for each video clip as the in/out for each clip would be trivial to code... If not they need to abstract out the rendering system and generalize, then doing what I described is trivial to code. Guess product managers have developers working on other things they deem higher priority.

Tried the latest release version of Resolve 14 (free version, will probably get the $299 version after this test). Added a bunch of video and audio effects to a Canon 1DX II clip (1080p, full frame): Pretty impressive performance on a 2010 MacPro and GTX980ti. Close to real time loaded with all these effects (easily realtime with basic effects / grading). Audio when not in the Fairlight panel is still 'clicky' after a few effects have been applied; guessing they're making video a priority and dropping audio samples to 'keep up'. A trivial linear interpolator could eliminate those annoying clicks; hopefully they'll fix that soon. Other than that, Resolve is looking really good. The stabilizer is faster than PP CC, and works much better. Their Film Grain and other locked/watermarked effects looked decent, will test them out after purchasing the app. 1DX II 4K 422 MJPEG also played butter smooth- first time I've seen this footage in real-time on this computer (state of the art PC plays these clips in real-time on GTX1080). Purchasing: when you click the Buy Now button, it takes you to the web, then shows a list of retail stores. It's also available in the Apple App store, however someone reported that the App Store version won't allow the use of 3rd party OFX plugins? Is there a dongle for the retail/box version? One thing I like about PP CC is the ease of use on multiple computers and OSes vs having to schlep around a hardware dongle and use up a USB port.

It's internet etiquette to not reveal movie spoilers without warning. The other thread is about 16mm and technical aspects vs. the story (this thread).

16mm is considered below 2K/HD digital resolution due to large film grain: http://www.cinematography.com/index.php?showtopic=14944 (without grain is similar to HD). In that thread they state there is no actual resolution for film, however if you shoot resolution charts indeed there is... The Mother! trailer looks especially soft on YouTube compared to other trailers, likely due to issues with compression since the grain is so heavy. One of the tricks used for softer / lower resolution cameras is to shoot mostly closeups, as was done with the Canon 5D3 H.264. Low resolution is especially visible in wide shots. Aronofsky used mostly tight / closeups where the soft/grainy look worked well for the narrative of Mother!. To get e.g. the GH5 looking like 16mm film, you'll want to reduce the resolution and add large noise grain (e.g. FilmConvert or 16mm noise scan).

GH5 => 16mm Patina Recipe: Reduce DR, esp. highlights. Nothing will be very bright Down-res to 720p or so, then perhaps scale back to 2K or 4K to add noise (via Nesting, etc.) Apply really chunky monochrome-ish organic noise (FilmConvert or 16mm film scans etc.). FilmConvert LUT for appropriate Kodak etc. film stock could also help (could then skip (1) & perhaps (2) above) https://www.premiumbeat.com/blog/achieving-a-super-16mm-film-look-when-shooting-digitally/

The film has an excellent, organic, abstract, low-res grainy look. The antithesis of pristine 4K! Some films, such as action/adventure/SciFi/Fantasy benefit from ultra clear ultra high-res imagery, and others such as horror/thrillers totally work with organic low-res grainy look: makes one 'less sure' about what they are seeing. More detail reply in this thread (spoilers; best to watch film first): https://www.eoshd.com/comments/topic/25995-aronofskys-mother-post-watch-discussion-spoilers/

@Don Kotlos: Abstraction is an understatement for this film! Pi and The Fountain were a little abstract, Mother! takes abstraction to a whole new level. I think that's why the reviews are polarizing. If one views this movie literally or at face value, it's a bizarrely terrifying horror thriller and ultimately makes little sense. When viewed as an abstraction, paying attention to the symbology, it works as quite an amazing experience. I was laughing during many parts of the film because I saw what he was doing, and it was amazing. Looking around others were horrified or puzzled. My fellow philosopher / psychology enthusiast friend, she was also giggling and got what he was doing. On one level some scenes appear to be absolutely surreal and absurd. However he's doing an expose on materialism, narcissism, selfishness, control freaks, introverts, extroverts, police-military power / powerlessness, insanely dysfunctional families, cults, mob mentality, and religion (mother nature & God, Adam & Eve / Garden of Eden, Cain & Abel, great flood etc.). And perhaps most importantly, it's all illusion, especially romantic love and how psychopaths, who have no natural ability to love or have empathy, take advantage of deep empaths, who have deep ability to love and get taken advantage of by psycho/socio-paths. And ultimately he's making fun of Hollywood! After watching the Fountain and Mother!, it appears Aronofsky has deeply studied psychology, religion, common (illusory) social constructs, and especially concepts of attachment and suffering from Zen/Buddhism. Truly a brilliant work with so many concepts compressed into a 2 hour film.

As I watched Aronofsky's Mother!, I noted how soft and noisy it was, figured 16mm film (it was: http://www.imdb.com/title/tt5109784/technical?ref_=tt_dt_spec). Lots of handheld, close follow shots, POV pan-reveals, and overall genius-level unnerving camera motion and editing. This is probably his best film to date: A+, 5/5 etc. Camera, color, editing, acting, story, lighting, music, sound, VFX, all top level work. He digs deep into the human psyche / ego / control / fear and doesn't let up 'once it starts' (you'll know what I mean when you see the film). For those who haven't studied psychology, it might seem incomprehensible. However while it works as an amazing horror movie even if it doesn't make sense, it's genius on so many other levels! Worth repeat watching to learn more, it's that good. Recommend not reading any reviews beforehand, and hopefully no spoilers are leaked in this thread. [edit: removed the trailer link, best to see it without any fore knowledge!]

I've used s3 for business projects: I don't think it would be worth the cost in your case, and certainly couldn't outperform setting up a local file server in your location since you have a 200Mbps fiber connection. Once your clients upload files to your local file server, you'll have access via your local network at 1Gbps (or 10Gbps if you have those NICs).

What makes it cool is ease of use for these advanced effects. Lots of creative possibilities! Next they'll add depth sensors so we can simulate any camera/lens in post

You may find that when you set deadlines, things get finished, and when you don't...

In order to plan the bacon-wrapped-shrimp party when's it gonna be done?

This is the real issue from that article: I could see Amazon/PrimeVideo acquiring Netflix in the future. Amazon makes money in other ways which could help deal with the internet/media monopoly. Netflix is making original content for survival as the media companies own everything else. Most indie content isn't worth watching for most people. What's needed is better search/matching tools for 'long tail' indie content that could be efficiently matched with micro markets. Ultimately, 'good content will become popular virally'. If people aren't sharing your content, it's not good enough (or you need to find your market where they'll share with others of similar interest). Once again our old friend Artificial Intelligence could be leveraged to provide better content matching with long-tail micro markets. Until we get our quantum entanglement people-owned wire-free and radio-free network (not communism/socialism, something else), these issues will persist.

Note that pretty much no one could guess the right camera order in the A7S II, C300 II, 1DX II side-by-side test on EOSHD: And even on the more 'working professional' oriented dvxuser: http://www.dvxuser.com/V6/showthread.php?346890-1DX-II-A7S-II-C300-II-Compared-Side-by-Side I'm currently selling a bunch of gear, GH4, A7S I, Panasonic and Sony lenses... These posts have me seriously thinking about selling the A7S II (along with Speedbooster and MB IV adapter), and getting an 80D or 5D4 as the light/travel camera to replace the A7S II. In this 80D vs. Alexa demo: I have a pretty good idea how to get the 80D looking more like the Alexa with a custom picture style and WB offset. The 80D looks a lot more like the Alexa than the A7S II, it has DPAF, and can use all my Canon lenses natively (with AF). The 5D4 might make more sense if the 1080p is similar to the 1DX II (color/look is more important than the 4K crop). If the A7S III has something close to Canon DPAF, it will be worth a look (especially since they'll likely improve color quality and IBIS as well). Until then I might let the A7S II and Sony lenses go too...

Yeah, that would be better; if the D16 is capturing an accurate image of real-life, then the resulting 3D LUT might still be helpful. Panasonic is listening to customers: on a GPU a 3D LUT operation can be very fast. Not clear if the GH5 has builtin 3D LUT hardware or GPU level processing cycles available.

Haha, good one! Wait, are you seriously taking Wes Anderson seriously?

Take your GH5 and take a picture of that image. Take the GH5 copy and the original into any tool which does shot matching. Perform the shot match operation. Test and tweak further to get both images to match as close as possible by eye. Save out the resulting 3D LUT. To generalize and automate this process, so the resulting LUT will work better in the general case, use machine learning to train a neural network to create an optimized LUT based on GH5 input and <target camera> output

Netflix Originals are most likely Sony or Red because of the "4K" requirement (which seems like a political thing against ARRI domination, and perhaps a reason behind Steve Yedlin's recent resolution test. Alexa 65 is rental only and not widely available). Sony pro video cameras have been traditionally more reliable vs. Red (which in recent time is also pretty reliable), and thus Sony is a good guess. And then I looked it up ;). Watching again on the desktop with a high-end 32" calibrated 4K Dell display (vs. on my iPhone when originally replied), it does have that Sony video look. It's suppressed mostly by using low-contrast/glow lens filters (see glow on very bright areas), expert lighting, and expert grading. The recent Sony camera which doesn't have that inherent video look is the F65. Using the F65 for a Netflix production is probably too expensive (storage) and maybe the camera size itself is an issue (It's dumb, but people choose gear based partly on how it looks. The F55 is a lot sexier than the F65. This was pointed out by Lucy's DP when they chose the F65 vs. ARRI and Red based purely on image quality (Red used for some shots)). Can I get results I like with Sony, sure, with a lot of work in post (A7S II): Simple camera-specific things that help make an image 'filmic or cinematic': High color fidelity: lots of colors and color variation. For the final 8-bit render, lots of color/tonal variation is used and maximized. This is especially important for skintones. If we remove color and view the image in grayscale, we look for tonal variation in light & shadow, and texture detail, especially in skin. High intensity white is rarely if ever seen. In a sense, filmic tends to be lower dynamic range than real life. Highlights don't clip very often, and when they do, it tends to be a smooth effect vs. a hard clip. Noise adds texture and even when it's subtle does something to make the brain think the experience is more organic, more analog, vs. digital. Perhaps acting like temporal dithering in a way. Motion cadence and motion blur- more subtle, though people notice differences in motion, some cameras are more pleasing than others. Then we can add low-contrast/glow filters, lighting (most important), color grading/look, story, performance, sound (especially ADR and Foley), etc. to get the full filmic/cinematic experience. Shallow DOF is not filmic per se, as some directors love deep DOF, and ultimately depends on the story and emotion of the shot. ARRI, Red, and Canon (in that order) tend to produce the most filmic look with the least effort (Black Magic is getting there, and once reliability and quality improves, will be a contender). Sony and Panasonic don't look like ARRI, Red, or Canon (F65 and Varicam come closest). We've all seen studio tests where a bunch of cameras are graded to match and it's hard to tell them apart. For actual productions, the cameras that produce the best results with the least effort, are the most forgiving of errors (exposure etc.), ultimately produce the best results over time. And we can see why ARRI dominates, and why ARRI in general looks better than all the rest. It just takes the least effort. Red looks pretty good lately but still hasn't matched ARRI (and also takes more work in post). In the next tier down, Canon takes less effort than Sony or Panasonic. When the C300 II was priced higher, the FS7 got a brief boost because of the slomo craze, however now that the C300 II price has come down, that's not the case anymore. Look on eBay and see all the FS7's for sale (a bunch last I checked) vs. C300 II (none last I checked). You can also see true value by studying used prices. The 5D3 held it's value for a very long time (and still does relative to similar Sony and Panasonic cameras). 5D3 raw looks better to me than the F55 (but not the F65!). So if you're letting us know the mystery camera was the F55, then all I can say is wow! A fun exercise would then be, how do we intentionally make footage look like video? Use 'bad' lighting- low CRI/TLCI, unflattering light positions etc. (low CRI/TLCI will help with (4) below) Clip the highlights and/or blow out the whites frequently, especially in skin Go crazy on noise reduction- little or no noise, especially in skin. Reduce texture as much as possible to get that plastic look Massively reduce the color space. Imagine an RGB cube, representing all the possible colors. Now compress/crush it so many original colors are now mapped to the same color (quantized). This is especially important for skintones: the less color variation/tonality the better. Underexposing and/or shooting in very low light and using aggressive in-camera NR can also help (both (3) & (4)), then bring it back up in post: low noise/detail and low color variation! Enhance Magenta/Green instability. Meaning, make the highlights tend toward magenta and the shadows tend toward green. Make it impossible to get natural skintones without needing advanced / secondary color correction Produce a motion cadence / motion blur that makes 24p look like 30- or 60p That said, really "old" video, as in tape, has a fun look of its own, as with Kung Fury: Still looks great, was shot on Canon 5D3 and Sony FS700 (slomo). The FS700 footage needed less "work" to look like video (again, had one for a few years, great slomo, (internal) color was a challenge (external recorders look better)). While I agree that just about any camera today can be made to look filmic/cinematic, ARRI, Red, and Canon (and even BM) make it easier than Sony or Panasonic. We each have our own priorities, and for me number one is skintones. I still use the Sony A7S II when I want to use something small and light (the 1DX II is really right there for usable low light with the A7S II). However, SOOC the A7S II footage will look more like video than the 1DX II (esp. with my custom Filmic Skin profile). In the studio with high quality lights, the A7S II, C300 II, and 1DX II can match fairly closely (would need secondary CC to fix A7S II lipstick): As mentioned many times before, comparing Canon (or ARRI) to other cameras that one thinks looks good, is a useful exercise. Using a Canon camera as a reference for Sony is what led Andrew to start making picture profiles for Sony to match Canon. If Canon wasn't 'better', why bother? ARRI and Canon aren't perfect, however so far they are tools which provide the most pleasing color/look (esp. skintones) vs. the other brands for the least effort. Is anyone selling Red, Sony, or Panasonic looks for ARRI or Canon cameras?

Looks decent. The F55 can be matched to Alexa with enough work (at least in the studio).

Happy Birthday Jasper! Lol how many did you have? Bacon wraps nicely around prawns too. <?>?</?>