tellure

That footage looks superb for $500. If you had told me that was from a GH5 or A7III I would have believed it.

Great shooting and editing. Love all the quick close up shots and speed ramping, keeps things flowing and pacing high. Would have maybe liked a few more slow sequences, and a better voice for the narration. Really impressive overall.

Yeah that does look great for a $400 camera. The stabilization is great and color and sharpness look good. Beautiful parallaxing with that epic mountain backdrop. Seems like hyperlapse is the main culprit in the earlier videos.. when so much of the image is changing frame-to-frame these low bitrate streams can't keep up.

Thanks for the links! Some of the 4K hyperlapse in those videos do look better, but the ones with a lot of detail are pretty similar to the Vimeo one, and some are worse. Of course some of this is no doubt down to YT and Vimeo's crappy bitrates, so maybe when we can get some original files we can see how much of this is the camera. 100% crop from 4K framegrab:

Damn the compression artifacts in that Vimeo video are pretty terrible. Sigh.. guess they must be recording their promo videos to an external recorder or something. 100% crop:

Editing Sony's native h264 X-AVCS is definitely slow for me even on a high-end PC / GPU combo in Premiere. Scrubbing isn't instantaneous and re-positioning the play head and pressing play takes a half second to a second to start playback. So like jonpais I also transcode now to ProRes proxies (1080p) to get fast editing.

Thanks for posting this, super impressive. I'm now planning on buying one, barring any issues that come up with a bit more research. It it surprising that we haven't seen anyone do software-based stabilizatrion with accelerometer data on a prosumer camera yet. Looking forward to that on the A74..

Sony Zeiss 35mm f/2.8 is pretty close.. https://www.bhphotovideo.com/c/product/1008123-REG/sony_sel35f28z_sonnar_t_fe_35mm.html?ap=y&gclid=Cj0KCQjwrZLdBRCmARIsAFBZllFBLn-0ZDc2k9aqFpGWSPxjzghSoOgptObYPup7MxiqO3_mAFrKESkaAq63EALw_wcB&smp=y

Great stuff. Tons of energy and impact in your shots. Also slo-mo drumstick flip was sweet .

Nice stuff stv, I thought the film effects and leaks were well done. Especially when the leaks were timed to the little bursts of static in the music (unless those were added in post?). I agree with anonim about the structure of the train travel breaking up the stops for travel moments was really effective. There were a few train station shots that lasted a bit too long IMO but overall well done.

I quite like the image quality in both these videos but the pacing is killing me.. in the first one with all the slow focus pulls from static background to a static foreground object (rock or piece of paper, etc.) are just too much of a snooze for me.. sorry, I couldn't finish. I feel like for shots like that to hold interest they need to have at least some interesting motion going on (camera or subject), or just phenomenal composition / framing / color / etc., like gallery-quality landscape photography. My suggestion would be to hold all the shots for way less time and try and tell a bit more of a story from the sequencing at least in terms of a progression between spaces or time frames. In the kid video I quite liked the tree climbing section at the beginning because it's all shot so tight and there's a huge 3D feel because of all the parallaxing of the branches and the body, but then it got pretty boring with just walking through fields and stuff so I couldn't finish it. I'm sure it's an awesome video for a parent though, since watching your kid play is endlessly entertaining. Sorry, don't mean to sound negative, I'm a total amateur here so not trying to come across as some kind of expert. And just so this post is not all critique I'll put some of my own vacation stuff out here to be cut up :). This video is from a trip to Granada, Spain a couple years back: There's lots of static backgrounds in this video too, which can be pretty boring, so I tried to keep the visual interest with a lot of camera motion, probably too much in fact (most of it is walking with the camera on a gimbal, hence the title), and trying to find some interesting people scenes in the environments. Tech specs: Sony A7R2, Zeiss 24mm f/1.8, Sony FE 55mm f/1.8, Sony 10-18mm f/4, PilotFly H2. In terms of making static landscape shots visually interesting, I'm a big fan of Camille Marotte's stuff (example below). He captures a lot of beautiful details in a travel context and often does a great job composing with very simple elements - e.g. the opening shots of this video with an out-of-focus arm resting on a boat, framing an ocean sunset or the sun flaring into the lens as it moves across the ripped edge of a tarp. And of course there's the people.. in travel videos I find we all yearn to see the people since so much of what defines the flavor of a location is the people themselves.

Super excited to hear that a compact version is possible; I'll definitely be in for buying one when you kickstart it. Given the size constraints most camera mfrs are working within it seems like it might be a long time until variable ND's appear inside the body of consumer mirrorless camera lines, so your approach of an external solution seems like it will serve this need really well. Any chance of a smartphone app to control it instead of another dedicated device? Would love to avoid having to carry another piece of hardware.

I love the functionality of this device and hate messing with my manual variable ND while shooting video. I hate to say though that I probably wouldn't buy one unless it was almost as small as a variable ND. Maybe a bit bigger radius and thicker but not nearly as big as the prototype there.. it would need to be small enough to fit into a camera bag while attached to a lens. I'm a hobbyist shooter though, I could maybe see a dedicated crew shooting with something that large.

I'm starting to think that the only way 4k60 could happen on the A7SIII is if they switch to H265. Roughly half the file size (or double the compression ratio) for the same quality. If they can get the BIONZ XI processor or whatever the A7III will use to do H265 4K encoding with around the same thermal profile as it did encoding H264 4K then in theory we could see 4K60 at Sony's favorite bitrate of 100Mbps (as much as we would all love to get higher bitrates).

I'm praying for 4k60 on the A7SIII just like everyone else but I'm not holding my breath. If they didn't even do it with the much larger body FS5-II what hope do we have of getting it on the A7SIII? Maybe we'll get a bump in the bitrate or something. Or maybe they'll make the jump to h265 finally and we'll get better quality with the same bitrate. And they can finally claim they bested the 4 year old Samsung NX-1.

It seems like the main message here is "If you learn the real skills of cinematography then the quality of your images will far eclipse all the high tech features, megapixels, and technical specs of the modern digital video age." And it's true - great craft and artistry will win out over tech specs. But I also think this is message overlooks some obvious realities about the nature of amateur and hobbyist videography - that tech specs often do increase image quality at relatively low cost. Sure I could spend many months and even years learning great portrait lighting techniques so I can light a interior people shot really well, or I could buy a 50mm f/1.4 lens, slap it on my high-ISO/low-noise full-frame 4K camera, do a basic rule-of-thirds framing with existing ambient lighting, press the touchscreen focus with face detect and boom, I now have a fairly professional looking shot with creamy bokeh and sharp eyes in focus. So sure it could be a much better image if I go much deeper into learning cinematography, beyond basic and overused techniques like point-of-focus shots. But the shortcuts and some of the tech specs that enable them can be very high bang-for-buck for us hobbyists and amateurs. Even more so if you're a vlogger or hobbyist that shoots run-and-gun or only with natural/available light under uncontrolled situations and you don't have the opportunity to do lighting setups or carefully framed shots. I think Boyle's message is still correct and can even inspire us shortcut-happy hobbyists (I am firmly in this category) to go deeper. But it does miss the point for a huge chunk of the videography audience, since few of us have the time or commitment to achieve the level of artistry he's calling for. Technology has enabled us to generate images that make us and our fans/friends happy, often with an investment that matches our commitment or resources. All of us are trying to maximize our return on investment, whether that's in time, money, or stress/difficulty spent learning a technique or tool, and technology has definitely escalated the return on investment ratio.

Minor note, but if you're shooting 4K and outputting to 1080 you can do the shaky-cam / stutter / re-framing stuff in post (albeit at the cost of a bunch of editing time). Might be challenging to get that 16mm hand-held film camera look unless you're practiced at it.

I have the same issue as the OP.. especially with those personal/family projects where there is no deadline, it's super easy to keep putting it off. Here are a few things I've found have helped me. Also, is this the most first-world problem or what? General life / process stuff: Find the music first and let it guide you. Usually you have an idea of the style of the video, and the tempo or mood you shot for, so instead of grinding through 10 days worth of clips it can be more fun to go trawling through some really cool music listening for the piece that works best. It does help to have a few clips to put up against the music candidates, but it's usually pretty quick to find a few good shots for that. Once you have the music (or a few top candidates) it can help clarify the tempo and pacing and sometimes certain sequences will start to take shape in your mind (e.g. an cool intro or a dramatic crescendo). The music can then act as a good motivator for starting to work through the clips because you're thinking about how the moments you're finding can fit into the different musical sections. Temporarily suspend some other personal projects/hobbies. All the other stuff in our lives can easily get in the way of starting a big project like this. Sometimes you need to give yourself a bit more breathing room. For example right now I'm taking a 2-week hiatus from guitar practice so I can get enough time to work on a personal vacation video project. Reviewing clips: Use markers in the clips to save the good moments. You mention reviewing clips and pulling any good moments directly on the timeline. I used to do this too but it's much faster to scrub through a clip and just mark the good in and out (with hotkeys of course) instead of trimming it on the timeline. It also gives you a permanent record of the good moments (saved in the marker data) instead of depending on that trimmed clip remaining on the timeline as the record of that moment. Also doesn't clutter up your timeline until you're ready to actually edit. Scrub through clips at high speed. Probably goes without saying, but there's no need to watch every clip from start to finish at regular speed. Most clips only have 1 or 2 good moments and you can usually spot them while scrubbing. Scrubbing is also better than skipping ahead as you'll see quick moments go by that could work. When you spot a good moment, scrub back a bit and then watch just that moment at full speed. Then mark the in and out, as above. Make sure you can scrub clips smoothly. If your camera source files don't scrub smoothly (like my Sony A9's X-AVCS h264 files, ungh) then make some proxies to be able to scrub quickly (I use 1080p Cineform). This will also speed up the editing process as it will remove all those little micro-delays when you re-position the play head or skip back a few seconds or do some scrubbing through the timeline. I waited way too long before doing proxies but they are totally worth it and very low-hassle nowadays (e.g. Premiere's ingest tool, only took me a few minutes to find the encoding settings and start encoding them). Remove useless or very low-value clips from the project. Again probably goes without saying but sometimes we review a clip there's nothing good there, but we're afraid of losing anything so we leave it there cluttering up the project. If you know there's nothing there then just remove it from the project as it will speed up the rest of the process and with over 1100 clips you'll almost certainly have more than you need.

Looks nice, has a cool moody 80's vibe. Very well-produced song too.

Now if we could just get an PC CPU or GPU that has some onboard optimizations for decoding X-AVCS h264, so I don't have to create proxies for fast scrubbing..

I agree, there could be a series of standardized tests developed for a gimbal, and it would be nice to have a consistent set of metrics to judge gimbal performance on. Probably a fairy significant effort to develop a legitimate, fair standard that would account the variety of conditions gimbals are used in and the variety of equipment that's mounted on them. Would be nice to see but I kind of doubt anyone will put enough time into doing it right. Seems like each gimbal manufacturer probably has some version of these tests already for testing their own hardware (like the camera mfrs do for dynamic range). If you're really gung ho about it, might be worth reaching out to them. Some mfrs probably already run their tests on competing hardware and if they know theirs is superior they might want to share their testing methodology so it can be replicated and generate good PR for them. Interestingly, I think one way to do this kind of testing would be with another gimbal that's attached to the gimbal being tested. The SimpleBGC software that many popular gimbals run is open source and you can also create scripts that create predetermined movements, like this guy below did for doing timelapses. Could probably script a series of different movements that would be used to apply various forces against the gimbal being tested, then measure a sensor or laser or something mounted to the test gimbal to see how much it controls for those movements. https://beyondthetime.net/3-axis-gimbal-timelapse/

This looks very much like gimbal micro-vibrations interacting with the internal IBIS. I've experienced similar things on my PilotFly gimbals with various Sony cameras. In my experience when it interacts with IBIS you get that kind of warping/jello effect as the IBIS is trying to adjust for the gimbal's extremely quick micro-vibration. My guess is that it's the combination of the super fast sensor movement combined with the rolling shutter that creates the warping/distortion of the image. I'm not familiar with the Crane software but if there's any kind of diagnostic mode where you can see a graph of the gimbal's movement you might be able to confirm that these moments are ones where the gimbal is doing a lot of vibration correction. You could also try disabling IBIS (and cranking up your shutter speed to eliminate any motion blur) to see if you can then see the gimbal movement more clearly.

The monitor / TV refresh rate does seem like it would be a huge factor here. You mention a 48Hz monitor; seems like a 120Hz monitor would also work (divides into 24 equally). Why does the video card have to be a Decklink though? Many video games provide options to lock the framerates at particular values and the consumer GPU's respect those settings. It seems like all that's needed on the video card side is the playback software outputting the video signal at the correct playback framerate to the card. Or if running in a window, then the desktop running at the correct framerate (which in Windows seems to be tied to the display's refresh). My display (LG 4K 31MU97) runs at 60Hz over DisplayPort but gets capped to 24Hz in 4K when connected over HDMI. I'm going to do some comparisons and see if I can notice smoother playback on 24p content when Windows and my video card (GTX 970) are capped at 24Hz.

Hopes destroyed for 4k60, 10bit, or higher bitrate on A7S3...

Damn.. there does seem to be a ton of stutter / choppyness in the motion in that video. Hopefully someone will do a side-by-side with the A73 vs. the A7R3 or even the R2 at various fps / shutter settings and nail down what is going on here.