OT: iphones as cameras?

<snip>

I know nothing of iPhones, but with an appropriate app, Android phones can be used as 'IP cameras' which allow live video view over WiFi. Don't know how good the sync is.

I've used Windows phones for this, but in different rooms and no need for sync. These phones are dead now.

Nor I! :>

In theory, a phone *could* inject an SMTPE timestamp into the video stream sourced from the GPS timebase. But, I've no idea if there are any *real* guarantees as to latency/jitter between the actual received time signal and the signal available at the application layer. Presumably, two phones would be able to lock to relatively tight precision (note that you also have to sync audio so the tolerances there are tighter).

I.e., instead of genlocking the cameras, let them each *report* their apparent timestamps and handle synchronization in post.

I saw the tail end of a TV ad that showed an iPhone in some sort of holder/frame to make it more presentable as a "pro" camera (I am assuming it wasn't just a prop). No doubt, pitching the quality of their video. But, conveniently ignoring all of the other issues involved in a shoot! (how do you route the video to a "monitor" so someone other than the cameraman can actually SEE what you are shooting?!)

And, using quality microphones (instead of the dinky mic in the camera).

And the inability to genlock.

etc.

Apparently, there's at least one product that purports to address the timecode issue. So, I sent out a request to friends/colleagues for "old iphones" in order to see how that works. And, what *other* devices I'll have to acquire for a COMPLETE solution.

[There's a reason to NOT upgrade kit -- you don't have to go through all these gyrations to find an *equivalent* solution to the one you already *have*! :< ]

....

This app can synchronize multiple iPhones for taking multi-camera videos.

It's also got WiFi... but why iPhones? There's myriads of old point-and-shoot autofocus/autoexposure cameras with analog video outputs, even some with HDMI, and you can just pipe as many as you want into a security-type DVR (up to four inputs is cheap, eight or more is available). And, they have a tripod-friendly mount.

Editing video clips together would be an offline operation, of course. One of my DVRs has four video but only two audio channels.

Yes, but I suspect BT has more focus on latency issues (used for audio) than the WiFi interface (which is typ used for bulk data)

My interest was sparked from adverts touting the quality of iPhone cameras. No doubt there are other digital cameras with comparable quality but iPhones are ubiquitous.

They are also very small/lightweight. E.g., it is tedious to set up my (ancient) cameras as they are large and relatively heavy (~10 lbs ea). iPhones could be mounted on mic booms or selfie sticks attached to tripods. I.e., *I* can possibly weasel out of having to attend these events! :> (not likely when you have a car full of kit that has to be uncrated, set up, cabled, etc.)

*Not* having to tie things together is an asset. Let each phone/camera act as an independent source and harvest the data after the shoot. This isn't really practical for an iPhone by itself as the mic isn't particularly directional, etc.

But, you need a way of *virtually* tying them together -- in time. If I was looking to design such a solution, I'd rely on the timestamps available from their GPS receivers -- which I assume is more precise than the (phone) "network time" that they present to the user. Embed this into the video stream "in some way" (ideally SMPTE time code as it is "portable") where it is affixed to specific frames (and audio).

[You might also want to look at the specifics of your DVR; many reduce the effective frame rate "per camera" as the number of cameras increases (to keep disk bandwidth constant). Esp those that do scene analysis]

Yes. You'd collect the video/audio/time streams from each camera and load them into the NLE, massaging the "composite" there. Being able to treat the streams as independent/self-consistent entities makes this easier. E.g., you can duck the audio from the camera watching the palette -- UNTIL the brush comes into scene (at which point, hearing the brush slop around the palette adds interest/realism to the presentation)

Another win of a more modern camera (over mine) is that you can take advantage of the increased resolution to do digital effects in post (zoom, pan, etc.) without losing much quality (compared to what my cameras could do). E.g., take the palette PIP window and explode it to full-screen when the attention shifts to that...

Oh, they can do that, too, on battery power. Date/time is stamped on stills, I'm not sure about video files...

That timestamp is certainly part of the security-DVR solutions, and if you used a POE type camera/DVR solution, the wiring is not very troublesome...

Yeah, thus the clap-stock used at scene inception in classic videography...

Good point. It still seems to me that a central DVR with periperal cameras is going to be less of a herding-cats exercise. I'd expect that any iPhone solution will break with some future (or be incompatible with some past) IOS version.

You need to have a timestamp per frame as the sorts of time stamps that a DVR would likely apply are just intended for human consumption (e.g., time-of-day... HH:MM:SS)

But their still wired together. The beauty of an iPhone (or any stand-alone digital camera) is that all you'd need to do is put it in a support, point it (using its screen/viewfinder as a guide) and press "record".

If you wanted to add another camera (e.g., to "watch" the audience's reactions), just point one facing outward.

Yes. Note the electronic version now has a timecode displayed in LEDs.

If (see previous product postings) each camera embeds an SMPTE time code in the video (audio), then *that* acts to tie their content together. So, no need to ensure each camera is turned on at the same time, etc.

My current solution relies heavily on "ancient" hardware -- simply to leverage my prior knowledge.

My original goal was to "scrap three cameras" (4th quarter is my annual equipment "re-visioning") in favor of something smaller, more convenient to lug around (my cameras aren't particularly heavy but they are incredibly bulky) while preserving that shoot capability.

Now, I'm becoming opportunistic: if I can come up with a solution that lets me bow out of the actual shoots (it's typ a 4-5 hour commitment on my part for an hour of presentation) so SWMBO can set things up by herself, that's a big win! Still leaves me with the post but at least that's an *interesting* activity...