Blackmagic Camera App - General discussion and Releases

Angle is mostly reserved for actual film based recordings.
In all other cases, it only complicates things (but it keeps beginners at bay, which some old f*rts may need as a buffer to real life My personal view. YMMV.
Regards

As an "old f*rt" I agree. That's who it's aimed at. Even so, I prefer shutter speed. It's more descriptive of what's going on.

Speaking of which, my Sony late 90s PD150 would do video shutter speeds as low as 1/4 sec. That's three stops more exposure available than at 30fps. Very useful, both technically and creatively. I wish my iPhone would do that.

Angle is mostly reserved for actual film based recordings.
In all other cases, it only complicates things (but it keeps beginners at bay, which some old f*rts may need as a buffer to real life My personal view. YMMV.
Regards

As an "old f*rt" I agree. That's who it's aimed at. Even so, I prefer shutter speed. It's more descriptive of what's going on.

Speaking of which, my Sony late 90s PD150 would do video shutter speeds as low as 1/4 sec. That's three stops more exposure available than at 30fps. Very useful, both technically and creatively. I wish my iPhone would do that.

I also count “meself” to be part of the still somewhat exclusive guild of “old farts” with an age of three score and a dosen coming January 2024 (no need to use these newfangled decimal descriptors cropping up everywhere these days

Regards and remember to have a bit of fun every day

P.S. Try what’s possible with RAW photo burst mode; also there has been an app, that actually allowed to adjust “DNG foto bursts” to be used as low to standard DNG’s per second, depending on phone hardware. Should be possible to use in Resolve as DNG sequence.

If it still exists is anyone’s guess. Can’t remember, if it was Android, iPhone or both.

BM's app can capture time-lapse starting from 2 frames and is still offering 360 degrees (for the old f*rts).
I wonder what that'll do? It's too sunny here right now to test, even with my variable ND.

And now to EDR: I could reproduce a clear limit by filming with 50 fps. At 1/80th of a second the hard wall at 80% is there, and at 1/96th the highlights extend. Hurray! So we can shoot with enough motion blur (for 50) and have more room in the highlights. It would be nice is others can reproduce this (ans maybe, test 60 fps for our American friends).

The old f*rts are free to calculate the numbers in degrees

P.S. Longer exposures are not possible in time-lapse.

Thanks for the detailed posting. I avoid Cinematic mode and use Standard mode. Apple’s ambitious and aggressive ‘cinematic’ theories may be good theories, but the execution may not be up to the task… yet. Some future processors with increased performance may achieve the desired results.

If you are referring to stabilisation, I see massive overshooting with Cinematic mode.

Thanks for the detailed posting. I avoid Cinematic mode and use Standard mode. Apple’s ambitious and aggressive ‘cinematic’ theories may be good theories, but the execution may not be up to the task… yet. Some future processors with increased performance may achieve the desired results.

Doesn't hurt, but this behaviour does illustrate a specific "start strain" exhibited by the Apple video "machine". I just haven't found a way to document my real suspicion, yet.

As I see it - without solid proof at this moment - it is NOT critical, what you decide to activate (modes, settings etc.) and In what combination. To some degree, the first frames - hugely varying number - will always show some minor or major variation in frame rate. Maybe also the last (few) frame(s) may be effected (not checked at all).

The real suspicion (and possible conclusion) is, that you probably will have a very stable, maybe even constant frame rate, if you cut out the first few seconds (one or two) and maybe the last two frames from a sequence (the jury is still out there), if start and stop was not affected by that action in other ways (touch introduced movement of iPhone).

When I cut out a few seconds from start, and restrict duration - whatever fits the moment - and then export the selected range I get rock solid constant bit rates. That state may be a result of the actual export process, or the actual state of the restricted size content. Hard to establish, without a tool enabling a process to inspect frame-by-frame behaviour - ideally frame timestamps - from start to end in a given video sequence.

So, still: "beware of gremlins", but my initial tests also indicate (not the same as proof) that if you exclude a few seconds from the beginning of a recording (and maybe a few frames from the end), you will have very stable material, maybe even a constant frame rate, in the "trimmed" part of the sequence.

This is easily handled. Always start recordings a bit early. In post exclude the first frames or seconds (if handheld required anyway), and Bob's your uncle. Still no proof, but sound indications, and if adjusting to these "oddities" are as simple as described, the problems are limited. If KNOWN!

Until real proof (or the opposite), this remains pure conjecture.

I'll return with a solid conclusion, when proof is available or my theory has been disproved.

Regards and remember to have a bit of fun every day

If you are referring to stabilisation, I see massive overshooting with Cinematic mode.

NO.

I'm just interested in the effect displayed, when recording is started. The Apple "video engine" is showing clear signs of strain, in both speed consistency and also revealing, that the initial default mode may need a lot of processing to get stable, when really strained (shown by the drastic alterations in exposure - try it).

Actual stabilisation is a completely different kettle of fish.

Even if you place the iPhone on a tripod, and activate the "Cinema Mode" stabilization, the same effect is displayed with dramatic deviations in frame rate (especially in manual mode, if large deviations from some default initial behaviour is used). A straining "processing engine", simple as that (otherwise it wouldn't take up to a second to adjust exposure from "deafult" (but completely wrong) to actual required state).

Cinematic mode stabilization is not ideal, overshoot can and will happen, if you're not carefull, but handheld outside after a few minutes in freezing cold, it does wonders for the stability of the footage. Even in tropic regions (you just have to put your hands in the freezer compartment a few minutes, and experience what that does to handheld stability

Every user can decide stabilization mode and requirements for themselves.

Regards and remember to have a bit of fun every day

For many decades of actual (mechanical, real film and real tape) production, has it not been the practice of cinematographers or directors to begin a shot with “Roll sound”… “Roll camera”… “Clap”… “Action!” and the crew operators and actors respond to each appropriately “Sound rolling”…” Camera rolling”… (pause), clap the clapboard… then begin the action! It only takes a few seconds but that’s how every shot of this week’s music video had everyone in sync without problems. Sure I had everything controlled by Tentacle Sync, but verbal cues really help the musicians and vocalists to focus, pardon the pun. Just a pleasure in post to see every clap perfectly aligned.

Seems like a plan to ensure successfully taking a shot even on a smartphone. Turns out it’s even more relevant than ever given the analysis you and others have undertaken on our behalf. Gratitude.

P.S. Longer exposures are not possible in time-lapse.

An unfortunate deficiency, IMHO. Perhaps addressable in the future.

In my early days of shooting with a video camera, I was frequently berated by editors for not allowing enough pre-roll on my shots. It was a holdover from my film days where you only rolled at the last possible second to save film.

Now, referring to kfriis' observations, it appears it matters more than ever to allow lots of "handles", especially when shooting fast developing, uncontrolled situations.

Would you believe on this week’s music video shoot, the producer criticized me for calling Cut after each song take was entirely over? The producer wanted me to keep rolling to capture the band’s reaction to their performance. Okay, did that for several seconds longer; I’m not sure people really want every word, cuss, gesture, and expression recorded for eternity. After all we’re not The Beatles.

We really decide after each take if it’s a Good Take or not. Even after a Good Take, we shot again a couple of times and sometimes got another Good Take. The Best Take of a song was after five previous attempts, some of which were Good Takes. Lots more fun when you’re not performing a set in front of a live audience.

Would you believe on this week’s music video shoot, the producer criticized me for calling Cut after each song take was entirely over?

Sounds like the producer has been watching Frederick Wiseman films Are you the director as well as the cameraman and sound recordist these days?

P.S. Longer exposures are not possible in time-lapse.

An unfortunate deficiency, IMHO. Perhaps addressable in the future.

It would be great if these could be done as integration of multiple images, just like Apple is doing it for low-light situations.

Now, referring to kfriis' observations, it appears it matters more than ever to allow lots of "handles", especially when shooting fast developing, uncontrolled situations.

Reminds me of the early days of U-Matic editing, where you always had a 'waterfall' going down the picture (from tape stretch) making the first three seconds unusable.

Frame duration results

I'v managed to collect frame duration results for the three test examples, I described earlier.

All times in seconds.

File: A001_12021136_C008
60 Hz Flicker compensation
Cinematic Mode
Total video size 73 frames

Initial durations in seconds.

The third frame (in this example) has an extreme duration of ~0.05 seconds (practically one 20 fps frame). The iPhone is clearly stressed here.

0.040017 **
0.040000
0.049983 *** BIG PROBLEM
0.040000
0.039983
0.040000
0.039983
0.039983
0.040000
0.039983
0.040000
0.039967 *
0.039983

The largest parts of the frames have durations of 0.040000 seconds and 0.039983 seconds, which is VERY near the ideal constant 0.040000 (probably caused by other activities in iPhone). Singular 0.040017 and 0.039967 seconds (sometimes paired) turns up. Last packets within “normal”:

0.039983
0.039983
0.040000
0.039983
0.039983
0.040000
0.040000
0.039967 *
0.039967 *

File: A001_12021139_C009
50 Hz Flicker compensation
Cinematic Mode
Total video size 78 frames

Initial durations in seconds

The fourth frame (in this example) has an extreme duration of ~0.05 seconds (practically one 20 fps frame). The iPhone is clearly stressed here.

0.040000
0.040017 **
0.040017 **
0.049983 *** BIG PROBLEM
0.039983
0.040000
0.039983

The largest parts of the frames have durations of 0.040000 seconds and 0.039983 seconds, which is VERY near the ideal constant 0.040000 (probably caused by other activities in iPhone). Singular 0.040017 and 0.039967 seconds (sometimes paired) turns up. Last packets within “normal”:

0.039983
0.040000
0.039983
0.039983
0.040000
0.039983
0.039983
0.039983

File: A001_12021147_C010
50 Hz Flicker compensation
Standard mode
Total video size 43 frames

Initial durations in seconds

First packets within normal. No stress.

0.039983
0.039983
0.040000
0.039983
0.040000
0.039983
0.040000
0.039967 *
0.039983

The largest parts of the frames have durations of 0.040000 seconds and 0.039983 seconds, which is VERY near the ideal constant 0.040000 (probably caused by other activities in iPhone). Singular 0.040017 and 0.039967 seconds (sometimes paired) turns up. Last packets within “normal”:

0.040000
0.039967 *
0.039983
0.040017 **
0.039967 *
0.039983
0.040000
0.039983
0.039983

I have placed the dataset on the files into one zip’ed EXCEL file:

https://www.icloud.com/iclouddrive/096I ... Qg#Results

The file may be deleted without prior warning any time AFTER new year 2023/2024.

Conclusion

These short test sequences indicate, that the use of Cinematic stabilization mode probably is the cause of initial stress for the iPhone. In both streams, the ~0.05 sec (~20 fps) sample indicates the start of adjustments away from “default”, and the first frame adjusting to real request follows soon after, but typically requires in the region of 25 frames to stabilize visually.

The very problematic frame is part of the clearly visible “not yet correct exposure” part of the initial frames.

There seem to be no “oddities” in the last frames.

Otherwise the stream (1/25s, 25fps requested) switches between the values 0.039983s and 0.040000s with a few samples of 0.040017s or 0.039967s here and there, sometimes in pairs. Deviation is very small for an environment, where all sorts of other, maybe important, processes are constantly active.

It’s not constant in the strictest sense, but I would not regards these figures as problematic for most productions. Where talking +17 and -33 microseconds up to minus 0.0825 percent periodic deviation from the ideal 0.04 second duration in these, short samples, where the phone had very little time to enter a stable state.

These data were obtained with iOS 17.1.2 (as far as I remember). There is no guarantee, that later firmware updates will show exactly the same. Behaviour, but that caveat applies to “real cameras” too.

Regards and remember to have a bit of fun every day

… Are you the director as well as the cameraman and sound recordist these days?

For Monday’s music video, yes. But never a dictator.

Anybody around here got the full version of Cinema P3 and compared it?

Yes. Ask away.

Regards and remember to have a bit of fun every day

Anyone notice that if you shoot in 23.976p on the Cinema P3 (I'm using iPhone 12 Pro Max), that the metadata in Resolve claims that it's 24p. Doing the same in the Black Magic app correctly reads as 23.976p. Do you folks see this with your more recent iPhones?

@ Steve
It's correctly flagged as 23.976 here from an iPhone 15 Pro.

@Kurt
Did you test "Extreme" too?

@ Steve
It's correctly flagged as 23.976 here from an iPhone 15 Pro.

@Kurt
Did you test "Extreme" too?

Thanks for checking, Uli! I suspect there is a bug there but I don't know for sure. It seems that there is no issue with the BMD Camera App so the iPhone 12 Pro Max.

I'm so sorry, Steve!
I didn't check in DR (which was busy at the time) but in MediaInfo, where it shows as 23.976.
DR is not reading the speed correctly, it shows 24 fps.
So, there's miscommunication between CinemaP3 and DR.

Thanks for checking further, Uli. Cheers!

Found another difference:

This is from CinemaP3:

CinemaP3.png (71.81 KiB) Viewed 606687 times

And this from BM Camera:

BM_Camera.png (72.75 KiB) Viewed 606687 times

Probably DR is looking at the max, not the indicated average, which seems kind of stupid.
Both apps also state "Frame Rate" 23.976 under "General".

@ Steve
It's correctly flagged as 23.976 here from an iPhone 15 Pro.

@Kurt
Did you test "Extreme" too?

I did today. In another environment (overcast, instead of sun, yesterday, so less contrast possible). Tendency is the same.

Standard Stabilisation
Billedratemodus : Variabel
Billedfrekvens : 25,000 FPS
Minimum billedfrekvens : 24,979 FPS
Maksimum billedfrekvens : 25,010 FPS

Extreme stabilisation
Billedratemodus : Variabel
Billedfrekvens : 24,955 FPS
Minimum billedfrekvens : 21,413 FPS
Maksimum billedfrekvens : 25,021 FPS

In this case, initial adjustment from "default" automatic mode to manual mode 25fps, 1/25s. 1000 ISO was less demanding, but the behavior was similar (starting off from automatic, near perfect over a series of frames gradually adjusting to correct).

It depends on the "initial chock" in difference to the "ideal" automatic settings (my theory), when cinematic or extreme stabilisisation is involved. You can "deduct" a problematic frame from the 21.413 fps minimum rate (roughly 0,046701 seconds instead of 0,040033 seconds for lowest frame speed). Calculated values, since my "tool" rigged on a Windows machine is not available right now (and even slower to crank up to speed, than the iPhone 15 Pro subjected to the "commandeering" from Blackmagic Camera App

"Current light" will have some influence, but it is a characteristic, that both Cinematic and Extreme need an initial number of frames, to switch from some default automatic mode, into purely manual settings. This is reproducible. That some automatic mode is a default fallback in a discovered stress situation, is natural for a consumer oriented product.

The good thing is, that recovery is swift (around a second - plus/minus), and after that initial "adjustment phase", the frame rate is remarkably stable. In any mode.

Expressed in microseconds, the shutter is nominal 0,04 seconds (1/25s) plus 17 minus 33 MICROseconds for the remainder of the file, where the majority of frames are at 0.040000 and 0.039983 seconds in the samples I have tested. In effect a worst case deviation from max to min frame rate of 50 microseconds.

My personal view is, that it would be nice to know, what accuracy the "constant" rate 25fps camera setting has in real life in "real cameras", that are NOT hard synced to a reliable, professional quality source. The iPhone values are probably both true and realistic (based on external time servers, where available), and I can certainly live with a "not-quite-constant" frame rate, that deviates at worst 1 in 1212,121... frames (as my measurements imply) - and as seen in the samples, the average fps for the "near constant" region after start is around 0.0399992 - a bit higher than 25 fps. Again based on the figures, I have extracted from other samples (excluding the first second or two, which is often necessary anyway).

My testing was indoor, in a living room (certainly held at a more or less constant 22C, nether subject to freezing cold outside or searing sun outside - at least today).

Regards

Maybe other cameras just insist that they are constant, but I've never seen an exposure variation over the first few frames, not even from big brand hybrids, like Canon, Nikon, or Sony.
I have encountered variable frame rates only from smartphones and game recordings.

My tests above had the stabiliser set to standard in both apps.

Maybe other cameras just insist that they are constant, but I've never seen an exposure variation over the first few frames, not even from big brand hybrids, like Canon, Nikon, or Sony.
I have encountered variable frame rates only from smartphones and game recordings.

My tests above had the stabiliser set to standard in both apps.

Does it matter, if you have to cut start and end from handheld material in most cases anyway?

You don’t need Cinematic or Extreme even Standard stabilisation when using tripod or gimbal, do you?

Regards

Maybe other cameras just insist that they are constant, but I've never seen an exposure variation over the first few frames, not even from big brand hybrids, like Canon, Nikon, or Sony.
I have encountered variable frame rates only from smartphones and game recordings.

My tests above had the stabiliser set to standard in both apps.

By the way: Are you absolutely sure, that most cameras record the first "available" frames, or do they postpone actual recording, until frame delivery (nothing to do with stabilization) and settings have stabilized?

If this is the case, the camera just hides any possible initial hiccups (just to offer a not totally excluded explanation If a dedicated camera "delays" recording a few frames worth of time, would you be able to notice, if you didn't look?

Regards and remember to have a bit of fun each day

Does it matter, if you have to cut start and end from handheld material in most cases anyway?

You don’t need Cinematic or Extreme even Standard stabilisation when using tripod or gimbal, do you?

1. No, not to me.

2. Standard is good enough even handheld if I don't drink too much coffee. Cinematic and Extreme loose resolution and I see some overshooting on sudden motion. So, yes, mechanical measures are still valuable.

Does it matter, if you have to cut start and end from handheld material in most cases anyway?

You don’t need Cinematic or Extreme even Standard stabilisation when using tripod or gimbal, do you?

1. No, not to me.

2. Standard is good enough even handheld if I don't drink too much coffee. Cinematic and Extreme loose resolution and I see some overshooting on sudden motion. So, yes, mechanical measures are still valuable.

Cinematic and Extreme are like "maximum extra" stabilization on "real cameras" - ALSO introducing side effects in form of heavy cropping in most cases. As often is the case, options are personal choices.

I wanted to DOCUMENT, what could happen, in Cinematic and Extreme, if stressed. The whole test scenario was designed to do that: Find "worst case" scenario, and what would happen in the following frames.

I documented, that around the first second COULD be of doubtful value - in MANUAL mode - but after the first second especially two, there were no visible side effects. The footage was near constant alll the way to the end. Most people do NOT care, what happens worst case the first one to two seconds, of a five minute shoot, IF they are prepared.

This is important, in any case, where you need maximum (handheld) stabilization, and need to capture something in the shadow - requiring manual mode - in bright summer daylight (unless you can turn off "heavenly light", which most can't). This is not unusual during travel, reportage etc.

As it turns out, if you plan for worst case - shoot as soon as SENSIBLY possible - chances are, that you get everything important in the best possible quality. Nice to know, instead of bickering on "a low minimum FPS", that as far as I could establish, only exists in exceptional cases (where minimum FPS is still true, but NOT representative for the footage, you get after a second or two - worst case!!!!).

You seem to have gotten the impression, that worst case equals best case, so here are best case (handheld and fully automatic mode):

No stabilisation (image jittery as hell):
Billedratemodus : Variabel
Billedfrekvens : 25,000 FPS
Minimum billedfrekvens : 24,979 FPS
Maksimum billedfrekvens : 25,010 FPS

Cinema mode (image rock solid):
Billedratemodus : Variabel
Billedfrekvens : 25,000 FPS
Minimum billedfrekvens : 24,979 FPS
Maksimum billedfrekvens : 25,010 FPS

You can complain, that you never use fully auto mode, but that is a personal preference, not really representative for the tens of millions new iPhone 15 Pro (Max) users the coming months.

My aim was to document, what happens in a specific case, that stresses the system, and how to avoid unnecessarily side effects. Worst case for situations, that demand maximum "help" in form of heavy stabilization, as far as I could test it. Nothing more.

If your camera is always mounted rock solid, it has no significance for you. For me, the opposite is true. The only reason I invested time in finding out, what actually seemed to happen.

Cinema mode has a tendency to overshoot, if you're not carefull (test it in real use, and you know the limitations before use - universal advice in photo/video), but "Standard" or "None" stabilization makes that argument superfluous, because my handheld video is then mostly unusable - especially if it's freezing outside, as is currently the case (or requires heavy stabilization in post, also cropping the frame dramatically - if footage can be at all saved!!!).

It's a choice between pest or cholera, if I want the footage at all. No other "common walkabout" alternative, or....?

If you never use stabilization, always use tripod and... and... always have all the time in the world to compose the video, you want (not what you can get away with), wonderfull for you. Your camera, whatever it is, will always have the best available conditions to work under.

That's not my standard use case.

I'll stop the thread here.

Regards and remember to have a bit of fun every day

Don't get me wrong, I really appreciate your scientific diligence.

Just a comment of the current conditions for photography/video in the high north of this globe (the "flat earthers" have the same problem up here :

Today: Mild overcast, -1C around mid day.

If the sun is fully visible (typically means down to -5 to -10C at night), the big lamp "up there" is max ~12 degrees over the horizon (55.4 degrees latitude, early winter).

That problem never exists, when I'm in Singapore.

If you can't move the object of interest, you have no alternative, but to accept an often very bright sky (preferably avoiding direct glare from the "hottest light" anywhere in the vicinity).

That was also why, I discovered the "odd conditions " (super low minimum frame rate), when doing videos.

Next question was: Was that avoidable? Typical deviations all through the footage? Hmm... only way to find out, is to rig up some tools, that were able to - at least - indicate, what turned out to be a simple approach.

Regards and remember to have a bit of fun each day

Another, so far unmentioned, problem with iPhone photography in winter is the fact that it's 100% touch screen. ie, bare hands/cold fingers. This not a common problem in Oz.

I'm aware that there are "cellphone gloves", but none of them have worked for me. I have a screen protector installed.

Another, so far unmentioned, problem with iPhone photography in winter is the fact that it's 100% touch screen. ie, bare hands/cold fingers. This not a common problem in Oz.

I'm aware that there are "cellphone gloves", but none of them have worked for me. I have a screen protector installed.

I can testify, that the touch screen on my Lumix S5 “real camera” isn’t that practical, whatever kind of gloves I use. Many buttons etc are hardly usable either, so my sausage-fingers need thawing from time to time

Regards and remember to have a bit of fun every day (brrr… or not)

I bought a rubber tipped cellphone stylus that seems to work. $5.

Looks like the Blackmagic Cam app will have a new competitor for professional video recording on a smartphone from the makers of Halide:

https://apple.news/AUVZGpyxqRvO5lZP0Wl2RyA

Kino coming in February 2024

Looks like the Blackmagic Cam app will have a new competitor for professional video recording on a smartphone from the makers of Halide:

https://apple.news/AUVZGpyxqRvO5lZP0Wl2RyA

Kino coming in February 2024

I suggested here about a week ago that Blackmagic have a look at what Halide is doing inside and outside the app. Like Blackmagic, these people really understand user interface and user engagement. They also go back a long way. One of the three main developers was with Twitter when, as he has put it, you could put the whole team at a single table in the lunchroom.

The video in your link is from Ben Sandofsky's personal YouTube channel. Very interesting that they intend to make this a bit akin to a reality TV show. I don't mean that as a criticism. It may be brilliant marketing. Sandofsky made a video about a trip to the Galapagos Islands earlier this year. It's worth watching just for his views on photography and phone cameras.

I can confirm my observations about the EDR kicking in.
It's not only clearly visible between 1/80 and 1/96 of a second when shooting in 50 fps, but also between 1/30 (the minimum) and 1/40 for 25 fps. The time that function needs should be longer than 7.5 ms and shorter than 10 ms then.
I'm not sure if it works for 60 fps if you keep the old f*rts rule of 180 degrees.

I can confirm my observations about the EDR kicking in.
It's not only clearly visible between 1/80 and 1/96 of a second when shooting in 50 fps, but also between 1/30 (the minimum) and 1/40 for 25 fps. The time that function needs should be longer than 7.5 ms and shorter than 10 ms then.
I'm not sure if it works for 60 fps if you keep the old f*rts rule of 180 degrees.

Thank you for the heads-up.

In Apple Log only? All codecs? In other settings too?

Regards and remember to have a bitlocker fun every day

It works like that in Apple Log, both for ProRes and HEVC. It's really great if you want to keep the old f*rts motion blur and just balance ISO with NDs, since you can redistribute values to a large degree, preferring highlights or shadows depending on the scene. Just for our overseas friends, I checked 60 fps too and it clearly kicks in between 1/100 and 1/120th of a second.

Remember, 10 bit is still not very much if your final result is intended for HDR and it helps a lot to shift values around on that log curve, plus using more bits in the highlights. I must admit that I like Apple Log a lot, it's much more realistic for 10 bit than Sony S-Log3, which is far too flat. It seems like Sony took it from the Venice and put it into their hybrids, so users can claim: "Look mom, my camera can do what the big guys do." Which is not true. Apple has tuned their log to make the most of the iPhone, it seems.

If you switch to HLG, something is happening between these shutter values too, but it's so uncontrollable that I didn't care to investigate further.

Oh, and CinemaP3 with its large waveform is excellent for seeing what you are doing there.

Just a small heads-up for the 50Hz AC majority in this world.

Instead of concentrating on 422HQ in all cases, I decided to try something different.

It turns out, that ProRES 422 2020 HLG actually handles well at 50 fps for internal recordings (at just below 1 Gigabit/sec) for my most frequent lens use case at 24mm. Deviation from ideal framerate is +/- 0.17% around nominal 50 fps (in this case, fully automatic, cinematic mode stabilization), locked white balance during recording (etc).

Code: Select all

Generelt
Fulde navn                               : /Volumes/980Work2TB/Project/422HLG50/A001_12081604_C001.mov
Format                                   : MPEG-4
Formatprofil                             : QuickTime
Codec-ID                                 : qt   0000.00 (qt  )
Filstørrelse                             : 2,83 GiB
Varighed                                 : 25sek. 642 ms
Samlet bitratemodus                      : Variabel
Samlet bitrate                           : 949 Mb/s
Billedfrekvens                           : 50,000 FPS
Kodet den                                : 2023-12-08 15:04:03 UTC
Tagged den                               : 2023-12-08 15:04:29 UTC
Skrivende bibliotek                      : Apple QuickTime
com.apple.proapps.manufacturer           : Apple Inc.
com.apple.proapps.clipID                 : A001_12081604_C001
com.apple.proapps.reel                   : 1
com.apple.proapps.scene                  : 1
com.apple.proapps.shot                   : 1
com.apple.proapps.isGood                 : 0
com.blackmagic-design.camera.dayNight    : Day
com.blackmagic-design.camera.environment : Interior
com.blackmagic-design.camera.dateRecorde : 2023-12-08T16:04:03+0100
com.blackmagic-design.shutterMode        : Speed
com.blackmagic-design.shutterSpeed       : 1/50
com.blackmagic-design.camera.iso         : 1664
com.blackmagic-design.camera.whiteBalanc : 3400
com.blackmagic-design.camera.whiteBalanc : -2
com.blackmagic-design.camera.aperture    : f1.8
com.blackmagic-design.camera.lensType    : iPhone 15 Pro 24mm
com.blackmagic-design.sensorFPS          : 50
com.apple.proapps.cameraName             : A
com.blackmagic-design.camera.guides.safe : 85
com.apple.quicktime.location.ISO6709     : +xxx+xxxx/xxxx
com.apple.quicktime.software             : Blackmagic Cam 1.2.00076
com.apple.quicktime.model                : Apple iPhone 15 Pro 24mm
com.apple.quicktime.creationdate         : 2023-12-08T16:04:03+0100

Video
ID                                       : 1
Format                                   : ProRes
Formatversion                            : Version 0
Formatprofil                             : 422
Codec-ID                                 : apcn
Varighed                                 : 25sek. 642 ms
Bitratemodus                             : Variabel
Bitrate                                  : 946 Mb/s
Bredde                                   : 3.840 billedpunkter
Højde                                    : 2.160 billedpunkter
Størrelsesforhold                        : 16:9
Billedratemodus                          : Variabel
Billedfrekvens                           : 50,000 FPS
Minimum billedfrekvens                   : 49,917 FPS
Maksimum billedfrekvens                  : 50,083 FPS
Farverum                                 : YUV
Chroma subsampling                       : 4:2:2
Skantype                                 : Progressiv
Bits/(Billedpunkter*Billed)              : 2.282
Strømstørrelse                           : 2,82 GiB (100%)
Titel                                    : Core Media Video
Skrivende bibliotek                      : Apple
Kodet den                                : 2023-12-08 15:04:03 UTC
Tagged den                               : 2023-12-08 15:04:29 UTC
Color primaries                          : BT.2020
Overførselskarakteristika                : HLG
Matrix-coefficienter                     : BT.2020 non-constant

Lyd
ID                                       : 2
Format                                   : PCM
Formatindstillinger                      : Signed
Codec-ID                                 : lpcm
Varighed                                 : 25sek. 640 ms
Source duration                          : 25sek. 643 ms
Bitratemodus                             : Konstant
Bitrate                                  : 3.072 kb/s
Kanaler                                  : 2 kanaler
Channel layout                           : L R
Samplingsrate                            : 48,0 kHz
Bitdybde                                 : 32 bits
Strømstørrelse                           : 9,39 MiB (0%)
Source stream size                       : 9,39 MiB (0%)
Titel                                    : Core Media Audio
Kodet den                                : 2023-12-08 15:04:03 UTC
Tagged den                               : 2023-12-08 15:04:29 UTC

Andet
ID                                       : 3
Type                                     : Time code
Format                                   : QuickTime TC
Varighed                                 : 20 ms
Billedfrekvens                           : 50,000 FPS
Time code of first frame                 : 16:04:03:00
Time code of last frame                  : 16:04:03:00
Time code, stripped                      : Ja
Titel                                    : Core Media Time Code
Kodet den                                : 2023-12-08 15:04:03 UTC
Tagged den                               : 2023-12-08 15:04:29 UTC


It's nice to have the option.

Now, what if... the BMC App allowed sorting the presets manually (in order most frequently needed).

Regards and remember to have a bit of fun every day

If Apple earbuds use Bluetooth, they might be selectable as audio inputs for BM Camera on the iPhone. Using my Bluetooth hearing aids in that capacity effectively gives me wireless microphone capability.

Can anyone with the camera app and Apple (or other Bluetooth) earbuds test this function?

Yes, you can connect them, but you can't control the gain. And then, they are not great, being made for phone calls, not for sound from a distance. Just 22 kHz sample rate and mono.

If you just need a guide track, the internal mics are good enough. If you want serious sound recording, get one of the other options already discussed around here.

The Time Lapse function of the Blackmagic Camera app is working well here. It delivers 4K 10bit 4:2:2 CBR MOV files direct to iPhone storage. No assembly required. Gotta love it.

I've intentionally chosen to record h.265 files. This 21 second file requires just 76MB of drive space.

From MediaInfo:

Video
ID : 1
Format : HEVC
Format/Info : High Efficiency Video Coding
Format profile : Format Range@L5@Main
Codec ID : hvc1
Codec ID/Info : High Efficiency Video Coding
Duration : 21 s 167 ms
Bit rate : 29.5 Mb/s
Width : 3 840 pixels
Height : 2 160 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 30.000 FPS
Color space : YUV
Chroma subsampling : 4:2:2
Bit depth : 10 bits
Bits/(Pixel*Frame) : 0.118
Stream size : 74.3 MiB (100%)
Title : Core Media Video
Encoded date : 2023-12-06 18:12:52 UTC
Tagged date : 2023-12-06 18:23:27 UTC
Color range : Limited
Color primaries : BT.2020
Matrix coefficients : BT.2020 non-constant
Codec configuration box : hvcC

Yes, you can connect them, but you can't control the gain. And then, they are not great, being made for phone calls, not for sound from a distance. Just 22 kHz sample rate and mono.

If you just need a guide track, the internal mics are good enough. If you want serious sound recording, get one of the other options already discussed around here.

Too bad about the earbuds. For voice recording, so far my Phonak hearing aids seem adequate. The microphones are apparently very high quality.

Wind resistance is poor, however. I don't think I'll resort to wearing a dead cat on my head to mitigate this.

Yes, you can connect them, but you can't control the gain. And then, they are not great, being made for phone calls, not for sound from a distance. Just 22 kHz sample rate and mono.

If you just need a guide track, the internal mics are good enough. If you want serious sound recording, get one of the other options already discussed around here.

Too bad about the earbuds. For voice recording, so far my Phonak hearing aids seem adequate. The microphones are apparently very high quality.

Wind resistance is poor, however. I don't think I'll resort to wearing a dead cat on my head to mitigate this.

Is it necessary, that you use a Bluetooth?

I’m thinking about “audio sync” in post, alternatively direct feed to iPhone. What are you aiming at recording? Scenario, distance, whatever?

I have two options in mind: The Shure MV88 (M-S) with built in headphone output, and direct digital feed to iPhone (Works with BMC App).

Alternative approach is using a Zoom F2(-BT). I use two BT versions (BT for optional control only). They come with lavalier mic (and dead Cat, as far as I remember - I use a better lavalier with same “sound” as my XLR pencil condensers). Direct headphone out possible too. Biggest advantage is bias out too, and -5.5 dBu max input, 48kHz 32-bit float (no need for and not possible to adjust recording input).

Automatically turns on - for instance - Røde VideoMic NTG and Sennheiser MKE400 too (easy on microphone batteries), and works well with Sennheiser MKE600 Shotgun. What I’m trying to say, is that microphone selection is very wide (dynamic or electret). Tiny. Belt clip standard. Battery powered. Micro SD for recording. Completely self sustained.

Easy sync in post. Have used two for years. Far better than my Røde wireless Go II (they clip easily on far lower inputs, than the Zoom).

Just to mention a couple of alternatives, that may help (the products mentioned may work as inspiration for your own perfect and different solution).

Regards and remember to have a bit of fun every day.

Is it necessary, that you use a Bluetooth?

No, just amazingly convenient.

Note that I'm now just a hobby user, no longer earning a living in this business. As long as the voice is undistorted and intelligible, that's all I care about.

I have no batteries and other stuff to carry with me and manage, no wiring up the talent (because I'm the talent) Nothing to do but choose the hearing aids as an audio input and I'm ready to go.

No input gain control is a moderate problem. Audio will clip if I yell loud enough. Faint sounds like distant birds are recorded quite well and with very low noise. As I said, the Phonak mics are pretty good.

I just wish the BM Camera app would record stereo. Then, I'd have binaural audio.

I have no batteries and other stuff to carry with me and manage, no wiring up the talent (because I'm the talent) Nothing to do but choose the hearing aids as an audio input and I'm ready to go.

...

I just wish the BM Camera app would record stereo. Then, I'd have binaural audio.

You would indeed have binaural and I'd really like to hear it.

Are you planning to put some videos on social media, e.g. YouTube? If so, I'll be subscribing.

I have no batteries and other stuff to carry with me and manage, no wiring up the talent (because I'm the talent) Nothing to do but choose the hearing aids as an audio input and I'm ready to go.

...

I just wish the BM Camera app would record stereo. Then, I'd have binaural audio.

You would indeed have binaural and I'd really like to hear it.

Are you planning to put some videos on social media, e.g. YouTube? If so, I'll be subscribing.

Done right, it’s really outstanding - if you listen on headphones. Otherwise, it is not that interesting.

The problem in this case is lack of distance to your own body.

First rule of good sound is minimum distance to source. Second rule of good sound is maximum distance to unwanted sounds (easier to isolate from unwanted sound in a studio, than in open air, but…).

Many, many years ago I bought the Soundman binaural microphone system (Klassik with A1). It has very high quality microphones, very wide frequency range and high sensitivity too in addition to low noise. Should be ideal…

Except your own body can be rather noisy, at very short distances. Normally you do not “hear” your body; your brain has an effective filter system, but - alas - that filter does not work on the binaural microphone input.

In order to get good binaural sound, you may have to train yourself to hold your breath and keep absolutely still for a minute or more, if you can. Two minuts tops, and you have to breathe; recording affected. Long, uninterrupted recordings are… ehh… breathtakingly hard to do with body mounted microphones.

These days, binaural recordings are sometimes done with an “artificial head” (extremely expensive Neumann solution comes to mind) or an Ambisonic setup, where you can decide microphone characteristics etc. in post, even convert to binaural on the fly (for monitoring on headphones or mixing typically).

Two typical relatively low cost solutions are Røde NT-SF1 (Soundfield) and the very cheap Zoom H3-VR with internal recorder (very sensitive to touch, handling and other noise). And then there are… cost skyrocketing.

My ten cents…

Regards and remember to have a bit of fun every day

All I did was write three sentences saying that I'm interested in what Peter is doing and that I'd like to know if he's creating a social media channel.

Next thing I know I get an eight paragraph lecture on binaural sound...

These days, binaural recordings are sometimes done with an “artificial head” (extremely expensive Neumann solution comes to mind) or an Ambisonic setup, where you can decide microphone characteristics etc. in post, even convert to binaural on the fly (for monitoring on headphones or mixing typically).

My ten cents…

Regards and remember to have a bit of fun every day

You mean the Neumann mike that's in my avatar? KU 100? This one?

ku100.jpeg (346.48 KiB) Viewed 601176 times

The one that I gave a covid mask to, and cannibalised one of Philippe Chenevez's Cinela Leos for to fashion a windscreen due to the mike's high sensitivity to air currents?

ku100 masked.jpeg (255.52 KiB) Viewed 601176 times

Is that ambisonic as in this Sennheiser AMBEO VR? Is that why I have Sound Devices's ambisonic decoder and use Dear Reality's VR spacialisation software?

sennheiser.jpeg (343.06 KiB) Viewed 601176 times

My ten cents…

No comment.

Regards and remember to have a bit of fun every day

After the 2 or 3 hundredth time, chalk scraping against a blackboard comes to mind.

Kids ....

Take it down a notch ... or three

Forgive the interjection ... for heavens sake ...its a phone app … nothing more
and limited by Apple's restrictions .... no one is going to change the world, win
an Oscar or find Nirvana with it ....

It is good but not astoundingly so … nice addition to one's choices ... I would posit
that you are better served with a BM 4K … hell even a digital Bolex will be profoundly more engaging
than this ....

I am pleased with Apple log ... nice addition but really a marginal improvement.

Just to be clear ... the forum has been overwhelmed with passionate emotive personal
gushing posts about the BM camera App … and its competitors ...

The bandwidth allocated to this peripheral capture device seems a bit excessive.

Sorry for the rant ... back to normal programming ... whatever that means.

Let the flames begin ....

… win an Oscar …

I'm not so sure. In 2011 Park Chan-wook and Park Chan-kyong got a Golden Bear for „Night Fishing“, shot on an iPhone 4. Seems to have something to do with art more than technical details.