Uli Plank wrote:Sigh, that's how things go. Same settings, same environmental conditions. Only another camera looking at the screen. It doesn't stop with HEVC. Not even after twice the time.
A bit like our cat, which doesn't do anything it shouldn't do if we are looking at it
I have suspicion, maybe even an "inspiration from above". What if... (more later)
Do you remember, if you went directly from starting the App into time lapse mode, without doing different recordings (or even another time lapse prior to "the real thing")?
Had you done work in the App, when you started the privious time lapse? Maybe moved the App from "the background" to the front, working on other things between using the App?
Is it possible to get MediaInfo data on the "failed" and on the "successful" time lapse? Do they differ in bitrate? How about the metadata, also saved into the file?
There seems to be an element of chance involvedI have looked into my roughly four hour time lapse, and surprise, surprise, something is NOT, what it should be. Usually I only use two HEVC settings - 20MByte/sec or 40MB/sec. But my recording was only at 28,1 Mbit/sec (~3,5 MByte7sec) according to MediaInfo in 2160p HEVC.
This should not happen. The file clearly states Cinema.P3 in the metadata. So… ehhh… hmmm… what if…
I remember to have used Blackmagic Camera App, and asked for 1 out of 10 frames, HEVC and active alert on dropped frames (maybe no influence, but….) I also looked at Apple Camera. But… sequence, settings etc. are - to be frank - a bit murky now.
The Cinema P3 recording has 23min 52s or 35825 frames. IF 1 in 10 was used, the duration was 358250 ordinary frames at 25fps, or 14330 seconds equaling 3 hours 58 minutes and 50 seconds. File size is only 5.05GByte (even if I had used 20MByte/sec, I should have produced around 160-200 megabit/sec for the resulting file of around - let’s say, give and take - a 32 GByte file (5 GByte * 180 megabitpersecond/28 megabitpersecond = 32).
How come, I did NOT get that?
A few very quick and dirty experimentsWhen I start the 10x Time Lapse AFTER selecting 20MByte/sec (I had setting at ProRES), I get typically between 24,4 and 24,7 Mbyte/sec as a result.
When I start the Cinema P3 10x Time Lapse directly (just to make sure, I rebooted the iPhone) I get 24.4 MB/sec. The last App accessed, was Cinema P3 in 10x Time Lapse.
When I reboot the iPhone, start the Blackmagic Camera in x10 time lapse mode, put that into the background, and then start the Cinema P3 10x Time Lapse, I get 5,4 Mbyte/sec.
When I close both the Blackmagic Camera and Cinema P3 App’s (swipe up to remove), start the Apple App, select “Time Lapse?” (In my language called “Tidsforløb” by Apple), put the App into the background, and the start Cinema P3 10x Time Lapse I get 4,5 MB/sec.
God only knows, what I had originally selected, used, switched to and from, BEFORE starting - or heaving Cinema P3 into the foreground - but it certainly affected the outcome.
AND I CAN AFFECT recording quality by simply starting Cinema P3, AFTER having had another App active, before “running” Cinema P3 in Time Lapse mode.
TheoryCinema P3 uses whatever settings may be active in the iPhone, AFTER being started. Whether another App has been active in the meantime or not, makes no difference. Cinema P3 seems to use “current state” (when brought into the foreground), whether that state was set by Apple Camera (in this example leading to 4,5MByte sec) or Blackmagic Camera (in this example leading to 5,4 Mbyte/sec) instead of the 24,4 Mbyte/sec (in this example), being the only Camera App started (never sent to the background).
In short, Cinema P3 seems to use whatever camera state exists, when brought into the foreground.
AdviceIF you ALWAYS select the Cinema P3 camera preset you want to use as BASE MODE for the Time Lapse (ProRES 422/422HQ, HEVC 40/20MByte, whatever), you get, what you ask for. Otherwise, you get was is currently actively set (by this, that or the other App, maybe even Cinema P3
Try selecting the required preset immediately before starting Cinema P3 Time Lapse.
My guess is, that a previous overheating not reproducible later may be caused by…. You’ve guessed it.
This may of course also explain why time lapse recordings ordinarily working as a treat, but sometimes failing miserably. The element of chance actually could be reproduced in the above examples.
Regards