Messages

Is there a difference in what type of power outage would cause a different reaction in ml?


Sent from my iPhone using Tapatalk

Here is my ML folder.
https://www.dropbox.com/s/x2slyxcjii90vfv/ML.zip?dl=0

What I'm seeing is either around the 15-16 hour mark or 4000+ shots the camera shuts off.

I didn't have to reenable modules so I assume it was regular crash or potential power outage or these power supplies are failing around the 15hr mark.

I ordered new power supplies I'll be able to test on Monday to rule it out.


Sent from my iPhone using Tapatalk

Hmm I don't have any of the cameras handy as they are at the place shooting right now and started new timelapse this morning.

I have a new theory though.

Looking at two different cameras the time lapses are stopping once they reach 9999.

However, I see now that this may be only one reason the camera stopped. As one camera turned off well before hitting this number.

When I go there tomorrow I'll swap the sd card. I'm sure the camera will turn off again.


Sent from my iPhone using Tapatalk

LiveView will heat up the sensor (more noise) and will double the number of shutter actuations if you are not using Canon's silent shooting mode (since the camera must return to LiveView after taking each picture).

It will also affect ETTR metering if you choose Always On (metering on the LiveView image instead of the previous still image). This may help if exposure changes a lot (as it will get a chance to try various exposure parameters until it settles), or may hurt in low light (it's hard to meter a long exposure from a 1/30s LiveView frame).

Great tips. I am not using live view on this project. I switched the cameras to silent mode as well.

Additional question that might be magic lantern related it might not be. I have three cameras running continuously for the next 22 days. Everyday I go and check the cameras and notice they are turned off. Is there anything in ML and the intravalometer that would turn the camera off? Seems to be happening after around 15hrs of shooting.

I have the camera set to not shut off on the canon menu - that wouldn't be this anyways, but just saying.

I was also wondering if it's the power source which is a Chinese knockoff dummy battery power AC adapter.

These 3 cameras are running 1.23.


Sent from my iPhone using Tapatalk

Definitely much better however if you don't mind me asking how exactly did you do the desqueezing process in Post? Did you do everything with a 1.67x? If so, I notice some (not all) were a bit stretched too much. Overall it looked fabulous other than these few shots.

Could that be due to shooting in different aspect ratios during the entire stay in Italy? I know it can be rather difficult to keep track of what we shot in and whatnot. Thanks for sharing once again @limey!

Yeh I think some of the clips were not crop mode, but just raw with a close enough aspect ratio. I scaled the height on all clips 166, but some clips at 166 didn't stretch it enough. I assume these are those clips that were recorded either in str8 raw, 5x, or 10x. So those clips stretched to 166 or more could be overstretched. Hope I'm making sense.


Sent from my iPhone using Tapatalk

I'll upload the frames (if I remember) Monday when back at the office.

It wasn't card rotation, because I run ML off the SD card and shoot to the CF. I leave the SD card there the whole time while rotating CFs.

Thanks for the help. Some other context to the video. Lens was my beaten up tamron 24-70, and I  graded all the clips individually in resolve.


Sent from my iPhone using Tapatalk

This was done over the course of say 10 hours. It was done using battery rotation and card rotation. When I had live view one for say one full battery session, then the next battery I didn't have live view on. After checking the footage, there is a change in frame size dimensions. If you are saying that is not something that ML should be doing, then maybe it could be attributed to lens drift or something?

New version is desqueezed.
https://vimeo.com/189956119

No dual iso for sure don't want to deal with the extra post processing.

I'm wondering if I should shoot live view or not. The timer idea is interesting if it works it could be huge.

It turned out that shooting sometimes with liveview on and liveview off was the culprit as the frame size is different. Not sure what settings caused that.

Additional info. These cameras will be largely unmanned for the month, however, I plan to check on them every 3 days. I'm wondering if a timer on the outlet would be a good idea since no activity will occur at night so no point firing shots.

If I used this timer, would it not allow the timelapse to restart when the power came back on? Unfortunately I don't have time to test it.
https://www.amazon.com/Enover-Programmable-Digital-3-prong-Appliances/dp/B0191ZGELS/

I have a lot of experience shooting timelapses with ML now for a few years. However, I'm about to embark on a month long timelapse. I think the cameras (3) will run 24 hours a day with plugin adapters.

What I want to know is will AutoETTR be the best method? I have done some timelapses where the AutoETTR looks like its failing and not adjusting the exposure properly. Particularly from daytime to Sunset. Is it something to just keep the faith in and let it be or will it need to be massaged?

Ok I'll scale up in premiere - which I started this morning before usual distractions took over.

If I scale up the clips in premiere to say 164....is that not the same as using the after effects script to desqueeze?


Sent from my iPhone using Tapatalk

Just a little bit more context to this video - since it was shot in Spring 2016 and I've done so much "work" related editing since then I didn't really remember details.

I graded in Resolve all the clips - I said above I might have used Lumetri - It was Resolve for sure. Recently for other stuff I've been using Lumetri which was my confusion.

Will give it a shot. Side note. I just recently shot a video in Mexico with the A7S II and I gotta say that I wish I shot it in ML Raw. It just looks better imo than the 8bit raw Sony puts out.

Hey limey,

     Great video! If transcoding clips in the future, try using Smart Import 2: https://www.magiclantern.fm/forum/index.php?topic=13354.0

     This will do the desqueezing automatically. If you go back to your project, try clicking scale to frame size on a clip. If not, change the height of the clips by around 1.6x. Meaning if the clips are 1600 x 540, then change 540 to 900.

     Hope this helps!

- Paul Harwood

Awesome. Thanks for the info.

Is the 3x crop supposed to be desqueezed? Was that the intention?

Otherwise you lose some material at the ends.


Sent from my iPhone using Tapatalk

Went to Italy this past spring and wanted to use it as a test for the 3x crop. Most of this video is using the 3x crop. All hand held aside from resting camera on objects at times.

https://vimeo.com/189956119

nk87,
Are you shooting two cameras at one time for the zoom shots? One camera wide, the other much tighter on the subject with enough room to zoom in post?

Looks like raw video sped up for zoomed tracking shots like the boats. No?


Sent from my iPhone using Tapatalk

Success. I searched for Kitchetof's name and found this link:
https://bitbucket.org/kichetof/lr_cr2hdr/downloads

Downloaded the beta 3 (pretty sure I was installing this from the earlier posts) and this time it worked!

I do have the plugin, but it says that its missing these items, however, I've installed them individually and still no work.

Maybe I have a different plugin - not Kichetof's?

Really having problems getting this to work on my mac pro. I have it installed on my laptop - Win. Anyone have a step by step?

If I type in which dcraw exiftool this is the output:
/usr/local/bin/dcraw
/usr/local/bin/exiftool

My lightroom has the plugin installed, but I get the error:
"This plugin (cr2hdr) need dcraw and exiftool.

Please install it before run this plugin"