Topics

Hi guys!

I would like to run a script, which would be enabled when the picture is taken with the intervalometer and the AutoETTR module is also in use. The idea is to force a delay (2-4 seconds) between the shots independently from interval time. I would like to use this gap for pan/tilt motion.

Could you please help me out for updating the previous script, that was this one. It's not working anymore on the latest nightly for 550D.

Code Select Expand

force_delay_menu = menu.new
{
parent = "Intervalometer",
name = "Force Delay",
help = "Add an extra delay to each interval",
min = 0,
max = 300,
unit = UNIT.TIME,
}

events =
{
intervalometer = function() msleep(force_delay_menu.value * 1000) return true end
}

Is it possible to send a "T" character through the camera's USB port with a proper USB cable to trigger the MX2 controller?

Please take a look at the TimeScapes forum, I'm not at home at this kind of engineering things.

I ask this because I would like to trigger the MX2 as the shutter closes, but my camera's hotshoe + Pixel TF-321 + PC sync cable isn't working well, the triggering happens as the shutter opens.

I did again an extreme curve comparision (At first I wrote about it on doom9 forum in june) and I would like to ask people to check it, if it's correct or not, because I'm not a video expert. (On that forum nobody checked it.)

The main idea behind this is maintaining the best quality from RAW (MLV) till the end (eg.: MOV, MP4).

To check it, please follow this:
01. Download the Advanced FrameServer and install it.
02. Make a preset in Media Encoder (Format: Advanced FrameServer, Color space: RGB24)
03. Download the AviSynth+ and install it.
04. Make an eg.: "E:\rendering\signpost.avs" file:

Code Select Expand

#24 fps source
#AVISource("E:\rendering\signpost.avi", audio=false).AssumeFPS(24,1)
#25 fps source
AVISource("E:\rendering\signpost.avi", audio=false).AssumeFPS(25,1)
#30 fps source
#AVISource("E:\rendering\signpost.avi", audio=false).AssumeFPS(30,1)

# ConvertToRGB24()

# 4:4:4
#ConvertToYV24()

# 4:2:2
#ConvertToYV16()

# 4:2:0
#ConvertToYV12()

05. Download the VirtualDub and the Gradation Curves Plugin.
06. Download one of the MLVFS Windows clients and setup.
07. Mount a MLV with the MLVFS.
08. Open the first DNG from the sequence with ACR, make color correction, open it in PS and save as 16 bit PNG.
09. Open the PNG in VirtualDub and add the Gradation Curves filter on it, make an extreme curve, something like this. (Export the curve, it will be used later too.) Click on show preview and make a sreenshot, save it on a PS layer.
10. Import the DNG sequence into AE (Project settings: 16 bpc, sRGB), save the project.
11. Open Media Encoder, use the previously mentioned AFS preset (02) on the AE project file, start the frame serving to eg.: E:\rendering. You will get an small AVI (eg. signpost.avi).
12. Open the AVI in VirtualDub and add the Gradation Curves filter on it, import the previously saved curve (09), again show preview, screenshot, save it on PS layer.
13. Don't stop the frame serving in Media Encoder.
14. Open the AVS file (comment # every line except one of the AVISource line with the FPS setting and save it) with VirtualDub and add the Gradation Curves filter on it, import the previously saved curve (09), again show preview, screenshot, save it on PS layer.
15. Open the AVS file and delete the comment sign before (4:4:4 setting) save it, open the file with VirtualDub and add the Gradation Curves filter on it, import the previously saved curve (09), again show preview, screenshot, save it on PS layer.
16. If you made everything correctly, then there will be only a negligible difference between the reference PNG frame and the YV24 (4:4:4) setting. Check and uncheck the layers in PS. The fully commented AVS or RGB24 settings will be the same as the PNG and AVI.

If you want to convert the sequence to MOV, MP4 or HEVC, then download the latest ffmpeg build and use it. Here are some example BATs, what I made:

ProRes 4:4:4 10bit:

Code Select Expand

ffmpeg.exe -i "signpost.avs" -c:v prores_ks -qscale:v 1 -pix_fmt yuv444p10 -filter:v unsharp=5:5:0.5:5:5:0.5 "prores_1080p_444_10bit.mov"
pause

x264 4:2:0 8bit:

Code Select Expand

ffmpeg.exe -r 25 -i "signpost.avs" -s 1920x1080 -c:v libx264 -b:v 25000k -minrate 25000k -maxrate 25000k -bufsize 25000k -x264-params keyint=25:min-keyint=1 -g 25 -pix_fmt yuv420p -filter:v unsharp=5:5:0.5:5:5:0.5 -r 25 "x264_1080p_420_8bit.mp4"
pause

x265 4:2:0 8bit:

Code Select Expand

ffmpeg.exe -i "signpost.avs" -s 1920x1080 -c:v libx265 -x265-params crf=1:vbv-bufsize=12500:vbv-maxrate=12500 -pix_fmt yuv420p -filter:v unsharp=5:5:0.5:5:5:0.5 "x265_1080p_420_8bit.hevc"
pause

If you have time please check this workflow till point 16.

Here is my extreme curve comparision, download the JPGs to check them and put them into a PS file, zoom in 200% or more.

I have a motorized dolly to get motion when I'm timelapsing and I just thought about something. Would it be possible to trigger the flash shoe after that ML saves the DNG into MLV?

It would be usefull to trigger the MX2 motion controller to move the dolly, please read the comments here and you will understand what I'm speaking about.

Somebody knows a way to convert CR2s (14bit) + their XMP datas (adjusments) directly to 4K ProRes (10bit, 444)... of course without colorshifts?

If you don't care about the images being exposure-matched in overlapped area (I assume you don't, otherwise you would have used M mode and --same-levels), just enable Post Deflicker, and meter each image as you wish (Av or ETTR).

Of course, in this case, the recommended way to use ETTR is on the SET key (and meter before each shot). The Always On mode should be used when the exposure is unlikely to change between shots (because it meters from the previous shot).

I tried this as a1ex suggested and it worked very well.

https://www.dropbox.com/s/mx8luiqxjb8tsh2/panorama_03_small.jpg (deleted)

Next time I will test the same, but combined with Dual ISO, let's see what happens.

I shoot my first sunrise timelapse in the last month with my Canon 550D (runing a nightly build from october) + Canon 18-55 mm.
The used functions: Auto ETTR + Intervalometer + Adobe XMP sidecar files.

I shoot my first sunset timelapse in the last month with my Canon 550D (runing a nightly build from october) + Canon 18-55 mm.
The used functions: Auto ETTR + Intervalometer + Adobe XMP sidecar files.

Lens: MF

Canon settings:
* RAW
* Image review: 2 seconds
* Max ISO: 1600
* Long exposure noise reduction: OFF
* High ISO speed noise reduction: Standard
* M mode + Liveview

ML settings:
* White balance: Sunny
* ISO: 100
* Shutter: 1/1000
* Aperture: F5.6
* Auto ETTR
- Trigger mode: Always ON
- Slowest shutter: 6 seconds
- Exposure target: -1 EV
- Highlight ignore: 0.1%
- Allow clipping: OFF
- Midtone SNR limit: 6 EV
- Shadow SNR limit: 2 EV
- Link to Canon shutter: OFF
- Link to Dual ISO: OFF
- Show metered areas: OFF
- Show debug info: OFF
* Exposure simulation: ON

* Global draw: ON
* Cropmarks: 16:9
* Histogram
- Color space: Luma
- Scaling: Log
- Clip warning: ON
- Use RAW histogram: Full histogram
- RAW EV indicator: ETTR hint

* Intervalometer
- Take a pic every: 10 seconds
- Start trigger: Leave menu
- Start after: 3 seconds
- Stop after: 400
- Manual focusramp: OFF

* Post deflicker
- Sidecar file type: Adobe XMP
- Deflicker percentile: 50%
- Deflicker target level: -4 EV

Could ML supported DSLRs control stepper motors through USB connected nanoMoCo board (nanoMoCo, MoCoBus)? I mean motion control on multiple axis. Do they have enough power in principle? It just came to my mind, because earlier I saw a timelapse video about focus ramping...

EDIT: Some additional information about nanoMoCo + connector board here and here.

Another future thought: The Canon 650D has a touchscreen, so it could be used as an input opportunity to use a keyframing option with curves... just like Dragonframe, but a simple version.