650D/T4i Raw Video

[spider]

It's normal.
Not changeable without deeper changes in Canon FW I think.

[Walter Schulz]

Liveview -> press INFO until MLs info screen appears. Done.

Ciao
Walter

[spider]

Liveview -> press INFO until MLs info screen appears. Done.

Isn't working for me. Sometimes it shows still up.

[jimmyD30]

Finally got the fast cards! I can report that I am getting 1280x720@24fps continuous RAW using SanDisk Extreme Pro 95 MB/s card. This occurs after about 15-45 seconds of recording with a write speed of 36.9 MB/s and idle time reaching over 200ms after a few minutes and then slightly over 300ms thereafter. Not sure how much format allocation unit size matters, but setting between 64KB-128KB seemed to work best.

Recorded for about 11.5 mins with file size of 25.7 GB on a 32 GB card when it shut down automatically (on its own).

One thing to note, in the past when I set the fps override to 24fps it would display actual fps of 23.973, now when I set it to 24fps it states an actual frame rate of 24.002, so I don't know what's up with that.

UPDATE Dec-14-2013: Using the latest nightly build (Dec-13-2013) the actuall fps is back to 23.973 (whether setting Desired fps to 24.000 or 23.976), so I guess that's good.

Also, on a side note related to the new MLV raw video format, on the 650D .mlv is a little slower than .raw recordings, to the point that it won't record continuously for 1280x720p @ 24fps, I can see the MB/s slow down jnto the low 36.x MB/s when in MLV recording mode, all else being equal.

[mlrocks]

I dont know what you mean.

I mean can ml do raw recording 18 mp at 1 fps without using the mechanical shutter? Thanks for all of your help and sharing. Great community.

[jimmyD30]

I mean can ml do raw recording 18 mp at 1 fps without using the mechanical shutter? Thanks for all of your help and sharing. Great community.

I think what you are asking is can ML do a complete 18 megapixel sensor (for 650D) write to the SD card at 1 fps in video/movie mode??? While probably technically possible, no ML version that I know of is presently setup to do that, all the versions I'm aware of basically use various video formats (e.g. 720p) and write it to the SD card at various frame rates, mostly maxing out at around 1280x720 @ 24 fps on the 650D.

Is this what you are asking? Or do you mean using the intervalometer in still picture mode with RAW image, if this is the case, then yes, I believe it's possible, but I'm not well versed in this area of ML, so you may want to re-post your question with more clarity and maybe someone else can answer it better than me.

[spider]

I mean can ml do raw recording 18 mp at 1 fps without using the mechanical shutter? Thanks for all of your help and sharing. Great community.

Not possible because ML only dumps the LV stream on the card.
And the resolution of the stream is as it is.

[jimmyD30]

Well, here's my first video shooting RAW on the 650D: https://vimeo.com/81964818

Workflow as follows: Shot RAW 1280x720 @ 23.976 fps continuous video with Canon 650D/T4i then used Pink Dot Remover v0.08 on RAW file. Used RAWMagic 1.0 (Mac) to convert RAW file to DNG files. Opened DNGs in Adobe After Effects which started Adobe Camera Raw where I did some minor color correction (mostly auto-this and auto-that). After that I rendered in Adobe After Effects as lossless (animation). Opened lossless file in Adobe Premiere Pro and added audio then exported as H.264 10 Mb/s for Vimeo upload.

Note: Although not part of the workflow, I also played the lossless animation file in QuickTime Player which converted it to ProRes 4444 for viewing on my Mac and I was prompted to also save the file as such.

[davidtlong]

Cool, what did you use for the sound.  I am playing around with a surround sound recorder.

dave

[jimmyD30]

Cool, what did you use for the sound.  I am playing around with a surround sound recorder.

dave

Being a last minute kinda thing I didn't have all my gear with me, so I shot with a kit lens and captured sound on my iPhone with a 4-track recording app called Audiostar on best quality setting using just the phone's mic. You can still hear the wind blow even after I reduced it quite a bit in post. The alarm ringing was for the crossing gates just behind me, so this was actual audio of the event. I was lucky enough to have the train engineer blow the horn as he was passing by me giving that Doppler effect usually associated with trains.

We had just gotten 13.5 inches of snowfall the day before in upstate NY, so I thought it would be cool to catch the train blasting through the new snow and I wanted to try shooting RAW, so it was a good enough excuse for me to get out there and do it :-)

[davidtlong]

Being a last minute kinda thing I didn't have all my gear with me, so I shot with a kit lens and captured sound on my iPhone with a 4-track recorder app called Audiostar on best quality setting using just the phone's mic. You can still hear the wind blow even after I reduced it quite a bit in post. The alarm ringing was for the crossing gates just behind me, so this was actual audio of the event. I was lucky enough to have the train engineer blow the horn as he was passing by me giving that Doppler effect usually associated with trains.

We had just gotten 13.5 inches of snowfall the day before in upstate NY, so I thought it would be cool to catch the train blasting through the new snow and I wanted to try shooting RAW, so it was a good enough excuse for me to get out there and do it :-)

Nice, What technique did you use to sync the audio?.   

d

[jimmyD30]

Nice, What technique did you use to sync the audio?.   

d

Syncing was pretty easy considering it wasn't a dialogue type scene or some such thing. I just pressed the record buttons on both the camera and phone at the same time and then the stop buttons at the same time also, this left me with a 47 sec video recording and about a 47.25 sec audio recording. So, overlaying the audio on top of the video was straight forward as I was working with basically the same length for both with both being recorded at very nearly the same time. I ended up stretching the audio a bit in post, so that it played when the title was showing at the beginning of the video.

Keep in mind, I had initially forgot that recording in RAW wasn't going to capture audio, so like I said previously, I was kind of scrambling at the last minute to pull this off and just used what I had immediatlely available, hence using the iphone for audio recording.

You mention you were looking to record in 'surround sound', will you be using a special device or a stereo mic or maybe two uni-directional mono mics facing in different directions?

[davidtlong]

Syncing was pretty easy considering it wasn't a dialogue type scene or some such thing. I just pressed the record buttons on both the camera and phone at the same time and then the stop buttons at the same time also, this left me with a 47 sec video recording and about a 47.25 sec audio recording. So, overlaying the audio on top of the video was straight forward as I was working with basically the same length for both with both being recorded at very nearly the same time. I ended up stretching the audio a bit in post, so that it played when the title was showing at the beginning of the video.

Keep in mind, I had initially forgot that recording in RAW wasn't going to capture audio, so like I said previously, I was kind of scrambling at the last minute to pull this off and just used what I had immediatlely available, hence the iphone.

You mention you were looking to record in 'surround sound', will you be using a special device or a stereo mic or maybe two uni-directional mono mics facing in different directions?

When I found out how easy it is to make surround sound audio files, yes two stereo microphones would work (two Iphones:)).  But I bought a ZoomH2n.  It has 5 microphones and for surround it records to two wav files (two channels in the front and two in the back).  Much easier than having two separate microphones as I really want to try field recordings/video.  One thing I want to do is the train video like you did with surround sound. 

I use Audacity to take the Zoom files into either 5.1 or 7.1 surround sound (SS).  (I did not know Audacity could do this and how easy it was). There is a nice video on the web about making 5.1 but once you do that 7.1 is a snap. I export either as SS wav or ogg or m4a audio files.  Ogg and m4a files are much smaller.  After "developing" the raw video (I use Photoshop to edit as well as render the video.  I used to use Quicktime for rendering.  I have compared output form both and I like PS so for me it so easy to do all "developing"  in one piece of software)  as  to I combine the audio and video files in PowerDirector (v. 12).  I sync via clapper board (hence my original question).  I really need to find a small finger size clapper board or make my own.  PowerDirector as far as I can tell through comparing video editors  is the only editor to be able to combine ss ogg and m4a audio.  The problem however is that PowerDirector does not assign the speakers properly for ss ogg or m4a.  For example the left rear speaker maybe assigned the center speaker channel.  I have worked out how PD maps the audio to the speakers and have sent this study to  Cyberlink (they make PD).  They are now working on this and hopefully have a fix soon.  WAV files work fine but as I said they are large files that can make the video large.

Hope this is of interest.

dave

[jimmyD30]

When I found out how easy it is to make surround sound audio files, yes two stereo microphones would work (two Iphones:)).  But I bought a ZoomH2n.  It has 5 microphones and for surround it records to two wav files (two channels in the front and two in the back).  Much easier than having two separate microphones as I really want to try field recordings/video.  One thing I want to do is the train video like you did with surround sound. 

I use Audacity to take the Zoom files into either 5.1 or 7.1 surround sound (SS).  (I did not know Audacity could do this and how easy it was). There is a nice video on the web about making 5.1 but once you do that 7.1 is a snap. I export either as SS wav or ogg or m4a audio files.  Ogg and m4a files are much smaller.  After "developing" the raw video (I use Photoshop to edit as well as render the video.  I used to use Quicktime for rendering.  I have compared output form both and I like PS so for me it so easy to do all "developing"  in one piece of software)  as  to I combine the audio and video files in PowerDirector (v. 12).  I sync via clapper board (hence my original question).  I really need to find a small finger size clapper board or make my own.  PowerDirector as far as I can tell through comparing video editors  is the only editor to be able to combine ss ogg and m4a audio.  The problem however is that PowerDirector does not assign the speakers properly for ss ogg or m4a.  For example the left rear speaker maybe assigned the center speaker channel.  I have worked out how PD maps the audio to the speakers and have sent this study to  Cyberlink (they make PD).  They are now working on this and hopefully have a fix soon.  WAV files work fine but as I said they are large files that can make the video large.

Hope this is of interest.

dave

I have to laugh, I just bought the Zoom H2n last night! What I especially like about it, in addition to all the awesome features at a great price, is that it uses full size SD cards, the same as my camera, not micro-SD like a lot of other recorders are using these days.

I would suggest not to worry so much about the WAV file size, especially in relation to the RAW video recording file size. And considering you want to capture audio as surround sound, the uncompressed WAV format will be far superior in quality than any compressed format.

Have you looked into a smart phone or tablet Clapboard app? I don't know how good they are, but some are free.

[Letni]

Hey people,

What would you choose rather in RAW MOVI ? 

(Canon 1280x720 50p)
1536x616 (23,980p) = 946176 PIXEL stretch to 1536x864 in post and scale it up to 1920x1080 24p

or

(Canon 1920x1080 25p)
1280x720 (25p) = 921600 PIXEL scale it up to 1920x1080 25p

Looking forward to your answers. 

[davidtlong]

I have to laugh, I just bought the Zoom H2n last night! What I especially like about it, in addition to all the awesome features at a great price, is that it uses full size SD cards, the same as my camera, not micro-SD like a lot of other recorders are using these days.

I would suggest not to worry so much about the WAV file size, especially in relation to the RAW video recording file size. And considering you want to capture audio as surround sound, the uncompressed WAV format will be far superior in quality than any compressed format.

Have you looked into a smart phone or tablet Clapboard app? I don't know how good they are, but some are free.

Interesting.  Hand not thought about that, the app, that is.  Actually in my test quality of the sound for the none wav files is good.

[mageye]

I know this is all slightly off topic but seeing as there are a number of people that are using the Zoom H2 I thought I would mention that I also have one, and as clunky as the thing is - it's a great little device.

I made a modification to it that allows you to hijack the points where the audio is fed to the recorder. It allows you to have four separate audio inputs (the H2 only has 2). The inputs are switchable between MIC/Line level.

My one has cables hanging from it with a a couple of connectors that switch out the onboard microphones when you insert alternative sources.

The mod was 'fiddly' but it does work. You just have to be willing to 'hack' your device a little. Believe me, I was nervous because I don't have millions of $$$$ to mess around with. Neither am I an expert in electronics. I was careful and methodical and thankfully I did it. It works, and as you can tell: - I love what it has allowed me to do.

There was a video on You Tube that inspired me to do it. With instructions. I think there is also more stuff around the internet too. Just google 'Zoom H2 mod'

Anyway if you are interested here is a little video for it: -

https://www.youtube.com/watch?v=N1vJq13ukrk

[sivencov_v]

Hello! I badly understand English text to translate, sorry.
Tell me, if you write the RAW video on my camera 650d, make such artifacts


how to solve this problem please help? or give a link to the topic where they talk about it?

once again sorry, I have not found a solution to this problem, due to the poor knowledge of the language

thanks for the help !

p.s. I have installed a version of ML v2.3.NEXT.2014Jan04.650D104

[Walter Schulz]

PinkDotRemover
http://www.magiclantern.fm/forum/index.php?topic=6658.msg73736#msg73736

[sivencov_v]

PinkDotRemover
http://www.magiclantern.fm/forum/index.php?topic=6658.msg73736#msg73736

thank you very much! everything works, artifacts disappeared!!!!!

you helped me a lot! ...and Google translator 

[Roger2004]

First test in 1280x720 - 24fps

Far from perfect (also the workflow was really annoying too) but would now love to have a 5dmk3 and playing with 1080 raw because coming from 1080 h264 to 720p even in raw is a bit of a shock...!

[gabolandia]

I Have an Error on my 650D (T4i) when i record RAW, it stopped immediately and after says: Early Stop (17) Please report this bug. What happened? Thanks.

[a1ex]

The raw recoriding module over-estimated your card speed a little, so it stopped earlier than normal. It might have squeezed a few more frames (1-2 seconds).

I added this message so I could get an idea about how often it happens.

[gabolandia]

The raw recoriding module over-estimated your card speed a little, so it stopped earlier than normal. It might have squeezed a few more frames (1-2 seconds).

I added this message so I could get an idea about how often it happens.

So i need a faster card, right ?

[Walter Schulz]

Depends. 650D is bottlenecked by a slow SD-card interface. You get about 41 MByte/s transfer rate. No more. Period.

Ciao, Walter