Recording setting for EEVblog ?

[JacquesBBB]

Hi Dave,

You may have alredy put it somewhere (in this case just provide the link) but if not, could you pleqse describe in detail the whole material and setting you are using for recording and mounting the EEVblog episodes.

They are of very good quality, including the sound, and I may wish to copy it for some MOOC educational project.

Thanks

[EEVblog]

Some info here, but outdated.
https://www.eevblog.com/2013/04/16/designing-a-silent-cheap-video-editing-pc/

[JacquesBBB]

Thanks Dave,
This is perfect  for the editing of the video.

Can you add some more details on

- Your current camera

- How do you record the sound ?  Is it  directly from the camera ?  With an external microphone ? Which one ?
 I saw that on some videos  with interviews, you use some additional microphones. Is it stil the case ?  Which microphones ?

[EEVblog]

Canon HFG30 for main bench, whiteboard, and field stuff. So almost all my video work.
Sony NEX-VG30 for mailbag talking head shots (it's annoying to use for anything else)
Internal mic on both cams.
Sennheiser EW 122-P G3 wireless mic for whiteboard and interview stuff. Interview stuff I mix with the internal mic as well.
Occasionally a Rode Videomic Pro on the Canon.
Rode Podcaster mic for voiceover.

[Fungus]

A good part of the secret is also in the lighting setup, those weird-looking red/black panels on the walls, the time spent trying out all the possible camera options, etc.

Making good recordings is much more than just going shopping. All Dave's equipment is over a year old and therefore 'obsolete'. There's much better hardware available in the shops now.

[JacquesBBB]

But its good to know that it is possible to achieve excellent results with a relatively modest equipment.

If you have any suggestions for a more up to date material, you are  welcome to mention it, but of course in the same price range.

[EEVblog]

A good part of the secret is also in the lighting setup, those weird-looking red/black panels on the walls, the time spent trying out all the possible camera options, etc.
Making good recordings is much more than just going shopping. All Dave's equipment is over a year old and therefore 'obsolete'. There's much better hardware available in the shops now.

I could make decent videos using my phone given enough light and fiddling around. My lab doesn't have nearly enough light, it's always a compromise.
Audio is more important than video. But it's not worth spending twice the effort to get a 10% improvement for example. Same with the video. Again, it's always a compromise.
It's much easier if you shoot the same type and style of video every time. Almost every video I do requires something different in terms of setup, so I'm always fiddling around.

[EEVblog]

But its good to know that it is possible to achieve excellent results with a relatively modest equipment.

As someone who makes a lot of videos, usability of my gear for my purposes is more important than the quality.
My Sony NEX-VG30 for example is a better quality quality camera than my Canon HFG30, but I use the Canon for 95% of my work. There are usability reasons for that.

And it's all the little things that make my videos what they are (which are not spectacular by any means, but more than good enough for the type of content). The angles I use, the framing of shots, the depth of field I set manually, the macro shots when needed (different lens), the transcoding, the white balance, getting the audio right in-camera etc.
And I tweak practically nothing in editing, maybe just the odd audio level here and there.

And then there is the willingness to release less than perfect material just to get the video out there.

[Cerebus]

And I tweak practically nothing in editing, maybe just the odd audio level here and there.

I quite often 'batch watch' several EEVBlog videos in a row and one thing I've noticed is that, although you largely do a good job of getting the levels right in a particular video, there's sometimes inconsistency in the absolute level between videos. I tend to watch them after 'er indoors has gone to bed and I can find myself scrabbling for the remote to turn the volume down as I finish one video and start the next and the levels jump up (or sometimes down).

It's not easy to get right. Back in the day, when I used to do sound recording for a living, we'd spend all our time on final master mixdowns watching a pair of VU meters and PPM meters like a hawk to get levels right and the studio's chief engineer (and owner) used to run 'line-ups', cals on all the meter bridges and tape machines, once a month. We'd still get shouted at sometimes by the vinyl mastering guys as they had to wrestle with what we'd sent them.

[ez24]

I can find myself scrabbling for the remote to turn the volume down as I finish one video and start the next and the levels jump up (or sometimes down).

I have this problem with all YT videos.  I am surprised YT does not set the levels.  I believe Dave uses Vegas.  Maybe it has (and he could use) a function called Normalize in Audacity.

Maybe someone knows how and what we could use to run videos through to "normalize" the audio levels?

I have set up a "chain" in Audacity with Normalize in the chain, and I always run my MP3's through it before I listen to them.  I do not know how to do videos.

Anyone know how to "normalize" the audio levels in videos in a batch method?

thanks

[Fungus]

I believe Dave uses Vegas.  Maybe it has (and he could use) a function called Normalize in Audacity.

Most video editing packages are rubbish for normalizing. Either they have a crappy preset or require an awful lot of manual work.

I save my audio out as a separate track and normalize/compress it in Audacity.

("compression" as in dynamic range compression, not data compression)

[Fungus]

It's not easy to get right.

Nope.

The real pro editors chop the audio track up into a gazillion pieces and normalize each one separately.

It's a real chore, only worth it for people who get paid by the hour.

[ez24]

I save my audio out as a separate track and normalize/compress it in Audacity.

("compression" as in dynamic range compression, not data compression)

How do you separate it? 

Compressor is my second most useful Audacity function (after Normalize).
It really helps OTR files (old time radio).  Took me years to find it 

It would be great if this could be done batch method (take out audio from video).

[Fungus]

I save my audio out as a separate track and normalize/compress it in Audacity.

("compression" as in dynamic range compression, not data compression)

How do you separate it? 

Compressor is my second most useful Audacity function (after Normalize).
It really helps OTR files (old time radio).  Took me years to find it 

a) Render the finished video with no data compression on the audio track.
b) Extract the audio as a wav file.
c) Process it
d) Re-merge it with the video from (a)

It would be great if this could be done batch method (take out audio from video).

You can probably write ffmpeg batch files for steps (b) and (d). I do it manually in my video editor (which is smart enough to not recompress the video during step (d)).

[EEVblog]

I quite often 'batch watch' several EEVBlog videos in a row and one thing I've noticed is that, although you largely do a good job of getting the levels right in a particular video, there's sometimes inconsistency in the absolute level between videos. I tend to watch them after 'er indoors has gone to bed and I can find myself scrabbling for the remote to turn the volume down as I finish one video and start the next and the levels jump up (or sometimes down).

Yes, there will be some variability between videos. But the levels inside a given video should be fairly consistent (I eyeball it, no auto leveling etc)

[EEVblog]

I have this problem with all YT videos.  I am surprised YT does not set the levels.  I believe Dave uses Vegas.  Maybe it has (and he could use) a function called Normalize in Audacity.

Normalise is not a solution the problem of variable audio level.
You can't just apply the normalise function to the entire audio track, it doesn't work. You have to do selected bits that are specifically low in level.
I already do this selectively in Sony Movie Studio were required, but the only real time I have to do that is if I deliberately move away form the mic (reaching for something and talking etc)

Compression on it's own on the whole track doesn't work work magic either.
For example:
Before compression:

After compression:


What's improved?
Nothing. You still have the soft bits and the loud bits. Pointless.

If you want a serious tool to level audio then you use the levelator. We use this on the Amp Hour radio show, and I very occasionally use it on interview video and the like.

[EEVblog]

The real pro editors chop the audio track up into a gazillion pieces and normalize each one separately.

Yep, and this is what I do in Sony Movie Studio and it's pretty trivial, just press S (split) in two spots and then Normalise and drag the clip level (it's almost always I need to amplify soft bits)
But I don't fuss, only when required.

[EEVblog]

I save my audio out as a separate track and normalize/compress it in Audacity.

("compression" as in dynamic range compression, not data compression)

How do you separate it? 

Compressor is my second most useful Audacity function (after Normalize).
It really helps OTR files (old time radio).  Took me years to find it 

a) Render the finished video with no data compression on the audio track.
b) Extract the audio as a wav file.
c) Process it
d) Re-merge it with the video from (a)

Correct. And I've occasionally done this in special cases and run it through The Levelator and re-import the audio.
No video blogger in their right mind who produces lots of content wants to dick around separating out audio and processing it separately. Only when you absolutely have to.
The key with editing is to do as little as possible, not create extra workflow steps.

[EEVblog]

Most people seems to misunderstand what the "Normalise" function is. It does not make soft parts of your audio the same level as your louder parts unless yu Normalise every single part with different level individually.
Take this audio:


If I level the whole lot it looks like practically the same, as it just take the largest peak and makes it -3dB (or whatever). It doesn't magically make the soft bits louder.
I have to manually select the soft bits like this and level them:




If your audio is filled with an hour of you moving toward and away form the mic (turning your head with a lapel mic can do this) with all sorts of levels, then your audio leveling has to be done on every single piece. it's a nightmare. Any smart video blogger or podcast host will avoid this at all costs.

[EEVblog]

Here is my edit for todays video:
1hr of material, and the only audio adjustment required is on half a dozen clips that were unusal in that I shot them on my Tagarno microscope with HDMI capture. Not my usual setup.
Maybe the odd tweak on a few tracks to drop the level were I got annimated and raised my voice, but that literally takes one mouse click and drag and 1 second to do.
Everything else recorded in camera and screen capture doesn't need anything.
And no, I don't use auto level on my camera, always manual audio level.

[Brumby]

And no, I don't use auto level on my camera, always manual audio level.

Auto anything can be problematic.  You should always aim to have control - which means paying attention to lighting, white balance and audio.  Focus is another issue where auto doesn't always give you what you want.

But on the audio - if there are a couple of seconds of silence then the last thing you want is to have the audio gain creep up trying to get a signal and find you get the rumble of the air con, a truck down the street and the flushing of a toilet at the other end of the hall.

Paying attention to all these things during the shoot is imperative for good quality footage to be captured in camera - which should be the prime objective.  Ideally, editing is the only post production process you should expect to be doing.  Enhancement is generally just too time consuming - and is to be avoided if at all possible.

[Bud]

Here is my edit for todays video:

Aha, I think i have a good idea what this video is going to be about. 

[EEVblog]

And no, I don't use auto level on my camera, always manual audi level.

Auto anything can be problematic.  You should always aim to have control - which means paying attention to lighting, white balance and audio.  Focus is another issue where auto doesn't always give you what you want.

Although I do typically use auto level when I'm recording on site and I don't have control of the situation.
That latest video with the solar install for example, I didn't know if the electrician wanted to be on camera till the last second, I didn't know what position I'd be in, what shots I'd get if any etc So I just set the camera to auto audio gain and be done with it.
Better to have footage with auto gain audio than miss footage because you are dicking around mics and levels etc, or the audio is too low and you don't pick up a candid off-camera remark or whatever.

[Brumby]

You should always aim to have control

Circumstances don't always lend themselves to the ideal - but by aiming to have control, you learn how things work and what is actually important - and can better judge the compromises that may be forced upon you.  Otherwise known as 'winging it', such a practice is far more likely to produce useful results when the fundamentals are understood.