Canon T3i - Audio Dilemma

Beatlesfan1225 · Jan 19, 2014

Hello everyone,

So as of late I've been struggling heavily with audio when shooting with my Canon T3i. Let me walk you through the situation -

1. Being a student, I don't require perfect audio, I just want decent audio. One thing I've spent much time with in the past, is syncing up audio from a separate audio device. This technique has gone terribly wrong for me, since it requires a level of precision and coordination amongst my crew that, honestly, isn't there. Not to mention, I've just had difficult with syncing and finding audio files in post production when I do this. Usually a boom guy or camera guy accidentally created extra files to deal with, so I end up sitting there for hours finding audio clips and video clips that work together. This is especially hard when filming the same scene a dozen times. So, I'm ruling this method out.

2. I've resorted recently to utilizing the onboard sound of the T3i. It's surprisingly good... as long as everyone talks loudly and there's zero wind...

3. The onboard settings of the T3i are very difficult for me personally, to understand. Even though it has "manual" control, you still get a mass amount of hiss when plugging in mics. (The hiss isn't due to the mics, it happens when the audio goes silent, and the camera thinks that something is there, so it boosts to compensate. This means that quiet scenes turn into hissy scenes.) So, plugging mics directly into the T3i has proven very difficult.

4. The T3i doesn't offer any sort of sound monitoring, which is scary when using a separate mic. One of my constant fears is that if the mic doesn't pick up a scene, dies, or even unplugs on accident, you lose whatever you just shot. This would be solved if the audio could be monitored, but it can't be.

So, the overarching question I have, is how can I achieve better sound and work around these factors that are proving difficult for me? If it helps, I own both a Rode Videomic and a Zoom H1, both of which are probably better audio than the onboard, but are difficult for me to use as per the items above.

Thank you so much for any help you can provide me, it's so appreciated!

Caidh Mor · Jan 19, 2014

Collaborate with your audio tech more, and use the H1 and Rode mic. As long as the both of you take the time to be on the same page as far as file organization goes, your precursor problems should disappear. Alternately, get someone to function as an assistant editor to do the file matching for you.

Beatlesfan1225 · Jan 19, 2014

Caidh Mor said:
Collaborate with your audio tech more, and use the H1 and Rode mic. As long as the both of you take the time to be on the same page as far as file organization goes, your precursor problems should disappear. Alternately, get someone to function as an assistant editor to do the file matching for you.

I may consider having someone else do the file matching for me, that is a good idea. Does anyone have any good ideas for how to go about making sure the video and audio on set stay together fairly well? Possibly documenting it as you go in some fashion?

What I've considered doing, is plugging the output of the Zoom into the input of the camera, and the Rode into the input. That way I could push the manual function on the T3i all the way down, eliminating the hiss for the most part, and boost the signal on the Zoom instead. If I use a splitter, I can also allow audio monitoring through the Zoom as well, which potentially could eliminate my issue of not being able to hear the audio as we shoot. Then I wouldn't have to do any file matching, because all the audio is part of the video files.

Does this sound like a feasible solution?

AcousticAl · Jan 19, 2014

Beatlesfan1225 said:
1. ... I just want decent audio.

2. I've resorted recently to utilizing the onboard sound of the T3i. It's surprisingly good... as long as everyone talks loudly and there's zero wind...

These two things are contradictory.

The internal camera mic from the T3i is not "surprisingly good," even if all other circumstances are ideal. Nothing can substitute for getting the mic off the camera and as close to the action as possible.

Beatlesfan1225 said:
One thing I've spent much time with in the past, is syncing up audio from a separate audio device. This technique has gone terribly wrong for me, since it requires a level of precision and coordination amongst my crew that, honestly, isn't there. Not to mention, I've just had difficult with syncing and finding audio files in post production when I do this. Usually a boom guy or camera guy accidentally created extra files to deal with, so I end up sitting there for hours finding audio clips and video clips that work together. This is especially hard when filming the same scene a dozen times. So, I'm ruling this method out.

What precision?

THIS has worked for almost 100 years and is still a standard tool with second-system sound. If you aren't using one, you should be.

It's simple:

Step 1: call for "roll sound" (mixer/recordist calls "speed" when the recorder is rolling in record mode).
Step 2: call "roll camera" (camera op calls "speed" when camera is recording)
Step 3: put the slate in full view of the camera and read aloud the scene and take numbers
Step 4: clap the sticks
Step 5: Action!

Instead of "ruling this method out," you should be adopting it as standard operating procedure. Furthermore, your sound person should be keeping a log sheet of files by name with corresponding scene/take info, to make logging in post even easier.

In post, you have not only the visual slate on each clip that indicates scene and take, you have the same information read at the start of each audio file, making logging and matching a breeze. As for sync, simply find the frame of video where the sticks come together, and the snap in the audio waveform, and match them up.

Beatlesfan1225 said:
3. The onboard settings of the T3i are very difficult for me personally, to understand. Even though it has "manual" control, you still get a mass amount of hiss when plugging in mics. (The hiss isn't due to the mics, it happens when the audio goes silent, and the camera thinks that something is there, so it boosts to compensate. This means that quiet scenes turn into hissy scenes.) So, plugging mics directly into the T3i has proven very difficult.

The problems you hear come from two issues: the crappy audio circuitry in the camera, and AGC (automatic gain control). Installation of Magic Lantern will allow disabling of the AGC. Nothing can fix the quality of the audio circuits, but taking out the AGC will prove a marked improvement. A dedicated, external audio recorder with boomed mic is always your better bet with DSLR.

Beatlesfan1225 said:
4. The T3i doesn't offer any sort of sound monitoring, which is scary when using a separate mic. One of my constant fears is that if the mic doesn't pick up a scene, dies, or even unplugs on accident, you lose whatever you just shot. This would be solved if the audio could be monitored, but it can't be.

Installation of Magic Lantern, plus a special cable from Sescom, will enable headphone monitoring from your T3i. It's not perfect, but it's better than nothing. However, again, in-camera DSLR audio is not the best or first solution.

Beatlesfan1225 said:
So, the overarching question I have, is how can I achieve better sound and work around these factors that are proving difficult for me? If it helps, I own both a Rode Videomic and a Zoom H1, both of which are probably better audio than the onboard, but are difficult for me to use as per the items above.

Get the mic off the camera and boom it to get it close to the actors.
Slate every take.
Keep written shot and audio loga.

Lucky Hardwood · Jan 19, 2014

If you have an iPad, there are many slate apps available. I type in the info for the take (including the audio file number from the recorder and the number of the memory card being used) and hold the slate in front of the camera when it begins recording each scene/take. I also call out all of that same information What this does for me is sets the slate info as the thumbnail on the video file. It makes matching things up much simpler. You could also just write all of the information on a sheet of paper and hold that in front of the camera when (record) is pressed. I think I paid $19.99 for the slate app that I use.

Sweetie · Jan 19, 2014

Being a student, I don't require perfect audio, I just want decent audio. One thing I've spent much time with in the past, is syncing up audio from a separate audio device. This technique has gone terribly wrong for me, since it requires a level of precision and coordination amongst my crew that, honestly, isn't there. Not to mention, I've just had difficult with syncing and finding audio files in post production when I do this. Usually a boom guy or camera guy accidentally created extra files to deal with, so I end up sitting there for hours finding audio clips and video clips that work together. This is especially hard when filming the same scene a dozen times. So, I'm ruling this method out.

Step 3: put the slate in full view of the camera and read aloud the scene and take numbers

Again, this is your method.

If your current crew cannot handle this simple method, you need to find better crew. If you cannot, find h44's threads where I tell him to be a better leader so you can find better crew. Then again, after more careful reading, it seems to me that you're just lazy or you don't have any workflow (plan to have your data sorted).

There is an easier method, though it's through the use of Timecode. You'll need both a camera and the audio that use it. Warning. They don't come cheap and you should know the previous method anyway in case things go wrong.

So, the overarching question I have, is how can I achieve better sound and work around these factors that are proving difficult for me? If it helps, I own both a Rode Videomic and a Zoom H1, both of which are probably better audio than the onboard, but are difficult for me to use as per the items above.

Please tell me you're a psychology student keeping video records of your experiments and that you're not a film student.

8salacious9 · Jan 19, 2014

im sorry but its simply not that hard to sync audio, i do it manually all the time even for hours of footage.

if you have masses of recordings then check the file info for what time the audio was taken and put it in a folder with recordings of similar time frames... not that hard its just called effort.

AudioPostExpert · Jan 19, 2014

Most of this has been said already by others in this thread but there's a few useful bits of information below, even for the experienced filmmakers on IndieTalk!

Beatlesfan1225 said:
2. I've resorted recently to utilizing the onboard sound of the T3i. It's surprisingly good...

It's not, it's surprisingly bad! In fact, it's difficult to find any equipment with lower quality mic-preamps and audio circuitry than that found in DSLRs. For this reason, any serious filmmaker, even serious hobbyist filmmakers, always use a separate audio recorder. This takes care of your questions 2, 3 and 4 and just leaves question 1.

Beatlesfan1225 said:
1. Being a student, I don't require perfect audio, I just want decent audio. One thing I've spent much time with in the past, is syncing up audio from a separate audio device. This technique has gone terribly wrong for me, since it requires a level of precision and coordination amongst my crew that, honestly, isn't there. Not to mention, I've just had difficult with syncing and finding audio files in post production when I do this. Usually a boom guy or camera guy accidentally created extra files to deal with, so I end up sitting there for hours finding audio clips and video clips that work together. This is especially hard when filming the same scene a dozen times. So, I'm ruling this method out.

Firstly of course, it's difficult to know what YOU mean by decent audio.

What AcousticAl said is ALL good advice but I'd like to expand on it a little. Slate is essential, as is a Sound Log but before you even get on to the set for the first time, you need to sit down in pre-production and consider file management. Not the sexiest aspect of filmmaking I'll admit but an essential one if you don't want to waste 10 times the amount of time in post than proper planning will cost you in pre. Making shorts is not so much of a problem, you may only have a few tens or maybe a hundred or so production sound files but in a feature length film there maybe many thousands and whatever problems you've had with shorts are multiplied many times over. If/when handing over the film to the audio post person/s, the result is going to be far more time and expense or, if it's a fixed fee audio post job, a much poorer quality audio post job! So, getting into the habit of doing it right now will serve you greatly further down the road.

First thing you want to do in pre-production is to come up with a sensible file naming convention and have the camera op and production sound mixer agree to stick to it! For example, in addition to the scene and take number, it's also useful to include the character/s' name/s in the file name, plus maybe an indication of location, whatever will be useful to the editors (picture and sound editors!). The order in which you put the information in the film name will obviously affect how it's indexed and sorted, which should make looking for files just a matter of scrolling down a list. In addition to the file name, there is a wealth of information in a number of fields which can written into the metadata of every bwav file. Some of it may be written automatically; sample rate, bit depth, time and date of file creation but there are other manual fields which can greatly help the searching and sync'ing during post. For example there are separate Reel, Scene and Take fields in a bwav's metadata. A consumer or prosumer audio recorder may not be able to write to these fields but you can download free bwav metadata editing software for your computer, so this info can written into each file at the end of the day's shooting. Combined with a good sound log, finding the required audio file/s should take just a second or two. BTW, even though there are no reels as such in today's digital world, reels are still used as a file management tool. So again, get used to using them! As Sweetie said, there's also a time-code metadata field (in fact there's more than one) and this can be even more useful as it not only helps to locate files but can be used to automatically sync them up as well. Although standard in professional workflows, the equipment required to generate and write this time-code information accurately to the audio files is probably well beyond your means/budget at the moment (as Sweetie also mentioned).

With regards to your crew, it doesn't sound like you have a crew, just some people who are willing to help with the odd thing here and there! A racing driver is not someone who can just drive very fast, a racing driver has to be able to; drive safely, monitor and manage engine, tyre, brake and fuel consumption and still drive very fast! If your Production Sound Mixer (or his/her assistant) is not able or willing to create an accurate Sound Log or do audio file management then they are not fulfilling their role and you need to make them fulfil it or find someone else who will!

G

Beatlesfan1225 · Jan 19, 2014

Sweetie said:
Again, this is your method.

it seems to me that you're just lazy or you don't have any workflow (plan to have your data sorted).

Please tell me you're a psychology student keeping video records of your experiments and that you're not a film student.

I'm a high school student, with an immense amount of work on my plate from a dozen different directions, I'm not lazy, I'm not a psych student, and yes, I have been accepted to, and will be going to film school. As an extremely busy student, I'm attempting to find a way to have this process flow a bit more seamlessly for myself. I'm well aware of how to use a slate, I'm attempting to find a technical work around to feeding audio directly into the camera.

I'm here to learn how to more effectively fix the issues for myself I addressed in the original post. If I was lazy, I'd throw in the towel and accept the sound quality the T3i is giving me. I'm here to learn.

I offered an idea of how I could possibly go about doing so in my second post in this thread, but it seemed to be entirely read over.

Beatlesfan1225 · Jan 19, 2014

AudioPostExpert said:
Most of this has been said already by others in this thread but there's a few useful bits of information below, even for the experienced filmmakers on IndieTalk!

It's not, it's surprisingly bad! In fact, it's difficult to find any equipment with lower quality mic-preamps and audio circuitry than that found in DSLRs. For this reason, any serious filmmaker, even serious hobbyist filmmakers, always use a separate audio recorder. This takes care of your questions 2, 3 and 4 and just leaves question 1.

Firstly of course, it's difficult to know what YOU mean by decent audio.

What AcousticAl said is ALL good advice but I'd like to expand on it a little. Slate is essential, as is a Sound Log but before you even get on to the set for the first time, you need to sit down in pre-production and consider file management. Not the sexiest aspect of filmmaking I'll admit but an essential one if you don't want to waste 10 times the amount of time in post than proper planning will cost you in pre. Making shorts is not so much of a problem, you may only have a few tens or maybe a hundred or so production sound files but in a feature length film there maybe many thousands and whatever problems you've had with shorts are multiplied many times over. If/when handing over the film to the audio post person/s, the result is going to be far more time and expense or, if it's a fixed fee audio post job, a much poorer quality audio post job! So, getting into the habit of doing it right now will serve you greatly further down the road.

First thing you want to do in pre-production is to come up with a sensible file naming convention and have the camera op and production sound mixer agree to stick to it! For example, in addition to the scene and take number, it's also useful to include the character/s' name/s in the file name, plus maybe an indication of location, whatever will be useful to the editors (picture and sound editors!). The order in which you put the information in the film name will obviously affect how it's indexed and sorted, which should make looking for files just a matter of scrolling down a list. In addition to the file name, there is a wealth of information in a number of fields which can written into the metadata of every bwav file. Some of it may be written automatically; sample rate, bit depth, time and date of file creation but there are other manual fields which can greatly help the searching and sync'ing during post. For example there are separate Reel, Scene and Take fields in a bwav's metadata. A consumer or prosumer audio recorder may not be able to write to these fields but you can download free bwav metadata editing software for your computer, so this info can written into each file at the end of the day's shooting. Combined with a good sound log, finding the required audio file/s should take just a second or two. BTW, even though there are no reels as such in today's digital world, reels are still used as a file management tool. So again, get used to using them! As Sweetie said, there's also a time-code metadata field (in fact there's more than one) and this can be even more useful as it not only helps to locate files but can be used to automatically sync them up as well. Although standard in professional workflows, the equipment required to generate and write this time-code information accurately to the audio files is probably well beyond your means/budget at the moment (as Sweetie also mentioned).

With regards to your crew, it doesn't sound like you have a crew, just some people who are willing to help with the odd thing here and there! A racing driver is not someone who can just drive very fast, a racing driver has to be able to; drive safely, monitor and manage engine, tyre, brake and fuel consumption and still drive very fast! If your Production Sound Mixer (or his/her assistant) is not able or willing to create an accurate Sound Log or do audio file management then they are not fulfilling their role and you need to make them fulfil it or find someone else who will!

G

I greatly appreciate this response, especially in contrast to the rest, being it's less belittling and more straight forward as to how I can work on this. I appreciate it. As a student who is about to go into film, I'm making simple short films, so I'm not lazy whatsoever, I'm just struggling with these aspects of creating films. And you are right on the money, my 'crew' is a rag-tag group of other students who I drag around to do whatever I can get them to do. So I essentially do it all on each production, which is a learning experience, one which I'm obviously still at the very beginning of.

I had no idea that I could use metadata and the timestamps on my files to help myself more effectively sort through the files for syncing; that is very helpful. I've also never considered vocally announcing the info on the slate (Yes, I do have one) for the audio equipment. Both of those could drastically help me in this arena, and I appreciate it.

What does everyone think of the possible solution I noted in my second post, though? I did a short test of it this morning and it seemed to be fairly effective.

Sweetie · Jan 19, 2014

Beatlesfan1225 said:
I'm a high school student, with an immense amount of work on my plate from a dozen different directions, I'm not lazy, I'm not a psych student, and yes, I have been accepted to, and will be going to film school. As an extremely busy student, I'm attempting to find a way to have this process flow a bit more seamlessly for myself.

I hear it all the time. I'm busy, I have no time to do it the correct way, please help me find out a quicker way to do it and still keep the quality, blah blah blah. Don't you think that if there was a better way, the professionals would already be doing it? Do you really think that a high school student who doesn't have a lot of spare time would really become the motivation for an entire segment within the filmmaking industry to change their workflow?

We all live busy lives. In my opinion, you have to make a decision. Do you want to learn to do filmmaking the correct way, or do you want to continue to spend time discovering how to design a workflow that produces inferior results.

Beatlesfan1225 said:
I'm well aware of how to use a slate, I'm attempting to find a technical work around to feeding audio directly into the camera.

The work around is really a decision. Either put up with crappy audio or do it the right way. Just decide one way or another. If you want good audio, you'll have to put in the hard yards or invest in (currently) very expensive hardware.

Beatlesfan1225 said:
I'm here to learn how to more effectively fix the issues for myself I addressed in the original post. If I was lazy, I'd throw in the towel and accept the sound quality the T3i is giving me. I'm here to learn.

I offered an idea of how I could possibly go about doing so in my second post in this thread, but it seemed to be entirely read over.

Your idea from your second post will help. I'm not sure why it's relevant, you seem insistent on continuing down a path of using camera as the audio recorder. It just won't end well and it's a bad habit to be getting into.

Alcove Audio · Jan 20, 2014

Using a slate and verbally slating has been standard practice since 1927. There's a reason why certain procedures haven't changed, because they work!

In the "digital age" there are a number of things that make life a lot easier. Although it may be painfully obvious, cameras and audio recorders name individual files, and this is a great advantage. Just imagine how easy it will be to match audio and video files when they all have very similar file names, not to mention the metadata embedded in the files. And if you have software assistance like PluralEyes it makes the process even easier as it will do a lot of the work almost automatically, leaving you only to tweak the problems.

It still is very important to maintain detailed sound and video logs. You have not reached that point yet, but in the future you may be handing off the editing of both the visuals and the sound to others. Having verbal & visual slates, and consistent accurate file names saves people like me hours of time. For me it comes when I do dialog editing. I can easily find the files that I need instead of having to listen to each individual file before assigning it.

So it makes sense to follow procedures from the very beginning of your career; you get into good habits that will save you (potentially huge) amounts of time and money when you get into larger scale projects. It also means that you will have the ability to translate to a professional situation much more easily.

And let's face it; would you like to pay me for 20 hours of digging through the audio files just to match them up? When you are on a limited budget (low/no/mini/micro budget is my niche) wouldn't you rather have me spend those hours editing, doing Foley, creating sound effects and otherwise improving the film rather than wasting my time doing "housekeeping."

Just one more thing... Most artistic endeavors are exercises in patience. I know that we all want to get to the fun stuff, but extensive, comprehensive preproduction combined with adherence to time-tested procedures will save you lots of time, money and frustration, not to mention improving your final product - and allowing you to actually have more fun.

Put this one down in the "Lessons learned" column.

And here comes one of my favorite quotes:

The difference between a hobbyist and a professional is that a hobbyist learns from his mistakes; a professional learns from the mistakes of others.

AcousticAl · Jan 20, 2014

Beatlesfan1225 said:
I offered an idea of how I could possibly go about doing so in my second post in this thread, but it seemed to be entirely read over.

Using the Zoom to feed the camera doesn't help your situation, except that it can provide reference audio to help with file sync later. BUT, that connection also eliminates headphone monitoring from the Zoom. Audio should always be monitored during production and never left to chance.

Aside from losing monitoring, it doesn't fix the AGC problem, or the issue that the T3i's audio circuits pretty much suck. I highly recommend that you install Magic Lantern (free, third-party firmware plug-in), as that will allow you to turn off the AGC. Still you'll be getting better sound by using a dedicated recorder and syncing files later.

Beatlesfan1225 · Jan 20, 2014

Alcove Audio said:
Just one more thing... Most artistic endeavors are exercises in patience. I know that we all want to get to the fun stuff, but extensive, comprehensive preproduction combined with adherence to time-tested procedures will save you lots of time, money and frustration, not to mention improving your final product - and allowing you to actually have more fun.

Put this one down in the "Lessons learned" column.

And here comes one of my favorite quotes:

The difference between a hobbyist and a professional is that a hobbyist learns from his mistakes; a professional learns from the mistakes of others.

I shall put this down as a lesson learned. On the project I'm shooting this week I'll practice keeping detailed logs and separating the audio from the camera. You're right, I should be practicing the right practice early on. Thank you for the detailed help.

AcousticAl said:
Using the Zoom to feed the camera doesn't help your situation, except that it can provide reference audio to help with file sync later. BUT, that connection also eliminates headphone monitoring from the Zoom. Audio should always be monitored during production and never left to chance.

Aside from losing monitoring, it doesn't fix the AGC problem, or the issue that the T3i's audio circuits pretty much suck. I highly recommend that you install Magic Lantern (free, third-party firmware plug-in), as that will allow you to turn off the AGC. Still you'll be getting better sound by using a dedicated recorder and syncing files later.

This process actually wouldn't eliminate monitoring from the Zoom, because I'm splitting the audio jack there - one outputting to the Canon, the other to the headphones for monitoring. You're right, though, it's still difficult to solve the ACG problem. I found that if I pushed the "manual" audio levels all the way down on the camera, and boosted them on the Zoom (which is receiving from the Rode) it seemed to do a decent job at getting rid of the ACG hiss, though. I don't intend to follow that method, but it did seem to do an okay job as an odd-ball work around to monitor the audio and reduce ACG hiss.

knightly · Jan 20, 2014

There is another bit as well... I heartedly encourage you to put in the effort to slate every take... and record with your on camera mic as well... so that either FCPx or pluraleyes can match up the audio from the two sources and attach the audio files to the video files FOR YOU! Best of both worlds... but we still slate for when the new fangled technology fails.

Chrisk · Jan 21, 2014

Not quite on topic but may open up some possibilities for the OP, but I don't totally the audio circuits on these DSLRs are totally useless. I have heard far worse, and of course, far better. Once you move away from the inbuilt mic things change very drastically.
As naughty as it may be I have recorded to my 550d (T2i) when self shooting. With magic lantern enabling manual control (which I believe the T3i may have anyway), a bastard, digital balanced input and monitoring I have plugged a wireless straight into the cam and had decent enough results for the circumstance. I'm sure Beatlesfan would be quite happy with similar results for his/her projects.
If anyone wants to listen:

https://www.behance.net/gallery/To-Kafe-ths-Chrysanthis/10961595

Bear in mind that if you cannot understand the dialogue it may not be sound quality issues