Tracks done, videos done, now how to upload high quality to Youtube?

[Folder]

How do y'all syncronize the video with the sound?

 

I've just started experimenting with my camera.

 

I have a splitter cable coming out of my mp3 player with one side going into the camera and the other side I use for monitoring thru my stereo.

 

I can line up all the camera audio tracks really well by zooming in on the waves in Sony Vegas. They are close enough that when I play them back I hear no phasing issues.

 

But when I import the original Wave file it looks slightly different than the camera waves do. The samples look like they are not in the same places than the 48khz waves so it is harder to get it to align properly.

 

My camera records audio ACC at 48khz but the master file that I'm importing into Vegas is an mp3 at 44.1khz. So now I'm thinking that if I convert the camera files to 44.1 they will look the same as the master file. Or maybe I should convert the master file to 48? But I thought that YouTube audio should be uploaded at 44.1 and my master is 44.1.

[UstadKhanAli]

 

I recorded this in a great studio called Sound Sanctuary , and it was tracked to Pro Tools with high quality mics, amps, and front-end outboard gear. But more importantly, it was mixed by Phil O'Keefe and mastered by Bill Plummer, which I think were the most important tools in achieving the sound.

 

 

Did you feel that Phil or Bill would want to be called a tool?

[blue2blue]

How do y'all syncronize the video with the sound? I've just started experimenting with my camera. I have a splitter cable coming out of my mp3 player with one side going into the camera and the other side I use for monitoring thru my stereo. I can line up all the camera audio tracks really well by zooming in on the waves in Sony Vegas. They are close enough that when I play them back I hear no phasing issues. But when I import the original Wave file it looks slightly different than the camera waves do. The samples look like they are not in the same places than the 48khz waves so it is harder to get it to align properly. My camera records audio ACC at 48khz but the master file that I'm importing into Vegas is an mp3 at 44.1khz. So now I'm thinking that if I convert the camera files to 44.1 they will look the same as the master file. Or maybe I should convert the master file to 48? But I thought that YouTube audio should be uploaded at 44.1 and my master is 44.1.

If that was the problem (the 44.1/48 kHz audio mismatch), the sync would be hugely off -- about 10%.

 

But Vegas attempts to and usually succeeds in allowing the user to mix video and audio codec formats freely and mix and match video clips of different frame rates and audio of different sample rates.

 

 

If it's just a little off, it's possible that video could be misidentifying as 30 fps when it was actually 29.97 (so-called 'drop frame' -- more info here: http://documentation.apple.com/en/finalcutpro/usermanual/index.html#chapter=D%26section=6%26tasks=true ) or vice-versa.

 

But it's also possible (and probably more likely, since no such mis-identification is necessary) that the crystal clock which controls your digital camera's timing is slightly off-sync from the crystal clock that controls timing in your computer's soundcard or other audio interface (which it sounds like you're simultaneously recording audio into -- if I understand what you wrote above).

 

[The impossibility of 'perfect' crystal tuning is the same reason one must set a single master clock source and slave all other devices when using multiple audio interfaces in audio recording. And why, when using separate audio and video recording in a professional environment, synch between camera and audio recorder must be maintained in a similar manner, by setting a single master clock source and slaving all other devices to it.]

 

Such a mismatch wouldn't necessarily be much -- but with lower quality/poorly 'tuned' (micro-precision machined or "shaved") clock chips, it could be -- I had a Nikon still-oriented cam made around 2005 or 2006 that also had a very poor attempt at video that -- when attempting to match audio from my 'pro' audio interface (a MOTU 828mkII) -- could produce a timing drift of roughly a second per minute -- about 1.66%.

 

And the onboard audio chips in many computer mother boards may be quite imprecise in their own timing. The audio chip in my computer's motherboard does track at a substantially different rate than my presumably better-clocked MOTU audio interface -- roughly 1/3 of a percent [0.33%]. It's not hard to see how, if two such imprecisely timed devices were substantially out of sync 'in opposite directions' such a relatively large timing drift [that ~1.66% drift I experienced between my MOTU's clocking and that of my Nikon camera] would be possible.

 

Unfortunately, if the latter is the case, I'm not sure there's much that can be done -- except 'hope' that one of them fails, allowing you to buy a better replacement.*

 

 

* But that is precisely what 'cured' my situation when attempting the same sort of slip-sync. One day, not all that long after the one year warranty, the thoroughly pampered Nikon camera refused to open its telescoping lens assembly on power-up. After a number of attempts to coax the relatively new Nikon back to life -- and some googling that revealed that it was a very widespread complaint about that model of Nikon, I bought a cheaper -- but far superior -- Cannon [which made me rather sad because I'd been a fan of Nikon in my college years in the early 70s, having owned and rather liked one of their Super 8 cam film cameras] and it's worked out quite well, even for such low tech slip-sync applications.

 

And, unlike the Nikon, the video image quality, which maxes out at the same 640x480/30 fps (but also offers 320x240/60 fps) and on the fly automatic exposure are quite decent; the onboard audio, while a bit tinny, is acceptable and can even be used for music with some corrective EQ -- as opposed to the audio on the Nikon, which not only sounded like it was 8 bit (if that) -- but which was a couple hundred milliseconds off sync from the picture.

[Jeff da Weasel]

 

How do y'all syncronize the video with the sound?

 

 

I do all my syncing "by hand" in my video editing application. If it's a music video like the one I posted earlier, the song remains static, which means I adjust the video so that the lip movements align with the audio. By the way, when making music videos, never, EVER try and "free flow" your lip sync. Make sure you have a playback device available while shooting so you can sing in perfect time with the actual song. Otherwise, it's bound to get screwed up.

[Jeff da Weasel]

Did you feel that Phil or Bill would want to be called a tool?

 

The sharpest tools in the shed, both of them.

[LRon Butterfly]

Did you feel that Phil or Bill would want to be called a tool?

 

It wouldn't be the first time.

[blue2blue]

I think being a 'tool' of the music is what an artist does. The music -- or the muses if you will -- is/are looking for the right tool. I aspire to being a ready and willing tool as much as possible.

 

[LRon Butterfly]

If you are on a mac, the Elgato turbo.264HD is a great help. It is a hardware accelerator (USB) that converts most any format to YT and does easy, quick uploads. There is a software only version that only does the conversion/upload.

[Folder]

 

never, EVER try and "free flow" your lip sync. Make sure you have a playback device available while shooting so you can sing in perfect time with the actual song. Otherwise, it's bound to get screwed up.

 

 

Don't know what you mean by free flow. I plug my mp3 player straight into the mic input on the camera and mime along while it's recording. There is no digital input and it records sound at 48khz.

 

I then go in and align all the audio waves visually in Vegas. The volume settings on the mp3 player and the camera stay the same for each instrument so the waves look virtually the same. I don't even look at the video just the audio.

 

It's when I import the master wave, the one that I plan on using as the final audio track that I have problems, it is at 44.1khz whereas all the camera tracks are at 48khz so when I zoom all the way in as far as I can go it looks different than all of the tracks that were recorded into the camera at 48khz.

 

This is the only way that I have figured out to do it without some kind of professional time code or something.

[Folder]

 

I do all my syncing "by hand" in my video editing application. If it's a music video like the one I posted earlier, the song remains static, which means I adjust the video so that the lip movements align with the audio. .

 

 

I'm trying to do it the other way around I guess, the main video is staying static and when I bring in the other parts I try to sync them by lining up their audio waves.

[Folder]

 

If that was the problem (the 44.1/48 kHz audio mismatch), the sync would be hugely off -- about 10%. But Vegas attempts to and usually succeeds in allowing the user to mix video and audio codec formats freely and mix and match video clips of different frame rates and audio of different sample rates.

 

 

Both the 44.1khz and the 48khz waves play back together in the same project at the same speed and pitch.

 

I am trying to sync them visually and aurally. I have no time code.

 

If one wave is slightly ahead or behind the other then they phase or sound like a flanger. I nudge them foward and backward until I get them as close as I can with as little phasing or flanging as possible then I know they are synced together. I mute all the camera tracks when rendering.

 

I have little problem syncing all the camera waves because they all look virtually the same. 48khz

 

The master 44.1 wave looks different when zoomed in (the sample points are in different places) so it is very hard for me to get it "exactly" in sync without phasing issues. This is the most important track to have synced because it will be the only one not muted during rendering.

 

Maybe I'm going about it the wrong way, I was just asking how other people do it.

[Folder]

 

-- I had a Nikon still-oriented cam made around 2005 or 2006 that also had a very poor attempt at video that -- when attempting to match audio from my 'pro' audio interface (a MOTU 828mkII) -- could produce a timing drift of roughly a second per minute -- about 1.66%.

 

 

I have figured out that the camera recorded audio files are slightly shorter than the mp3 audio.

 

Over a three and a half minute song the camera's AC-3 48khz audio is 3 or 4 tenths of a second shorter than the master 44.1 mp3 track.

 

I assumed that all digital recorders had perfect timing and that any recorded audio files of the same source would be the exact same length. Unlike tape recorders which can have speed variations.

 

Or maybe AC-3 files are different lengths than mp3 files?

Sony Vegas will not import AC-3 audio files only video files with AC-3.

[Jeff da Weasel]

 

I'm trying to do it the other way around I guess, the main video is staying static and when I bring in the other parts I try to sync them by lining up their audio waves.

 

 

Man, that's like mixing using waveforms. Don't even look at the waveforms. Your people viewing the video won't see waveforms. Use the song as the guidepost and sync the video by ear/eye to the lip motion. The end result will always be more natural.

[pressthekey]

Decided to go for a new video camera. Any advice or recommendations appreciated or see my other thread. Thanks.

[blue2blue]

I have figured out that the camera recorded audio files are slightly shorter than the mp3 audio. Over a three and a half minute song the camera's AC-3 48khz audio is 3 or 4 tenths of a second shorter than the master 44.1 mp3 track. I assumed that all digital recorders had perfect timing and that any recorded audio files of the same source would be the exact same length. Unlike tape recorders which can have speed variations. Or maybe AC-3 files are different lengths than mp3 files? Sony Vegas will not import AC-3 audio files only video files with AC-3.

It's all dependent on the clocks involved (any camera, audio interface, or other digital converter will have a clock of some kind which is the reference for AD and DA -- they're mostly crystal but some very high end converters actually have atomic clocks that are reputedly extremely accurate) when transitioning between the 3DW and the digital realm. Crystal clocks are 'tuned' by shaving the crystal to a certain mass and shape. Variances in that process result in variance in resonant frequency of the crystal. Also worth noting ambient temperature affects crystal clocks -- although I suspect that the variance is not great, but I really don't have a grip on what kind of percentages we're talking about.