ZOOM H4N and d90

Coco Bermudez

Coco Bermudez

Well-known member
Has anyone already covered this topic? I am planning on getting a ZOOM H4N to use with my d90...my question is...will it be out of sync once I drop the ZOOM H4N audio files on the timeline and try to sync it with 24p video straight out of the d90?

I've searched around this forum but have not found an answer to this
 
Why would it be out of sync? I mean i can see a million reasons it could but why are you worried? I worry about sync when i shoot wind up 16 mm, not when everything is digital. :-)
 
It's an understandable concern, since the H4 has this problem, but yeah, in the digital realm you can fix this quite fast. Nevertheless, I expect it to be in sync to begin with, precisely because it's digital... ;)
 
Im not exactly sure what the best way is to achieve the sync since the audio i believe is recorded at 23.976 and the video is exactly 24p. Either slow the audio down or speed up the video?
 
It depends on the nle, audio is just audio with no inherent framerate. There's always some combination of timecode, timebase and framerate that gives you correct sync, even though it can be hard to find. If your nle works like it should there should be no difference in sync if you conform the video to 23.976 and edit on a 23.976 timeline or if you do native 24.
 
My question comes out of pure ignorance about recording audio. I've really never had to think about it just because I just plug a boom, a lav or whatever into a video camera and there it is...you got sound.

In the video world (as long as I know) when I set my camera and shoot 60i with audio and drop it into my FCP tiimeline, the video and audio are always in sync. When I shoot 24p with my video camera and record audio and drop it into a FCP...there it is again...audio and video are sync'ed...never had to think about it...maybe the camera compensated for frame rate and adjusts the audio to be recorded in sync with video frame rate? I don't know.

Now that i want to get a better audio solution for my d90 I find everyone raving about this and that solution. I see people at the Canon 5d markII having problems shooting audio and video and then converting it to 24p and having all out of sync with various recording devices. To my attention came the ZOOM H4N...but at what frame rate does it record? Does sound even have a frame rate? I am not understanding something here....

-----------------------------------------------

oops..just saw Mattsand's replyy that audio has no frame rate....interesting and more confusing
 
audio has a "frame rate" of 48.000 fps. if recorded at exactly that and played back at exactly the same it will run at the same speed as the sound that was recorded. the same goes for video, if it's recorded and played back at exactly the same framerate it will run at the exact same speed. now there are several problems. some recorders might not run at exactly the right speed, wind up film cameras as i mention can run at anything from 23 to 25 fps when the dial is set to 24, for example, and analog tape recorders have the same problem. in the digital world there's no reason a recorder or playback device shouldn't always be perfect, but then there's the issue that we often record faster or slower to match ntsc and pal frame rates, 24 fps film is usually played back at 23.98 or 25 fps, and the same speedup or slowdown has to be applied to the audio as well. some nle's do this automatically, which can be great but can also be disaster if you're trying to compensate yourself as well. some audio recorders can do this automatically for example, and some video cameras that say they are 30 fps are really 29.97 fps, some record 29.97 but the nle can only play it back as 29.97, and so on. i don't know if this made you any wiser but it should give you an idea of what the problem is and what it isn't.

/matt
 
If its not sync with smpte timecode, its not really sync'd. However, you can get it real close. Some devices will drift. This is because the two devices have different internal clocks moving at different speeds. Hence the reason many using DSLR cameras most Slow the audio file in FCP to 99.9%.

And yes audio may have different frame rates, Ex:

44.1 Khz | 16 Bit |@ 30fps say for a Record
44.1 Khz | 16 Bit |@ 29.97fps for a TV Show
48.1 Khz | 16 Bit |@ 24fps for Film
etc..

The only thing you should keep in mind is Devices like Zoom H4n do not give you the option for Varable framerate. Sound Devices 702T can, but set you back $2,500. Hope that helps.

Cheers,
MrMoe
 
mrmoe said:
If its not sync with smpte timecode, its not really sync'd
Click to expand...
oh come on, if that was the case who says the smpte generator doesn't drift as well? counting the number of samples or the number of frames gives you an equally valid and exact timebase. but yeah, smpte has the ability to skip frames and catch up since it's continuous, but that's rarely an issue for maintaining sound sync.
the two devices have different internal clocks moving at different speeds. Hence the reason many using DSLR cameras most Slow the audio file in FCP to 99.9%.
Click to expand...
not really. as long as you play back the material at the same speed it was recorded it will stay in sync. of course if you record at 24 fps and play it back at 23.976 you need to slow down the audio accordingly, but it has nothing to do with clocks. 24 and 48.000 are already as different numbers as they come, right?

/matt
 
In answer to the original poster's request:

Super 8, this a classic problem with the all the Zoom (Samson-Phonic = nasty) hand-held digital recorders, I've been using mine for about 3 years and although the device is incredibly handy, it's inner clock runs a tad fast which means you have to compress the audio on the timeline ever so slightly. The lag in the audio starts to become apparent at around 1.5 minutes into sync'd playback.

Here's the funny thing; my little Iriver T30 can sync up to video for a good hour before there is any drift, so the mind wonders, why, if Iriver can get this right on a pidly $50 recorder why Zoom can't do the same on a device costing 9 times the price!!
 
Last edited:
thanks all for the responses....lots to think about...what a mini headache!

Evro said:
Here's the funny thing; my little Iriver T30 can sync up to video for a good hour before there is any drift, so the mind wonders, why, if Iriver can get this right on a pidly $50 recorder why Zoom can't do the same on a device costing 9 times the price!!
Click to expand...

hmm...good point...I would take the $50 dollar solution!
 
mattsand said:
oh come on, if that was the case who says the smpte generator doesn't drift as well? counting the number of samples or the number of frames gives you an equally valid and exact timebase. but yeah, smpte has the ability to skip frames and catch up since it's continuous, but that's rarely an issue for maintaining sound sync.

not really. as long as you play back the material at the same speed it was recorded it will stay in sync. of course if you record at 24 fps and play it back at 23.976 you need to slow down the audio accordingly, but it has nothing to do with clocks. 24 and 48.000 are already as different numbers as they come, right?

/matt
Click to expand...

Matt sorry to disagree. Smpte time code standards were created for that reason alone. Ask any Professional Sound Supervisor, Recording Engineer, Sound Recordist, Mixer or Audio Expert who has worked on theatrical films, tv and records. Proper sync is with Timecode and clocks have everything thing to do with it.

However, I will agree with you that will live in a digital age and you can always, cut, slip, nudge things done to the subframe and get good sync. When you use timecode, you never have to slow done or speed up anything. It just works! Perfect sync without spending extra hours or money.

It would be like me arguing that FPS doesn't matter in Video. In Audio, FPS does apply and is just as important for locking sync. Sympte timecode = locked sync.

I grew up in studios syncing everything from, Otari 2-4 Track recorders, Multiple Studer A824 24 track 2 inch machines, DAT w/ timecodes, Protools, as well as using crystal sync devices for the machine rooms with video decks, Digibetas, D2, 3/4 machines, etc. One device is the master, the rest can be slaves.

Please don't misunderstand me, The Zoom H4n is a great little device and for the price, its a steal. I would recommend it in the digital age. Hope that helps.

Cheers.
 
Last edited:
Evro said:
In answer to the original poster's request:

Super 8, this a classic problem with the all the Zoom (Samson-Phonic = nasty) hand-held digital recorders, I've been using mine for about 3 years and although the device is incredibly handy, it's inner clock runs a tad fast which means you have to compress the audio on the timeline ever so slightly. The lag in the audio starts to become apparent at around 1.5 minutes into sync'd playback.

Here's the funny thing; my little Iriver T30 can sync up to video for a good hour before there is any drift, so the mind wonders, why, if Iriver can get this right on a pidly $50 recorder why Zoom can't do the same on a device costing 9 times the price!!
Click to expand...

The manufacturer's of both internal clocks may be based on the same algorithm.
 
I don't have to ask anyone since i already own a recording studio and have worked with audio for many years. I'm not a sound guy but i know the workflows perfectly. I'm sorry if a rhetorical question was too much for you to handle, I never said there was anything wrong with timecode, just that if you can make a clock generator run in sync you can make anything else do that too, it's the exact same issue. Anything that makes two things run at the same speed, be it a sync block, pilotone, smpte, or simply an accurate digital clock, along with a common sync mark, be it a clapper slate, a 2-pop, a sharpie mark on the tape, or whatever, is enough and if you're gonna argue against that i promise you will lose. I don't have statistics but i'm willing to bet that a large majority of all sound films ever made were done without timecode for at least part of the production if not all.
 
The zoom h2 holds sync fine btw. Are you sure it's not an ntsc speed issue? That would be consistent with noticing a drift after a minute or so, since the difference is around one frame per 40 seconds. Even though it's true that many sampler clocks drift it's very very rarely that much for no reason.
 
mattsand said:
I don't have to ask anyone since i already own a recording studio and have worked with audio for many years. I'm not a sound guy but i know the workflows perfectly. I'm sorry if a rhetorical question was too much for you to handle, I never said there was anything wrong with timecode, just that if you can make a clock generator run in sync you can make anything else do that too, it's the exact same issue. Anything that makes two things run at the same speed, be it a sync block, pilotone, smpte, or simply an accurate digital clock, along with a common sync mark, be it a clapper slate, a 2-pop, a sharpie mark on the tape, or whatever, is enough and if you're gonna argue against that i promise you will lose. I don't have statistics but i'm willing to bet that a large majority of all sound films ever made were done without timecode for at least part of the production if not all.
Click to expand...

Matt, with all respect, I'm not looking to argue.

I was just pointing out to poster who seemed confused, that proper sync is with sympte timecode and clock devices do make a difference. Things don't magically just perfectly sync up all the time.

Yes, many films have been made without timecode, but so many hours have been lost fixing it in post, instead of just being creative. Now we have timecode to help us.

Alot of folks are having problems with syncing H4n, but understandble so. Its a recording device with out timecode. Kinda like D90 with no manual controls or timecode for that matter. :)

As always...Do what works for you. With all respect. Lets move on. Cheers. Have a great weekend.
 
Last edited:
Just to reiterate-

My testing confirmed for me = To sync H4n audio with the D90 you need to conform the video to 23.976 and edit on a 23.976 timeline, (I used FCP, others may vary). This recipe gave me perfect sync for the full 5 minutes of D90 recording-time goodness that Nikon has been so generous to give us.

But don't take my word for it. Test it out. Pretty easy.

PEACE.
 
1. Hey Tim Joy, your answer on the GH1 thread (http://www.dvxuser.com/V6/showthread.php?t=169872) concerning your method for conforming the video to 23.976 (via MPEG Streamclip) brought me to a second question. When you mention changing the timeline frame-rate to 23.976 do you mean using the 23.98 preset for a newly created sequence (FCP won't let you change the timebase after a sequence is created)? If you are talking about sequence presets, are you changing, in the "advanced tab", the Frames per second to custom and inputting 23.976 (in the Compression settings) or are you just leaving it at 23.98 in Sequence settings?). (The thing is that there is no "interpret footage" option in FCP as there is in After Effects...).

2. That Zacuto video is just like every Zacuto video, when it's over you end up wanting to buy this little piece of adapter that when you check in the website... costs $190! Having said that, and changing from sync to mount, what are the options for making a hotshoe adapter like that, or a cheapter adapter in general? I see pictures of DLSRs rigged like that in posts, but I simply don't know where these hot shoe mounts come from. Can anyone point in a good direction?

Peace to all.

Update:

Ok, I was googling it wrong, digging around with "hot shoe mount" (instead of "hot shoe adapters", which was only showing me the opposite of what I wanted), I actually found these options:

http://www.adorama.com/VDMALCDMT.html

http://www.markertek.com/Digital-Vi...tor-Mounts/Delvcam-Monitor-Systems/D489.xhtml

http://www.markertek.com/Video-Equi...pters/TecNec-Cables-Connectors/TE-LBRAC.xhtml

http://www.dvcreators.net/hama-delvcam-hot-shoe-mount/

So, my question now is: does anyone know of better options?
 
Last edited:
You must log in or register to reply here.
Back
Top