HDV audio

Emanuel said:
Can be a compressed audio system like HDV acceptable to digital filmmaking@big_screen?

This article gives the answers: http://www.sundancemediagroup.com/articles/HDV_Audio_comparisons.htm
Click to expand...

Doug and I have had quite a time discussing this. He and I will continue to disagree to some degree. While we agree that good audio should be done on a double system instead of using HDV audio, I disagree that it is as usable as he does.

His charts and graphs do not sway me. He explained to me that his tests were done by pointing an AT shotgun at a speaker. Not a good shotgun and I don't remember what the speaker was. He was playing tones through the speakers and recording them with the mic and then comparing the waveforms.

Several problems there for me; the mic, the speakers and recording tones is pretty useless because real audio has transients tones don't.

He ends up contradicting himself by the end of the piece.

If the same source were recorded at 24-bit 48 kHZ on a Schoeps with a good preamp, you'd hear a big difference between that and 384 kbps stereo that's about 1/5 the data of a 16-bit , 44.1kHz file.

When he made the comment that he was working very closely wih Sony as strength for his argument, I had another take on it.

Even the tech at Sony to whom I spoke about the matter said that for the best audio, you should double record when shooting HDV.

We got rather heated about it and the string was removed from that forum.

Regards,

Ty Ford
 
Thanks very much for your useful report, Ty Ford. I have a problem: I'd like shooting HDV instead HVX (uncompressed audio) and to record on camera without any other source (dialogues, for sure) it will be a requisite.
 
I understand. Direct recording in the miniDV mode that's 16-bit 48kHz is acceptable. A good mixer with a good preamp is also a good idea. For simple shoots a Sound Devices MixPre does a great job.

Regards,

Ty Ford
 
Emanuel said:
Can be a compressed audio system like HDV acceptable to digital filmmaking@big_screen?

This article gives the answers: http://www.sundancemediagroup.com/articles/HDV_Audio_comparisons.htm
Click to expand...

Hi Emanuel!

First off, the article is really not answering that question... it seems to be addressing the issue of whether using MiniDisc for a double system is worthwhile - i.e., comparing ATRAC compression to MPEG I Layer 2 compression. In the article, they're both clearly nowhere near the quality of PCM.

Further, I find the article a little problematic and misleading. On the issue of converters, for example, thae author says that the only thing better would be a Rosetta or Echo... but I find that hard to believe. Especially when one considers that conversion quality is dependent on the analog portions of the signal path... I'd guess ANY piece of decent outboard gear would do a better job than what's crammed into the messy electronic world of the camera internals. Ditto for preamps, etc. And, unfortunately, as his waveforms clearly indicate, we can't ignore the role of compression. It makes a big difference. Can you get away with it for dialog - maybe, for some voices, on some situations... but I wouldn't bet on it.

So: It may be correct that MiniDisc is no better than HDV audio. But that DOESN'T mean that compressed audio is OK. And now, there are so many low-cost options that record uncompressed (such as Flash-based and MD-HD and even the new iPods) that I find the comparison pointless.... if you care about sound quality, you'd use an uncompressed device for a double system, not MD.

-Barry
 
Hi Ty and Barry!

Thanks Barry for your enlight advice and thanks again for your care Ty! I was@your_website and I downloaded your clip where you have a good performance as cinematic character! My compliments!

Well, I see who you are. So I must present myself, as well.

Now, I'm working as arthouse digital filmmaker@europe where I'm developping some projects to film-out@big screen. My filmschool experience is the directing as major & editing as minor fields. About sound I had only one-year experience@NAGRA & mini-disc for video purposes (in those times were introduced@ESTC LINK where I studied) and some 16mm film-out option.

Concerning your tip, Ty, do you are talking about one sole piece, isn't it? That is LINK1 LINK2

Is it the best affordable option to my needs?

Is it strictly necessary its use in all situations? Or is it possible to have good audio only with the Z1U manual options, for instance?

Barry (ullanta) gives other uncompressed possibilities but as I said, my point it's to have the sound at same time than image on camera. It's easier...well, if acceptable even to 35mm film-out purposes@big_screen. If not, HD100 or Z1U cannot be more a possibility rather the HVX uncompressed four channels...

What can I have as alternative? Flash-based, MD-HD or even the new iPods? Any suggestions?

(sorry my ignorance but the sound definitely it's not my field...)

Thx in advance!

- Emanuel
 
Well, if you want HD with the best in-camera audio you can get, the HVX is clearly the way to go for now.

I read somewhere (and I'll try to post it tomorrow if I can find it) that the JVC HD100's "ProHD" format actually records uncompressed audio as well as compressed... but that retrieving that audio is problematic. The poster had been told by JVC techs that, after capturing the HDV video/audio from tape, the tape could be played back and captured again IN DV MODE to capture uncompressed audio. Sounds strange to me... but something to look into, maybe?

For a double system however, there are many possibilities at many price levels. Tell us a bit more about your needs, and what you plan to shoot, and how muchg you can spend, and we can give you some good ideas. There are devices small enough to prevent you from feeling the impact of a double system, and bigger ones that may simplify your syncronizing.

I personally tend to syncronize by hand. I do a rough lineing up of audio and video by waveform (using the camera audio waveform as a rough guide), then hand-tune it to what works best. This is essential in a lot of what I do (primarily classical music performance), because the mics are often far from the camera anyway, and speed-of-sound differences become very important in such things as fast piano passages. It's doable. For film-style work, a clapboard helps of course!
 
My HVX problem...well, you probably know...picture noisy. So, regarding HD100, I read this: LINK

Simple syncronizing will be appreciated even because I wouldn't like to work with clapboard. Handy-doc-work-like, do you understand me? That's why audio on-camera would be great. Do you think that compressed audio to dialogues it will not be enough?

If so, what's the most affordable solution?

According what I can follow from Ty's opinion, uncompressed audio to dialogues sound capture it will be acceptable. Of course, maybe with a good mixer with a good preamp, as he adviced. Don't you agree?
 
tyreeford said:
I understand. Direct recording in the miniDV mode that's 16-bit 48kHz is acceptable. A good mixer with a good preamp is also a good idea. For simple shoots a Sound Devices MixPre does a great job.

Regards,

Ty Ford
Click to expand...
After seeing again your post I notice that you are talking about mini-DV mode (16-bit 48khz), well, I'm talking about HDV where the sound is compressed in any of the possibilities: JVC HD100, SONY Z1U, CANON HL-H1. What's your superior advice about the best audio capture on-camera solution regarding these three models?

Thanks in advance!
 
Well, you basically should try a camera out in the kind of situation you'll be in, ands see what you think of the sound...

And here's a super-crappy solution that may or may not help depending on the camera's signal path: get an HD-MD recorder or anything else cheap that will record uncompressed (M-Audio 24/96, whatever... there's a lot...) And hook that up to the RCA audio outs of your camera. Record everything this way; only look to the uncompressed recording for parts of the audio that don't sound good in the compressed version...

Of course, you should test this to see if sound recorded this way is significantly better....
 
I'm extremely wary for two reasons;

1. On-camera mics are too far away to sound good, frequently pickup the sound of the shooter and the camera mechanism.

2. The audio capture on HDV is 384 kbps stereo MPEG1, layer 2. That's about 1/5 the audio data of 16-bit, 48 kHz audiothat has a data rate of about 1.5 Mbps.

Regards,

Ty Ford
 
ullanta said:
Well, you basically should try a camera out in the kind of situation you'll be in, ands see what you think of the sound...
Click to expand...
The question is I must to have a decision without this possibility to a comparative test.
 
tyreeford said:
I'm extremely wary for two reasons;

1. On-camera mics are too far away to sound good, frequently pickup the sound of the shooter and the camera mechanism.

2. The audio capture on HDV is 384 kbps stereo MPEG1, layer 2. That's about 1/5 the audio data of 16-bit, 48 kHz audiothat has a data rate of about 1.5 Mbps.

Regards,

Ty Ford
Click to expand...
1. When I speak on-camera sound record I don't mean to use on-camera mics because of the same reasons that you notice. I want to say to record on-camera at same time adding to the image capture, using the same device. That is, the camcorder itself for the image & sound together and not in separate devices.

2. For dialogues capture so much compressed signal can be considered acceptable to 35mm film-out purposes at big screen or not? -- it's my major doubt...

...if not despite the article above-posted?

I'm referring HDV audio and not DV audio, unfortunately.

Thanks in advance!
- Emanuel
 
I have worked on audio recorded on the Sony Z1 and can tell you at least from the Sony stand point, they are still putting $3.00 worth of audio equipment in their video cameras. All else being equal, compressing audio will always give you worse results then uncompressed. This is critical when orginally recording your audio. With my results i have found the room tone to be completely damaged and inconsistent with the Z1 audio. I would compare the audio quality similar to what you get running your audio through an analog wireless.

The article you linked is written by a guy who is an fan of HDV so understand the source is bias. Secondly, it says little as to the quality of compression of the audio. The correct comparison would be to record several speakers (with different freq ranges), to the Z1 in HDV mode while recording on a 2nd Z1 in standard DV mode with uncompressed audio.

You will probably find some significant lost in extreme voices but not much in the rest. Sony has made it quite clear they are not concerned with recording audio on their cameras so I would not expect the audio in DV mode to be anything I would consider acceptable (judging from their PD 150/170 cams). The other 2 HDV manufacturers I have not heard their audio yet and might be better in what they put into their audio before the compression.

To get quality audio you are going to want to record double system with any of these cameras. The TC in/out of the Canon makes this easy. With uncompressed audio you can easily record TC onto the audio tracks to use for sync in post, not sure if this is possible with HDV compressed tracks.

Cheers
 
I don't know anyone that can tell 384kbit 2 channel encodings from 16bit original audio cd input. And that's in my system which Includes Dynaudio BM15A actives, ATC 50A and Bryston 4st/Yamaha NS10. The ATC is extremely precise and reveals every single bit of detail. It gives justice to any recording and is an extremely good monitor. A lot better than most theater Dobly/DTS installations and certainly better than the typical HD tv:)

384kbits is practically the same with uncompressed by itself.

If the audio is going to be processed, things change a little, but at 384kbits it is pretty safe. At 128kbits, after processing and reencoding to mpeg for distribution it is pretty bad. At 224kbits I can barely hear artifacts is some forms of audio processing (extreme dynamics mostly). But 384 will do the job.

Thinking logically, the rate is 25mbits. 384kbit mp3 only saves a few kb/sec. To be more precise, 16bit uncompressed 48khz stereo would be 192kbytes/sec and mp3 is 48kbytes/sec. So they save 144kbytes/sec or 1.152mbits/sec out of 25mbits. And they do need more logic to compress/decompress the signal. Do you think that an engineer would choose 384kbits if it wasn't good for the job only to save 4.6% of the total rate? The engineers decided that audio does not loose anything important at 384kbits, and this would increase the logic cost only by little and provide that little extra rate to the video which definetely needs it in HDV. These things are decided by professionals.

If HDV video was audio, it would be 128kbits quality in best case.

Bottom line. Don't worry about compressed audio:) You wouldn't record an extremely important musical performance at 16bits and with those ADs anyway. It is for dialogue, ambience and some effects. And it does than fine even compressed at 384kbits.
 
Thinking logically, the rate is 25mbits. 384kbit mp3 only saves a few kb/sec. To be more precise, 16bit uncompressed 48khz stereo would be 192kbytes/sec and mp3 is 48kbytes/sec. So they save 144kbytes/sec or 1.152mbits/sec out of 25mbits. And they do need more logic to compress/decompress the signal. Do you think that an engineer would choose 384kbits if it wasn't good for the job only to save 4.6% of the total rate? The engineers decided that audio does not loose anything important at 384kbits, and this would increase the logic cost only by little and provide that little extra rate to the video which definetely needs it in HDV. These things are decided by professionals.

If HDV video was audio, it would be 128kbits quality in best case.

Bottom line. Don't worry about compressed audio:) You wouldn't record an extremely important musical performance at 16bits and with those ADs anyway. It is for dialogue, ambience and some effects. And it does than fine even compressed at 384kbits.
Click to expand...


OTOH, thinking logically: 16-bit, 44.1 stereo is about 1.5 Mbps. That's about FIVE TIMES the data rate of 384 kbps. If you can't hear the difference on your monitors, try using better mics and preamps.

Regards,

Ty Ford
 
You must be kidding:huh: Forget mics and preamps. Do a 384kbit mpeg yourself from a commercial cd. Do you hear any difference? There have been many studies on the subject of mpeg audio compression. Done by scientists and in very large groups of people and with ABX blind tests. We know 384kbit stereo is extremely close to 16bit quality. It works even in tests performed by audiophiles from various sources (hifi magazines etc). You can't ignore all this and tell me to check my preamps or my ears:)

The problem with cameras is the low quality analog to digital comnversion. There is distortion and there is noise. They are 20-40dbs more noisy than a good external recording solution. It is something that we must expect from integrated designs that are just an addon to a video camera. The last thing anyone should worry about is mpeg compression at 384kbits.
 
limo991 said:
The problem with cameras is the low quality analog to digital comnversion. There is distortion and there is noise. They are 20-40dbs more noisy than a good external recording solution. It is something that we must expect from integrated designs that are just an addon to a video camera. The last thing anyone should worry about is mpeg compression at 384kbits.
Click to expand...
But regarding HDV cameras is there (low quality) analog to digital conversion? Isn't it 384Kbits stereo MPEG1?
 
In the case of the JVC HD-100 there are 4 audio streams captured. When in HD mode it captures not only the HDV audio but also the uncompressed PCM audio in stereo. Soooo, if there were to be any problems with the HDV audio, the PCM is there for the taking as well.

take care guys...
 
You must log in or register to reply here.
Back
Top