Can the DVX audio be improved if the camera is recording the sound?

And by the way Barry, not only have I seen people post 44khz, but many folks have posted that the DVX has "CD quality sound'. As you know, CD quality sound it 16/44.1. DVD quality is 16/48.
 
Sorry that was MY bad... I was simply making a general point that if you think a CD sounds great then the DVX is capable of matching that level of sophistication... sorry 'bout that guys.

I didn't think I'd spark a feud.

PS... I've had several Canons and only one Panasonic... so this isn't a DVX versus other stuff thread... it's a "why can't you get Hollywood sound with just your cam and a mic... I think you can" thread.
 
I never took it that way. It's my fault, I just did a cursory search because I remembered seeing it and when I found on or two references I went with it because I was too tired to do any real research...

Matt
 
Matt- I've had the extreme and everlasting joy of using a Sound Devices 302 mixer to mix sound for a DVX production. While I didn't do a controlled comparison, I didn't notice any major increase in audio quality by using the 302 into the line input. I don't think the DVX pre's are dirty enough to notice a big difference. Maybe with a low output mic, but with a hot mic--I doubt it. Being a sound mixer/boom op without a mixer is painful, but more for the general setup than the final sound quality of the DVX. Nothing like banging heads with the camera operator just to check levels.
 
Don't know where they're getting this 44.1 stuff from. I mean, you can look right on the viewfinder at the two modes you choose between: 32khz or 48khz. No mention of 44.1. You can look at the spec sheets in the manual; it says 48khz. Or you can import the files and look at them in a sound editor -- 48khz.

So this 44.1 is baffling. Again, no DV camera *ever* has recorded 44.1khz audio, as far as I know. It just isn't done. Hasn't ever been done, as far as I know.

And, again, yes the DV format provides for 44.1, and if you firewire'd a DV file with a 44.1khz audio track, the DVX would record it to tape (as would an XL2 or any other DV camera). But that's the only way to get 44.1 into a DV camera; otherwise it's all either 32khz/12-bit or 48khz/16-bit.

If you encounter anyone passing on the 44.1 rumor, take the time to correct 'em, 'cause it's simply not true.
 
Sorry for all the misunderstandings... just to get this issue back to center... tell me where my logic is flawed here.

Get a quiet, dialog driven movie and listen to it on headphones... now use a nice mic and record directly into your DVX (or XL2) in an acoustically excellent environment. Okay now play THAT audio back using the same headphones. Can you hear that much difference in quality? I don't think you will... honestly.

So if you do this and you can't hear that much difference in quality... meaning the clarity and realism of voice, seperated by clean silence... and overall detail... then where would you make the judgement that the DVX has anything less then AMAZING audio?
 
Hey Matt,

Did you go back and check the review of the DVX100 in 2002 on DV.com? I hope I'm not totally off base, but it seems that you are asking if a line level input will give you "substantially" better sound than just giving it a mic signal. Jay Rose said in his review on the sound of the DVX that there was basically a 2db change in dynamic range between a line level and a -50dbu mic input. Not a huge difference IMHO.

Tim
 
All,

I can throw together a little shootout in my studio and post the results for all to hear. I would like your input on what combinations of things might be of interest. I was planning to test this anyway. I will use a Neumann KM84i as the source mic, since it is fairly close to a mic that would be used to capture dialog. Here were my thoughts on a few test setups.

1) KM84 -> mic pre on the DVX100a This would be the baseline setup

2) KM84 -> Neve 1272 Preamp -> DVX100a line in. This would emulate high quality external preamplification (like a field mixer) while using the recorder in the DVX.

3) KM84 -> Neve 1272 Preamp -> Digidesign HD192. Most people would probably consider this a high end external recording scenario.

I would note that it is fairly tricky to pull off these types of comparisons fairly as even small differences in output level can skew the perceived "quality" of the results. My prediction based on gut feeling is that the preamps may result in a noticable difference for some people, and that the differences in the recorder alone (DVX100a recording vs Protools HD recording) will not be that obvious. To keep things easy I will record the audio into Protools at 48/16. It is my opinion that 24 bit source for dialog will really not provide any distinguishable difference for this test.
 
This will be very interesting!!!! I know (or at least, I feel) that the different audio chains would make a perceivable difference with music. I don't know how big a difference it will make with spoken dialog though.... I can't wait to listen!
 
Greetings, all, from a latecomer...

I have extensive experience in simultaneous recording a) to DVX/External mics and ProTools, b) to DVX/Line in and ProTools (from different preamps) and c) to DVX/Line in and ProTools (from the same preamps). This is primarily classical music, so demands may be higher than yours.

By far, the greatest distinction comes from bypassing the preamps of the DVX. They are pretty hissy in comparison to even low-end (but good) dedicated preamps. This is indeed noticeable!

Converter differences are comparatively minor, and for dialog likely completely irrelevant.

As for sample size, 16 vs. 24-bit - in most location settings, it really isn't likely to make a difference. If you're on a soundstage or in a studio, or micing things so close that ambient noise really drops away, yes; but for typical "DVX" situations, not a big factor. (Higher SAMPLE RATE - say 96K - has a greater effect, reducing high-end brittleness).

So, for most effective, inexpensive, and convenient sound improvement, go for the preamps...

-Barry
 
ullanta said:
Greetings, all, from a latecomer...

I have extensive experience in simultaneous recording a) to DVX/External mics and ProTools, b) to DVX/Line in and ProTools (from different preamps) and c) to DVX/Line in and ProTools (from the same preamps). This is primarily classical music, so demands may be higher than yours.

By far, the greatest distinction comes from bypassing the preamps of the DVX. They are pretty hissy in comparison to even low-end (but good) dedicated preamps.

Converter differences are comparatively minor, and for dialog likely completely irrelevant.

As for sample size, 16 vs. 24-bit - in most location settings, it really isn't likely to make a difference. If you're on a soundstage or in a studio, or micing things so close that ambient noise really drops away, yes; but for typical "DVX" situations, not a big factor. (Higher SAMPLE RATE - say 96K - has a greater effect, reducing high-end brittleness).

So, for most effective, inexpensive, and convenient sound improvement, go for the preamps...

-Barry
Click to expand...


Your comment pretty much mirrors my experience in general. Good preamps seem to have the most bang for the buck in the recording chain. Hopefully the sample files will be helpful when evaluating how to best improve recordings.
 
Yes, samples are always useful, and it's generous of you to be putting them together!

I also wanted to add: The other weak point of the DVX audio system is monitoring! Anything you can do to improve your on-set monitoring will also likely lead to one of the biggest possible improvements of your audio. So, if you can monitor from your preamps, all the better. Even a little Boostaroo or something to crank up the headphone output can be useful... as can good passive-isolator headphones like the HN-7506 or Extreme Isolation units.

A small mixer (like, say, a Yamaha MG10/2) can provide improved preamps and monitoring, as well as the ability to hook up and mix a number of mics. Frankly, I don't have much experience with the video-film oriented units like SDs... but they tend to be priced at a premium (though they offer the benefit over audio-orineted units of battery operation, which is not a factor for me in general, but may be for you).
 
I do not use the sound of the DVX as a sound track because the internal mic records the cam noise. (It's worst when you play with the zoom control)l. Normally, I record the sound with a DAT and use the sound of the DVX as a sound reference (for the slate).

The sound recorded with the DVX is 16 bits (or 12 bits depend) and with the DAT, I can record at 24 bits, so the sound quality is better.

But, if you can't rent a DAT, the best if you rent a mixer (personnally, I use a Shure FP33) and you can control the sound.

ElvisNTSC
 
ElvisNTSC said:
I do not use the sound of the DVX as a sound track because the internal mic records the cam noise. (It's worst when you play with the zoom control)l. Normally, I record the sound with a DAT and use the sound of the DVX as a sound reference (for the slate).

The sound recorded with the DVX is 16 bits (or 12 bits depend) and with the DAT, I can record at 24 bits, so the sound quality is better.

But, if you can't rent a DAT, the best if you rent a mixer (personnally, I use a Shure FP33) and you can control the sound.

ElvisNTSC
Click to expand...

There is only one DAT machine that I know of that records 24 bit, and it's an expensive studio machine. There are no field DAT machines that record 24 bit, they are all 16 bit. Maximum resolution on a DAT machine is 48Khz, 16 bit. There is a noticable difference between 16 bit and 24 bit, namely dynamic range, and less crappy sounding high end, like silverware jingling together, and that sort of thing.

-Dave
 
Dave_Fisk said:
There is only one DAT machine that I know of that records 24 bit, and it's an expensive studio machine. There are no field DAT machines that record 24 bit, they are all 16 bit. Maximum resolution on a DAT machine is 48Khz, 16 bit. There is a noticable difference between 16 bit and 24 bit, namely dynamic range, and less crappy sounding high end, like silverware jingling together, and that sort of thing.

-Dave
Click to expand...

Well, depends on your definition of "field DAT"... at least a few small rackmount systems will record 24-bit.

As I said, I really don't think the 24 vs. 16 bit difference willl be significant in realistic/low budget situations; the extension of dynamic range is downwards, and will likely be well below the noise floor. As for the high end, that's much more a product of sample rate than sample size.
 
tigeba said:
All,

I can throw together a little shootout in my studio and post the results for all to hear. I would like your input on what combinations of things might be of interest. I was planning to test this anyway. I will use a Neumann KM84i as the source mic, since it is fairly close to a mic that would be used to capture dialog. Here were my thoughts on a few test setups.

1) KM84 -> mic pre on the DVX100a This would be the baseline setup

2) KM84 -> Neve 1272 Preamp -> DVX100a line in. This would emulate high quality external preamplification (like a field mixer) while using the recorder in the DVX.

3) KM84 -> Neve 1272 Preamp -> Digidesign HD192. Most people would probably consider this a high end external recording scenario.

I would note that it is fairly tricky to pull off these types of comparisons fairly as even small differences in output level can skew the perceived "quality" of the results. My prediction based on gut feeling is that the preamps may result in a noticable difference for some people, and that the differences in the recorder alone (DVX100a recording vs Protools HD recording) will not be that obvious. To keep things easy I will record the audio into Protools at 48/16. It is my opinion that 24 bit source for dialog will really not provide any distinguishable difference for this test.
Click to expand...

Thank you SO much for stepping up to the plate for us! If all the final files are normalized then there will be no difference in level and the test is fair. It's easy to do... and I wouldn't sweat the tiny difference that may be left over.

Again, if 1/20th of a turn on the volume knob is enough to fool people into thinking the cheap route is better then the high-end route then just kick up your levels in the cam 10% and save yourself $4K... if people can't sort the quality aside from a little difference in levels then kick that high-end gear to the curb. :grin:

Seriously... if you do that test you mentioned I'd be very grateful... and if you need a place to host the files I'll do it. :thumbsup:
 
Well for all practical purposes this thread is dead... at the hands of tigeba.

As my brother says when he sees a june-bug spinning around on his back outside the back door... "sooner or later you gotta' turn off the porch light or go out and step on 'em".

Thanks for turning out the porchlight... (or stepping on me). :cheesy: Really... thanks!
 
You must log in or register to reply here.
Back
Top