There have been a couple of threads on the low output level of the supplied shotgun mic (being low indeed - nobody questioned that), as well as the overall quality of the V1's audio section (disputable - some said it's extremely poor with the rolloff frequency being as high as some 1200 Hz, others strongly denying).

Well, I guess I can shed some more light on this, with a first-hand experience I've just had. Some of you may remember my posts on how well the live music performance in a church came out for me recently, using the Edirol CS-50 stereo mic:

http://www.dvinfo.net/conf/showpost.php?p=709631&postcount=1

In fact, I was pleasantly surprised with the sound quality - until, that is, I received the sound track recorded independently by a radio station crew. Well, the two tracks' main differences are:

- the camera recording is much louder (which denies accusations on the whole audio section, not just the mic, being too low)

- the radio recording is by far reacher in lower frequencies (which confirms the claims of the V1 not being very good at producing flat freq response).

I was able to equalize the V1 audio track in Vegas to sound almost identical to the radio professional recording, so it's not that bad, after all. However, it can only work provided the audio is free from any handling noises, hums etc - otherwise, gaining up lower bands by more than 10 dB over quite a considerable bandwidth would inevitably bring the noises up, rendering the footage useless; the bottom line being that the V1 lacks in lower freq response for sure.

Just take a look at how much I had to equalize!

- the radio recording is by far reacher in lower frequencies (which confirms the claims of the V1 not being very good at producing flat freq response).

Wouldn't this be more of an issue with the microphone? Granted, I don't know the technical intricacies of what goes on when the V1 writes audio data to tape, but I would think it would write what it "hears".

It would be a shame to have a camera like that where a high quality microphone could possibly be useless.

Of course the microphone is an important link in the chain, but I have a lot of references on the internet that this model (Edirol CS-50) is renowned for being capable of 50Hz-20kHz +/-3dB, so it isn't likely to be the weakest link in this regard.

I'd like to stress though that the overall sound quality is good enough to get equalized very easily, and after bass boosting is no worse than the professional audio recording I'm comparing it to. I wonder, how do other camera of this class compare?

Piotr, thanks for sharing this information.

Wouldn't it stand to reason that boosting the low end would also amplify any background noise present in the audio chain? Background noise that would otherwise be at the normal level were the bass roll-off not present? If I'm reading the EQ graph correctly, you had to boost the low end audio by as much as 14 dB.

Yeah, that's exactly what I stated in the original post. But then again, if there really is an unusually high bass roll-off that weakens the bass content of music being recorded, for instance - it'll also weaken the handling noise, won't it...Who knows, perhaps this is the purpose of the roll-off? Cause otherwise, I can't imagine why couldn't the V1 offer a flat freq response, when even the chaepest audio recording devices can?

Well, yes, a low-end rolloff filter would reduce handling noise. What I meant by background noise, however, was any noise inherent in the camera's audio circuitry and processing.

Really, though, I don't understand such a big rolloff. Like you said, why can't it have a flat response as is the case with even cheap audio recorders? Is such a rolloff natural with any current audio DSPs? Or is this something that would have to be added purposefully? If so, why wouldn't it be switchable by the user on such a pro broadcast camera?

The inclusion of pro audio features such as XLR inputs would tend to suggest that the V1U should handle audio decently. As such, the inclusion of an unswitchable high-pass filter doesn't make sense to me.

Were all of microphones placed next to each other? My first guess would be that the difference is explained by each microphone's placement within the room (e.g., relative to walls etc.) and not by the microphone's or V1's frequency response. Even a small change in position can cause a large change in low frequency response in many rooms.

Piotr, so that we could eliminate as many variables as possible, would it be possible for you to run a test? My suggestion would be to feed an audio test signal into the camera, first with the XLR inputs at line-level, then mic-level. I would suggest recording with the camera in DVCAM/DV mode so that the audio is recorded in uncompressed PCM format. Then you could capture the video and analyze the audio frequency response.

Were all of microphones placed next to each other? My first guess would be that the difference is explained by each microphone's placement within the room (e.g., relative to walls etc.) and not by the microphone's or V1's frequency response. Even a small change in position can cause a large change in low frequency response in many rooms.

Yes, the two radio mics were placed 2 meters away from where my tripod was located. But let me express my reservations for one more time: I've no idea if and how the radio recording was processed before I got it on a CD; it's quite possible it was also equalized in one way or another! One thing is certain: the frist time I heard it, I knew it was better in the bass department than my own. However, this is by no means any proof that the V1 has indeed a roll-off at 1200 Hz by -14dB!

However, this is by no means any proof that the V1 has indeed a roll-off at 1200 Hz by -14dB!
Whatever the specifics of what may be happening in the V1U's audio system, you did say in your first post that "the radio recording is by far reacher in lower frequencies (which confirms the claims of the V1 not being very good at producing flat freq response)."

Would you be willing to try the test I suggested to try to discover exactly what the V1U's audio frequency response is?

it's already been done. More than once. Using both test tone generators and noise.
The FUD post that is here on DVInfo.net contains some of that discussion and result.

Christopher, if you're reading and accepting of that information, then the only test result that will likely satisfy you is one you take yourself.

The V1u, given a flat tone, at a standard level, will reproduce that tone. Given the variables of a microphone, room tone, and whatever other variables may be introduced isn't a test, it's a guess.
Putting in an oscillator sweep produces fairly accurate reproduction up to about 16k and down to approx 80Hz. This is typical of compressed formats.

Douglas, thanks for the response. Could you point me to the discussion you are referencing? I must have missed it somehow when I did a search.

The FUD post that is here on DVInfo.net contains some of that discussion and result.

Could you point me to the discussion you are referencing?

Indeed, if there is a FUD post here on DV Info Net, then it needs to be withdrawn from public view. We're working hard to make this site a FUD-free zone, so if that sort of thing is happening here, it should be reported so it can be excised. Thanks in advance,

My apologies to all, the FUD post had already been removed from public view (and forgotten).
The thread had been closed by one moderator and moved by another.
The edited version of the original thread may be found here (http://www.dvinfo.net/conf/showthread.php?t=98736)

Whatever the specifics of what may be happening in the V1U's audio system, you did say in your first post that "the radio recording is by far reacher in lower frequencies (which confirms the claims of the V1 not being very good at producing flat freq response)."

Would you be willing to try the test I suggested to try to discover exactly what the V1U's audio frequency response is?

Christopher,

Sorry I didn't respond earlier. No, sorry - I have no testing facility or time for it; it's my eyes and ears I always trust. The picture from my V1E is gorgous, and the audio can be equalized to sound exactly like a pro recording. In fact, this has been the point of my posting: be aware that - should you expect to get some hi-fi music on tape - you're gonna have to play with it in post. The good thing is it can be done, and I'm happy with it. BTW, should I trust those many who say the V1 audio is also too low, I'd probably have turned the gain manually all the way up, not leaving enough room for equalization. I didn't (it was manual, but merely at 6-7) - and the recording is very loud, no distorsion, clipping etc. And possible to equalize, so what's the big deal?

With regards to your suggestion on any low-frequency inherent noise: no, there isn't any. Mind you, the music I recorded was quiet, chamber music (just violin and guitar), and yet - after boosting the basses up with as much as 14 dB - no hum or anything similar is audible. Yes, there was a very low frequency noise in my audio (the mic was on the working camera, and the DR60 drive was attached to it, too), only audible when using a high-end home theater system with LFE on and a subwoofer; you can see on my graph I cut it away as well (below some 40 Hz). If the V1 audio was that prohibitive on basses as claimed by the FUD posts mentioned above, nothing of this kind would ever get to tape!

just returned from my wife's book launch - held in a museum in sydney uni.

a. beautifully lit cases and exhibits, hardly any other lighting. well, the cases were brilliant, people too dark, split the difference, ended up at 12db gain, a little noisy in the shadows, but pretty good otherwise. i have yet to find a client (not another pro) who will complain about 'noise' in a low lit shot. all due thanks must go (presumably) to all those reality tv shows shot with mini cams in dark cars / rooms that 'educated' their audience to the fact that night-time, low light situations usually end up grainy. first comments on this project are - what noise?

b. audio - just as piotr points out - the audio is adequate. i shot with a me80, manual and turned up to about 7. no added gain. got home, eq'd and fiddled a bit (not much), and i have perfectly acceptable speeches, background chatter, glasses breaking (not the cases thank goodness!), etc.,

i still hate the linking of ch 1 and 2 together - i would have preferred to have run ch 1 at auto, and 2 at my setting. i feelsafer that way....

but other than that - all this talk of low-light 'horrors', and useless audio, seems to be from people shooting black cats in dark rooms, and recording audio either next to a monitor stack, or from across the road from their subject.

leslie

My apologies to all, the FUD post had already been removed from public view (and forgotten).
The thread had been closed by one moderator and moved by another.
The edited version of the original thread may be found here (http://www.dvinfo.net/conf/showthread.php?t=98736)
Hmm, unless I’m missing something that thread doesn’t have any test results. Also, it isn’t about the low end frequency response of the V1.

Christopher, if you're reading and accepting of that information, then the only test result that will likely satisfy you is one you take yourself.
If I owned a V1 I would be happy to test it and post the results.

Actually, the reason I’m so interested in this subject is because I’m considering purchasing a V1. It appears that it would be a good replacement/upgrade for my VX2000 in many respects. However, I like to research and consider big purchases and not take a blind leap. I’d really like to know what I’m getting before I drop 4k+ on a camera.

the V1u, given a flat tone, at a standard level, will reproduce that tone. Given the variables of a microphone, room tone, and whatever other variables may be introduced isn't a test, it's a guess.
Putting in an oscillator sweep produces fairly accurate reproduction up to about 16k and down to approx 80Hz. This is typical of compressed formats.
The RightMark test results I’ve seen of the V1's audio response were from a test conducted with the camera in DV mode. Thus, the audio was recorded as uncompressed PCM and there was no chance of MPEG audio compression altering the recorded signal. The signal was fed directly into the camera through the XLR jacks in line-in mode - no microphone was used. As such, microphone characteristics and room tone could not have colored the results.

The RightMark Frequency response measured under these conditions shows the V1's audio response beginning a steep dive around 600 Hz. It’s in a shallow decline before then, but 600 Hz is where the decline really begins to enter a steep curve.

I looked at some test results on RightMark’s site for audio equipment, and even an old multimedia (i.e., non-pro level) Creative SoundBlaster Live 5.1 sound card has a frequency response that is strikingly flatter than the results for the V1.

If you’ve posted test results that refute these, then my apologies as I missed them.

Indeed, if there is a FUD post here on DV Info Net, then it needs to be withdrawn from public view. We're working hard to make this site a FUD-free zone, so if that sort of thing is happening here, it should be reported so it can be excised. Thanks in advance,
My apologies if it sounded as if I was interested in FUD. The reason I asked for the link to the thread Douglas Spotted Eagle mentioned was that he seemed to be saying it contained test results showing the true frequency response of the V1.

I want to make it clear: I’m not trying to be hard headed. I’m not interested in FUD. I’ve seen the test results and hard numbers on one side. If there are test results and hard numbers that contradict these then I would really appreciate seeing them.

Sorry I didn't respond earlier. No, sorry - I have no testing facility or time for it; it's my eyes and ears I always trust.
No problem, I understand. Thanks!

Christopher, I understand perfectly your being anxious about an investment in a new camera that might not fulfill your expectations, but believe me - I'm the same; I spent a couple of months trying to make up my mind on which way to go - the Canon A1 (which I tested for 6 weeks) or the V1 (which I had for 2 weeks before the Canon, and returned because of the early 25p problems). Considering all pros and cons, I have chosen the V1 after all. As you can see in my sig, I also had a CCD-VX3 Hi8 camcorder before I entered the digital world. I can assure you the V1's audio is no worse than that of the other two. Yes, with the supplied mic it's a bit low - but still adequate; with a hotter mic it doesn't show any flaws in my opinion. The only thing lacking is the ability of a single mic feeding both channels with different levels throught single input (CH1) - but you can use an Y-cable to feed them separately as a work-around.

I have also seen the measurements showing a dramatic dive of response to the left of a threshold frequency as high as 1200 Hz; have no idea what was the purpose of the person publishing them because - as I said - I trust my own ears, and I could hear it easily should it actually be a V1's 'feature'.

When you analyze the graph I posted, please note I have also boosted the high frequencies - I did it to "lighten" the overall sound atmosphere of a live recording in a rather noisy venue (a church), with plenty of reverbs present. This might be the reason of my first impression that the other recording was reacher in basses, as it was absolutely flat. - when I have time, I might try to remove the high freq boost, adjust the loudness of the 2 recordings and only then compare them in the bass department again. I'll keep you posted.

Can I respectfully suggest a little study of the science of acoustics?

Two matched microphones even a few inches apart recording into identical gear in a very 'live' environment such as a church can produce dramatically different results. The purists working in the field of live acoustic recording spend way more than the price of a V1 just on a microphone and then spend considerable time (and no small amount of experience) getting microphone placement correct. They do nothing to the recorded sound either apart from matrix decoding.

Spending money on a good let alone excellent mic and pluging it into a camera, any camera, is going to give variable results. Microphone performance can be affected by several factors, termination impedance and the quality of the phantom power. That's before the signal hits the camera. From then on it's amplified and fed through A->D converters built from commodity components. That's not dissing the camera manufacturers, they're doing good engineering, they know anyone who wants top shelf audio isn't going to record in camera anyway and the cost burden of decent let alone top shelf audio would more than double the retail price of the camera and you'd still be stuck with 16 bit PCM or mpeg-1 anyway.

For me the only thing missing on cameras audio sections is enough bottom end cutoff, I've more than once wished for a good roll off of 18dB/octave or more. Would have helped when untalented talent start banging on desks, kicking my mic stand or thumping my mic to see if it's on.

To compare what got recorded by a microphone placed 'somewhere', into a camera, to what was recorded by guys with lots of experience and gear who were focussed on only the audio is a pointless exercise. Spending some time learning from the people who made that 'other' recording would be a better way to go.

And if you're really serious about audio, spending some money on decent plugins wouldn't be such a bad move either. The plugs that ship with Vegas aren't entirely useless but they're certainly not stellar either. Of course if it's recorded right in the first place very, very little to nothing is the best plug of all.

Very true, Bob :)

Let me stress though, that in my original post of this thread, my intention was not to compare apples to oranges, and criticize apples for not tasting like oranges do. On the contrary; as a follow-up to my previous post about the generally positive church recording experience, I wanted to confirm the sound recorded by V1 can at all be comparable to professionally recorded audio. Really not much was needed to make them sound identical...

Really, though, I don't understand such a big rolloff. Like you said, why can't it have a flat response as is the case with even cheap audio recorders? Is such a rolloff natural with any current audio DSPs? Or is this something that would have to be added purposefully? If so, why wouldn't it be switchable by the user on such a pro broadcast camera?

The inclusion of pro audio features such as XLR inputs would tend to suggest that the V1U should handle audio decently. As such, the inclusion of an unswitchable high-pass filter doesn't make sense to me.

Well, in the light that Bob has just shed on the subject (particularly his remarks on accidental, low frequency noises spoiling recordings), and taking into consideration the V1 is by design more suited for ENG run&gun type of work rather than recording musical performances - I'd tend to suspect the evident bass roll-off placed relatively high could be there on purpose...

Fair enough,
I'm still a bit interested in this though, not so much from the V1 point of view but why the two recordings ended up so different. Of course you may never know if what they'd done to their recording had any bearing on their sound compared to yours. Just that you seem to be applying a aweful lot of Eq that could have side effects.

What I'm also thinking about is what kind of sound you're after, one that sounds good or one that's accurate.

Edit:

I just went and looked up the specs on the mic that you're using. I'm wondering if that has any bearing on what you're hearing. Certainly not the kind of mic I've had anyone recommend for your application. Typical mics are X-Y mics such as the Rode NT4 or (if you've got a wad of cash) the Soundfield mic.
In this kind of environment what the mic records is I think a bit more complex than you can judge just looking at response curves, they're measured in anechoic test rooms and a church is a vastly different place unless you're close micing.

What I'm also thinking about is what kind of sound you're after, one that sounds good or one that's accurate.

Both:)
Seriously though (well, still not quite), the former takes precedence with me. Imagine guitar and violin in a venue like a church; which would prevail? The violin of course! Now, due to reverbs present, what do you do to quickly clean it up? Boosting trebles, right? OK - that's what I did; the result being even further domination of the violin over guitar.

Now, to complicate things even more (and make them interesting), imagine the guitarrist happens to be your brother-in-law. He says: hey, I canot hear myself loud enough; my instrument got dominated by the violin (and the violinist happens to be one of the most prominent Polish violin player)...

Then we get hold of the radio recording; what does my brother-in-law say? -Hey, this sounds a lot better to me, at least the guitar is audible! So what do you do to please him? Boost the basses! By as much as only possible before distorsion starts (hence 14 dB in the graph posted...).

On the other hand, the radio crew had of course two high quality mono mics, one of them directed lower towards the guitar, the other one - higher to where the violin sound emanated... They could have balanced the two channels very easily (no equalization was done, though, AFAIK).

So, this has been what, and why, I did to the original V1 audio. Just a short sample of the result: http://rapidshare.com/files/48330102/V1_in_church.wav

I just went and looked up the specs on the mic that you're using. I'm wondering if that has any bearing on what you're hearing. Certainly not the kind of mic I've had anyone recommend for your application. Typical mics are X-Y mics such as the Rode NT4 or (if you've got a wad of cash) the Soundfield mic.
In this kind of environment what the mic records is I think a bit more complex than you can judge just looking at response curves, they're measured in anechoic test rooms and a church is a vastly different place unless you're close micing.

Absolutely Bob - the CS-50 mic by Edirol is just a general-purpose stereo shotgun. And yet, giving acceptable results even for music recordings.

Well I've listened and looked. (and sorry Chris this is getting WAY off anything V1).

The recording is mono, or so close it doesn't matter. That's both by my ears and phase meters. That means we've lost the ability to localise the sound, the violin and the guitar are mixed together in the one place. In a stereo recording yes the violin might be a lot louder than the guitar but our ears and brains can separate them. Furthermore, with a stereo recording (or dual mono like the radio guys were doing) you have chance to adjust the balance.

Worse still, well for me at least, that recording has no sense of 'place'. The listener should be able to close their eyes and 'see' the church but all the cues are missing. But, not only is the acoustics missing that gives the sense of place (which is what would happen with close micing) but we've lost the the sound of THE guitar and THE violin, we're kind of left with A guitar and A violin.

Did you by any chance have your mic set to Focus instead of Wide?

A couple of tips:
1) Eq in this situation is a very blunt tool, I've found I can lift the guitar out of the mix a bit using a multiband compressor to get more attack on the guitar. Not a dramatic improvement but worthwhile.
2) In general when using Eq it's better to turn down what you don't want than to turn up what you do want, at least when you want drastic changes. This way you can avoid clipping.
3) Your project prefs might be wrong, the sample is 44.1K, not a big issue but best to avoid resampling, even when playing back on the timeline it's using up CPU cycles.

Thanks Bob for your remarks, I really apreciate it.

Yes , I did use focus instead of wide in purpose - I couldn't get close enough to the 'stage', and wanted to eliminate as much reverb and surrounding noise as possible. Of course this is the reason it sounds almost monoaural, but let me note that complete separation of the two instruments would be wrong, as well. Nevertheless, at the next opportunity I'll try the wide setting.

Or, better still, use my new cables with all possible connectors/adapters to get a proper sound through the Line inputs:)

PS. Bob, can you recommend some reading on audio recording/postprocessing? You know - for technically capable, but not specializing in the field, guys like myself...

I just went and looked up the specs on the mic that you're using. I'm wondering if that has any bearing on what you're hearing. Certainly not the kind of mic I've had anyone recommend for your application. Typical mics are X-Y mics such as the Rode NT4 or (if you've got a wad of cash) the Soundfield mic.
In this kind of environment what the mic records is I think a bit more complex than you can judge just looking at response curves, they're measured in anechoic test rooms and a church is a vastly different place unless you're close micing.

Bob, would the AT835ST be a better choice than the Edirol?

http://www.audio-technica.com/cms/wired_mics/b92dc794916f0fa7/index.html

Some good info here:
http://www.xowave.com/doc/recording/mic-pair.shtml
and perhaps more than you'd want to know here:
http://www9.dw-world.de/rtc/infotheque/stereo/stereo_recording.pdf

The AT mic you're looking at has the same problem as the one you've got, well not so much the mic as what you think it can do for you. Take a look at the polar plot in Narrow. Nothing, it's a mono hypercardiod. No doubt a very good one but that's not the way to get a good stereo image.
I've just finished editing a short brass and organ piece for a client, didn't shoot it myself but the client / conductor /composer tells me they used his AT825 plugged into the camera, sounds pretty good, it sounds like you're in the cathedral. The Rode NT4 is also a good mic for this kind of recording, the AT825 is a little light at the bottom end for big pipe organs, we might get a DPA 'reference' mic that goes down to 10Hz for this as well.

It's a bit of a problem venue as it's in the middle of the Sydney CBD and the organ is half way up a side wall, that's where they also had the brass players. The client told be today he took a lot of trouble walking around the venue listening for the best place for the mic while the musicians warmed up and rehearsed. Now, it was pouring rain when they recorded this. Sure you can hear the rain and the outside traffic noise and the odd clang and bang, no doubt if there was an audience there'd be the odd cough and splutter from them in the mix too. But it's a live recording, the listener will cope with this, they know it's not a studio recording, that's what the video bit is for, to tell them where they are and the sound should enhance that experience. For this kind of thing the sound is the show, the video is just there really so they're not looking at a blank TV.

Now I know you've got an issue with putting a mic where it should go but a bit of lateral thinking might solve your problems. Get a cathedral pole, it's a very tall stand with a long boom. For the event in question it can go behind the alter with a few shot bags on the base. The mic goes in front of and and above the performers. Record into a standalone recorder, leave enough headroom so clipping is impossible and just leave it running while you worry about the video. Or else get a long 5 pin XLR cable and lots of gaffer tape and run a lead back to your camera.

Anyways, hope this gives you some ideas. This isn't a simple thing to master and I'm no expert. One thing I learned long ago though was for acoustic recordings don't think about frequency, think about wavelength. You start to see the problems is a different light.

Thanks again Bob for the invaluable hints. I admit I mainly treated the event as a test for the video side (I knew it would be not very bright inside the church); I didn't bother too much about the audio because I knew in advance it would be recorded by the radio guys, so I'd bee getting a CD from them, anway.

For the next time though, I'll get better prepared.

In the case of guitar violin exposed here, i think only a multitrack recording (with a mic close to each instrument) can do the deal.

I think you should not involve the camera in a problem that is only related to audio recording, especially if you set the level of quality to "professional".

Purchase a multitrack firewire audio mixer that can record several tracks to a laptop and then you can eventually remix as per the taste of the audience.(one DVD for the violin guy and one for the guitar guy, so everybody is happy).

On my own, i use the ALESIS multimix firewire (12 tracks, but the software provided is limited to 8 track ).

This discussion would be the same if you put a small light on the camera and then compare with professional lighted studio and say "the camera is not keeping up in picture quality while compared with shot taken by a professional team".

The funny part being that, after all, my in-camera recorded sound actually is almost keeping up - which might put the radio professional in a not so good light, but anyway :)

Thanks Giroud!

Christopher, I understand perfectly your being anxious about an investment in a new camera that might not fulfill your expectations, but believe me - I'm the same; I spent a couple of months trying to make up my mind on which way to go - the Canon A1 (which I tested for 6 weeks) or the V1 (which I had for 2 weeks before the Canon, and returned because of the early 25p problems). Considering all pros and cons, I have chosen the V1 after all. As you can see in my sig, I also had a CCD-VX3 Hi8 camcorder before I entered the digital world. I can assure you the V1's audio is no worse than that of the other two. Yes, with the supplied mic it's a bit low - but still adequate; with a hotter mic it doesn't show any flaws in my opinion. The only thing lacking is the ability of a single mic feeding both channels with different levels throught single input (CH1) - but you can use an Y-cable to feed them separately as a work-around.

I have also seen the measurements showing a dramatic dive of response to the left of a threshold frequency as high as 1200 Hz; have no idea what was the purpose of the person publishing them because - as I said - I trust my own ears, and I could hear it easily should it actually be a V1's 'feature'.

When you analyze the graph I posted, please note I have also boosted the high frequencies - I did it to "lighten" the overall sound atmosphere of a live recording in a rather noisy venue (a church), with plenty of reverbs present. This might be the reason of my first impression that the other recording was reacher in basses, as it was absolutely flat. - when I have time, I might try to remove the high freq boost, adjust the loudness of the 2 recordings and only then compare them in the bass department again. I'll keep you posted.
Thanks, Piotr. You've obviously put the V1 through it's paces. Good to know it has lived up to your standards.

For me the only thing missing on cameras audio sections is enough bottom end cutoff, I've more than once wished for a good roll off of 18dB/octave or more. Would have helped when untalented talent start banging on desks, kicking my mic stand or thumping my mic to see if it's on.
You could always engage the "Wind" filter.

For me, at least, I would prefer to get the audio as clean as possible. If there's low frequency noise that needs to be removed, I can do that easily enough in post.

If a camera manufacturer wants to add a low end roll-off to a camera's audio, that's fine with me - so long as it can be turned off. Imagine what it would be like if your camera had a non-defeatable "feature" that crushed shadow detail.

Well I've listened and looked. (and sorry Chris this is getting WAY of anything V1).
It's still about audio. ;)

Really, though, it's fine with me. I'm learning from your discussion.

I have also seen the measurements showing a dramatic dive of response to the left of a threshold frequency as high as 1200 Hz; have no idea what was the purpose of the person publishing them because - as I said - I trust my own ears, and I could hear it easily should it actually be a V1's 'feature'.

The person that published that "scientific examination" of the audio on the V1 had purchased a greymarket cam that had a problem. He was tweaked because he had no warranty and thusly developed an issue with Sony. This was one area in which he had some knowledge, and therefore published the "data."

After chasing my tail and testing for a day and a half, I realized what had taken place and felt the fool for having been sucked in. Owning 6 V1's, having had content (and audio) recorded on them broadcast on CNN and other outlets, coupled with roughly 3 dozen training products using audio from the cam in addition to my own tests based on the heavily slanted "tests" published leads me to be as happy with MPEG audio as anyone could be. Bob points out some excellent considerations, although they are not relevant to the original discussion. Bottom line? What do your ears tell you, and what do your eyes see?

The AT mic you're looking at has the same problem as the one you've got, well not so much the mic as what you think it can do for you. Take a look at the polar plot in Narrow. Nothing, it's a mono hypercardiod. No doubt a very good one but that's not the way to get a good stereo image.
Actually, the AT835ST is a stereo shotgun microphone. It has two capsules, one with a line cardioid polar pattern and one with a figure eight polar pattern. Each capsule's polar pattern is shown separately on the AT835ST page (http://www.audio-technica.com/cms/wired_mics/b92dc794916f0fa7/index.html). The line cardioid is "Mid" and the figure eight is "Side."

The mic's stereo image is generated by a built in Mid-Side decoder. The M-S decoder can also be disabled so that each capsule's signal can be recorded separately, one on each stereo track.

Now, obviously, being a stereo shotgun mic, the AT835ST isn't going to give you the same stereo image as a non-shotgun stereo mic like the AT825. Which one you choose should depend on your needs.

If you want to know more about the AT835ST's audio characteristics and how it performs, you can look at the Electronic Musician magazine review (http://emusician.com/mics/emusic_audiotechnica_atst/).

in addition to my own tests based on the heavily slanted "tests" published

Douglas,
Did your tests reveal the true frequency response? Thanks. Just curious.

Bottom line? What do your ears tell you, and what do your eyes see?

Douglas, I'm not sure whether these were just rhetorical questions - because I, for one, have already answered them many times: I am satisfied with my V1E, both the picture and sound. Having said that, I can hear some bass deficit - but certainly not as dramatic as "those measurements" imply.

Douglas,
Did your tests reveal the true frequency response? Thanks. Just curious.

Like Lee, I believe some hard figures from an independent and competent source would only help those who - like Christopher - would like to make as educated decision on buying their new camera as only possible.

Actually, the AT835ST is a stereo shotgun microphone. It has two capsules, one with a line cardioid polar pattern and one with a figure eight polar pattern. Each capsule's polar pattern is shown separately on the AT835ST page (http://www.audio-technica.com/cms/wired_mics/b92dc794916f0fa7/index.html). The line cardioid is "Mid" and the figure eight is "Side."

The mic's stereo image is generated by a built in Mid-Side decoder. The M-S decoder can also be disabled so that each capsule's signal can be recorded separately, one on each stereo track.

Now, obviously, being a stereo shotgun mic, the AT835ST isn't going to give you the same stereo image as a non-shotgun stereo mic like the AT825. Which one you choose should depend on your needs.

If you want to know more about the AT835ST's audio characteristics and how it performs, you can look at the Electronic Musician magazine review (http://emusician.com/mics/emusic_audiotechnica_atst/).

No doubt in Normal the mic does a pretty good job, my comment was directed to what it does in Narrow. As I said (or hope what I said was understood as), what's wrong isn't the mic. It's the idea that sound is like like light. Move back from the subject and you need a longer focal length lens to get the shot framed the same. The same doesn't apply to sound. Certainly we can build very narrow angle pickup microphones but that only works over a decreasingly narrow part of the spectrum plus lots of things happen to the sound as it travels and reflects off walls. Phased array mics do a very good job of side rejection but once you get so far away from the sound source that you need that much side rejection any chance of getting a stereo image is gone.


Kind of interesting thing is we've got a Sanken CMS 10, an even more expensive stereo shotgun than the AT. Some time ago it went out in the field to record much the same as Piotr was recording. Same mistake was made. The mic was placed way back in the venue and switched to Narrow and the user complains "It's not stereo!".

The AT and Sanken mics are made to be used as versatile on camera mics. Outdoors in Wide they'll give you good stereo sound. Indoors switched to Narrow they'll give you very good side and rear rejection, great for speech in a 'live' room or if you're an ENG guy covering a live event good enough pickup for news. At a pinch in the right place in Wide / Stereo they'd no doubt do a pretty good job too, although as the article referenced does say there's better (and probably cheaper) mics that'll handle that role better.

Christopher, thanks for the article link. It helped me realize that - apart from the possiblity of switching the MS matrix off - my Edirol offers the very same capabilities that the AT835ST does, only for half the price...:)

Bob, do you mean that the focus (or narrow) mode simply switches the S capsule off, leaving you basically with a mono directional shotgun? If so, then I made an awful mistake indeed. I was trying to narrow, and not disable, the stereo ambience...

EDIT: Bob, actually I'd aprreciate your answer on the question above, because I'd like to avoid mistakes like this in future. The microphone's manual doesn't say it effectively works as mono when switched to narrow (FOCUS), but - using my ears - I can confirm your observation there is very little, if any, difference between the L and R channels.

No doubt in Normal the mic does a pretty good job, my comment was directed to what it does in Narrow. As I said (or hope what I said was understood as), what's wrong isn't the mic. It's the idea that sound is like like light. Move back from the subject and you need a longer focal length lens to get the shot framed the same. The same doesn't apply to sound. Certainly we can build very narrow angle pickup microphones but that only works over a decreasingly narrow part of the spectrum plus lots of things happen to the sound as it travels and reflects off walls. Phased array mics do a very good job of side rejection but once you get so far away from the sound source that you need that much side rejection any chance of getting a stereo image is gone.
Ah, understood. Very good point.

Christopher, thanks for the article link. It helped me realize that - apart from the possiblity of switching the MS matrix off - my Edirol offers the very same capabilities that the AT835ST does.
You're welcome!