Sony Head phones for Sony PMW EX1

setting gain at the camera doesn't introduce noise where a mixer won't. unless you've set the gain incorrectly.

line won't be a mic level turned down. doesn't matter what camera.

having a mixer is a great idea if there's someone to operate it. personally, if i don't have an experienced sound person, i'd rather monitor and ride levels myself at the camera. but i earned my bread as a sound man for a number of years.
 
MPB said:
setting gain at the camera doesn't introduce noise where a mixer won't. unless you've set the gain incorrectly.

line won't be a mic level turned down. doesn't matter what camera.

having a mixer is a great idea if there's someone to operate it. personally, if i don't have an experienced sound person, i'd rather monitor and ride levels myself at the camera. but i earned my bread as a sound man for a number of years.
Click to expand...

You're talking about "setting" and I was referring to "riding" levels.
Setting levels at the camera won't introduce noise in and of itself, unless you turn up the EX1 gain past 6. You haven't heard the noise? There is a noise floor on the EX1 and riding the level dials (which is what I said) on the EX1 will give you a changing noise floor. Not cool for post work. Using a MixPre's preamps and going line-in is much cleaner, especially if indeed the EX1's Line-in is a separate circuit. The MixPre is has a lot of gain (60db) and is pretty dang quiet.

So in a level riding situation, the MixPre is the better choice, primarily for audio quality, but also, have you noticed the lag on the EX1 when you change the audio level? It's a dial that digitally tells the computer to turn up the level. It's not quick enough for fast changes, and the ex1 dials (being flat and only partially exposed for thumb adjustments) are not even designed to be constantly changed. The mixPre has proper dials you can put 2 fingers and a thumb on, and every move is instant because it's analog.

And believe it or not, many "Line Level" devices are indeed only the "Mic Level" with a pad engaged. That was the case with my Sony VX2100 (a popular pro-sumer camera) until I had the entire mic preamp bypassed with the Greg Winter mod. I've mixed plenty of live shows and albums, and I know a good dial. The EX1 dials are not ready for prime time, HOWEVER the EX1 IS a very clean audio capturing device that indeed one can go straight in to and get a great recording. I do it that way too, but I'm not fooling myself that the MixPre to EX1 isn't the higher quality / user friendly way to maximize your signal. MixPre / Line-in you are still only using ONE preamp per channel. The only difference is more control and better limiting and low-cut options.
 
oh, jeez. ;-)

hmm. any increase in gain has the potential to raise the noise floor. maybe you're describing a problem with the ex1 that isn't present in the ex3. i certainly haven't noticed unusual noise levels on the ex3's i've used or owned. anyway, if you're levels aren't set properly at the camera and you pound a bunch of gain into it from the mixer, you will obviously be raising the noise floor.

fast changes? he's not shooting verite. and my impression was that there isn't a sound person. so, if the choice is giving an inexperienced boom op a mixer to operate as he/she hold a boom, and has to make "fast changes," or to monitor levels and make adjustments at the camera, personally it seems a lot safer for me to handle at the camera. but that's just me. as far as getting higher quality sound, that requires a sound person. and you're right, they usually come with a mixer.

and you might pad line down to mic, but not the other way around.
 
I was talking about Mic IN vs Line IN. On many cameras there is a Mic in / Line in switch. And the line in is only the Mic-in padded down 20db. The MixPre only sends out a Line level signal, which is much hotter than a mic level signal. Recording a line level signal gives you less noise because either you skip the mic preamp on the camera all together, or at least your camera's mic preamp level is 20db lower resulting in a higher signal to noise ratio.

I get what you are saying if one is going from a mic into a mixer's preamp, then going from that to a camera/recorder's MIC preamp. That's 2 preamps to potentially add noise 2x. But if indeed the EX1's Line input is a true line level input, separate from it's mic level input, I would wager that the preamps on the MixPre are quieter than the EX1's.

I wouldn't say ANY increase in gain will add noise. Potentially, ok. But not if the gain is coming from a clean source. Not all preamps are created equally. And I bet if you turn the EX3 levels up too high you will hear hiss. I suspect that they are the same preamps between the EX1 and EX3. I'm just putting forth the statement that the MixPre has better preamps than the EX1/3. That's all. Of course you can do it without a mixer. I'm just saying you can do better if you are so inclined. Not trying to start a pissing contest here. I'm just speaking from my experience recording both ways.
 
ok, i gotchya. pre-amps are good, in the hands of trained sound people.

but you have your mic and line backwards. :-) you can pad line down to mic. if you put a pad inline with a mic signal, what you get at the other end will not be line.
 
My last post said, "the line in is only the Mic-in padded down 20db".

I never said to pad down a mic signal. I said that some mic preamps in cameras and recorders are padded down (at the factory with a switch) to receive line level signals. For a "Mic" input to receive a ""line" level signal there must be an attenuation.
 
I think the majority would agree with you chad...

But, I go for simplicity myself here as well... Maybe cuz I'm a low budget novice... I put the mic right into the camera set the levels one line to 6 one to 5, if what I hear is low at some point in the take, I would yell "retake, talk louder please...", or "come on, work with me boom operator..."
 
Yes, for a lot of stuff, I do a mic straight into ch1, set to 6, and (this is important) set the trim to get the level in the ballpark. You can turn up the trim and not add the noise that comes from the post-preamp levels. It's when I want that extra level of quality/control that I throw in the mixer. I start with -41db Trim, and if that is too loud I go down to -35db or -29db. It all depends on what the source is or how close to the mic it is. I also have both channels recording, with one a tad lower just in case I clip, the lower channel will be safe. But I can't remember ever clipping on my EX1 yet because it has it's own limiter that does a nice job.
 
Chad's concept is from the older television era, when s/n ratio on camera audio was so bad that microphone self-noise was insignificant by comparision. However, today that model is reversed. The mic is the noisiest link in the chain (even the Neumann U87a has a s/n of only 82dB) and the concept of 'gain staging' a recording chain today means setting the gain of the recording system such that the mic self-noise is just below the mic preamp / a/d converter self-noise.
Gain riding is for 1970s radio disc jockies and for listeners who enjoy hearing the LP turntable rumble brought up to 100% modulation during the lead in grooves.

My new school of thought is to record an ACCURATE snapshot of the sound and, starting with an accurate master, add distortion later, by using compressors, if I know my destination is going to be a 17" plastic Daewoo television. I recently used this technique at a rally and it worked great, because I was able to pull up the soft-spoken persons at 30', yet have the crowd cheers not clipped. In the master, the speech was at -56dB for the sofest talker, while the cheers and clapping approached -1dB on the camera footage. By the time I was done in Premiere, I had reduced the dynamic range to about 12 dB and it sounded very clean and not a trace of audible hiss.

The EX1/3 is not a PD150, folks. You can do things with the audio that you wouldn't dream of with analog linear track on VHS, or on a PD150 with it's 40dB s/n preamps.

Gain riding and AGC is for sound like auto iris and auto focus is for video. No serious camera operator would use them. And no serious audio engineer would use compression on the master recording. Safe the destructive distortion for post production. 'Tis better to work from pristine master recordings.

Technical note: today's cameras use the same signal chain for Mic and Line levels--they simply switch in attenuation by CMOS switching circuits (electronic, not mechanical switches). Programmable op-amps are the present technology, and are adjustable by digital control for a wide variety of gain levels.
 
basspig said:
Gain riding and AGC is for sound like auto iris and auto focus is for video. No serious camera operator would use them. And no serious audio engineer would use compression on the master recording. Safe the destructive distortion for post production. 'Tis better to work from pristine master recordings.
Click to expand...

Well we've gotten way off of headphones here, but it's a good discussion none the less. Here's my take...

Signal to noise ratio is still alive and well, even in the modern digital age. And I don't think it's accurate to categorically say that "the mic" is the nosiest link in the chain. Not all mics are the same, as not all mic preamps are the same.

I wouldn't put "gain riding" in the same boat as AGC Mark. And "gain riding" only means that the mixer op is making conscious decisions to bring up the level of something too quiet or bring down something too loud before it gets recorded. AGC is automatic and brings up noise in quiet passages. If what you were saying is true about mixers not being needed today then large budget productions would not have a person working as a mixer. The proper way is to have a boom op, then a mixer op, going into a recorder of some sort. The only reason people go straight into the camera is because they can't afford the budget to do it the proper way.

If you turn up someone in post from a recording made at a tea bagger rally, you're not going to hear noise the same as on a quiet set. In a rally you have all sorts of random noise from the world covering up the hiss you would get if you turned up, say, a woman whispering on a closed set with the on-set mixer level too low, or no mixer was used, so you have a mic level going into a 600.00 recorder, or a camera. Preamps can be a weak link in the noise chain if they are cheap. A good mixer will bring the signal level up more cleanly than many recorders used by indi, and even more pro film / video makers. I've done post for a few films, and countless TV spots and ads, and I deal with this stuff all the time. Not just theoretically, or as a hobby. Hundreds of scenes where my job would be easier if the (mostly nonexistent) mixer op would have just added some clean gain at the mixer stage, before it got to the recording device. I am working on a film now and dealing with all of this stuff. Thank god for iZotope RX!

Don't believe me. Ask Francis Ford Coppola if he uses a mixer, even with only one mic. Or more realistically ask someone like Ty Ford, or one of the talented guys on DVXUser that do location sound for a living. Or go check out Taper op and try to say that a H4n's preamps are as good as a Sound Device's mixer.
 
Chadfish said:
Don't believe me. Ask Francis Ford Coppola if he uses a mixer, even with only one mic. Or more realistically ask someone like Ty Ford, or one of the talented guys on DVXUser that do location sound for a living. Or go check out Taper op and try to say that a H4n's preamps are as good as a Sound Device's mixer.
Click to expand...
That's a bit rediculous to compare a Hollywood production to the low budget EX-1 projects that we're all doing, I don't think anyone here is arguing on that level. We might use the same mics, but they aren't using SD302 mixers. Most often they aren't even going into the camera and are using a separate recorder.

I think the argument is more along the lines of, can you get the same quality sound using a bare EX-1 with the mics going straight into it,
vs,
using a mixer WITHOUT a separate recorder and THEN going into the EX-1.

My opinion is that a mixer such as the SD302 isn't going to give better sound than going straight into the EX-1. Can you ride levels with the mixer? Sure you can, but as long as you're getting good levels during your dialogue recording, and you can hear what you are getting on the fly so you know if the quality is good or not, then you're fine.

If there's still any argument, record some dialogue using both methods and post it. I bet we can't tell them apart, and I'm a betting man.
 
Last edited:
precisely.

and Lucere implied there was no sound person, which is an important part of getting good sound. handing someone a mixer isn't going to make up for that. in fact, if they don't know how to use it, it could seriously complicate things. it might not, but i'd take the safer, simpler bet, personally.
 
Since I've yet to hear really GREAT audio on any Hollywood production to this day, I'm going to remain skeptical that having a mixer and a union 'sound' guy feather bedding the film shoot is really the best way to go.

While someone gain riding the levels during shoot can make a lazy editor's work easier, modern equipment has such a low noise floor that once you match the gain level of the preamp to the mic's self-noise, there is no further s/n to be gained by increasing the gain in the mic circuit. At that point, you'd just be losing headroom at the loud end, in exchange for a little more loudness at the soft end.

I don't intend to sound arrogant, but 30 years before I started shooting on video, I was making sound recordings of orchestras, nature, wildlife, weather events and voice over work. I designed some of the world's finest audio sound systems over that span of time, and I've a BSEE in electronics engineering since the '70s. I've designed, wrote specifications, tested and verified performance of amplifiers, pre-amps, compressors, limiters, stereo multiplex hardware for broadcast use and the list goes on. What I'm saying, between the lines, is that I have an unusually-intimate relationship and understanding of the electro-acoustic issues of sound recording and reproduction. Most folks in broadcasting only have a basic grasp of what they learned in college from one narrow perspective. They can't 'see' electrons flowing in the circuits they use, because they've probably never designed a mic preamp in their life, much less know how one works and what causes noise, what the effects of high open loop gain vs. feedback-free circuitry has on inter-aural delay that is so important to that sense of 3D soundfield, etc.

Sound recording is really quite simple, if you understand the circuitry in the systems you're using, and know how to 'gain stage' all the blocks in the chain so that you utilize the maximum dynamic range. There is only one optimum gain setting for something like this. The rest degrades dynamic range. The ONLY valid reason why you might gain ride is that if you're editing on a 16-bit editor, you would slightly minimize digital quantizing errors. Having more digital steps to work with by having more gain, can be an advantage, although it's debatable whether anyone can here the difference, with modern dithering algorithms and 64-bit float editors.

I strongly feel that the correct way to set the levels is to anticipate the loudest possible SPL and set the sensitivity/attenuation so that a sound of that level peaks near -1dBfs on a peak responding meter.

I have been recording chamber ensembles a lot lately, and have found that -35dB attenuator setting on the EX3 with condenser mics having -37dB sensitivity works out very well, providing levels that don't need to be normalized in post, yet clipping nothing. Gain riding a concert would be a very bad thing. The only situation where gain riding would be appropriate may be something like a press conference during live broadcast. I wouldn't do it in any other situation.


Sure, I can get better dynamic range using my studio condensers into a 24-bit MotU 896, and the measured noise floor on that is down around -136dB over the audible range, so I can make a recording outdoors in the woods in the still of night, normalize the recording to -12dB and hear the pollen falling onto the leaves in the forest in early June. But in a dialog for film situation, such a low noise floor is extreme overkill. And for the chamber ensembles, darned if I can tell the difference between the EX1 audio with my studio condenser mics, and the MotU 896, because the lowest ambient noise dB is in the low 40s. The EX1 is super-adequate for any conventional recording situation involving dialogue, music and ambient sounds in an urban environment. As long as you're not trying to isolate a mouse fart at 200 yards, nothing special, hardware-wise, is needed.

A lot of this business of needing a separate sound guy is vestigial with today's technology. It's a leftover from the old days when equipment was so fragile that you had a 20dB window to work in between distortion and noise, and in the heavily-unionized world of film and television, any sound guy has a vested self-interest to convince the bean counters of his need to continue to exist. I know that sounds blunt, but there's no other way I can put it.
 
most of my work isn't union. and i've done plenty of one-man-band. plenty. but if i'm shooting verite, or some other event that i have no control over, and there's no sound person, i really don't want to hear a producer or director complain about rf hits or clothing noise. i'll certainly do what time and circumstances allow, but if the camera mic isn't good enough to cover problems, you should have hired a sound person.

and, my camera mic is a cmit5u, if that gives you any indication of the importance i place on sound.
 
Last edited:
This is a valid point. When shooting a movie, a good boom pole operator and someone to take responsibility for monitoring audio to make sure that things like clothing noise, wind noise, etc. is not a problem is still a much desirable component of crew.

On the subject of wireless mics, I'm exploring more ways to employ these in possible filmmaking scenarios. Concealment without clothing noise is the challenge. But the sound pickup from actor-worn wireless mics can be quite good.
 
chasing clothing noise on hidden lavs is one of the things that ensured i was going to keep going after the camera work and leave the audio gigs behind. ;-) i hated that part. that, and the jokes about feeling up the talent/subject. but i did develop a great appreciation for what a good location sound recordist does.
 
One thing that I think is missing from the discussion is the 'sonic quality' of the pre amp. Why do commercial recording studios have preamps that are thousands of dollars each? Its definitely not because of the low noise floor (trust me, I have worked with some delicious vintage tube preamps that are down right noisy).

I think we can all agree that when a manufacturer creates a $5k or $10k, or $100k camera; there is so much focus on optics, codecs, sensors and other 'important' and 'sell-able' features that there can be a tendency to skimp on 'lesser important' and 'lesser marketable' features. The majority are not going to buy a $5k-$10k camera on its world class preamps alone, if it also has a very poor lens, sensor and codec. Beyond pre-amps, we could also have a big and rather unfruitful discussion on the quality of the A/D converters (but I digress).

I know we are talking about 'weak links' in the signal chain, however, there is something to be said about using a high quality preamp on a high quality mic and feeding a recording device a line-level signal. Even if its just to add a little pre-emphasis in the 12khz range to pull out later, smoothing the hiss of a lower grade line-level preamp.

BTW - I favor AKG headphones, better isolation than the 7506s IMHO. 171s and 271s are great.


Jason
 
Last edited:
Wow, thanks so far everybody for the amazing wealth of information. Definitely a lot to contemplate here. The shoots a very low-budge directly to DVD horror and mostly takes place indoors with a pretty small cast. I'll be on camera and there will be a boom and lighting person, but that's about all our tiny budget allows.

Please excuse my lack of audio experience. Is general consensus nix the mixpre for this particular shoot and be confident that a bare EX1 with a good shotgun mic will acquire pretty good audio, and as basspig said- anticipate the loudest possible peaks of dialogue and set the camera's audio dials to a certain point below that? Or will it be important to put the mixpre between the shotgun mic and camera? Thanks again!
 
oh, you wanted a practical solution? :D

we did get a little distracted, but i think there was general consensus that in this type of scenario there probably won't be a tremendous or discernible difference in audio quality merely because of the decision to feed or not feed the camera via a mixer.

given the limits of money and crew that you've described, assuming that when you refer to the use of the shotgun, that that will be on a pole used by your boom op and that you will be the person monitoring sound at the camera... yes, you should follow the method outlined by basspig and others where you set safe levels based on the peaks likely for that shot or take.

as suggested, when you have only one source (mic) going into the camera, assign that source to both camera audio channels, and stagger the input levels to provide yourself a safety. for instance, having set channel 1 at what you feel to be your optimal level, set channel 2 slightly lower. that way if channel 1 gets too hot, channel 2 has a chance to still be in range.

bear in mind that you should be setting levels appropriate for each take. if there is a particularly quiet bit of dialogue right before someone shouts, you're going to have to try to address that. don't set the levels at the beginning of the day and forget about it. as you're doing run-throughs or rehearsals you're going to have to ask your actors to give you an idea of what to expect audio-wise so you can set your levels before the first take. having said that, your first take is likely to be when you really find out what they're going to do. actors, understandably, save their real energy til the camera rolls. if they surprise you, or if you're not sure if a take worked for audio, play it back before moving on, just as you would for camera.

i prefer not to have to be responsible for audio when i'm shooting, but it frequently doesn't work out that way. don't rush yourself. on set, at hour 14 for the day, no one wants to re-take because there may have been iffy audio. too bad. unusable audio will make everyone miserable. and pissed. at you. so get into a good routine from the start and follow it.

good luck!
 
Last edited:
You must log in or register to reply here.
Back
Top