• Wondering which camera, gear, computer, or software to buy? Ask in our Gear Guide.

Do field recorder limiters, reduce quality?

I was told before to use the limiter on my field recorder to prevent clipping. However, now that I have had a lot of practice, I feel ready to record with it switched off, to have the highest quality recording. But does the limiter reduce quality at all? I have tried recording on both, and turned the amplification up high to compared and I cannot hear a difference in quality in post. The noise floor seems about the same, and voices seem to sound the same.

Is the difference very subtle, or more noticeable on theater speakers perhaps? I have the FR-2 LE.
 
It all depends - as usual - upon the recorder/mixer, the limiter (and the rest of the signal chain) and the skills of the operator. A great limiter will be almost unnoticed; a crappy limiter (or the wrong settings on a good limiter) could possibly affect the signal negatively.

If your skills have improved (chuckle, chuckle, chuckle) you can adjust the limiter for a higher threshold; most professional PSMs have the limiter on to take care of sudden, unexpected transient peaks.

As always, every situation is different and requires the PSM/boom-op to make judgements on the fly. If you have to ask the question at all you probably aren't ready yet.
 
Look, H44; once you get into "which is better, A or B?" there are thousands of opinions. I haven't played with the FR-2 versus other units enough to have a real feel for the details.

And does it matter? You're "stuck" with the FR-2, for now. If you eventually get a quality mixer - Sound Devices, for example - then you'll notice the difference in the mic pre's, etc. That's provided you do everything correctly, of course.
 
I have tried recording on both, and turned the amplification up high to compared and I cannot hear a difference in quality in post. The noise floor seems about the same, and voices seem to sound the same.

Huh? That's not possible unless you've got a broken unit. As you turn the gain up, the noise floor (and everything else) should get higher (louder)! As you continue raising the gain and hitting the limiter harder and more frequently the noise floor should continue to get louder while the signal (dialogue) should not. This is obviously bad as at best you are reducing the SNR (Signal to Noise Ratio) and at worst are reducing the SNR, introducing additional distortion from the mic-pre's and maybe from the limiter itself.

The FR-2 LE is a decent unit for the price, I would say it's more at the budget end of the professional range rather than a prosumer range product like the Zoom's or Tascam DR's. I don't know the FR-2 LE well but as Alcove stated, the effect of a limiter can be anything from none at all to quite severe, depending on the quality of the limiter, how it's set and how hard you're hitting it. Most good PSMs in my experience have the limiter engaged most of the time. However, they have the threshold set high, the mic pre's set relatively low, very rarely ever hit the limiter's threshold and therefore have few if any limiter artefacts. In effect, having the limiter disengaged or having the limiter engaged but never hitting the threshold should result in identical recordings.

Is the difference very subtle, or more noticeable on theater speakers perhaps?

More noticeable on theatre speakers compared to what? Theatrical sound systems have a wider dynamic range than good pro music studios or very good home cinema sound systems, very significantly more dynamic range than budget music studio or mediocre home cinema systems and massively more dynamic range than laptops or cheap computer sound systems. Even assuming no mic-pre or limiter distortion, the reduction in dynamic range and SNR caused by exceeding the limiter's threshold will be more noticeable on a system with more dynamic range.

The negative effects of a limiter are almost always preferable to the alternative of clipping (digital overload), so in your case, I would advise you not to record with it disengaged. Just practise not hitting the limiter's threshold!

G
 
Sorry for the confusion. I meant that I turned up the amplification in post, not while recording. In did it in post to compare limiter, vs. not engaging limiter.

However, I too find it to be identical. Why should I have the mic pre's set relatively low? I usually find I have to set them at least two thirds of the way, in order to hit -12 decibels on normal dialogue, depending on the acoustic resonance of the location.
 
I meant that I turned up the amplification in post, not while recording. ... I usually find I have to set them at least two thirds of the way, in order to hit -12 decibels ...

Huh? I'm more confused now than before your "explanation"! If you're peaking at -12dB and your limiter threshold is set at say -3dB then you're never hitting the limiter, not getting any limiting and therefore there should be absolutely no difference at all. And, how does turning up the amplification in post affect the limiter in your field recorder? Let alone affect a limiter in your field recorder which isn't actually limiting anything?!

Why should I have the mic pre's set relatively low?

The lower the gain setting on a mic-pre, the less noise/distortion it adds to the signal.

G
 
I thought that maybe having the limiter on in general may produce more noise, so I just turned up the amplification in post, to compare noise levels. I wasn't comparing clipping. Thanks for letting me know about the lower the gain. I already new that, but one of the actors I am recording in a shoot, has a quiet voice so I have to turn her up high. If I keep it low, like the other actors, you will probably have to turn it up too much in post, thereby raising the noise floor. But I turn her up until she reaches -12 db. On this particular shoot though, the director wants really wide shots, so I literally have to boom with the mic about five feet away though, with her, which is causing me to do that though.

On my recorder, there are two settings. 'Limit', and 'Limit + 12db". Is their one that is better than the other?
 
Last edited:
My logic is, is that I was told that ALC is bad cause the noise is turned up, when it's selected. However, the limiter is an option under 'ALC' on my recorder. So if the limiter is an ALC option, I do not want it to turn up the noise. I was asking out of caution, just in case it could turn up the noise.
 
My logic is, is that I was told that ALC is bad cause the noise is turned up, when it's selected. However, the limiter is an option under 'ALC' on my recorder. So if the limiter is an ALC option, I do not want it to turn up the noise. I was asking out of caution, just in case it could turn up the noise.

That's not logic. The question stands. What's the logic behind you thinking that the limited would "turn up the noise"?

I really want to know your thought process on asking that question.
 
I thought that maybe having the limiter on in general may produce more noise ...
What's your logic behind that deduction?

There is potentially some logic to H44's assumption, unfortunately though it's not the logic he has applied, no great surprise there! :facepalm:

The potential logic depends on how and where Fostex has implemented the limiter. If they've used an analog limiter in the chain, IE. After the mic-pre's but before the analog to digital conversion (ADC), then routing the signal through the additional analog limiter circuitry (even if the limiter isn't actually limiting the signal) will add more noise. Unless the analog limiter circuitry is poorly designed, it's very unlikely that this additional noise will be audible above the natural noise floor of the recording (environmental+mic+mic-pre noise floor). If the limiter is instead implemented after the ADC stage, IE. It's a digital limiter, then what comes out of the limiter should be bit perfect identical with what goes in, provided no limiting is actually occurring.

My logic is, is that I was told that ALC is bad cause the noise is turned up, when it's selected.

As suspected, H44 has applied incorrect logic! It's true what you've been told, you should not engage the ALC, but what has that got to do with the limiter? Turn the ALC OFF and the Limiter ON!

Thanks for letting me know about the lower the gain. I already new that, but one of the actors I am recording in a shoot, has a quiet voice so I have to turn her up high. If I keep it low, like the other actors, you will probably have to turn it up too much in post, thereby raising the noise floor.

Huh? If you have to turn it up in post you will of course also be turning up the noise floor by the same amount. However, if you turn up the gain on your mic-pre's you are still turning up the noise floor by the same amount! What's the difference? The difference is: 1. You're likely adding more mic-pre noise from the higher gain setting and 2. You're moving the noise floor of the signal further away from the noise floor of the recorder. As PSM, it's your job to balance these two facts, have the mic-pre's set as low as possible without the recorder's noise floor noticeably adding to the signal.

It's your duty to inform the director or AD, preferably during rehearsals, if an actor is mumbling or speaking at a significantly lower volume than the other actors. The consequence in post, if roughly the same dialogue levels are required, is that when this actress speaks the noise floor will unacceptably jump up in volume. The solution to this is adding more noise (room tone) to the other actors, applying more noise reduction to the actress' lines or a combination of both, all of which will degrade the sound quality. You increasing the gain on your mic-pre's during filming will not change the situation. By far the best solution is to avoid the problem by having the actress deliver her lines louder, which is why you need to inform the director asap. The boom op can also potentially help the situation a little (by getting the mic closer to the actress) but this is obviously dependent on the framing requirements and can cause other dialogue editing/mixing difficulties.

On my recorder, there are two settings. 'Limit', and 'Limit + 12db". Is their one that is better than the other?

I'm not absolutely certain what that "+12dB" means. My guess would be an extra 12dB of gain, for recording low level signals but you should check this in the manual. If I'm correct, having this +12dB option engaged will allow you to set the gain on your mic-pre's 12dB lower which will lower the amount of noise they produce. However, this additional +12dB gain stage will obviously also add it's own amount of noise. Whether this +12dB gain stage adds more or less noise than having your mic-pre's set 12dB higher will depend on how high you've got your mic-pre's set and how well the +12dB gain stage has been designed/implemented. I would recommend that you run some tests, to identify the mic-pre gain setting beyond which you get less noise by engaging the +12dB gain stage (and lowering your mic-pres by 12dB).

G
 
Okay thanks. I am not engaging the ALC, but the limiter, is an option under the 'ALC' options. Under ALC there are options such as 'on', 'off', 'limit', and 'limit +12'.

If my way of telling if there is more noise was illogical, then how do you judge the quality of the of the limit? I cannot tell a different by listening regularly between 'limit', and 'off'. I cannot tell which one has more noise, because I can't hear additional noise. So if turning it up in post to hear it better is illogical how do you tell?

The manual does not say exactly what limit +12 means but I as well, assume it means you get an extra 12db. The manual also does not say whether or not it is a digital or analog limiter. I asked a PSM who I have worked with before, but he said he wasn't sure, and says the manual sucks and does not go to into the technicalities as much.
 
Last edited:
I am not engaging the ALC, but the limiter, is an option under the 'ALC' options. Under ALC there are options such as 'on', 'off', 'limit', and 'limit +12'.

A Limiter automatically limits the peak levels, so engaging the Limiter is a partial form of Automatic Level Control (ALC). I assume setting ALC to ON, rather than to Limit, will automatically control the gain as well, much like AGC on some cameras.

If my way of telling if there is more noise was illogical, then how do you judge the quality of the of the limit? I cannot tell a different by listening regularly between 'limit', and 'off'. I cannot tell which one has more noise, because I can't hear additional noise. So if turning it up in post to hear it better is illogical how do you tell?

If you cannot hear a difference it is because you are not hitting the limiter or because your monitoring system simply can't resolve the difference. However, if you hit the limiter hard enough it should be noticeable even on a fairly poor/cheap monitoring system.

Here's the tests I would run:

1a. Setup a consistent sound source, preferably one which has not already been limited (!), an analog clock would probably work OK. Set your ALC to OFF and setup your mic close to the clock, close enough that you need a low mic-pre gain setting to reach a -12dB peak. Once you've got a -12dB peak level, record 15 secs or so.

1b. Without moving the mic at all from test 1a, whack the gain up by about 16dB-18dB, so every tick is easily exceeding the clip point (0dB) and set ALC to Limit. You should not now be clipping anymore, record another 15 secs or so.

2a. Set ALC to OFF. Setup the mic quite some distance from the clock, maybe 3m or so, until you have to turn the mic-pre up to say 80% to get the ticks peaking at -12dB. Record 15 secs or so.

2b. Without moving the mic at all from test 2a, lower your mic-pre so the ticks peak at about -24dB. Set ALC to Limit +12. Make sure your ticks are now peaking at -12db, adjust your mic-pre if not. Record 15 secs or so.

Now you can compare in your NLE:

1. Compare 1a and 1b by increasing the level of 1a and decreasing the level of 1b until the level of the noise between the ticks sounds the same loudness on both. The ticks in 1b should be quieter than in 1a and they may also be distorted in some other way, caused by the limiter. By repeating this test, hitting the limiter less hard, you can determine the effects of the limiter; how clean it is, at what point it starts reducing dynamic range, etc.

2. 2a and 2b can be compared in your NLE directly, without any changes, listen out for the difference in the amount of noise. You will want to repeat test 2 (2a and 2b) by moving the mic closer and further away from the clock, so the mic pre (with ALC set to OFF) needs to be set at different gain settings to achieve -24dB ticks. For example, with the clock set far enough away that the mic-pre has to be set at 95%, you should very clearly hear that 2b has less noise/distortion than 2a. The idea behind repeating this test is to find the mic-pre gain setting at which 2b has less noise than 2a (when 2a has 12dB of gain added in post). You will know in future that when on set and you need to set your gain beyond that point, you will get cleaner results using Limit +12 rather than just Limit.

G
 
Last edited:
(ALC as in AGC doesn't 'increase noise' it increases gain when the signal is low, resulting in a higher noise floor. Cause and effects. Learn that difference.

If ALC is off, why do you think it affects anything?
Your logic is from a different universe.

I'll leave the rest of this discussion to the sound department :P )
 
If ALC is off, why do you think it affects anything?
Your logic is from a different universe.

Not so much on this occasion because on the FR-2 LE the settings for ALC are not just ON or OFF. On this unit ALC has 4 potential choices: ON, OFF, Limit and Limit + 12. I recommend that H44 does not select ALC=OFF but instead either Limit or Limit + 12, depending on the results of the tests above.

G
 
Where the limiter settings would affect the topside of the audio, not the noise floor.
Right?

Indirectly a limiter does affect the noise floor because by reducing the peaks (topside) you are also reducing the signal to noise ratio. But even without this indirect relationship with the noise floor, just by reducing the peaks you are obviously having an "affect" on the recording when you hit a limiter.

G
 
Back
Top