Topic: What is the best part of the rig to get most of the gain from ?

[fandelive]

Hello,

my current rig consists in a ST-9100 preamp and an Edirol R09-HR.
I'd like to know which one is the best to get most of the gain from.

Thank you

[nedstruzz]

Hello,

my current rig consists in a ST-9100 preamp and an Edirol R09-HR.
I'd like to know which one is the best to get most of the gain from.

Thank you

I've always been a firm believer in proper gain structure. 

Here is my suggestion.

Set the gain on both of your pieces of equipment to roughly the same level so that neither piece of equipment is giving more or less gain than the other. 

Think of it like water running through your cables.  You don't want the 9100 open all the way just for the R-09 to squash down on the flow of signal.   Concurrently you don't want too little signal coming from the 9100 and then have the R-09 wide open.  You want nice smooth flow of signal/water. 

Now some may disagree with this method and say you want 90% of your gain coming from the 9100 and this may be true but test both methods out and see what you like. 

[yug du nord]

I have no personal proof to offer....  but there are theories too that suggest that it's best to milk as much as you can out of your ADC......  but that might depend on the actual ADC in certain rigs.

Ned's theory is your safest bet IMO. 

[stevetoney]

A third theory is to put most gain on whichever piece of gear is the quietest and/or has the lowest noise floor.

A fourth theory is the put most gain on the piece of gear that you like the sound flavor of the most.

Some people have said to rely the most on the preamp gain since that's what it's designed for, but the R-09HR is known to be very good too.

...take your pick which answer serves your goals the best because they're probably all 'right' answers.

EDIT TO ADD:  Regardless of what you decide for your gain mix, I'd make sure you're aware of the unity gain setting on the R-09HR and set your R-09HR accordingly.  While it's not a requirement to stay below it, you risk having distorted recordings (without realizing it until you get home) on the louder live shows if you set the R-09HR above the unity gain setting.  I think unity gain is something like 45 or 50.  Personally, when I had an R-09HR, I never let it go over 45.

[fmaderjr]

I think unity gain is something like 45 or 50.  Personally, when I had an R-09HR, I never let it go over 45.

Chris Church tested the R-09HR and posted that unity gain was about 38 "in most situations".

[datbrad]

I agree with Ned. You want to have your gain stages evenly distributed, not over driving or under driving any one stage. I know it will work with one component giving most of the gain, but at least for me I can almost always hear in recordings when a weak preamp output has been cranked up by the recorder (thin & scratchy), or when a too cranked preamp has been throttled down by the recorder (saturated & muddy).

BTW, the water example Ned used is a very good analogy. I have read educational material for electrical theory that uses water flow principals to teach with. It really does help it make more sense when you start talking about volts, amps, watts, and impedance when plumbing analogies are presented.

[fmaderjr]

I agree with Ned. You want to have your gain stages evenly distributed, not over driving or under driving any one stage. I know it will work with one component giving most of the gain, but at least for me I can almost always hear in recordings when a weak preamp output has been cranked up by the recorder (thin & scratchy), or when a too cranked preamp has been throttled down by the recorder (saturated & muddy).

BTW, the water example Ned used is a very good analogy. I have read educational material for electrical theory that uses water flow principals to teach with. It really does help it make more sense when you start talking about volts, amps, watts, and impedance when plumbing analogies are presented.

This may be absolutely correct for high gain preamps, but Chris Church recommends setting the recorder at unity gain to start (about 38 for the HR) and the ST-9100 at 90-100% of full gain (assuming the clipping light isn't coming on, and the show would need to be really loud for that to happen). Then if more gain needed you can get it from the recorder. This is how most  members here that have posted about it use the 9100/HR combo. Remember the ST-9100 has only 20 dB of gain. It is not going to "throttle" the recorder's line in, especially with the recorder set to unity gain to start. 

[datbrad]

I agree with Ned. You want to have your gain stages evenly distributed, not over driving or under driving any one stage. I know it will work with one component giving most of the gain, but at least for me I can almost always hear in recordings when a weak preamp output has been cranked up by the recorder (thin & scratchy), or when a too cranked preamp has been throttled down by the recorder (saturated & muddy).

BTW, the water example Ned used is a very good analogy. I have read educational material for electrical theory that uses water flow principals to teach with. It really does help it make more sense when you start talking about volts, amps, watts, and impedance when plumbing analogies are presented.

This may be absolutely correct for high gain preamps, but Chris Church recommends setting the recorder at unity gain to start (about 38 for the HR) and the ST-9100 at 90-100% of full gain (assuming the clipping light isn't coming on, and the show would need to be really loud for that to happen). Then if more gain needed you can get it from the recorder. This is how most  members here that have posted about it use the 9100/HR combo. Remember the ST-9100 has only 20 dB of gain. It is not going to "throttle" the recorder's line in, especially with the recorder set to unity gain to start.

With many of the commercially made preamps that average around 60db of total gain, when running one close to it's max gain capability, you can have increased THD in the pre, as well as overloading of the line inputs on the recorder. I also recall that the Beyer MV100 was only a 20 db gain preamp and I remember it sounded terrible cranked to the max, even though it was not too hot for the recorders.

However, if Chris Church has recommended running his very hot into the R09, he clearly designed it to do so and what you say makes total sense. The universal answer to the OP's question in the subject line for most recording setups is what Ned said.

If you are running a V2 or Sonosax cranked near the max, it would not be a good thing for most recorders I know of, definately.

[fandelive]

Thank you all for those very instructive bits.

Chris Church recommends setting the recorder at unity gain to start (about 38 for the HR) and the ST-9100 at 90-100% of full gain (assuming the clipping light isn't coming on, and the show would need to be really loud for that to happen). Then if more gain needed you can get it from the recorder. This is how most  members here that have posted about it use the 9100/HR combo.

That's a good way to start, thank you for the tip !

Regardless of what you decide for your gain mix, I'd make sure you're aware of the unity gain setting on the R-09HR and set your R-09HR accordingly.  While it's not a requirement to stay below it, you risk having distorted recordings (without realizing it until you get home) on the louder live shows if you set the R-09HR above the unity gain setting.

Thank you, I wasn't aware about the unity gain at all !!
Does that mean that if I set the ST-9100 at 90-100% of full gain and the Edirol R09-HR above 38, there's a risk of distortion ?

My mics are 4.7k (Chris Chruch) modded MM-HLSC-1's. I made a test yesterday at home in front of a speaker. I pushed the volume high, but it was definitely not as high as an average rock concert would be.
With the ST-9100 set at 90-100% of full gain and the Edirol R09-HR set at 38, the results were barely audible.
I had to crank up the input level of the Edirol at 80 (100%) to get good results (and no distortion).

I got my MM-HLSC-1's modded because they couldn't handle loud hard rock concerts bass and bass drums.
They were prefect for David Gray, but I got distorted results with Skunk Anansie or Nickelback for example. Still, the input volume never reached the 0dB limit in my Skunk Anansie or Nickelback tapes and the peak indicator never lighted on the Edirol during recording. So the problem was not about a bad input level setting I guess.

Tonedeaf : when you're talking about possible distortion with gain set above unity on the R09-HR, is it a general statement made on the average loud rock shows sound level or has it more to do with the rig parts capacity to handle loud bass / bass drums ? (I mean, whatever if the input levels settings are good and never reach the 0dB peak).

I'm asking this because you're talking about distorted recordings (without realizing it until you get home) and that's exactly what happened before my mics got modded. (didn't try the modded mics + ST-9100 preamp yet).

[fmaderjr]

Thank you, I wasn't aware about the unity gain at all !!
Does that mean that if I set the ST-9100 at 90-100% of full gain and the Edirol R09-HR above 38, there's a risk of distortion ?

tonedeaf is much more knowledgeable than I am about the general principles involved in recording, but I suspect that if the concert has a low enough sound pressure level that you are not getting enough gain with the ST-9100 at 100% and the HR at 38, you will not be risking distortion if you raise the HR's level quite a bit as long as the meters are staying under 0 dB. However if the concert is quiet enough, there is a point where you could be adding audible noise (hiss) to your recording from the HR's preamp. I don't know where that point is, but I'm guessing you could go over 60/100 (maybe well over) without adding noise. The older R-09's pre was decent until you raised if above 25/30 and then it started to get very noisy. The HR's pre is supposed to be much better.

If it takes a lot of gain from the HR to get the levels even to peak at -12 or so, I wouldn't raise the HR's level any further though. With 24 bit, you can add the extra gain in post without adding audible noise. I generally aim for peaks no higher than somewhere between -12 and -6 anyway, so that I don't need to worry about a sudden volume peak causing clipping (and so I don't need to stress myself out due to having to constantly monitor my levels).

It is my understanding that if your mics themselves are not distorting (as yours did before the modification) and your preamp is not distorting (check the clip light) you are not going to get distortion from the HR when it is set to unity gain and the meters are staying under 0 dB.

[runonce]

Without regard to any specific device....

You want to get max/safe gain as soon as possible in your signal chain.

Some get enough gain with no preamp -  so only 1 gain stage. (mic > line)

Some use two gain stages. Preamp > Recorder gain. You want to get the gain at the first stage...Ideally the second should be unity - or something specific for that device. (despite the term, it seems "line in" isnt a totally reliable/universal spec)

If you undergain your 1st stage and boost in a second stage...you also increase the noise from the first device by the same amount.

[datbrad]

Without regard to any specific device....

You want to get max/safe gain as soon as possible in your signal chain.

Some get enough gain with no preamp -  so only 1 gain stage. (mic > line)

Some use two gain stages. Preamp > Recorder gain. You want to get the gain at the first stage...Ideally the second should be unity - or something specific for that device. (despite the term, it seems "line in" isnt a totally reliable/universal spec)

If you undergain your 1st stage and boost in a second stage...you also increase the noise from the first device by the same amount.

This^

(btw, there are universal standards for line level. Generally there are devices with unbalanced consumer line inputs at -10db sensitivity, and professional balanced line inputs at +4db sensitivity. Obviously, this 14db difference in sensitivity is significant, and something that must be considered as a place to start when getting distortion on line inputs, with recording levels that do not go over zero. Most recorders today have active electronics between the physical input and the gain control, which is why they can overload so easily with unbalanced line inputs that are -10db consumer line level.)

[stevetoney]

Thank you, I wasn't aware about the unity gain at all !!
Does that mean that if I set the ST-9100 at 90-100% of full gain and the Edirol R09-HR above 38, there's a risk of distortion ?

tonedeaf is much more knowledgeable than I am about the general principles involved in recording, but I suspect that if the concert has a low enough sound pressure level that you are not getting enough gain with the ST-9100 at 100% and the HR at 38, you will not be risking distortion if you raise the HR's level quite a bit as long as the meters are staying under 0 dB. However if the concert is quiet enough, there is a point where you could be adding audible noise (hiss) to your recording from the HR's preamp. I don't know where that point is, but I'm guessing you could go over 60/100 (maybe well over) without adding noise. The older R-09's pre was decent until you raised if above 25/30 and then it started to get very noisy. The HR's pre is supposed to be much better.

If it takes a lot of gain from the HR to get the levels even to peak at -12 or so, I wouldn't raise the HR's level any further though. With 24 bit, you can add the extra gain in post without adding audible noise. I generally aim for peaks no higher than somewhere between -12 and -6 anyway, so that I don't need to worry about a sudden volume peak causing clipping (and so I don't need to stress myself out due to having to constantly monitor my levels).

It is my understanding that if your mics themselves are not distorting (as yours did before the modification) and your preamp is not distorting (check the clip light) you are not going to get distortion from the HR when it is set to unity gain and the meters are staying under 0 dB.

Thanks for the compliment fmaderjr, but I'm not more knowledgeable, just more than willing to share my opinion.      Ton of difference!  LOL.

For the sake of the OP, here's what I've learned (from others here on TS.com) that you want to be aware of.  Yes, the R-09HR has a unity gain point...which apparently is right around the 38 or 40 mark.  Some people have in the past got home from a loud rock concert and their preamp set quite low and was never distorting, so they figured they had a good recording just from watching their levels.  But then when they got home, they discovered that their recording was distorted and couldn't figure out why since their preamp levels were low...and this could be true even when the recorder levels were peaking say at -12db.  The reason was that the distortion happened at the recorder side. 

So, by knowing where unity gain is on the R-09HR that can help prevent this from happening.  That said, I do think fmaderjr is right that it's OK to go above unity gain, but it's probably also safe to say that the louder it gets the more there's a chance of issues from the recorder side.  OTOH, as fmaderjr says above, I also have always felt safer to leave my levels peaking low (and raise them in post) than going over unity gain on the R-09HR.

[runonce]

Without regard to any specific device....

You want to get max/safe gain as soon as possible in your signal chain.

Some get enough gain with no preamp -  so only 1 gain stage. (mic > line)

Some use two gain stages. Preamp > Recorder gain. You want to get the gain at the first stage...Ideally the second should be unity - or something specific for that device. (despite the term, it seems "line in" isnt a totally reliable/universal spec)

If you undergain your 1st stage and boost in a second stage...you also increase the noise from the first device by the same amount.

This^

(btw, there are universal standards for line level. Generally there are devices with unbalanced consumer line inputs at -10db sensitivity, and professional balanced line inputs at +4db sensitivity. Obviously, this 14db difference in sensitivity is significant, and something that must be considered as a place to start when getting distortion on line inputs, with recording levels that do not go over zero. Most recorders today have active electronics between the physical input and the gain control, which is why they can overload so easily with unbalanced line inputs that are -10db consumer line level.)

I agree and understand the consumer/pro level difference. A pitfall for sure.

Im referring more to the cheaper handheld recorders that have no obvious or intuitive way of setting it to unity.
And odd control systems and undefeatable settings.
For example I have to dial my mixer down a few notches to patch a H2 versus a JB3...at their presumptive "line in" settings.

[fmaderjr]

Some people have in the past got home from a loud rock concert and their preamp set quite low and was never distorting, so they figured they had a good recording just from watching their levels.  But then when they got home, they discovered that their recording was distorted and couldn't figure out why since their preamp levels were low...and this could be true even when the recorder levels were peaking say at -12db.  The reason was that the distortion happened at the recorder side. 

As you know, this could also happen if the mics couldn't handle the sound pressure levels (like unmodded AT853's and some of the less expensive mics). I take it you ruled out that possibility after reading the posts?

Knowing nothing else about the circumstances, I would have guessed that the mics would be the most likely source of the problem. According to guysonic, the HR's line in is supposed to be able to handle a very hot input without clipping and I wouldn't have thought that raising the gain (with the meters still under 0 dB) would cause it to clip.

If I'm wrong about this, I would certainly like to know it. I don't know anything about this from experience. I rarely have my recorders set much above unity gain because I can get most of the gain I need from my pre.

[Todd R]

With many of the commercially made preamps that average around 60db of total gain, when running one close to it's max gain capability, you can have increased THD in the pre, as well as overloading of the line inputs on the recorder. I also recall that the Beyer MV100 was only a 20 db gain preamp and I remember it sounded terrible cranked to the max, even though it was not too hot for the recorders.

However, if Chris Church has recommended running his very hot into the R09, he clearly designed it to do so and what you say makes total sense. The universal answer to the OP's question in the subject line for most recording setups is what Ned said.

If you are running a V2 or Sonosax cranked near the max, it would not be a good thing for most recorders I know of, definately.

Just to be more clear on all this clipping/overloading/brickwalling stuff:

There are several places where you can have overloading leading to distortion --

- at the mics (mics cannot handle the SPLs)
- at the input of the preamp (cannot take that hot of a signal from the mics)
- at the output of the preamp (cannot provide as high of a level of signal on its outputs after applying gain)
- at the input of the recorder (cannot accept that hot of a signal from the preamp)

The MV100's problem (I used to own one, and modded it to fix this problem) isn't that it cannot take a hot signal coming from the mics, it is that it cannot actually provide an output signal that is 20db above the input signal it is accepting (the third distortion type in the list above).

So it was designed to take a hot input, which it can, and it doesn't overall output too hot of a signal for the recorder as Brad notes (so none of the 4th type of distortion above) -- it just can't perform the minimum 20db gain function for all input signals that meet the max input spec.  Probably anecdotal evidence on my part, but I think this is pretty unusual -- overall just a crappy design on Beyer's part. I haven't come across this much -- generally it seems that manufacturers will insure the integrity of their design such that if the minimum gain is applied and the maximum input level is not exceeded, the preamp will not brickwall on the outputs.

Anyway, that's a long winded aside -- really just saying that the MV100 IMNSHO was a crap-ass design, and it shouldn't be used as a metric of how to set gain more generally. 

[Church-Audio]

Hello,

my current rig consists in a ST-9100 preamp and an Edirol R09-HR.
I'd like to know which one is the best to get most of the gain from.

Thank you

Get the most from the preamp  then get the rest from the R09 thats the point of the preamp. To reduce the need for gain from the R09 to get a better signal to noise ratio.

[fandelive]

If it takes a lot of gain from the HR to get the levels even to peak at -12 or so, I wouldn't raise the HR's level any further though. With 24 bit, you can add the extra gain in post without adding audible noise.

I didn't even knew that thing about 24 bit !!
This thread has been very usefull to me so far and I thank you all for that.

I'm gonna record in 24/48 (always did that way since I own the R09-HR), set the CA-9100 preamp at 90/100%, the Edirol at unity gain (38) and add the extra gain in post.

Thank you very much.

[Church-Audio]

If it takes a lot of gain from the HR to get the levels even to peak at -12 or so, I wouldn't raise the HR's level any further though. With 24 bit, you can add the extra gain in post without adding audible noise.

I didn't even knew that thing about 24 bit !!
This thread has been very usefull to me so far and I thank you all for that.

I'm gonna record in 24/48 (always did that way since I own the R09-HR), set the CA-9100 preamp at 90/100%, the Edirol at unity gain and add the extra gain in post.

Thank you very much.

Well I dont think I would do that. I would run the preamp at 90 to 100% then increase level on the recorder until you are at about
-10 or so then normalize in post. If you are sitting at -30 or so thats a whole lot of gain to make up in post. I dont think that will be the case but its better to shoot for a target then to try and boost later on. -10 Is a good place to start it leaves you lots of headroom.

Chris

[fmaderjr]

Yes, follow Chis' advice posted above and you should never have a problem.

[Todd R]

Without regard to any specific device....

You want to get max/safe gain as soon as possible in your signal chain.

Some get enough gain with no preamp -  so only 1 gain stage. (mic > line)

Some use two gain stages. Preamp > Recorder gain. You want to get the gain at the first stage...Ideally the second should be unity - or something specific for that device. (despite the term, it seems "line in" isnt a totally reliable/universal spec)

If you undergain your 1st stage and boost in a second stage...you also increase the noise from the first device by the same amount.

Generally, this seems right, though I'd like to learn a bit more about it.

For the bolded part, I think this is a little off. If you undergain the first stage and add gain at the second, you increase the noise associated with the first stage (the noise from the preamp itself) by the amount of gain you add at the second stage.  But there isn't any difference to the inherent noise of the mics (or the noise in the venue itself).  For the mic noise, I don't think it would be any different if you ultimately wanted 40db of gain whether you applied 40db at the 1st stage and 0 at the 2nd vs applying 20db at the first stage and 20db at the 2nd -- either way the mic noise is boosted by 40db.

This is where I'd like a better understanding from what runonce is saying -- if the noise of the preamp is absolutely swamped by the mic noise level and the level of noise in the venue being recorded (as in many db's below), does it make much/any difference if the total gain is applied equally across two stages vs being applied 100% at the first stage?

Say that the noise level of the preamp is -100db, and the noise level of the combined mic's inherent noise and noise in the venue is -70db, does it matter if you apply a second gain stage with say 10db of gain?  You've raised the noise level of the preamp induced noise from the first gain stage by 10db, but that noise component was already 30db less than the noise from the mics -- isn't that boosted level of noise from the preamp stage buried from the mic anyway?

[Church-Audio]

Without regard to any specific device....

You want to get max/safe gain as soon as possible in your signal chain.

Some get enough gain with no preamp -  so only 1 gain stage. (mic > line)

Some use two gain stages. Preamp > Recorder gain. You want to get the gain at the first stage...Ideally the second should be unity - or something specific for that device. (despite the term, it seems "line in" isnt a totally reliable/universal spec)

If you undergain your 1st stage and boost in a second stage...you also increase the noise from the first device by the same amount.

Generally, this seems right, though I'd like to learn a bit more about it.

For the bolded part, I think this is a little off. If you undergain the first stage and add gain at the second, you increase the noise associated with the first stage (the noise from the preamp itself) by the amount of gain you add at the second stage.  But there isn't any difference to the inherent noise of the mics (or the noise in the venue itself).  For the mic noise, I don't think it would be any different if you ultimately wanted 40db of gain whether you applied 40db at the 1st stage and 0 at the 2nd vs applying 20db at the first stage and 20db at the 2nd -- either way the mic noise is boosted by 40db.

This is where I'd like a better understanding from what runonce is saying -- if the noise of the preamp is absolutely swamped by the mic noise level and the level of noise in the venue being recorded (as in many db's below), does it make much/any difference if the total gain is applied equally across two stages vs being applied 100% at the first stage?

Say that the noise level of the preamp is -100db, and the noise level of the combined mic's inherent noise and noise in the venue is -70db, does it matter if you apply a second gain stage with say 10db of gain?  You've raised the noise level of the preamp induced noise from the first gain stage by 10db, but that noise component was already 30db less than the noise from the mics -- isn't that boosted level of noise from the preamp stage buried from the mic anyway?

Its always best to get the most gain from the preamp with the least self noise.. No matter where in the chain that preamp is. Getting gain at the first stage only really helps if you have long cable runs between your recorder and the preamp.. But in the case of the 9100 the way it was designed was to have high headroom and 22db of gain. With the noise floor being below -100db Unweighted... broad band. It makes the most amount of sense to get the gain with the 9100 first. Other preamps I cant comment on.

Chris

[aaronji]

(btw, there are universal standards for line level. Generally there are devices with unbalanced consumer line inputs at -10db sensitivity, and professional balanced line inputs at +4db sensitivity. Obviously, this 14db difference in sensitivity is significant, and something that must be considered as a place to start when getting distortion on line inputs, with recording levels that do not go over zero. Most recorders today have active electronics between the physical input and the gain control, which is why they can overload so easily with unbalanced line inputs that are -10db consumer line level.)

As I understand it, there aren't really universal standards.  Particularly with the consumer stuff, there is quite a bit of variation.  Even with the pro stuff, there are different standards (such as ARD at +6 dBu).  Also, those nominal levels are -10 dBV and +4 dBu, so they are actually ~11.78 dB apart (not 14)...

[datbrad]

(btw, there are universal standards for line level. Generally there are devices with unbalanced consumer line inputs at -10db sensitivity, and professional balanced line inputs at +4db sensitivity. Obviously, this 14db difference in sensitivity is significant, and something that must be considered as a place to start when getting distortion on line inputs, with recording levels that do not go over zero. Most recorders today have active electronics between the physical input and the gain control, which is why they can overload so easily with unbalanced line inputs that are -10db consumer line level.)

As I understand it, there aren't really universal standards.  Particularly with the consumer stuff, there is quite a bit of variation.  Even with the pro stuff, there are different standards (such as ARD at +6 dBu).  Also, those nominal levels are -10 dBV and +4 dBu, so they are actually ~11.78 dB apart (not 14)...

The 2 standards we face are for gear intended for sale and use in the USA, and the +6 level is a standard in Europe. It's just like the difference in AC power standards. Regardless, my understanding is that both dBV and DBu are measures of voltage where each unit of measure represents 1 milliwatt of gain, only difference being the impedance. Even if it is only effectively a 12db difference, that is more than double in actual "loudness". This means that a balanced XLR line is made to accept twice the signal level that an unbalanced consumer line input can, which is the main point I was making.

[it-goes-to-eleven]

With 24 bit, you can add the extra gain in post without adding audible noise.

I didn't even knew that thing about 24 bit !!

That's because it isn't true.

[fmaderjr]

With 24 bit, you can add the extra gain in post without adding audible noise.

I didn't even knew that thing about 24 bit !!

That's because it isn't true.

I suppose if the 24 bit signal is low enough, you will add noise that is audible, but not nearly as much as if the signal was 16 bit. Is that why your are saying not true, I guess I can't argue with you. I don't know the technical stuff as well as you may know it. I was just trying to give advice that should help a beginning taper make a good sounding recording with no danger of distortion and with a minimum of stress while making the recording.

I think peaking at -12 dB to -6 dB at most is good practice for a new taper (if not all tapers) when recording in 24 bit. There will be no danger of peaks distorting, they won't have to stress about constantly monitoring their levels, and their recordings will sound fine when boosted. However this a a hobby we are all in for fun, so we should all use whatever method works well for us. This method has been recommended by many here and it has worked well for me.



 

[it-goes-to-eleven]

With 24 bit, you can add the extra gain in post without adding audible noise.

I didn't even knew that thing about 24 bit !!

That's because it isn't true.

I suppose if the 24 bit signal is low enough, you will add noise that is audible, but not nearly as much as if the signal was 16 bit.

It is the same.

For example..  If you increase the gain in post processing by 10 dB, the noise floor will increase by 10 dB.  It does not matter whether it is 16, or 24 bits.

This reminds me a little bit of the Spinal Tap conversation - except it is 24 bits and not 11

I think peaking at -12 dB to -6 dB at most is good practice for a new taper (if not all tapers) when recording in 24 bit.

I think we always need to emphasize what sounds best in actual listening, not what we assume sounds best, or what specs might lead us to assume.  And when listening, we need to consider the limitations of our playback.

Any discussion of peak levels is incomplete without mentioning average levels.  Source material varies greatly.   A bluegrass quartet is going to have a very different spread between average levels vs. peak, then rock through a PA.    -12 is a fairly low peak.  Is it too low?  That depends...

[fmaderjr]

Any discussion of peak levels is incomplete without mentioning average levels.  Source material varies greatly.   A bluegrass quartet is going to have a very different spread between average levels vs. peak, then rock through a PA.    -12 is a fairly low peak.  Is it too low?  That depends...

Ok, now I understand where you're coming from. Thanks for explaining why you disagreed and I certainly can't argue with you.

I love bluegrass, but most of my recording has been fairly loud rock and roll so in those cases the average levels are going to normally be  reasonably high when you're peaking at over -12 dB (and often I get to -6 dB) so I haven't had to think much about peak levels.

For example..  If you increase the gain in post processing by 10 dB, the noise floor will increase by 10 dB.  It does not matter whether it is 16, or 24 bits.

But isn't it true that in 24 bit the noise floor can be low enough that raising it 10 dB wouldn't be audible? All I care about is what I can hear.

[datbrad]

Venturing off topic along with you guys, I have found that in practice, it does not sound the same with 24bit to have maximized the bit depth where the peaks are hitting close to zero during the recording, versus running soft levels at the show peaking below -6 and then boosting them close to zero in post. Theoretically it should be true, but my experience has been that it's not.

I am not talking about just noise. I am talking about the overall quality of the signal itself. I experimented last year running one set low, peaks between -12 and -6, and the next set much hotter, peaks between -4 and zero. I boosted the first set in post up to match the second set that was left as it was recorded, dithered/resampled both to 16/44.1, and they were not identical sounding. The second set where the levels were optimal to begin with sounded much better to my ears, smoother, cleaner, or something along those lines.

Except in very select situations where the source is wildly dynamic, or in cases where the situation does not afford you the ability to easily access your deck to adjust, I think it's missing the mark to run soft levels when you are in an open taping situation. I sure don't want to walk out of a show with 16bits worth of information when I had a 24bit recorder. This is just my opinion, of course. To each his own..........

[fmaderjr]

To each his own..........

Yeah-we all should do what makes us happy. My recordings that peak at -6 sound fabulous to me and I'm never again going to worry about keeping my levels near 0 and worrying about clipping.

Since you're apparently a skillful enough recordist to be able to do that, more power to you. I need to be able to keep things simple or recording is not enjoyable for me (as long as the results sound good to me).

You probably have better ears than I do also. I'm getting up in years, so I'm sure I wouldn't hear any difference if I recorded closer to 0 dB anyway. After all, a good Hi-SP minidisc recording sounds pretty much the same to me as a good 24 bit one.

[aaronji]

The 2 standards we face are for gear intended for sale and use in the USA, and the +6 level is a standard in Europe. It's just like the difference in AC power standards. Regardless, my understanding is that both dBV and DBu are measures of voltage where each unit of measure represents 1 milliwatt of gain, only difference being the impedance. Even if it is only effectively a 12db difference, that is more than double in actual "loudness". This means that a balanced XLR line is made to accept twice the signal level that an unbalanced consumer line input can, which is the main point I was making.

"Universal" is universal, not "gear intended for sale and use in the USA".  Many people on this site are located outside of the US.  Additionally, as the input of your recorder is fixed in terms of impedance, dBu and dBV are still ~2.22 dB different.  I understand your point, but on a logarithmic scale that's a noticeable difference...

Venturing off topic along with you guys, I have found that in practice, it does not sound the same with 24bit to have maximized the bit depth where the peaks are hitting close to zero during the recording, versus running soft levels at the show peaking below -6 and then boosting them close to zero in post. Theoretically it should be true, but my experience has been that it's not.

That's been my experience too, especially on quiet parts in music with a broad range...YMMV and all...

[datbrad]

To each his own..........

Yeah-we all should do what makes us happy. My recordings that peak at -6 sound fabulous to me and I'm never again going to worry about keeping my levels near 0 and worrying about clipping.

Since you're apparently a skillful enough recordist to be able to do that, more power to you. I need to be able to keep things simple or recording is not enjoyable for me (as long as the results sound good to me).

You probably have better ears than I do also. I'm getting up in years, so I'm sure I wouldn't hear any difference if I recorded closer to 0 dB anyway. After all, a good Hi-SP minidisc recording sounds pretty much the same to me as a good 24 bit one.

I would not call myself skillful, but since I spent over 15 years recording at 16bit, I have gotten pretty good at optimizing my levels as close to zero as possible without going over. I guess it's more intuitive to me to run strong levels that are what I want in the end result and do nothing more than covert to 16/44.1 in post. And I am on the topside of 40, so my ears are certainly not as good as they were 20 years ago. Nevertheless, I feel there is something to be gained from working hard at the show to optimize levels.

[fandelive]

I would run the preamp at 90 to 100% then increase level on the recorder until you are at about -10 or so then normalize in post.

So I'm back from the show and I just wanted to thank you all for your priceless advices.
I couldn't set the volume during the first act as it was an acoustic/solo performance.

I started running the preamp at 100%, the Edirol at unity gain (38) and checked the levels during the first song.
My right mic hit the peak several times so I just reduced the preamp gain using the knob until -6dB.

As a result, I got a great tape with no distortion

[fmaderjr]

So I'm back from the show and I just wanted to thank you all for your priceless advices.
I started running the preamp at 100%, the Edirol at unity gain (38) and checked the levels during the first song.
My right mic hit the peak several times so I just reduced the preamp gain using the knob until -6dB.
As a result, I got a great tape with no distortion

This forum is an incredible source of info for new tapers and for tapers unfamiliar with the best way to use equipment they've just bought.

When I was flying blind I used to get distorted recordings at times and not even know why. Now, even if I'm using equipment I've never used before, that virtually never happens.