Topic: Suggestions for upgrade

[DSatz]

As a long-time audio professional I'm appalled at the number of people on this forum who advocate peak recorded levels of -12 dB or even lower. Peaks on digital recordings should go as high as possible without quite reaching 0 dB. If you're not sure you can judge the maximum sound levels that will occur, you might choose to aim initially for peaks at (say) -6 dB on the theory that louder moments might occur after that. But if so, the point is still to get peak levels of at least -6 dB and hopefully somewhat higher.

Or if you have a recorder whose meters aren't peak-responding (examples: the old Sony portable DAT recorders TCD-D7 and -D8), then sure, compensate by aiming for lower maximum levels (about -4 or -5 dB for those Sony models)--but again the point is that if you get -4 or -5 dB maximum levels on those crummy meters, without triggering the "PEAK" indicators, you very likely were recording peaks of about -1 dB in reality, which is damn near perfect.

Even in cases of great uncertainty, if your only way of being sure not to hit 0 dB is to aim for -12 instead, then let's please not deceive ourselves or one another; that's a tactic, and the real hope is still to get the highest possible maximum levels below 0 dB as the result of that tactic.

A recording that actually peaks at -12 dB isn't better than one that peaks at, say, -3 or -2 dB. There's no virtue or advantage in maintaining substantial headroom and never using it. Then it's not headroom any more, but wasted dynamic range. Headroom is useful only if it's a potentially available part of the dynamic range, i.e. if there's at least some chance of using some of it sometimes.

--best regards

[stevetoney]

^ As a pro, you're usually afforded fairly ideal conditions to record in...ac power, cordoned off zones free from foot traffic, adequate lighting, tables for your gear, etc.  Tapers aren't pros and field conditions we encounter might dictate a different recording strategy, depending on the specific recording scenario.  The -12db to -6db setting that's often suggested by experienced 'field' tapers is a 'set it and forget it' strategy for tapers that typically record loud music in special situations where they have their recorders tucked away in their pockets, or in a bag to prevent being spilled upon, and frequent monitoring is either not possible or not desirable.  By setting the recorder this way, you don't run the risk of going over if the FOH guys run the levels up or if the dynamics of the show changes significantly.  With 24bit technology, this is considered favorable option to setting limiters, especially if loud music is being recorded (you wouldn't hear the noise floor that you're amplifying when you bump levels in post).  With new recorder technology providing wi-fi capability of monitoring and control from a remote or a cellphone, perhaps this strategy will no longer need to be used.

[Gene Poole]

With 24bit technology, this is considered favorable option to setting limiters, especially if loud music is being recorded (you wouldn't hear the noise floor that you're amplifying when you bump levels in post).

^^^^This.

In the old 16 bit days, it was definitely tough to hug the line, but people seem to think that 24 is only 1.5 times 16, but this is incorrect.  Each extra bit doubles your headroom.  24 bit has 256 times more headroom than 16 bit.  Why risk clipping when you've got all that room to work with (unless you're in an ideal studio setting).

[2manyrocks]

I appreciate seeing this issue discussed. I've been burned recording events live without the opportunity to carefully set levels or change them during an event, and it has made me conservative on setting them.  However, I am not totally comfortable with this, either.  I suspect I'm not getting the best that could be had?

[Gene Poole]

You definitely want to take advantage of that extra bit-depth if you can (and you can and still keep a safe distance from the ceiling).

[hoppedup]

I've been using the dual recording feature on my DR-60D and I love it. It records the second source at 6db lower than the first. Takes out much of the guess work when setting levels.

[bombdiggity]

Personally I tend to aim for -6 or a little higher.  Most of the stuff I record is very quiet with a lot of transients and I tend to keep an eye on my levels.  Most rooms and music I see I know what to expect. 

My only real complaint with the R-44 is that 6db step mechanism in the gain (which is not smooth if you change it and turning the input down much at all by the fine trim will leave your signal fried in the pre-amp stage).  That is the big limiting factor to recording higher there for me, so I'm either -6-ish or a bit irritated to be -10/12-ish until I can bump it between songs. 

With the Sony I'll run closer as it is quite easy and smooth to adjust a bit on the fly. 

On the unpredictability if I have to fix a stray clap from the audience with Izotope I'm OK with that (I don't want to clip music but don't want to peak -12 or lower to be that sure about it). 

[stevetoney]

^ I don't have an R-44 anymore, but I hear you on that.  (I'm seriously thinking about buying another one).  I'll tweek if the mood hits me and then make mental or actual notes about where I made the changes so I can go back and smooth out those spots, if necessary, when I'm mastering.  More times than not though, once a show starts, I'm just as happy to never touch anything even if I'm kinda low because I like knowing there aren't any self made tweeks on my raw recording.  It's easier for me to do this with the type of recording I do though.  I agree with you...it's a bitch getting levels set optimally at quiet shows with large swings in dynamics. 

[carlbeck]

As a long-time audio professional I'm appalled at the number of people on this forum who advocate peak recorded levels of -12 dB or even lower. Peaks on digital recordings should go as high as possible without quite reaching 0 dB. If you're not sure you can judge the maximum sound levels that will occur, you might choose to aim initially for peaks at (say) -6 dB on the theory that louder moments might occur after that. But if so, the point is still to get peak levels of at least -6 dB and hopefully somewhat higher.

A recording that actually peaks at -12 dB isn't better than one that peaks at, say, -3 or -2 dB. There's no virtue or advantage in maintaining substantial headroom and never using it. Then it's not headroom any more, but wasted dynamic range. Headroom is useful only if it's a potentially available part of the dynamic range, i.e. if there's at least some chance of using some of it sometimes.

--best regards

This, it also surprises me as well to hear people recommend peaks as low as -12 db unless they are recording in a very low profile situation. I took a break from recording for 5 years & when I came back I read as much as I could here to catch up. The general opinion was the -12 db peak & I thought that was odd....I had taped in 24 bit & DSD before my hiatus without ever hearing that thought, we usually just aimed for -4, sometimes -2 peaks just as we had with DAT so I figured must be something new all the cool smart kids are doing.... After 5 or 6 shows (typical rock shows, taping in the open) I was not at all pleased with my recordings no matter how much I bumped them in post, they just felt flat & lifeless. Finally I just decided to run my rig as I always had in the past, run a little hot, get as close to -4, sometimes -2 as possible without going over & I started to enjoy my recordings again. Bingo, I was happy, my recordings came to life, I didn't clip, I didn't die & no huge dynamic swings rained on my parade. So to each their own, I've always loved this community, I learn every time it read a post around here so I'm just offering a suggestion of what works for me. Give it a shot before you upgrade your equipment, the worst that can happen is that you'll clip a recording possibly ruining it but you'll learn in the process & that's what counts.

[Marshall7]

That's why commercial CD's are compressed and then boosted so everything is right on the edge of clipping.  Apparently it "sounds better that way".

http://en.wikipedia.org/wiki/Loudness_war

[carlbeck]

Loudness war has little to do with peak level in a live recording.  If you never clip your recorder while recording a source with less than 70dB dynamic range (which, ch will be true of practically all concerts), there should be no difference in a recording with -2dBFS and -12dBFS source, when both are subsequently normalized to the same level.

Unless as you helped me determine in another thread that the ADC isn't receiving the full range due to some prosumer decks input gain stage at the mic preamps. In my case with the DR680 the full bit depth not being utilized at less than - 12 db.

[carlbeck]

Quote from: Jon Stoppable on September 24, 2014, 11:24:37 PM

This depends on the operating level of the recorder, and specifically how they decide to handle the analog input stage.  Realize that nearly every ADC chip has a maximum input of about 1VRMS, which is 0dBV (+2dBu).  So to interface with 'pro' levels, the input has to be padded before the ADC.  Or, if the input level control of the device allows, the sensitivity of the input can be set to any operating level.

tinybox is basically a -10dBV device with a bit of extra headroom (+13dBV).  Its noise performance is maximized a little above 20dB gain, so the noise performance is a little worse at +18dB, but for a loud source this just really doesn't matter.  If a quiet source is being recorded, I'd recommend using high (+32dB) gain as required.

The DR-680 is probably structured like a 'pro' device, so you need 12dB more gain to hit the same level at the ADC as a recorder that only operates at the 0dBV max level of the ADC.  That probably means that even at its nominal 0dB setting, the DR-680 is actually padding the input to the ADC.

A typical condenser mic is around -40dBV/Pa, so a not-very-loud concert will have average level of that, but peak level of up to -20dBV.  Given that the maximum peak input of an ADC chip is 0dBV, you shouldn't need more than 20dB gain.  There are reasons to use a higher operating level for signal transmission, but in a typical taper situation they wouldn't really apply.

This is why I designed tinybox as a -10dBV device:  it is structured exactly to the task at hand, which is recording concerts using condenser microphones into recorders operating at -10dBV.  This saves quite a lot of power over the 'pro' level approach of using more gain and a higher operating level, only to pad that level at the ADC.  Every 6dB increase in operating level doubles power consumption, so a 'pro' level input will use four times the power of a 'consumer' level input.

Jon I believe you are correct. When I first started using my DR680 I ran the external preamps & therefore the recording levels quite low as everyone had suggested (-16db on the horrible display, with peaks slightly higher) but I found the recordings flat & lifeless even with a healthy bump in post gain as everyone has suggested. It wasn't until I started to run my preamps at higher gain settings that I found the recordings sounding as I had expected. My point is that yes, I believe you are correct in the assumption of the DR680 requiring higher gain levels more than likely due to operating at pro specs, ie padding the input to the ADC. As always your input is appreciated & an asset to the community, thanks.