Taperssection.com
Gear / Technical Help => Recording Gear => Topic started by: Dutchman1101 on July 22, 2007, 09:10:59 PM
-
So I'm pretty sure I'm moving to a 722 and wanted to know what you guys do to get your levels close to 0db as possible. I noticed that the last 5 segments go from 12 to 0db whitch means that it goes 12, 9, 6, 3 & 0 I assume ??? Do you let the 0db LED light up every once in a while?
How do you guys set them? I can see this is the one chink in the 722's armor for me.
Harrison
-
These decks do NOT clip gracefully at all, which is why you find everyone pretty much running them around -6db peaks. Some run even lower, but I shoot for -6 myself, and occasionally it will hit the -3 light. I set my clip indicator lights at -2db, so if I see them flashing, I back off my levels immediately.
-
These decks do NOT clip gracefully at all, which is why you find everyone pretty much running them around -6db peaks. Some run even lower, but I shoot for -6 myself, and occasionally it will hit the -3 light. I set my clip indicator lights at -2db, so if I see them flashing, I back off my levels immediately.
what firmware are you running? ive got 2.something and have clipped without audible artifacts many times. i believe it was mainly older firmware that had the gnarly artifacts
i run mine to peak around -3, basically so the 2nd to last set of Red LEDs hits on the peaks maybe the last reds hitting once during the show, but i try to avoid that. generally i trigger the peak lights on the left column every so often, never really had an issue with the meters
-
i thought that was higher than i wanted...
i thought we wanted peaks around -10, with absolute peaks around -6
-
i thought that was higher than i wanted...
i thought we wanted peaks around -10, with absolute peaks around -6
yep
(assuming that we are talking 24bit here)
we are (talking 24 bit), m0k3. many thanks. and i am using one for the first time tomorrow (a loaner)
-
i thought that was higher than i wanted...
i thought we wanted peaks around -10, with absolute peaks around -6
???
that thinking is basically only to avoid clipping, but there is no reason that you need to or should run levels significantly lower for 24bit recordings
i thought that was higher than i wanted...
i thought we wanted peaks around -10, with absolute peaks around -6
yep
(assuming that we are talking 24bit here)
can you elaborate moke?
-
that thinking is basically only to avoid clipping, but there is no reason that you need to or should run levels significantly lower for 24bit recordings
I thought it was the other way around? I thought there was basically no reason to run hotter levels than that at 24 bit.
-
These decks do NOT clip gracefully at all
Absolutely incorrect. Otherwise I'd be howling about it as a serious bug.
The 1.46 release truncates the clipped waveform as expected and without issue. Some later versions had problems with clipping. Folks say the newer versions don't have clip issues but I'm stickin with what has worked well until I have a good reason to upgrade.
As to peaks... Many of us apparently feel the recordings generally sound better when peaked -12 to -6 vs. closer to 0. It would make an interesting comp.
is this for the 722 specifically? running line-in or phantom or does it matter?
-
It doesn't matter whether it is line or mic in, it's the level. I agree that they clip better than they used to, but I would prefer to have no clipping at all. I so want my Tascam meters back, with the max digital numeric readout. But, as I said, I usually run peaks that occasionally hit -3db, and set my clip indicators to -2db, because I would rather back off a little at that point.
-
I have my clip LED's set to -3db myself. I usually peak at the next set of LED's after the -12db lights, and I usually have to add about 5-6db gain in post MAX :) works for me and sounds great!
-
I have my clip LED's set to -3db myself. I usually peak at the next set of LED's after the -12db lights, and I usually have to add about 5-6db gain in post MAX :) works for me and sounds great!
that is what i am doing now. run the 722 lower and boast in post.
-
I was searching around for a thread on this... I just upgraded my firmware recently and have been having a hell of a time with the levels since then. Last night I ran what I thought was really low - after having this recorder for almost two years - and once or twice during the show got surprised by a 0 peak.
I changed my peak light because the last 3 shows I've taped have been like this. I used to keep it at -1, but I found I have to trick myself now, so I changed it to -3 to give me a feeling of more headroom at the top.
I say we all keep on SD to rework some of this: http://www.sounddevicessupport.com/viewtopic.php?t=515
-
If you are going to work on SD about metering issues, demand a margin display like the DAP1 had.
That is still the single most useful monitoring tool I've ever used and you know what happened while you were away in the can.
-
you don't need to run anywhere near that hot.
That's the thing, I didn't mean to be. My average levels were hovering around -12 to -9. I thought I was well in the clear.
If you are going to work on SD about metering issues, demand a margin display like the DAP1 had.
That is still the single most useful monitoring tool I've ever used and you know what happened while you were away in the can.
Exactly! That's the wording in that SD forum thread (on screen margin indicator).
-
I was searching around for a thread on this... I just upgraded my firmware recently and have been having a hell of a time with the levels since then. Last night I ran what I thought was really low - after having this recorder for almost two years - and once or twice during the show got surprised by a 0 peak.
I changed my peak light because the last 3 shows I've taped have been like this. I used to keep it at -1, but I found I have to trick myself now, so I changed it to -3 to give me a feeling of more headroom at the top.
I say we all keep on SD to rework some of this: http://www.sounddevicessupport.com/viewtopic.php?t=515
you don't need to run anywhere near that hot.
Indeed. Further up thread someone asked for an explanation of why you should set you levels lower. I know that this has been discussed at length elsewhere but it bears repeating in this thread. The simple reason that you do not have to run your levels hot (peaking at near -1 or 0dB) on 24bit is that the format already has a usable S/N ratio (dynamic range) much higher than most playback systems can reveal to you any way. Whereas the 16 bit format has a S/N of 96dB, 24bit has a S/N of 144dB. For 24bit, this works out to 6dB/bit. So, if you are peaking at -6 to -12dB, you are still producing a recording that is far more dynamic than a 16bit recording and exceeds the abilities of many playback systems. This means that you can leave lots of headroom, normalize while post processing and have no real noticable effects on dynamic range.
-
What he said. 8)
-
I thought I would revive this. I've had my fair share of "learning experiences" with the 7XX series boxes. I just ran a 744t last night for Black Mountain. I ran 24/48 - with peaks at 3 db on the 744t (average was probably -6 to -9). I am amazed by the high levels of distortion. Last night when I listened to it in playback, I thought it might be the compression/limiting of the venue PA (running a small PA too loud in a small concrete box).
What's up with the distortion? Is this normal for a 744t. I guess I've learned that the hottest I want is a -6 peak, but more likely -9 or -12. I haven't had a chance to open the files in wavelab yet and take a look. I'm kind of frustrated by this to say the least.
Any thoughts or insight appreciated.
I just thought I'd add some details:
I ran DPA4021 > 744t (mic in). I'm pretty sure they use a lot of compression in the PA at the small cruddy Meridian Red Room in Houston. There was not a lot of variance in the meters during songs, so I'm assuming it was compressed to hell. Still I think it may have been the recorder, because I don't remember the show sounding as bad as the recording.
-
Everyone talks about peak levels but I suspect avg rms levels are every bit or more important in finding the 'sweet spot'.
For example... I recorded a no-PA jazz show recently, mk4 > 722. Recorder gain was 35dB and the highest peak (drums or horn) was -3.60dB. Avg rms was -30.49. It sounds nice. Given a do-over I might have run 28-30dB of gain instead.. But I'm not certain that would have improved the result.
-
Any idea whether the distortion is caused by the PA or recorder with the levels I ran at?
PS - that's pretty low RMS level - but I guess you'd definitely have a huge dynamic range for that. I bet the RMS level on these ones is -12.
-
Not sure.... Was there a quiet song where you could compare?
-
I run 4021 > 722 and my levels sometimes peak at 0 at some point during a show. I have never had any distortion. Do you have the mic gain range set to normal or low? You need to be on low gain range or you may get distortion regardless of your peaks.
-
Do you have the mic gain range set to normal or low? You need to be on low gain range or you may get distortion regardless of your peaks.
I have read that SD says high and low are exactly the same with the only difference being the range available from the knob.
-
I run 4021 > 722 and my levels sometimes peak at 0 at some point during a show. I have never had any distortion. Do you have the mic gain range set to normal or low? You need to be on low gain range or you may get distortion regardless of your peaks.
The choices are low and normal. I had it set to low.
It sounds like this might be a PA (too much compression/limiter issue). Thanks for the input people. There's nothing I hate worse than making a cruddy recording.
-
I think this link has been posted before but it seemed applicable to this thread.
http://messageboard.tapeop.com/viewtopic.php?t=38430
-
Ladies and gentlemen - after re-reading the manual. I discovered this:
Page 15:
When limiters are engaged, audio on channels 1 and 2 is limited to -6 dBFS.
This advice on page 8 is also helpful:
27) Microphone Input Limiter LEDs
Illuminates orange when limiting is occurring on the microphone input. If constantly lit, the microphone input is being hit with too "hot" a signal. Reduce the input sensitivity until limiting occurs infrequently.
http://www.sounddevices.com/download/guides/744t_en.pdf
The limiter was definitely on - and I remember watching those orange lights flicker on and off constantly. Problem solved - too bad I lost a night worth of solid music. :'(
-
Why run the limiter at all?
-
The limiter is great for being on stage at a jazz show. It helps keep those drum solos or horn attacks from clipping, especially when the music is hovering around -18.
-
Why run the limiter at all?
It was on when I borrowed the box. Good call though - I'll probably turn it off in the future.
-
The limiter is great for being on stage at a jazz show. It helps keep those drum solos or horn attacks from clipping, especially when the music is hovering around -18.
Cool, I know a lot of people used to run the limiter on their Mini Me units with good results but I have not seen to many people run it on the 7xx series boxes.
Seems like a good reason to run it for those jazz situations.
-
at least you figured out the problem ;D
ive run the limiter when stealthing before but prefer to just run more conservative now and not bother