where logically should 0dbfs (digital max) register on a analog VU meter

[fatoldgit]

I know that a digital meter measures a different metric than an analogue VU meter but am looking to align my software based VU meters to the most logical point.

I think there are only two logical alignments:

- that 0dbfs = 0db on a VU meter
- that 0dbfs = +3db on a VU meter

I tend to favour +3db cause it indicates that the digital signal is "hot"

Not very critical in the scheme of things but any comments on this appreciated

Not really worried about the academic discussion re pro-audio standards (such as 0VU = +4dBu etc).... just what makes sense when looking at the dancing needles when playing back my digital files.

Thanks

Peter

 

[sam_adams]

Our good friend, Zed Brookes, has a nice little article—that has a nifty click-to-embiggen graphic—that should answer your question.

 

[DVDdoug]

I'd say it's best not to use VU meters with digital. It's an old analog standard that's a compromise between peaks, speed, and average, from a time when it wasn't as easy to show peaks on a meter.

Peak meters are probably the most useful, but depending on what you're doing you might also want RMS or LUFS meters.

Peaks are also important in broadcasting because it's illegal to over-modulate. They aren't so critical with analog tape because tape soft-clips before it saturates, and often there is some headroom.

Preamps and other analog electronics normally have tons of headroom and the VU meter is more of a loudness meter.

Normally, dBFS can't go positive (in integer formats) so you should never "try" to go into red. at all. From that logic, 0dB would be 0dB on both scales, but you'd never see the meter go over 0dB and you're not using the upper 3dB of the meter at all (if you're monitoring the digital level or the DAC output).

 

[kemmler3D]

I think the second option makes more sense, but I don't have much intelligence to add to this one.

 

[Rja4000]

Vu meters represent average level.
So you're supposed to align them with recommended average level, which is -18dBFS (EBU) to -20dBFS (SMPTE).
Most common level is +4dBu.

Overload led usually indicates +6dBVU to +10dBVU, which is what you want to be only happening once in a while and for a short period.

With modern productions, which are highly compressed, that may be quite low, though.
So if you just want to set the VU meters level for eye pleasure, you may want to have them aligned for -10dBFS or so.
Ideally, you'll have a rotary level adjustment knob in that case.

 

[Waxx]

Most pro audio equipment uses the calibration that 0dBVU is -18dBFS on the meters, but it's a choice of the builder of the equipment, not a fixed point. Home audio equipment is a lot less consistant in this. And strange enough, there is no real standard for this as far as i know.

 

[fatoldgit]

Thanks for all this.

As I say... its not about conforming to the strict standards (such as that 0dBVU is -18dBFS)... we all love VU meters but was wanting to make them somewhat useful even given I am playing back digital files.

@DVDdoug basically hits the nail on the head with "a VU meter is more of a loudness meter" which is really what I want... make the VU meters provide some meaningful feedback while looking pretty.

And @Rja4000 states that "Overload led usually indicates +6dBVU" so hedging my bets, my hunch of using +3dbVU is "good enough" to get the needles moving PLUS showing a hot rip CD visually.

Thanks again,

Peter

 

[DonH56]

The VU (volume unit) is defined: 0 VU = 1.228 Vrms (which is +4 dBu) across a 600-ohm load or 2.5 mW. It also had defined risetime of 300 ms, which allowed a true VU meter to register 99% of a 1 kHz sine wave applied for 300 ms.

dBFS is dB relative to full scale output so depends upon the device.

 

[sergeauckland]

In Europe, the EBU spec recommends that 0dBFS corresponds to +18dBu, and that 'peak operating level' is variously +6dBu in Europe or +8dBu in the UK but as measured on a Type 1 PPM in Europe or Type 2 PPM in the UK. That allows some 10-12dB headroom below 0dBFS. You can see this most clearly on the BBC's digital streams where there's always plenty of headroom above peaks, and they seldom go over -10dBFS.

CDs and other streamers like Radio Paradise or Spotify however use CDs as they're supplied, so regularly hit 0dBFS, as that's what's on the track being broadcast.

AS VU meters don't indicate peaks, if you want to see your VU moving satisfactorily, I can only agree with those above who suggested a variable control, and you adjust it until it looks right.


S.

 

[GXAlan]

Not really worried about the academic discussion re pro-audio standards (such as 0VU = +4dBu etc).... just what makes sense when looking at the dancing needles when playing back my digital files.

Thanks

Peter

If you are not worried about accurate discussion, I would put 0 db dancing display as -3.01 dBFS

if your dancing needle doesn’t go over 0, there is no risk of intersample overs.

Intersample Overs in CD Recordings

Intersample Overs are a Common Occurrence in CD Recordings We have frequently used Steely Dan's Gaslighting Abbie from Two Against Nature in our listening tests. This is a spectacular CD recording with lots of dynamics and a low noise floor. Nevertheless, in a little over 5 minutes, this track...

benchmarkmedia.com

 

[DonH56]

If you are not worried about accurate discussion, I would put 0 db dancing display as -3.01 dBFS

if your dancing needle doesn’t go over 0, there is no risk of intersample overs.

Intersample Overs in CD Recordings

Intersample Overs are a Common Occurrence in CD Recordings We have frequently used Steely Dan's Gaslighting Abbie from Two Against Nature in our listening tests. This is a spectacular CD recording with lots of dynamics and a low noise floor. Nevertheless, in a little over 5 minutes, this track...

benchmarkmedia.com

A mechanical meter, or one meant to emulate a mechanical meter, won't indicate peaks. Back when I was building VU meter packages for folk I used to include a peak detection circuit that drove a little LED (fancy new red light) to show when a peak was over a user-defined threshold occurred.

 

[Cbdb2]

A mechanical meter, or one meant to emulate a mechanical meter, won't indicate peaks. Back when I was building VU meter packages for folk I used to include a peak detection circuit that drove a little LED (fancy new red light) to show when a peak was over a user-defined threshold occurred.

And that threshold could be as high as 18db over 0VU depending on the headroom of the gear.
To answer the OPs question, set it so 0dbfs is +3db on the VU meter, if your not using it to record. Most music is so compressed its almost always near 0dbfs so the above setting should always bounce your meters around 0VU.

 

[fatoldgit]

And that threshold could be as high as 18db over 0VU depending on the headroom of the gear.
To answer the OPs question, set it so 0dbfs is +3db on the VU meter, if your not using it to record. Most music is so compressed its almost always near 0dbfs so the above setting should always bounce your meters around 0VU.

yeah, I probably erred in my original post by not stating the VU meter use was for non critical purposes.

90 % of my music is recorded between 1957 (dawn of stereo) and 1980 (so doesnt suffer compression) and of the remaining 10%, very little is modern enough to have suffered compression.

So my non-critical use of VU meters does play a minor role in showing me when an album, which might be relatively loud on playback is actually potentially compressed (i.e. relatively little needle movement) rather than just simply being loud.

I can then, if interested, run some DR tests later on to confirm. If its bad then the album gets deleted.

Peter

 

[Philbo King]

I know that a digital meter measures a different metric than an analogue VU meter but am looking to align my software based VU meters to the most logical point.

I think there are only two logical alignments:

- that 0dbfs = 0db on a VU meter
- that 0dbfs = +3db on a VU meter

I tend to favour +3db cause it indicates that the digital signal is "hot"

Not very critical in the scheme of things but any comments on this appreciated

Not really worried about the academic discussion re pro-audio standards (such as 0VU = +4dBu etc).... just what makes sense when looking at the dancing needles when playing back my digital files.

Thanks

Peter

It's important to note VU is used to meter analog audio, while dBFS meters digital audio numeric sample values. So it is an apples to oranges comparison.

Generally 0 VU (analog) is defined as 1.228 Vrms driving a 600 Ohm load with a 1KHz continuous sine wave. It has a 300 mSec rise time. (In other words, peaks are 'trimmed off'.) Back in the old days (pre-1960), a VU meter reading zero value was used was calibrated to indicate the 1% THD distortion threshold.

In pro studios 0 VU is often set to -18 dBFS (digital) for a continuous 1 KHz sine tone. This is done accomodate high instantaneous peak values without exceeding 0 dBFS when copying analog magnetic tape into a digital recorder.

In practice, you can set it at whatever level works for you; nobody will call the dB police...

 

[Cbdb2]

yeah, I probably erred in my original post by not stating the VU meter use was for non critical purposes.

90 % of my music is recorded between 1957 (dawn of stereo) and 1980 (so doesnt suffer compression) and of the remaining 10%, very little is modern enough to have suffered compression.

So my non-critical use of VU meters does play a minor role in showing me when an album, which might be relatively loud on playback is actually potentially compressed (i.e. relatively little needle movement) rather than just simply being loud.

I can then, if interested, run some DR tests later on to confirm. If its bad then the album gets deleted.

Peter

Compression is nothing new, music in the 60s was compressed. Most was listen to over the radio and so needed it. It just wasnt compressed to death.
Set your VU so +3 is at 0dbfs and see what happens, then adjust it to your liking.

 

[kchap]

Our good friend, Zed Brookes, has a nice little article—that has a nifty click-to-embiggen graphic—that should answer your question.

Excellent summation. I wish I'd known about that article earlier. It would saved a lot of paper and ink; metaphorically speaking

 

[stoo23]

Personally, I think there is Still a quite good argument for using VU meters.
I also always liked the BBC PPM style of meters, technically referred to as Quasi peak programme meter (QPPM) see here: https://en.wikipedia.org/wiki/Peak_programme_meter

I recently sent these 2 x articles to someone else struggling to understand the differences with Old (Analogue) and New (Digital) metering, that may be of interest.
VU Meter: Just How Useful Are They?
VU Meter: The Best Kept Secret Of The Pros

 

[sergeauckland]

Personally, I think there is Still a quite good argument for using VU meters.
I also always liked the BBC PPM style of meters, technically referred to as Quasi peak programme meter (QPPM) see here: https://en.wikipedia.org/wiki/Peak_programme_meter

I recently sent these 2 x articles to someone else struggling to understand the differences with Old (Analogue) and New (Digital) metering, that may be of interest.
VU Meter: Just How Useful Are They?
VU Meter: The Best Kept Secret Of The Pros

Perhaps with pop music that's more or less heavily compressed, there's an argument for VU meters, but for 'real' music. i.e. uncompressed acoustic music, I don't see any value whatsoever with a VU meter.

Furthermore, for transmission, i.e broadcasting, it's peak deviation that matters, not average levels, so VU meters are of no use.
For digital recording, it's peak level that matters, so peak meters are required, not some notional -18dBFS VU meter.
The only value I can see for VU meters is with analogue tape, where high quality machines like Studer or Ampex running high level tape could stand a certain amount of over modulation, and anyway, when running line-up tapes, it's easier to set levels with a VU meter than with a PPM (not that it's that hard even with a PPM)

Considering that VU meters were designed for telephony, they're best left there.

S.

 

[stoo23]

Perhaps with pop music that's more or less heavily compressed, there's an argument for VU meters, but for 'real' music. i.e. uncompressed acoustic music, I don't see any value whatsoever with a VU meter

Pardon ?? ... Really ??

Considering that VU meters were designed for telephony, they're best left there

Yeah,.. well, we simply Won't think about or recall, the inumerable decades of ALL kinds of Music recorded using VU meters then,.. shall we ??
Let alone ALL the huge and arguably famous console manufacturers like Trident, Neve, API, Quad Eight, Helios et al,.. that all included 'Telephony Meters' on their consoles,... even Modern DAW's have VU meter options (and NOT merely for going to Tape), let alone All the Meter 'Plug-ins' that abound,... Methinks you have completely,.. Missed the Point,.. sorry.

 

[sergeauckland]

Pardon ?? ... Really ??

Yeah,.. well, we simply Won't think about or recall, the inumerable decades of ALL kinds of Music recorded using VU meters then,.. shall we ??
Let alone ALL the huge and arguably famous console manufacturers like Trident, Neve, API, Quad Eight, Helios et al,.. that all included 'Telephony Meters' on their consoles,... even Modern DAW's have VU meter options (and NOT merely for going to Tape), let alone All the Meter 'Plug-ins' that abound,... Methinks you have completely,.. Missed the Point,.. sorry.

VU=Virtually Useless!



S.