Skip to main content

The more I mix, the more I know how important monitoring and metering is.


The most important application using a Spectral Analyser is the visualization of frequencies and levels found in music or speech. What you see is what your hear! The Analyser shows levels and frequencies even at the edge of the human ear's abilities. The visual display helps to train your ears, and avoids serious mistakes when mixing to the master tape. Usual studio monitors won't let you hear frequencies below 100 Hz. Simply look on the Analyser to see what's going on in the underground!

Reading the display is not easy for novices, because of the huge amount of information that it shows. But after some training you'll agree that this tool is a precious help in every day work.
More notes on Analysis of Music, Special Display Modes, Sound Measurements, Special Applications and Noise Signals can be found in the extensive online-help of Totalyser.

Totalyser: The all-in-one analysis tool

RME gives The Totalyser to us for free so if you've missed what its all about, this link is for you.

http://www.rme-audi…"]RME: Support TechInfo[/]="http://www.rme-audi…"]RME: Support TechInfo[/]

Attached files

Topic Tags


TheJackAttack Mon, 04/16/2012 - 10:09

The issue with the visualization is when some engineer decides that the whole spectrum needs to be utilized at all times like that so called "mastering engineer" that we threw out of here a couple years ago. If the analyzer looks like brick then it isn't going to necessarily sound good. Of course this is just my opinion.

audiokid Mon, 04/16/2012 - 11:39

Well, its a good opinion John and one I definitely am much more aware of now than I was last year. Thanks to you indeed!

I never did follow that thread, was that part of the JP22 thread?

I use my meters to stay around -18dBFS and/or 0dBVU so the transients have plenty of room. I never used to but do now no matter what style of music.

But the world of Rock and Roll and Electronic music is so different when it comes to preserving the transients eh? The transients are valued less and actually get in the way. In fact with my experience, a snare designed for a lot of Pop music works better brick walled than one with all the transients unscathed.
If I am mixing a pop track and don't take those transients out of a snare, I will never get a pop sound and the track will never make the cut or grab the attention of the pop ear. You would have to turn the snare completely down in order for it to work and get any level out of it.

I definitely think everyone should spend time recording classical music. It is a serious reality check on transients and very educational.

The loudness war is forcing us to do terrible things to audio but on the other side of the coin, its part of our current sound, thus why the Transient Designer is such a useful tool.

I feel for the ME. I'm guilty of this all too and it wasn't until I spent time recording classical music like you, John, Wow, do I realize how important those transients are now!

I'm on a deep learning curve to better understand this all and how different kinds of music ( or sounds) need to preserve them and how others not so much. I used to think transients were our enemy and the more we could remove them the louder the track would be, thus, better Mastering and a better sound. But I have been in the Techno world for most of my life so this is all new to me.

Volume impresses people. A sure way to get a referral is to get a track louder than the kids on the block can do. So, these metering tools should help get as close to the mark no matter what style, I guess. Sequoia 12 has upgraded their metering so I'm doing some home work.

Things I'm discovering:
The less complicated music, the louder you can get it without it sounding like crap. Where as Rock and Roll, much more a challenge. And something I'm trying to understand but think I get it, is, electronic music can get really loud and it doesn't sound like the transients are even clipped off. I guess there isn't much of that information in electronic music but I didn't realize that all these years. And what a challenge blending acoustic music with electronically designed music. So we look for glue to blend this all together.

Digital music isn't very complicated in comparison the acoustic music. There are so many more colours to acoustic music and therefore, much harder to play the loudness war without us knowing something very wrong is happening. Wow...

Our current state of music is evolving this direction for a lot more reasons that maybe we are paying attention too. But I think I'm getting it. Maybe the new generation is actually following the club sound because it does sound better, brick walled at low amplification compared to complex acoustic music struggling to compete in the online world. Electronic music sounds much better compressed than acoustic music does. The digital world is really effecting more than we know.

that's my daily rant.

RemyRAD Tue, 04/17/2012 - 10:35

I have found the metering and analyzers in Adobe Audition to be more than adequate. You have the phase scope. You have the spectrum analyzer. You have numerous ways in which to tweak and view those along with the waveform display or spectral display. That along with your Peak and hold responding meter and metering. Why anyone needs more than that, I have no idea? So for just a couple of hundred bucks you not only have great program to utilize you have thousands of dollars of analyzers to view. I still have my 100 MHz Tektronix scope which I no longer need to use for phase display. I don't even need inexpensive real-time spectrum analyzer. I have VU meters on my console and peak meters in my software. It's all any engineer really needs even mastering engineers.

I have mastered just about everything in the recording realm.
Mx. Remy Ann David

RemyRAD Tue, 04/17/2012 - 16:00

There are actually some meters designed to indicate ± phase variances. But what we really used to use back in the day was an oscilloscope. You would plug the left channel into the vertical input and the right channel into the horizontal input. This would then produce a (and my spelling is miserable so I'll probably get this wrong) LISAGOU pattern from the lower left to the upper right at a 45° diagonal to the vertical and horizontal. All of the random phase from instruments in the left & right channels along with the random phase from reverb, would then show up as something of an oscillating fuzz ball. Adobe Audition has this feature built in. And it includes 3 ways to view this pattern along with a normalization feature to expand the display. This phase scope method was extremely important to disk cutters at the record lathes. Simply because the record lathe could not successfully cut a signal that was completely out of phase.

Meanwhile, back on the studio level, we would always be hitting our Mono button to make sure nothing was canceling out and/or disappearing. Though I know plenty of engineers with gold records on their walls that never bother to check their Mono. Go figure? And they've got college degrees in this field. That simply tells me, they weren't taught much. Or, because they came from the digital age where it really didn't make any difference. Especially since movie soundtracks and their associated sound effects frequently utilize an out of phase signal to make sound Rush behind you from the in phase signal which appears to come from the front. Great for car wrecks and explosions, rockets, gunfire, etc.. So really that's all you're missing.

Back in the day when I worked for NBC radio, most of our music was no longer coming off of records. Instead, they would transfer records to NAB cartridges. The NAB cartridge was actually the early automotive 4 track cartridge. That cartridge had a big hole in the bottom where the pinch roller would come up through the deck plate into the cartridge. Later, they stuck a hard plastic pinch roller in the cartridge and that became the 8 track cartridge. Along with narrowing the track widths by 50%. Plus, they also put the head on a movable assembly to move it up and down. The NAB cartridge was actually a three channel recording format. Two channels were utilized for the music and a single channel had cue tones that could flash a light bulb and then also stopped the cartridge at the beginning of the song so it would already be cued up and ready to go again at the press of a button.

The biggest problem with these NAB cartridges as is with analog cassettes was the fact that regardless of the tape guides, tape would continue to skew. And when listened to on a monaural car radio or tabletop radio, you would hear the most God awful phasing. I had suggested a simple fix for that while at the same time a company called Pacific Recorders introduce an NAB cartridge machine utilizing this simple fix. The fix was to encode the left and right channel audio to left minus right & left plus right or, MS as we know it today. So when the tape would skew, the stereo width would merely go from wider to narrower and there would be no phasing to be heard. The tape in NAB cartridges ran at 7.5 IPS and on the car stereo 4 track cartridges, it would run at 3.75 IPS. The same speed was utilized for the 8 track cartridge. And that could actually sounds a heck of a lot better than a cassette running at 1.75 IPS but the noise was rather awful and there was no Dolby utilized in that format. Especially since the tape was of a standard oxide type and it was highly lubricated with silicone since it was a continuous loop and would have to pull out from the center of the pack.

It may also be interesting to note that FM stereo is also transmitted not as left and right but as MS a.k.a. Mono & Difference (L+R, L-R) and is then decoded in your FM radio to bring you beautiful stereo sound. The same for analog television stereo sound. Today it's all a digital interleave.

Where is Mike Berry when I need him?
Mx. Remy Ann David

mberry593 Tue, 04/17/2012 - 16:07

Wow, there is just so much to say here.

Let's start with preparing shows for TV. You MUST use a Dolby LM 100 and maintain dialog between -26 & -28 or run the risk of the embarrassment of having your program rejected by the network. Discovery here in Silver Spring has been notorious for throwing out material. Here is their current specification:

Note : "DCI also requires that program dialogue levels be analyzed using a Dolby LM100 broadcast loudness meter."

There are a lot of people complaining about this and I have a few reservations about it myself but I have to applaud them for taking a stand at the horrible state of TV audio these days. Things are really bad and it seems like they are getting worse every day.

New subject

I hope you are not involved in a loudness war, but if you are please be sure to use an oversampled meter. The Trillium Lane Labs Master Meter that is bundled with PT is just fine for this.

New subject

These days I am not paying much attention to any metering! Of course, I am not doing TV programs.
When I first got involved with computer audio recording, I decided that I didn't like the on-screen peak metering. I built a panel of traditional VU meters, calibrated them to 0 = -20 dBFS and was happy.
Later I became aware that no one else was doing this & I really couldn't defend it either. I still liked the idea of mechanical meters so I built a box with BBC style PPM meters calibrated to PPM 4 = -20 dBFS and was once again happy for awhile. My justification was that these were more peak responding than the VU meters & therefore more likely to keep me out of trouble.
But what trouble? When recording at a 24 bit depth, it is very unlikely to get into any trouble. (I almost always do 24 bit/48 kHz (how about that, I didn't type kc!!)). I now just set levels conservatively in the PT channel green area when tracking and I don't worry about it after that. I just listen for loudness. CAVEAT: You must calibrate your monitoring for 85 dB to do this but that's really easy and cheap to do & I hope that everyone here has already done it.

New subject

There is one area of audio where listening doesn't work. That's live TV. The audio engineer has to listen to the director and may actually be involved in pre-listening to upcoming feeds. In that case, a loudness responding meter is necessary and for all of the faults, the good old VU meter comes out on top for that IMO. ...of course as always YMMV!

mberry593 Tue, 04/17/2012 - 16:19

@Remy: Deep historical trivia......WGMS actually did the sum/difference cart trick on their ITC machines long before Tomcats. A problem with this is that mono S/N ratio is degraded. With a full wall-to-wall stereo signal, it doesn't matter but with a pure mono signal, there is nothing on the difference track. Most material is correlated and therefore closer to mono than wide stereo so levels on the difference track tend to be low and the overall S/N is lower than if you put good level on both tracks. Some people fooled with cheating the difference level up which helps but can be a little dangerous. Others I understand tried to do noise reduction. The problem with noise reduction of those days (we are talking about the 1970s) was that both the DBX & Dolby systems exaggerated any frequency response errors. On a nice open reel machine running at 15 or 30 this wasn't an issue but the cart machines ran at 7 1/2 and had barely acceptable high end. The low end was also a mess but that is a very long other discussion.....NR just made all of those problems worse.

RemyRAD Tue, 04/17/2012 - 21:24

Thanks Mike! I used to work with Mike from 1981 till 1984 when I got transferred into television. He was our chief maintenance engineer at NBC radio, DC. He is so smart. He puts me to shame. I'm almost embarrassed but only almost. He knows I'm a nut but he also knows I'm not bad at this either.

I didn't know that WGMS tried that out? Yes it does have the ability to create extra noise when working with Mono only stuff because nothing is on the difference channel and it will be adding its own noise to the monaural channel. But that's better than phase skewing in my book.

Give me noise or give me death
Mx. Remy Ann David

mberry593 Tue, 04/17/2012 - 23:00

kmetal, post: 388236 wrote: Hey mberry just wondering why you calibrated your VU meters to -20dbfs instead of -18? was it just to leave some head room?

In the US, SMPTE standard SMPTE RP155: -20dBFS = 0

In Europe, EBU R68: -18db dBFS = 0

In the UK, BBC standard (sorry, I don't know the number) -18 dBFS = PPM 4

Do whatever you want, they are only 2 dB apart and it really doesn't matter at all except for interchange.

You will find some old Avid equipment that is set for -14 dBFS = 0. That's getting too hot for me.

mberry593 Thu, 04/19/2012 - 17:41

niclaus, post: 388309 wrote: And i also know some folks, here in France, who still use the -16dbfs = 0Vu...
But other than them, here it's -18=0, as mberry said...

I'm happy to find someone here from France. Could you please say a few words about ORTF metering? We all know about US VU, UK BBC PPM, Nodic, etc.

[[url=http://[/URL]="http://en.wikipedia…"]Peak programme meter - Wikipedia, the free encyclopedia[/]="http://en.wikipedia…"]Peak programme meter - Wikipedia, the free encyclopedia[/]

But I don't know anything about practices in France and as you can see the Wikipedia article refers to a VU meter with a +2 reference.


niclaus Fri, 04/20/2012 - 01:52

Well, for the last 15 years, the only meters i used were VU, Peak DIN (DIN PPM) and full scale.
The reference level was such as +4dbu=0Vu=-9PeakDin=-18dbfs

But I know that those old meter such as used in the old days were different, and were, in fact, calibrated so the reference level was +2VU. But i never saw anyone using that ref level. But this probably explain why on the first digital machines, you had the possibility to calibrate @ -16dbfs, since 0VU=-18dbFS, +2VU was -16dbfs...

Again, i never used such a ref level...

For TV, for the last 15/10 years, the reference level was -18dbFS=0VU=+4dbU=-9PPM, and the real judge was the DIN PPM, and we were not allowed to go farther than 0PPM (DIN).
For the last 4/5 years, some stupid guy who obviously didn't understand the relationship between PPM and FS told everyone not to accept mixes that would go farther than -9dbFS, since he probably thought that -9PPM was corresponding to 0PPM, but he forgot to think about the rising and release time of those 2 different meters...
Anyway, we then had to deliver mixes that were crushed beyond salvation (because to go to 0PPM when you limit @ -9DBFS, you have to crush things), and it also allowed people to crush things so their program was louder than the one next to it... It did not make any sense anymore.

TV mixes waveform started to look like music waveform... You know, the good old brick...

For the last couple of years things changed and we were asked to use the dolby media meter as a reference, and mix @-25(dialogue level) while still limiting @-9dbFS... Since Dolby was not willing to give their algorithm away, the same comition (CST/Ficam) decided to use another loudness measurement and went for the EBU128 recommandation.
Around the same time, analog TV disapeared, and the 0PPM limitation with it...

Since january first 2012, we deliver mixes @-23LKFS (+-1), loudness range inferior to 20LU and max true peak @ -3dbFS... (using the same reference levels, +4dbu=0VU=-18dbFS=-9PeakDin)
Wich make much more sense than what we've been using for the last decade, but it still has some flaws (like on old movies for example).
So, now, we use Peak DIN, VU, DBFS, Loudness meter... (most of the time i only use Peak Din and Loudness meter)

For theatre release, well, when your room is calibrated @ 85/83dbC, you can do whatever you want... but people are still using the PPM meter (phase meters and lissajous) to check levels... (and optical soundtrack (analog LtRt on the side of the print) is still limited around 0PPM).

I hope i was clear, and not too boring...