TranceAddict Forums (www.tranceaddict.com/forums)
- DJ Booth
-- VU Meters
VU Meters
Alot of new DJs are asking some good questions regarding equipment, mixing, and techniques, but I see very few posts about using your VU meters, which is very important in harmonizing and maintating a steady level of output during mixing, and especially during your transitions.
You vet DJs out there, lets post some comments and techniques regarding the usage of VU meters and how it affects mixing.
Thanks in advance.
I couldn't function without them and are wonderful.
BUT
Don't totally rely on them for loudness. Some tracks can be really powerful but not that loud at all.
Doesn't really make sense I know but it's true.
very good question
i myself am trying to learn how to keep equal levels by using the VU. it makes the mix so much smoother when the level is kept the same. it is a bit harder to do this on an average mixer (i have a vestax pcv 002) but still possible.
what i try and do is this:
- adjust gain on channel so it is sitting mainly at 0db and every so often peaking to +3 db
- when bringing in second channel, make sure combined vu doesnt hit more then +3 db in total. this is much harder then it sounds (for me anyway). so i basically use the volume control to adjust it so this is the case. i dont think it matters if the volume is a bit higher then normal during a mix, but you have to be careful. not sure if this helps 
any expert have an opinion
| quote: |
| Originally posted by raaven I couldn't function without them and are wonderful. BUT Don't totally rely on them for loudness. Some tracks can be really powerful but not that loud at all. Doesn't really make sense I know but it's true. |
from the production point of view I can say that some tracks are mastered better than others and the maximum dB might be set high, but the rest of the track may be low in volume ...and vice versa. SOme tracks may stay at a steady 0dB no matter what sounds they make and others may scatter. Poorly mastered tracks will most likely scatter. and poor quality recordings will stay even.
When I've mixed the few time I have so far with an actual mixer (@ friend's place) I've used the VUs to keep the master signal no higher than +2 or +3dB.
If your mixer has the feature, you can listen to both signals in your headphones before you transition, so that you can hear and match the VUs before fading. I switch the process sometimes too and use the VUs to match both tracks. sometimes I even cue into the middle of the track quickly so that I know it will match volumes.
Personally I don't guide by dB because it can't always match to actual volume. You may find annoying highs or hurtful lows that burst the VU meter. It's really a game of comparison and good judgement.
SgtFoo is correct... if you follow the uv meters, they can sometimes show incorrect info, cause of certain frequencies bieng inbalanced in the recording. Sometimes, i may not have time to listen to both tracks long enough to determine and do proper adjustments, so i may use my uv meters to give me a basic guide... then from there as i bring in the track with my crossfader, i can better judge what changes i have to make on the fly.. but it takes a bit of experience and trial and error, but after a while you can master it.
I play both channels in my headphones and match loudness like that usually. The VU is great but it isnt accurate enough for me.
Excellent thread. I agree with everything that's been said here. I usually keep my VUs up a little higher, from +5db, to just BARELY sometimes peaking the +8 light. this is because my mixer has a stupid 6db cut on the outputs that i cannot adjust. if i play at 0db, it sounds weak, no matter how loud i put it.
Also, some records are definitely louder than others, even though the VU won't show it. I have one Art of Trance - Madagascar (Kumara mix) that is so friggin loud, i have to let it peak at about 0db (while the other tracks are at +6) or it's overpowering. I think it has something to do with the frequency range your mixer uses to record VU. anyone know what the standard is? what i'd REALLY like is a VU meter that is wider, so i can see the peaks on individual frequencies.
like this:
even better a 3D display, with X Frequency Y Amplitude Z Time:
kinda like this:
The fact is there are different sorts of meters. And usually what you see are not VU meters...
Real VU meters are like those you see on old equipment, with the needle moving left and right. They are designed to respond the same way as we hear (our ears need some time to perceive a sound as loud). The reaction time is about 360 ms. You can test this scientifically. If you play two short burts of sound that are of the same volume, but the first peak is VERY short, and the other is longer, you'll have the impression that the second sound is louder. That's because our ear needs to adapt first. They tried to imitate that on a VU meter. So if the needle is high, you'll really perceive it as a loud sound.
Most actual mixers have led meters that respond much faster. So they'll react much more heavily on transients we don't really hear as louder. Yet those meters don't react as fast as professional peakmeters (like RTW's) or digital peakmeters (look at the meters in a recording prog like soundforge and compare with your mixers meters). The only dj mixers I know that have those super fast responding meters are the Dateq mixers (the high end ones).
When you're dealing with VU meters, the ideal soundquality you'll get from your mixer is when the loudest parts of the signal are averaging around 0 VU. So peaks above are accepted.
Those superfast responding peakmeters are designed to give you the transient levels also, and those don't reflect the loudness. It's best to peak them at 0dB (like in broadcast appliances).
The meters you find on your mixer are usually between those extremes. So you will find the best spot is average around the 0dB, but the peaks won't go that high. Max +3dB occasionally I'd say.
It also depends on how much compressed your music is. Very dynamic signal can have louder peaks. On the other hand if the music is squashed to death, you'll lower the volume (because the overall loudness will be higher on the second one).
Why all this scientific mumbojumbo. Just to prove I know my theory 
Don't trust the meters, trust your ears. Use the meters only to roughly set your initial gain, and then forget about them...
Your ears are MUCH more sensitive when it comes to loudness more than anything else.
As for John Smith, the reference in audio is always 1000 Hz. But most meters react according to a specified curve (Fletcher Munson, to reflect how we hear). That's why you see the meters move less when you cut bass etc (if the reference was absolute 1000, you'll only see changes when you affect midrange).
And we had that discussion already about your so called 6 db cut. It's probably because your mixer dulls the sound somehow (that's because of the way they designed the mixer probably). Playing at those high levels will only add distortion. Distortion is nothing more than adding harmonics. That's what makes you think it sounds better.
I know there can be a big difference. When I was starting out I used a crappy budget mixer. That sounded already pretty good I must admit. But once I got my Xone, the difference became clear.
I compared both (with the -10 dBV output, as the budget mixer didn't have a pro +4 dBu output). In a controlled recording, there was no difference between the levels. But when running a frequency analyser it became very clear.
High grade mixers use high grade components. And those don't deteriorate the sound as much (or the just do, but in a good way, like valve mixers adding some warm fuzz). The Xone just has a much cleaner sound, and leaves more in the mid and high frequencies. The budget mixer just loses a lot high frequencies due to mediocre components/layout.
That's what's happening. It has nothing to do with a cut in level. But more in frequencies. Driving the mixer in the high levels will add distortion (and like I said distortion are harmonics). And that makes you think it sounds better.
You'll see there will be a similar effect when you would mix around 0dB with your mixer and insert an aural exciter on your master output.
very interesting Thy... nice piece of information. What is strange though... is how the problem becomes where sometimes the UV meter shows to be lets say at -1.. but then the track is much much louder than another track.. happens with records all the time...i guess its how the track is 'compressed' no the record... as you mentioned.
I use a Rane mixer and its overall accurate with my prog like cool edit and sound forge w UV metering.. only exception is when i have a problem as i mentioned above....
for instance.. if i have record A and its showin on my mixer to be -1db... but lets say the record sounds much more louder than that... and i look at my prog (soundforge or cool edit), usually the prog will be peaking in, while my mixer is not...
just to further prove your theory that the programs are more accurate.
Yup either compression, or there are much transients. Some (most?) of these transients are too fast to be detected by your mixer meters (and by your ears also, but they also have the faculty to add up). The digital peakmeters in programs are much faster (they go as fast as your soundcard allows, buffers and all taken into account). Those meters can detect transients that are only a couple of samples long. Your mixer meters will generally detect peaks that last more than a couple of milliseconds (do a simple calculation, cd quality 44.1 kHz = 44.1 samples in one ms. See the difference?).
I always try to keep the +3Db. That's ideal for my system, not too quiet, but still no danger for the system. But, some records do push it to +6 sometimes.
Actually, I think I don't have to worry too much for clipping of my speakers, cuz on my first amp of 80W, I have 100W speakers, and on my second of 150W, I have 200W speakers. So, I think the chances of clipping are rather limited. => I don't watch my VU meter every 5 seconds...
DJ Thy had a lot of good points to listen too. I've got a couple to add to that! First and foremost remember that the amount of bass (how hard it hits and how low the frequency goes as well as the span of frequency range) is the majority of the LED signal on the mixer volume levels. Given that, the bass is also the most noticeable sound when it comes to volume changes. This is because the bass beat in every song is usually the same exact beat repeated over and over and it is consistent throughout the whole song (except for breaks etc). And not only in timing, but in volume too! Most of the highs and mids fade in and out constantly as the song progresses. Therefore making the bass beat the same volume on both tracks is one of the goals of making a good transition.
There are two problems with this. First, not all bass beats have the same depth. By depth I mean the span of frequency range that it hits. Some beats might be 50Hz to 100Hz for example, while others may be 75Hz to 90Hz and still others may be 10Hz to 110Hz. Beats are not a single frequency; they are a combination of frequencies and the associated harmonics of those frequencies. The lower the frequency goes and the larger the range is, the louder it will be, and hence the more it will spike the volume when the beat hits. The reason for this, is that in order for a speaker to drive a bass beat, it must move the actual speaker the same distance as the wavelength of the bass beat (wavelength is just the inverse of frequency). Since sounds with lower frequencies have longer wavelengths, it stands to reason that you must push the speaker further out to reproduce these sounds. In doing so, you will need more power to drive the speaker, and hence the power gets converted into loudness.
The second problem is that if you set the volume of a track that has a very short depth and/or high frequency range on the bass beat, you might potentially clip the higher frequency end of the track because you have to turn up the gain on the track to get the LEDs to hit 0Db.
I correct these problems by using the average volume of the track at its loudest point, not just the bass beat. My live track should already be cued up to the average volume from the last mix, but I find the average volume of the cue track by spinning it to the middle of the track and finding a chorus (usually the loudest part of the track by far). Then I set this level to peak at 0Db.
This solves both problems in one for two reasons. First, this will set the maximum volumes of the two tracks to equal each other. Secondly, the track is much louder during the chorus of the track, while the beginning of almost every trance track I spin starts off with nothing more than a bass beat. Since you are eliminating all of the other instruments, the INITIAL volume that you mix in with on the cue track will actually be slightly lower than the volume of the live track.
You may be saying this is not good; I have the problem now that there is a volume drop when I mix. But you are also forgetting that most tracks end the same way they begin...with nothing more than a bass beat and maybe a hi-hat etc. In general the absolute volume of the bass beat will be constant throughout the whole track (and therefore the transition too if you adjust it correctly), let's say -2Db. This means that when the other instruments come in, they don't overload the mixer and clip. IE) The total volume of the track will go no higher than +2Db, but you don't want it exceeding about +4Db otherwise you will start to clip depending on your mixer�s headspace.
Another thing that you have to be aware of is that unless you mix in the very beginning of one track with the very end of another, the above doesn�t really apply to the transitions that are mixed in at an earlier time. The live track will be going full blast and since you don�t want to loose energy, you mix it into a part of the cue track that also has an equal amount of energy. Now, the outcome of how your gains are set is the same, but the reasoning is different. The reason being, that the volumes are now increased to near their maximum instead of near their minimum.
In summary, with practice, this technique is good because your bass beats will have matching volumes. Don't forget, you need to give your dancers a break too, and during the transition, it is just that--a transition! If you phase the tracks up right, the chorus of the live track should end and 0, 4 or 8 (usually anyway) beats later, the cue track should hit its chorus. This way, the drop in volume you may experience (from the elimination of other instruments) is minimized because as the entire track structure is changing, the volume drops are less noticeable because the listener is listening to other things. You don't want to loose the energy from the track, but you also do not want to give your audience a heart attack by not giving them any lull in the music!
But not to get too far off topic, the main point this thread brings about is that of gain structure, which delves into sound engineering. With each piece of equipment that you buy, you should get a technical sheet with all of the specifications of each output. This tells you what volume range you should play in. You will see something along the lines of this: Master Out 1 (RCA)����.0dBV (1V)/1kOhm. This states that the mixer�s output is optimal at 0dB, as shown by the master VU meter. This is equal to 1 Volt into the input, which will meet 1kOhm of impedance. The rest I won�t go into too much detail, but to summarize things, you should match up the output on your mixer with an input on your amp that has the same voltage and impedance values. If you do this, then you will have a much cleaner sound going through your system and hence coming out of your speakers. But in order to take advantage of this, you shouldn�t set your volume gains to anything much over or under the 0dB mark. If you get this as close to 0dB as you can, then you will be optimizing the output of your mixer. You also want to make sure that you are doing the same thing with each of your TTs/CD decks etc. If you start overloading the signal on your input channel, then you will send out a distorted signal to your amp, which sends it to the speakers and the more pieces of equipment there are in this chain that have a signal volume that is not optimal, the worse and worse it gets. I�m sure that you know from practical experience that there isn�t really a whole lot to worry about because mixers usually have a large headspace to work with, but the further you go into it (and remember that the dB scale is logarithmic) the worse things get�even if you can�t hear it yourself.
Another point to make is that your ears are going to be your main instrument for volumes. The whole spiel about bass beat depth etc in the beginning of this post doesn�t help you actually match the volumes because you can�t crank numbers and let math help you make a perfect volume match. You have to use your ears as judges of volume. If you want to be cautious, then set the gains a slight bit on the high side, then when you increase the channel volumes to transition (this means that you have to mix with the channel volume sliders and not the crossfader) simply listen to the volumes and judge how high you should raise the volume�obviously you won�t raise it to 10 most of the time.
Anyway, I�m kind of re-reading some of this and I don�t know how much sense I am making, so I will stop now. I�ve got sinus congestion and it�s affecting how I think and write!
wow, thanks guys. And, i know your right. My last recorded set is distorted.
download here if you are interested: DOWNLOAD (i look forward to your scathing critiques of my distortion)
I'd like to hear a DJ Flesch VS DJ Thy set! (or even just one from either of you)
Good posts everybody....keep them coming.
Also, I started this thread so that people can ask qestions as well regarding monitoring with VU meters, not just dropping excellent knowledge like every is doing right now.
How are the VU meteres on the DJM 500? Are they accurate enough to complement your ears for judgement during your cue? Plus, when you Normalize a file on lets say Cool Edit or SoundForge, does clear any uncessary outlying peaks and drops and "smooths" out the entire mix?
| quote: |
| Originally posted by JohnSmith I'd like to hear a DJ Flesch VS DJ Thy set! (or even just one from either of you) |
I'll send it to digitallyimported.com and tranceairwaves.com and any others that I can find around.
| quote: |
| Originally posted by dJohn Good posts everybody....keep them coming. Also, I started this thread so that people can ask qestions as well regarding monitoring with VU meters, not just dropping excellent knowledge like every is doing right now. How are the VU meteres on the DJM 500? Are they accurate enough to complement your ears for judgement during your cue? Plus, when you Normalize a file on lets say Cool Edit or SoundForge, does clear any uncessary outlying peaks and drops and "smooths" out the entire mix? |
compression is done at the end of the process???
you are the sound engineer guy on the board,so im not desputing that.. but if you add compression to make it louder, wont it make certain peak points peak past 0db? If you add compression, wont it be applied to the whole selected wave right? SO then, if you have passed the 0db point, wouldnt we want to normalize to -1db last?
BTW..are you going to majoring specifically for audio engineering? Or is it a more broad major?
Im asking cause i would like to start schooling for audio engineering, but in my state of Florida, here in the USA, there arent many schools that offer a program for sound engineer. And the good ones that do, will run over 100k for the whole thing. So its on the expensive side...
When a tune is created, compression can be used on individual instruments (or groups of them) to make them sit better in the mix (or give them a more particular sound).
But there will be compression at the end stage also (that's the mastering part).
I think you don't understand the compressor completely when I see your explanation.
A compressor basically does one thing : it attenuates peaks that go over a certain treshold.
So you set a treshold, for example -20 dB. Every part of the signal will be attenuated. How much? That's what the ratio is for.
Let's say you set a ratio of 1:5. This means that for every extra 5 dB louder than that -20 dB, the compressor will only output one extra dB instead of 5. Everything below the -20 dB will remain untouched.
So another example. Let's say your average signal is always below -20 dB (just an example keep that in mind). Suddenly you have a peak (short or long doesn't matter) at -5 dB (so 15 dB over the threshold). Would there be no compression, the output would be that peak at that -5 dB. But with our 1:5 ratio, the compressor would output a peak of -17 dB. For each five dB it only adds 1 dB (15:5=3).
The attack and release just dictate how fast your compressor will react.
As you can see the basic of a compressor doesn't make it louder, on the contrary it makes the overal signal weaker!
That's where the make up gain kicks in. With that gain you can bring the signal back up to the level you want.
Let's take the example again. Let us assume you want your signal to peak max at 0 dB (example huh). Without the compressor you could only add 5 dB of gain, because that peak was at -5.
With the compressor active, you could add 17 dB of gain, as your loudest peak was a lot lower.
The result is that your peaks don't go higher than the limit you set for yourself, but the overal signal will be louder...
As for my studies, audio engineer, I'm going pretty specific (right now I already have a diploma of audio technician which is broader). But you're right, it's damn expensive.
^^^
So to conclude, I must imagine that the whole purpose of compression, and the reason you would add it, is to pretty much eliminate peaks and to make the whole wave or source evenly level. And as you said, we add gains and make necessary adjustments to bring up the recording to our desired db level (usually loud ass possible without going past -1db).
Also, in reference as mentioned somewhere in this post before, about some records very low at beginning,then having crazy peaks, etc. While there others that stay overall consistant throughout the record, that must be that one was poorly mastered (primarily compressed wrong or not compressed enough ), while the other one was mastered properly.
Well, everything now makes sense as far as compression goes. Pretty cool stuff.
Thanks for the useful information, very helpful...
Ricky
Basically yes, but a compressor can be used more creatively than that.
Believe it or not, but actually the most important controls are the attack and release time.
Again, example time. Take a regular bassdrum. When you have a bassdrum hit, it mainly consists of two parts : a loud attack (the kicker hits the membrane), and the membrane decay which is much lower in volume than the short peak.
With the attack time, you will decide how much of this kick attack will be allowed to pass before the compression kicks in. With the release time you'll decide how long the compressor will keep working. Depending on your settings you can completely change the sound of that drumkick. You can lose the attack but get a fuller body into the sound, or on the other hand you can get a crunchier attack.
With attack and release you can do great things, but setting them wrong can make things worse pretty fast. Like setting a high ratio, with a short release will result in pumping (you really hear the compressor going on and off) which is usually not what you want (but it can be used as an effect though).
So you see, level maximizing isn't the only thing you do with compressors. By changing some parameters, you can change how something sounds, not how loud it's going to be.
Powered by: vBulletin
Copyright © 2000-2021, Jelsoft Enterprises Ltd.