Gearslutz (
-   Newbie Audio Engineering + Production Question Zone (
-   -   Voxengo Span shows different value each time (

ikamy 28th February 2017 02:22 PM

Voxengo Span shows different value each time
Hi folks

I cant trust to my voxengo span plugin because each time I play some part of my mix i get different peak level value

Each time I hit the reset button and play the part
Then I stop after finishing

Waiting all levels go down and again repeat the process

Sometimes I get -1.5 , sometimes -2 sometimes even 0.0 !!

When I play it on loop without stopping I got more weired results

Maybe I use it in a wrong way
My Version is 2.9

Zargg 28th February 2017 03:18 PM

Hi. I cannot say I have seen this myself.
My guess is that you start your playback / loop start / end when the audio is not at "zero crossing", and that SPAN jumps when playback / loop.
Others may know better.
All the best.

stinkyfingers 28th February 2017 06:06 PM

I have seen this behavior and the zero crossing explanation is what I always assumed. Force of habit now to just click the reset button after pressing play. Definitely helps to be mindful of your start/stop points. also sometimes if you have certain processing going on they will show a "spike" of level on startup but then it will settle. dynamics processors and filters can do this, maybe other efx too.

ikamy 28th February 2017 07:19 PM

thank you

reset button after playing may help
another question about span is do you guys stop after playing your audio to get the peak level or playing it on loop ?

stinkyfingers 28th February 2017 08:21 PM

i tested Span's (numerical) peak meter in Reaper using various tones and noise shapes and it seems to be pretty consistent. it gives a reading that corresponds to the source audio file's peak level, regardless of where i stop and start playing.
i would guess it has something to do with the plug-ins (efx and/or instruments) you are using, or a "buggy" behavior with Span and/or the DAW. i admit that once in a while Span has done some strange things on me in Reaper, but i don't know who's to blame.
there is never a guarantee that everything will always work how it should on every system in every situation, so it's good to test with a more scientific method to be certain. if you keep coming up with different values for something you know should be "x dBFS" then you know you have to start your detective work.
try using audio files with known levels. test tones. but use an actual audio file, not the tone generator, because that can show fluctuation in real time.

wrgkmc 28th February 2017 08:41 PM

Span has a slow response and isn't super accurate when it comes to riding leaks, There are much better tools for that.
Span gives you a general frequency analysis without the waveform jumping all over the place like it would in real time
which is actually a benefit for what its used for.

As far as your last question I'm guessing you're asking about the combined peak gain level of the entire waveform?
Or are you selectively looking at a specific frequency within the wave form.

If you're trying to find the combined level try this Sleepy Time meter for that. This thing is killer for that kind of stuff. It also lets you check the side and middle levels as well as adjust them as needed. Sleepy-Time DSP VST Plugin Archive! (Free Download)

I should note, its the RMS levels that are the most important to monitor. Most RMS meters are purposely measured slowly to come up with the most reliable number. Peak transients are so fast and sporadic its tough for a computer to respond fast enough to accurately capture an accurate number of them to give you a useful reading. You're better off viewing the wave in a histogram which has grid lines to show you the actual min's and maxes. RMS or Average levels are much more reliable with a moving meter because you have complete waveforms working there. You could simply use a little math and add 29.3% to the Root mean square level (RMS) value and get a consistent peak level that way.

stinkyfingers 28th February 2017 09:04 PM


Originally Posted by wrgkmc (Post 12473439)
Peak transients are so fast and sporadic its tough for a computer to respond fast enough to accurately capture an accurate number of them to give you a useful reading.

This is not true. If it can't give you a peak value, it can't give you a sample.
I have never seen a digital peak meter that did not give the proper sample peak value, bugs aside.
+/- a few tenths of a dB or so anyways. Some will round.

ikamy 1st March 2017 12:56 AM

so based on your comments I think I cant be sure if I have a project mixed with both instruments and wave sample with fx on both of them

I just can be sure with only 1 mixed file with no effects

anyother plugin worth buying and better than this ?

stinkyfingers 1st March 2017 01:14 AM

i assume you are talking about the numerical peak values that Span displays and not the spectrum view ?
if that is the case, i have found Span to be accurate as can be with it's measurements/values.
if you play something, and Span gives you the peak value, you can be assured that is the peak value (+/- 0.01 dBFS). unless of course there is a bug somewhere.
it will even show # samples clipped, which is also accurate.
are there better meters ? of course. it all depends on what you are looking for.
which trivial value would you like to have ?

if you are talking about the spectrum view and it's accuracy, there are a lot of variables to that and some background knowledge of how they work to set them properly for the occasion. but in short, i think Span is the best plug-in analyzer. everyone will say a different one. that should tell you something. so pick the one with your favorite colors.

hello people 22nd August 2019 05:54 PM

Does anyone know how to get the peaks to hold in the current free Voxengo Span?

Years ago it would hold peak levels across the frequency range and they'd stay as a line as you played through the track.

All I can get it to do is display the graph/ line in real time...but nothing is 'held'.