Hi Vojtech,
I was wondering, if you could add Max Infinite minus Avg Infinite curve to analyzers? To your multiband range in particular. It would help to judge a signal's dynamics.
Also, could you add a threshold to Avg Infinite, so it's not falling down when there is no any signal. At -100dB, perhaps.
Thanks,
Vitaly.
Max Inf less Avg Inf curve for analyzers
-
- KVRist
- 219 posts since 2 Jan, 2003
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122 - KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Hi Vitaly,
I'll check about the threshold, but the thing is, it would NOT be average anymore! Imagine you have just a single pop (consisting of all frequencies), and then some actually useful signal, it's likely that the frequencies that are actually not present at all would be most present on the analyzer. So it would need to take maximum from the whole spectrum. Hard to say if it would even be useful then, because the averaging does that for you anyway.
As for the Max inf - Avg inf, quite frankly this sounds waaaaaay tooooo scientific... I'm not even sure it would do what you think it would as different parts of the spectrum react differently.
I'll check about the threshold, but the thing is, it would NOT be average anymore! Imagine you have just a single pop (consisting of all frequencies), and then some actually useful signal, it's likely that the frequencies that are actually not present at all would be most present on the analyzer. So it would need to take maximum from the whole spectrum. Hard to say if it would even be useful then, because the averaging does that for you anyway.
As for the Max inf - Avg inf, quite frankly this sounds waaaaaay tooooo scientific... I'm not even sure it would do what you think it would as different parts of the spectrum react differently.
-
- KVRist
- Topic Starter
- 219 posts since 2 Jan, 2003
Hi Vojtech,
How about a classic gate effect controlling the pause for Avg Inf? In that case, a spectrum won't be changed. Only one controller (threshold). I could do it myself with the Follower but analyzers don't have modulators.
Max Inf - Avg Inf thing. Well, I'm already doing this math by my eyes and you can imagine the pain
Perhaps, it's not an ideal solution and yet it works. The reality is we have no clue of a signal's dynamics in freq-cy domain. I've done some experiments and, to be honest, I was very surprised. I hope, it's not too difficult as you've already implemented all modes.
Cheers,
Vitaly.
How about a classic gate effect controlling the pause for Avg Inf? In that case, a spectrum won't be changed. Only one controller (threshold). I could do it myself with the Follower but analyzers don't have modulators.
Max Inf - Avg Inf thing. Well, I'm already doing this math by my eyes and you can imagine the pain
Perhaps, it's not an ideal solution and yet it works. The reality is we have no clue of a signal's dynamics in freq-cy domain. I've done some experiments and, to be honest, I was very surprised. I hope, it's not too difficult as you've already implemented all modes.Cheers,
Vitaly.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
HI Vitaly,
my question was more like why? It shouldn't matter for averaging.
Sorry about the pain
. I'll check about that, but I don't really know whether it would be simple nor useful. ... had a little pause... ok, so it wasn't hard to implement, done already . But if it is useful, that I don't know... doesn't look to me honestly. Well, just wait for the update 
my question was more like why? It shouldn't matter for averaging.
Sorry about the pain
. I'll check about that, but I don't really know whether it would be simple nor useful. ... had a little pause... ok, so it wasn't hard to implement, done already . But if it is useful, that I don't know... doesn't look to me honestly. Well, just wait for the update -
- KVRist
- Topic Starter
- 219 posts since 2 Jan, 2003
Hi Vojtech,
The problem with the current Avg Inf mode is, it calculates the silence, as well. As the result, the graph is always too low. And even if I refresh an analyzer, every new gap in a signal pushes the graph down again. To avoid that, I use the Pause button. It's not convenient, but it works.
Regarding the use of 'Max Inf - Avg Inf'. It will help to figure out the dynamic "activity" of the signal. Yes, it's not the true dynamics analyzer and yet let me explain how I see its use. (If you don't mind I'll call it an "Activity" analyzer).
So, by looking at it a user can easily see at which parts of the spectrum a signal is more 'active' dynamically and at which less. Based on that, s/he sets bands frequency in a multiband compressor.
Next, the plug-in will use the average of Avg Inf within each band (or Avg Inf - Max Inf) for setting band's Threshold and the average of 'Max Inf - Avg Inf' within each band for setting band's Ratio. That is, the higher 'Max Inf - Avg Inf', the higher the Ratio. All what's left are Attack, Release etc. Hey, you could build completely new multiband compressor where much of guessing job is done for a user. I realize it sounds scientific, but this is only for you. For us, users, it would be just a single button - 'Auto setting'. You could also apply all that to SpectrumDynamics.
What I'm trying to find is the easy way to set a multiband compressor for controlling a signal's dynamics. The way we use it, today, is rather controlling the signal's tone (like eq). We're thinking in terms of each octave's colour (low mids, high mids, presence etc).
But it's only a theory in my head and it's probably too much for one update. Let's have a look at the new mode, first. I'm so excited
Cheers,
Vitaly.
The problem with the current Avg Inf mode is, it calculates the silence, as well. As the result, the graph is always too low. And even if I refresh an analyzer, every new gap in a signal pushes the graph down again. To avoid that, I use the Pause button. It's not convenient, but it works.
Regarding the use of 'Max Inf - Avg Inf'. It will help to figure out the dynamic "activity" of the signal. Yes, it's not the true dynamics analyzer and yet let me explain how I see its use. (If you don't mind I'll call it an "Activity" analyzer).
So, by looking at it a user can easily see at which parts of the spectrum a signal is more 'active' dynamically and at which less. Based on that, s/he sets bands frequency in a multiband compressor.
Next, the plug-in will use the average of Avg Inf within each band (or Avg Inf - Max Inf) for setting band's Threshold and the average of 'Max Inf - Avg Inf' within each band for setting band's Ratio. That is, the higher 'Max Inf - Avg Inf', the higher the Ratio. All what's left are Attack, Release etc. Hey, you could build completely new multiband compressor where much of guessing job is done for a user. I realize it sounds scientific, but this is only for you. For us, users, it would be just a single button - 'Auto setting'. You could also apply all that to SpectrumDynamics.
What I'm trying to find is the easy way to set a multiband compressor for controlling a signal's dynamics. The way we use it, today, is rather controlling the signal's tone (like eq). We're thinking in terms of each octave's colour (low mids, high mids, presence etc).
But it's only a theory in my head and it's probably too much for one update. Let's have a look at the new mode, first. I'm so excited
Cheers,
Vitaly.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Hi Vitaly,
got it, we'll see then. I'll put there some overall threshold, but that's about it. As for what you are trying to do, personally I'd say, use MSpectralDynamics, that's what it was originally for after all
got it, we'll see then. I'll put there some overall threshold, but that's about it. As for what you are trying to do, personally I'd say, use MSpectralDynamics
, that's what it was originally for after all