The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Gear for sale     Latest  Trending
...if the level in a bus gets above 0 dB
Old 16th June 2019
  #31
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by RedBaaron View Post
[...] Probably there is some other way.
Sure there is:

...Just manually turn it down until you get the level you want.
.
Old 16th June 2019
  #32
Lives for gear
 
RedBaaron's Avatar
Quote:
Originally Posted by 12ax7 View Post
Sure there is:

You just turn it down until you get the level you want.
In the context of vst effects that are there in the channel strip for each track (which was my hypothetical situation), lowering the track fader only works without degrading the signal if the effects are processed post-fader rather than pre, or if a particular effect doesn't "mind" the higher gain levels (like trim plugins, for example).

The problem is that post-fader is not the default behavior in all DAWs. So in Studio One, for example, which only has Post-Fader on the Main Buss, if the source .WAV file is way too hot upfront, I either have to use 'Decrease Volume' on the wave file or use a Trim plugin as the first effect in my channel strip before, say, adding non-linear eq or saturation. Lowering the fader for that channel would only lower the resulting output volume after all the damage was already done there in the chain. Many daws like Cubase have gain trim pots, and a few like Samplitude let one move all the effects to post-fader, which in essence means that the channel fader alone suffices to get the volume down before it enters into any effects

In the case of auxillaries, pre vs post fader considerations may also apply, but in general, I agree it's pretty much straightforward by comparison: one need only reduce source track's volume by lowering their faders.

So depending on the daw and where the effect is (in a track vs. an auxillary), the faders alone might not do the job without degrading the source unnecessarily. In Samplitude, I'd probably suggest simply getting in the habit of making all the effects post-fader, which can easily be adjusted there in the effects popup window just by highlighting an effect and using the up and down arrows, if memory serves me.
Old 16th June 2019
  #33
Lives for gear
 

Quote:
Originally Posted by 12ax7 View Post
Yes!
This is indeed EXACTLY WHAT IT IS!

...(And for the life of me), I have NEVER been able to understand why people believe semantics to be such a "bad thing"...


In some cases people don't like complaints or interjections on semantics because it doesn't further the discussion at hand. Like in this case. It doesn't matter if you think that there is no such thing as "realtime" in digital, what matters is if there's a relevant difference between two solutions to solve the OP's problem and whether or not my use of words made the distinction clear.

I guarantee you even relatively new engineers could probably follow what I was saying and understand the difference between the two options.

Engineers call realtime processing "realtime" for a good reason.

Quote:
Originally Posted by 12ax7 View Post
...Hmmm, maybe it's because they just don't know what "semantics" means (oh, the irony!).

It would seem that the prevailing sentiment today is: "Dictionaries are for people with no imagination!"
Quote:
Originally Posted by 12ax7 View Post
.
If all it took was imagination to define words and phrases then we wouldn't need dictionaries, so that complaint of yours makes little sense.
Old 16th June 2019
  #34
Lives for gear
 

Quote:
Originally Posted by RedBaaron View Post
What I was getting at is that the the menu command with in Samplitude is called 'normalize' and can be used to lower the overall gain by choosing a target percentage (say 10% instead of the default 100%). I'm not sure of another way to do it directly to a file in that DAW, since I use S1 nowadays. Probably there is some other way.
It really shouldn't be done to the file though. That was my point.
Old 16th June 2019
  #35
Lives for gear
 

Quote:
Originally Posted by RedBaaron View Post
In the context of vst effects that are there in the channel strip for each track (which was my hypothetical situation), lowering the track fader only works without degrading the signal if the effects are processed post-fader rather than pre, or if a particular effect doesn't "mind" the higher gain levels (like trim plugins, for example).
You're not really degrading the signal by changing the level using trim, clip-gain/event-volume or faders. I think it's worth separating the issue of what level changes does to the signal by themselves versus what plugins do to them.

Quote:
Originally Posted by RedBaaron View Post
if the source .WAV file is way too hot upfront, I either have to use 'Decrease Volume' on the wave file or use a Trim plugin
I'm getting the feeling that by "wave file" you're talking about grabbing a "line" on a clip/event/region on the timeline, and you're thinking about that as a "wave file" because the clip/event/region displays a waveform.

So just to be super clear; a wave file is something quite different, and "normalizing" typically changes that wave file whereas clip-gain/event-volume etc does not typically change the file on the drive.

Quote:
Originally Posted by RedBaaron View Post
In Samplitude, I'd probably suggest simply getting in the habit of making all the effects post-fader, which can easily be adjusted there in the effects popup window just by highlighting an effect and using the up and down arrows, if memory serves me.
I hope you don't mind that I disagree on this. I really think it's better to get in the habit of adjusting levels before plugins using the other methods and then using the faders to set the output levels post-inserts when mixing. I think that's a more traditional and more common workflow so there'll be less confusion when interacting with others this way.
Old 16th June 2019
  #36
Lives for gear
 
RedBaaron's Avatar
Quote:
Originally Posted by mattiasnyc View Post
It really shouldn't be done to the file though. That was my point.
It wouldn't be my first choice, although It's probably a worse idea in some daws than others. In Studio One, it keeps a volume curve on each file, visible there on the wav file, which you can drag up or down to increase the volume. I don't think it has to create a new file each time that's adjusted. In the past, I'd sometimes hit control + a to select all tracks and lowered them all at once with that feature--I tip I got from one of their help videos. I don't fall back on that option though, because if I'm having to reduce the volume, I'm going to want to know how close I'm getting to 0VU in the process, so I prefer just to load up an instance of VUMT.

Some of the other daws have options to apply effects "destructively" and I would probably avoid that option altogether, if as you implied earlier, it might lead to a total re-rendering of the file.
Old 16th June 2019
  #37
A resounding NO to @ RedBaaron . Do not make a habit of putting vst effects post fader. That would mean changing the sound of the track every time you change it’s volume!

If the level level into an effect is too high, use a trim before it.
Old 16th June 2019
  #38
Lives for gear
 

Quote:
Originally Posted by RedBaaron View Post
It wouldn't be my first choice, although It's probably a worse idea in some daws than others. In Studio One, it keeps a volume curve on each file, visible there on the wav file, which you can drag up or down to increase the volume. I don't think it has to create a new file each time that's adjusted. In the past, I'd sometimes hit control + a to select all tracks and lowered them all at once with that feature--I tip I got from one of their help videos. I don't fall back on that option though, because if I'm having to reduce the volume, I'm going to want to know how close I'm getting to 0VU in the process, so I prefer just to load up an instance of VUMT.

Some of the other daws have options to apply effects "destructively" and I would probably avoid that option altogether, if as you implied earlier, it might lead to a total re-rendering of the file.
As I said; in most DAWs I've used "file" refers to something on a drive, not on a timeline. On the timeline we see "clips", "events", "regions" etc.

---

I just checked, and at least some people seem to refer to the above as "objects".
Old 16th June 2019
  #39
Lives for gear
 
RedBaaron's Avatar
Quote:
Originally Posted by mattiasnyc View Post
You're not really degrading the signal by changing the level using trim, clip-gain/event-volume or faders. I think it's worth separating the issue of what level changes does to the signal by themselves versus what plugins do to them.
Right. What I meant was that the plugins themselves will often degrade the quality if you're going into them too hot. And if you just lower the faders, and the effects are inserted Pre-Fader (the default behavior in Studio One, for example), that does not actually lower the volume of the signal going into the plugins until after the effects are applied. For example, compressors typically have a threshold volume target (often adjustable) at which processing occurs. So a hotter track will get more compression, and perhaps more artifacts. So if I found in S1 that some compressor effect on my track was distorting it, it wouldn't do me much good to try the lower the fader.


Quote:
Originally Posted by mattiasnyc View Post

I'm getting the feeling that by "wave file" you're talking about grabbing a "line" on a clip/event/region on the timeline, and you're thinking about that as a "wave file" because the clip/event/region displays a waveform.

So just to be super clear; a wave file is something quite different, and "normalizing" typically changes that wave file whereas clip-gain/event-volume etc does not typically change the file on the drive.

Well, that's fortunate that it doesn't cause a re-render. I had no idea whether it would re-render the .wav file in Samplitude. That's why I said "God I hope not." You mentioned " If that is a non-realtime process where you create new files in the process then I would highly recommend against that" and I was just saying I think that's a valid point, though it hinges on that speculative 'if' statement.

I'm understand there are some differences between normalizing and reducing the volume, although I can't say I heard a difference when normalizing to small target percentage in Samplitude. I've heard differences in normalizing to 100%, in say, Sound Forge back in the day. At any rate I didn't find another way of doing it directly. It's possible it just isn't named exactly right, or maybe I just didn't hear the difference. Either way, the OP has access to the software and can check that himself.




Quote:
Originally Posted by mattiasnyc View Post

I hope you don't mind that I disagree on this. I really think it's better to get in the habit of adjusting levels before plugins using the other methods and then using the faders to set the output levels post-inserts when mixing. I think that's a more traditional and more common workflow so there'll be less confusion when interacting with others this way.
NOt at all. (Freedom of speech, rock and roll and all I use Studio One so I prefer a trim plugin, as mentioned because of the metering. Others might not want to mess with a plugin so in that context, making the plugins post fader will at least allow a convenient way to back down the signal when it gets too hot. I think Sammy also has an input gain trim knob though. So there's that option. Functionally, as long as the gain is reduced before it hits the effects, I don't have a preference.
Old 16th June 2019
  #40
Lives for gear
 
RedBaaron's Avatar
Quote:
Originally Posted by thedberg View Post
A resounding NO to @ RedBaaron . Do not make a habit of putting vst effects post fader. That would mean changing the sound of the track every time you change it’s volume!

If the level level into an effect is too high, use a trim before it.
Well, that's the flipside of the coin. The advantage would be that you can easilly lower the gain if it gets too hot and starts distorting (like on a buss where you later send more stuff to). The disadvantage would be that once you do that, if you need to raise the volume later on, yet don't want more aggressive processing, you would then need to increase the output gain on the last plugin chain rather than increase the fader volume.

Functionally, effects can work the same way and equally well with either pre or post fader. It's just a question of whether the signal coming into them is too hot. However one avoids that, it accomplishes the same.
Old 16th June 2019
  #41
Lives for gear
 
RedBaaron's Avatar
To the OP: welcome to Gearslutz where even a seemingly simple question turns into a confusing game of "telephone."

Since I don't have all day to spend on this, I'll just refer you back to my original post: apart from the issues of signal levels coming into and out of the converter (a separate but related issue), what fundamentally matters to the sound quality is the signal level coming into each effect. There are numerous ways to keep it from coming into it too hot and degrading the sound, but moving the faders isn't always one of them (depends on whether it's pre or post fader). One of the simplest, tried-and-true methods is toa use trim plugin that also has an accurate VU metering before the effect. You'll have to double-check, but I believe the Samplitude mixer view also has an input trim knob that should accomplish the same--albeit without the advantage of knowing whether it's under 0VU.

Last edited by RedBaaron; 16th June 2019 at 11:32 PM..
Old 16th June 2019
  #42
Lives for gear
 
RedBaaron's Avatar
Quote:
Originally Posted by mattiasnyc View Post
As I said; in most DAWs I've used "file" refers to something on a drive, not on a timeline. On the timeline we see "clips", "events", "regions" etc.

---

I just checked, and at least some people seem to refer to the above as "objects".
Ok, some imprecision in my wording I suppose. How about: " In Studio One, it keeps a volume curve on each event, visible there on the visualization of the event." That should be clearer hopefully.
Old 16th June 2019
  #43
Here for the gear
 
yesyes's Avatar
"Now you're confusing me, and maybe some other people too. You don't typically use an Aux bus "on" tracks, you send or output from tracks/channels to it."
Sorry that was just due to the fact that I am not a native speaker, "on" tracks is what I meant.

Lowering the level of the AUX send and raising the threshold...I´m gonna try that next.
mattiasnyc: Don´t be sorry for all the text, it was very interesting for me - thanks a lot!

Ok, I checked out the pre/post fader options, I never really knew how to handle that. This is what worked best for me:
Put the AUX sends in the channels to prefader and start out with a send level of 0 dB. If I get way to much effect I just lower the volumefader of the AUX bus so that I get just a little too much effect. Than I turn down the AUX level send in the channel as much as desired.
Hope that makes sense. I do it like that because I believe the send level to the AUX bus is ok.
Alternatively I could leave the AUX bus volume fader at say 0 dB. But when I tested it, I had to turn the AUX send level in the channel down to -20 dB to get the same amount of effect.

Is it right that a plugin sounds better if it is fed with more power versus -20 dB? Because that´s what I think....I mean maybe they don´t all behave in the same way but talking about levels, it is probably not a good idea to feed a plugin with very low level, right?

In Samplitude I can switch every single AUX send effect pre/post in every channel. Means I can even do: Aux1 = pre, Aux2 = post.

Anyway I liked the pre send option more than the post one. If you go the post send way and change the volume fader of the channel it automatically changes the sound of the plugin. So for me pre send seems easier.

RedBaaron you said: "if you have a chain of effects, you could still be raising the gain in between effects, so you often need to check the metering before each one and adjust accordingly"
How do you know what level you want to feed the next plugin with? Is it experience? I mean like you know compX sounds well if it is fed with a certain dB level? If so a trim/metering plugin is a good idea I guess.

I always hated what limiters plugins did to my sound. I have now discovered Limiter6 and I love that one. But I read the manual before using it and watched a video - that helped a lot!

Normalizing a file in Samplitude can be done non destructive and I use that function sometimes.

Thank´s to you all and your very informative and inspiring postings!
I appreciate that very much!
Have a good day!
Old 16th June 2019
  #44
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by thedberg View Post
[...] If the level level into an effect is too high, use a trim before it.
I think that it is ALWAYS a good idea to use a trim at the first opportunity (if levels do not meet with what is needed for the gain-structure upon which you have decided).
.
Old 16th June 2019
  #45
To the OP:
I think this discussion got out of control and became confusing rather than clarifying. My advice:
1. Follow the guideline on the graph in post #6
2. Use default pre/post settings to start with (they are default for a reason)

Then stop thinking and start making music. Plugin levels are far, far less important than your art!
Old 16th June 2019
  #46
Lives for gear
 
RedBaaron's Avatar
Quote:
Originally Posted by yesyes View Post
[B][I]

RedBaaron you said: "if you have a chain of effects, you could still be raising the gain in between effects, so you often need to check the metering before each one and adjust accordingly"
How do you know what level you want to feed the next plugin with? Is it experience? I mean like you know compX sounds well if it is fed with a certain dB level? If so a trim/metering plugin is a good idea I guess.
I use the VUMT plugin. I try to trim the gain to get it back under 0VU before going into any other effect. There are various types of metering in DAWs. VU is actually an analog measurement. Metering in DAW's is long and confusing topic, but the bottom line is that VUMT does what I need. If you use chains of effects, there really isn't a better option I know of. Very light on CPU and inexpensive also. I do go out hotter to limiters and just use my ears there.

An about the whole pre vs. post fader thing: it doesn't matter which you use so long as you know how it works and don't overload the input , assuming that because you've lowered the output on the buss or aux channel, you're going to eliminate the distortion. That's true if the extra gain/distorition is being generated mostly from within the aux or buss effects chain, but not if you're just coming into it too hot to begin with.

Once you realize the difference in how post and prefader operate, and that plugins generally work best when they're not overloaded, then you can figure out what works best for your workflow, as far as where to put the plugins. As far as pre vs post goes, it could be more convenient one way or another. For example, someone mentioned post-fader changing the sound if you move the fader. That's true, but I always freeze my individual tracks to free up CPU, so it wouldn't be even a minor inconvenience for me--once frozen I could raise the fader without it changing the sound . The only thing I'd ever have effects stay on would be busses or auxes. There it was kind of convenient to be able to back off the fader a little when necessary . That was before I discovered VUMT was a more reliable way to guarantee I didn't do too much processing than just relying solely on my ears.

Anyways, those are all just workflow preference matters. I only mentioned the difference in pre vs. post behavior because it sounded like from your description as if yo had the effect on pre-fader and were lowering the fader, wondering why it was still sound distorted/overloaded. Disregard that if it's making things confusing in anyway. The relevant takeaway is: just don't go into your plugins too hot, into or out from the computer too hot, and you'll be good. I use 0VU to determine I'm not coming into plugins too hot, but different daws have different metering options, and others might prefer other ways.
Old 16th June 2019
  #47
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by RedBaaron View Post
[...] what fundamentally matters to the sound quality is the signal is coming into each effect.
Well, I think you are fundamentally correct in this observation.
...However, that does NOT mean that gain structure doesn't matter (especially from the standpoint of workflow!).

There are other (sometimes more important) factors:

Things can be ruined by levels hitting the initial A/D conversion too hard. (You'll never get back that loss.)

...Or any time when something goes back out to analog (the pooch can be permanently screwed).

But the good news is that (from a workflow perspective) working in digital is really a "gainstager's dream" (compared to how we used to have to do it when we only had about 70 or 80dB of dynamic range to work with on magnetic tape)!

...But also you have also additionally raised yet another very important issue:

Quote:
Originally Posted by RedBaaron View Post
[...] There are numerous ways to keep it from coming into it too hot and degrading the sound, but moving the faders isn't always one of them
This is very good advice!

This is crucial to "sane workflow" for many reasons, not the least of which is that because if you don't take this approach, you might just end up with faders in ridiculous positions, in that its a whole lot easier to use faders in the mix when they tend to be kinda close to the "target level" when they're somewhere close to "0". (It's a bitch at mix-time when you're trying to adjust a fader that's sitting at -24db.)
.
Old 17th June 2019
  #48
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by mattiasnyc View Post
[...] If all it took was imagination to define words and phrases then we wouldn't need dictionaries, so that complaint of yours makes little sense.
Well, I think that what I am saying here DOES make sense (though I don't really have any complaint).

Who ever said that "all it took was imagination to define words"? (It certainly was not me!)

I was simply making an observation, and one that is really the EXACT OPPOSITE of your perceived construction:

My observation was that communication though language depends largely upon a common understanding of the meanings of words (also known as "semantics"), rather than things like personal opinion or "imagination".

When there is no common ground as to these conventions, communication necessarily suffers (as often happens).

Forgive me (if you will), but I was just trying to correct inaccuracy (in order to advance the discussion at hand).
.
Old 17th June 2019
  #49
Lives for gear
 

Quote:
Originally Posted by 12ax7 View Post
when there is no common ground as to these conventions, communication necessarily suffers (as often happens).

Forgive me (if you will), but I was just trying to correct inaccuracy (in order to advance the discussion at hand).
There is common ground on this though, and it is that "realtime" is used when talking about DAWs and thus digital audio.

Pro Tools:
Realtime bounce vs. Offline bounce
Realtime processing vs. Audiosuite processing

Nuendo:
Realtime export vs. not
Plugins vs direct-offline-processing

..and so on. The word "realtime" is used counter to "offline" processing in a bunch of large DAWs, by engineers and by Avid, Steinberg, Magix etc themselves. It's an established term.

So, no, not inaccurate as far as I can see, exactly because there is common ground as to this convention. If you think there's something fundamentally wrong about this that is important to point out then you should really start a thread about it (I disagree with you btw, regardless of semantics as it relates to this thread).

PS: Can you stop with the 'indents'? It makes it a [email protected]#$ to quote and a bit annoying to read.
Old 17th June 2019
  #50
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by mattiasnyc View Post
There is common ground on this though, and it is that "realtime" is used when talking about DAWs and thus digital audio.
Well, not really...

...Because I happen to disagree (and therefore do not share this so-called "common ground" of which you speak):

I still maintain that NOTHING in a DAW has EVER happened in real time!
(And never will).
.
Old 17th June 2019
  #51
Lives for gear
 

Quote:
Originally Posted by 12ax7 View Post
Well, not really...

...Because I happen to disagree (and therefore do not share this so-called "common ground" of which you speak).

I still maintain that NOTHING in a DAW has EVER happened in real time!
(And never will).
.
Is this where we start discussing what the word "common" means then? Because hint: "the exception" doesn't make something "common"...

Go pick up a manual and look for the term, or look in forums. You'll find it. Plenty of times. Start here...
Old 17th June 2019
  #52
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by mattiasnyc View Post
[...] Because hint: "the exception" doesn't make something "common"...
What "exception" are you talking about? (Your word, not mine.)

Are you claiming that anything has ever truly happened "in real time" within a DAW? (Or maybe something else I don't understand?)

I was only only trying to point out that when you said "There is common ground on this though, and it is that "realtime" is used when talking about DAWs and thus digital audio.", I had a disagreement thusly:

This (so-called) "common ground" of which you speak does not exist, because others (like me) disagree!

If I disagree, then I am obviously NOT on "common ground", because I maintain that you can either have latency (as is ALWAYS the case with digital audio), or you can have "real time".

...But you cannot have both!
.
Old 17th June 2019
  #53
Lives for gear
 

I'm pretty sure most people here understand perfectly well what "realtime" means when juxtaposed with "offline", "audiosuite" etc. If you don't understand what people mean when they use that word in the relevant contexts then "engineering 101" for you.

Your objection is 100% meaningless in this thread because it changes nothing for the OP (other than add noise).
Old 17th June 2019
  #54
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by mattiasnyc View Post
Your objection is 100% meaningless in this thread because it changes nothing for the OP (other than add noise).
I don't have any objection.

I'm only making observations here.

I'm just trying to point out the meanings of the words people are using.

If that makes what I say "meaningless", then I guess we're all just doing nothing but inconveniencing a whole lotta electrons!
.
.
Old 17th June 2019
  #55
Lives for gear
 
12ax7's Avatar
 

Just to get back to the point I was trying to make in the first place:

All of this "real time" talk only came about after I responded to this post:

Quote:
Originally Posted by mattiasnyc View Post
If that is a non-realtime process where you create new files in the process then I would highly recommend against that.
...And the reason I first commented on it is that I just can't figure out how it would make any difference (as to quality) whether any digital process happened "in real time" or not.
.
Old 17th June 2019
  #56
Lives for gear
 

Quote:
Originally Posted by 12ax7 View Post
Just to get back to the point I was trying to make in the first place:

All of this "real time" talk only came about after I responded to this post:

...And the reason I first commented on it is that I just can't figure out how it would make any difference (as to quality) whether any digital process happened "in real time" or not.
.
What do you call a process that isn't offline/audiosuite yet processes audio in a DAW? Apparently we don't call it a realtime process, so in Nuendo and Cubase what is a non-offline audio export called? And what do you call the use of a plugin in Pro Tools that isn't an Audiosuite process?

The word I and other people would use is "realtime".

What word do you use for that?
Old 17th June 2019
  #57
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by mattiasnyc View Post
What do you call a process that isn't offline/audiosuite yet processes audio in a DAW? Apparently we don't call it a realtime process, so in Nuendo and Cubase what is a non-offline audio export called? And what do you call the use of a plugin in Pro Tools that isn't an Audiosuite process?

The word I and other people would use is "realtime".

What word do you use for that?
...Okay, fine. (Sorry):

You just happened to stumble upon a pet peeve of mine.
(It don't make either of us "bad people", or anything!)
...So moving along:

Can you please explain WHY you recommend against this "non-realtime process where you create new files in the process"?

I just don't understand why you think this would necessarily be a bad thing.
.
Old 17th June 2019
  #58
Lives for gear
 

Quote:
Originally Posted by 12ax7 View Post
I just don't understand why you think this would necessarily be a bad thing.
.
- It increases used disk space

- It often means clips/events/regions get consolidated and the ability to use "handles" disappear (i.e. a pain when re-editing). In other words the beginning and end of a region can no longer be extended.

- It needlessly does processing to the actual (new) audio file in which that processing is then written. If people start boosting / cutting the same piece of audio multiple times in succession using this method in fixed point processing there's actually a risk of degrading audio a bit. It's inherently different from a realtime floating point operation.

- Many DAWs (PT) won't allow you to undo this action on individual clips/events/regions. The undo-queue is sequential across the entire DAW, so if you execute an Audiosuite process, then do some midi editing, then move some faders in the mixer, the only way to undo the Audiosuite process would be to hit the "undo" function until it's undone which will go backwards through that process undoing mixer moves first, then the midi editing, and then Audiosuite. So you end up losing the other processing as well (Nuendo won't suffer from this).

I believe I said most of the above already.
Old 17th June 2019
  #59
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by mattiasnyc View Post
[...] If people start boosting / cutting the same piece of audio multiple times in succession using this method in fixed point processing there's actually a risk of degrading audio a bit. It's inherently different from a realtime floating point operation.
Ah...

...There it is!

...So (as far as preserving "quality"), this really has more to do with manipulation within a fixed-point environment than with any effect caused by the process happening "in realtime", eh?

...I mean, the same potential "damage" would happen with fixed point anyway (no matter how much time had passed in the process), right?
(If this is correct, then I guess I've found the point of my confusion.)
.
Old 17th June 2019
  #60
Lives for gear
 

Quote:
Originally Posted by 12ax7 View Post
Ah...

...There it is!

...So (as far as preserving "quality"), this really has more to do with manipulation within a fixed-point environment than with any effect caused by the process happening "in realtime", eh?

...I mean, the same potential "damage" would happen with fixed point anyway (no matter how much time had passed in the process), right?
(If this is correct, then I guess I've found the point of my confusion.)
.
You have to look at the context.

The context was "normalize".

"Normalize" for the longest time, as far as I can recall even, meant exactly what I said;

1. find peak
2. measure difference between peak and desired peak (normalized value)
3. raise/lower all of the file by that amount

and all of that happened offline, non-realtime, resulting in new audio files.

Since most people still record and store digital audio as fixed point data you end up with fixed point processing.... but that wasn't the bigger issue, the bigger issues were all of the other points I made.

I've said this many times now. I'm not sure I can make it clearer. It's just bad practice in my opinion to use "destructive" offline normalization in 2019 when we don't really have to do that. There are better options.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump