Go Back   Cockos Incorporated Forums > REAPER Forums > REAPER Q&A, Tips, Tricks and Howto

Reply
 
Thread Tools Display Modes
Old 12-17-2021, 02:12 PM   #81
DrFrankencopter
Human being with feelings
 
Join Date: Jan 2011
Location: Ottawa, Canada
Posts: 289
Default

Quote:
Originally Posted by ashcat_lt View Post
Trim envelope sits at the same point in the chain as the Volume fader. It is always post FX and basically doesn’t do any good in this situation. PreFX Volume envelope would work, and if you put it in its own lane, it will have a knob in the TCP.

I agree with Tod that Item/Take volume is probably the “best” place to handle these things.

None of that actually helps between plugins, of course.
Trim envelope, post FX, is 'exactly' what I want to be able to adjust.

My (ideal) general workflow is sort of like this:
1. Adjust clip gains/Item vol to balance out performance variations...like evening out a word here & there. Done using ears, and not meters.
2. Set up a static mix using the trim knob, and all faders at 0 dB (or maybe something like -10 dB to give move upward room). This is done using ears...just like you do on a mix (e.g. Kick drum is way higher than the banjo, or whatever), except you balance the instruments with the faders all in one spot.
3. Now move onto the mix itself...plugins, FX, automation, etc.

At this point, assuming you don't do drastic level (output vs input) changes with the plugins you insert your faders will show you directly how your mix is evolving relative to your starting point. 0dB on the fader now actually means something. Plus, this puts the fader into it's highest resolution area which makes subtle adjustments much easier/faster.

Now I have a baseline... 0dB on my fader gets me back to my reference point.

Also handy if you 'mix' while you track. Need a slightly different monitor mix while tracking a part. No prob, just grab the fader and get the part tracked. When you're done, shove it back to 0 dB. Want to decide on different instrument balances while going through the tracking process; just re-adjust your trims.
__________________
RME TotalMixFX Actions for Reaper here: https://stash.reaper.fm/v/29339/reape...MixOSC_x64.dll
DrFrankencopter is offline   Reply With Quote
Old 12-17-2021, 02:29 PM   #82
ashcat_lt
Human being with feelings
 
Join Date: Dec 2012
Posts: 6,511
Default

That’s fine if it works for you, I guess, but in context of this thread, each track is still going to be hitting it’s plugins at different levels. That trim envelope can help you keep your faders near 0, but doesn’t help any of the other aspects of gain staging that we’ve been discussing.

To the extent that keeping my faders near 0 matters to me, I usually figure that out in the plugin chain somewhere.
ashcat_lt is offline   Reply With Quote
Old 12-17-2021, 02:38 PM   #83
Pashkuli
Human being with feelings
 
Pashkuli's Avatar
 
Join Date: Jul 2006
Location: United Kingdom, T. Wells
Posts: 2,255
Default

Quote:
Originally Posted by Tod View Post
That's not a bad idea, one volume control that can be toggled "Trim" or "Volume".
Not bad but idea, though might lead into confusion.
Simple input Gain for each track by default would be nice.

For so many years, I've been wondering why in Reaper TCP and MCP there is no Trim\Gain knob or slider. Something like this:





Should not be that difficult for the devs to implement it.



__________________
♦ video → .: Pashkuli Keyboard :.
♦ instagram → @pashkuli.keyboard
Pashkuli is offline   Reply With Quote
Old 12-17-2021, 02:50 PM   #84
DrFrankencopter
Human being with feelings
 
Join Date: Jan 2011
Location: Ottawa, Canada
Posts: 289
Default

Quote:
Originally Posted by ashcat_lt View Post
That’s fine if it works for you, I guess, but in context of this thread, each track is still going to be hitting it’s plugins at different levels. That trim envelope can help you keep your faders near 0, but doesn’t help any of the other aspects of gain staging that we’ve been discussing.
Totally understood...and didn't mean to muddy the waters. In the context of this thread, if gain staging your plugins is important to you then yes it needs to be done at the start of the plugin chain (or even before), and managed as the signal passes from plugin to plugin by setting the output knob control of each plugin in the chain appropriately.

It would be pretty easy to set up a template that has a gain knob as the first plugin. This might be a suitable (and free) candidate: https://www.sonalksis.com/freeg.html

And, it's already set up with 0 at -18 dBFS...

Unfortunately, what isn't possible (yet) in Reaper is pinning a plugin to the bottom of the chain, now easily accessing the trim volume control for Post FX leveling.
__________________
RME TotalMixFX Actions for Reaper here: https://stash.reaper.fm/v/29339/reape...MixOSC_x64.dll
DrFrankencopter is offline   Reply With Quote
Old 12-17-2021, 03:58 PM   #85
Pashkuli
Human being with feelings
 
Pashkuli's Avatar
 
Join Date: Jul 2006
Location: United Kingdom, T. Wells
Posts: 2,255
Default

Quote:
Originally Posted by DrFrankencopter View Post
It would be pretty easy to set up a template that has a gain knob as the first plugin. This might be a suitable (and free) candidate: https://www.sonalksis.com/freeg.html

And, it's already set up with 0 at -18 dBFS...

Unfortunately, what isn't possible (yet) in Reaper is pinning a plugin to the bottom of the chain, now easily accessing the trim volume control for Post FX leveling.
Oh, that is true. Reaper does not have post-fx plugin section on tracks. I think Cubase and Samplitude have it.

This -18dBfs does not mean anything if I can't have a plugin or chain of plugins, to allow me to push up the RMS to

-18···-17···-16 | dBfs

and to see my peaks getting compressed, saturated and gently EQd.

Been trying this for a while, still no satisfying result.
Either get too much pump, too much distortion but the worst is the clipping is there... so have to use clipper (somehow mild and oversampled to the max) and if that does not do it... limiter at the end but it squares (almost) the result at the limiter's ceiling level.

With analogue do not have to even think about this thing. Most of this I get almost for free.
__________________
♦ video → .: Pashkuli Keyboard :.
♦ instagram → @pashkuli.keyboard
Pashkuli is offline   Reply With Quote
Old 12-17-2021, 04:04 PM   #86
Judders
Human being with feelings
 
Join Date: Aug 2014
Posts: 10,551
Default

The trim on items makes a trim knob on tracks redundant. Not that it would hurt if it could be hidden.
Judders is offline   Reply With Quote
Old 12-17-2021, 04:19 PM   #87
ashcat_lt
Human being with feelings
 
Join Date: Dec 2012
Posts: 6,511
Default

Quote:
Originally Posted by Judders View Post
The trim on items makes a trim knob on tracks redundant.
A hypothetical new PreFX trim knob would be mostly redundant, yes, except it wouldn’t help in like live input situations, or for VSTi tracks, or any number of other cases. It’s also a little weirder when you’ve got multiple items on the same track that all want different amounts of gain. Honestly, the “easy mode” for PreFX gain staging of audio items is to select items and use the SWS loudness normalize actions.
ashcat_lt is offline   Reply With Quote
Old 12-17-2021, 04:19 PM   #88
DrFrankencopter
Human being with feelings
 
Join Date: Jan 2011
Location: Ottawa, Canada
Posts: 289
Default

Quote:
Originally Posted by Judders View Post
The trim on items makes a trim knob on tracks redundant. Not that it would hurt if it could be hidden.
Not quite...for at least the following reasons:
1. You can have many items on a single track...wouldn't it be easier to just set one knob instead of changing volumes on each item? And what happens if you decide you want to make them all +1 dB louder?
2. Item trim is Pre FX....what if you want to control the overall volume going into your fader (post fx)? Currently your only option is to use the output volume of the last plugin in line. Sure, you can probably map that to a control on the TCP, but if you add a new plugin to the chain you need to re-do the mapping.

The funny thing is that there is already a post FX trim function in Reaper...and it has an envelope, but the Dev's never assigned a knob/slider to it, so you can only adjust it using envelopes.
__________________
RME TotalMixFX Actions for Reaper here: https://stash.reaper.fm/v/29339/reape...MixOSC_x64.dll
DrFrankencopter is offline   Reply With Quote
Old 12-17-2021, 04:20 PM   #89
Pashkuli
Human being with feelings
 
Pashkuli's Avatar
 
Join Date: Jul 2006
Location: United Kingdom, T. Wells
Posts: 2,255
Default

Quote:
Originally Posted by Judders View Post
The trim on items makes a trim knob on tracks redundant. Not that it would hurt if it could be hidden.
Not so fast... You can split items and have multiple independent trim knob on those items.
The trim\gain knob on the TCP and MCP respectively will act as a global trim for all items in the track.
For example on chopping vocal in phrases (separate items) before hitting the track with a compressor.

It's been more than a decade actually and still no trim knob.
__________________
♦ video → .: Pashkuli Keyboard :.
♦ instagram → @pashkuli.keyboard
Pashkuli is offline   Reply With Quote
Old 12-17-2021, 04:23 PM   #90
Pashkuli
Human being with feelings
 
Pashkuli's Avatar
 
Join Date: Jul 2006
Location: United Kingdom, T. Wells
Posts: 2,255
Default

Quote:
Originally Posted by DrFrankencopter View Post
The funny thing is that there is already a post FX trim function in Reaper...and it has an envelope, but the Dev's never assigned a knob/slider to it, so you can only adjust it using envelopes.
Where is that, please? I must have missed it.

But nevertheless a knob or post-FX slot (last in chain) section would be even better! And a TCP\MCP trim knob, of course.
__________________
♦ video → .: Pashkuli Keyboard :.
♦ instagram → @pashkuli.keyboard
Pashkuli is offline   Reply With Quote
Old 12-17-2021, 04:29 PM   #91
ashcat_lt
Human being with feelings
 
Join Date: Dec 2012
Posts: 6,511
Default

Quote:
Originally Posted by DrFrankencopter View Post
...and it has an envelope, but the Dev's never assigned a knob/slider to it, so you can only adjust it using envelopes.
Actually if you have that envelope enabled and visible in its own lane, there IS a knob in in the TCP. I’m not sure that helps much, but...
ashcat_lt is offline   Reply With Quote
Old 12-17-2021, 04:40 PM   #92
Judders
Human being with feelings
 
Join Date: Aug 2014
Posts: 10,551
Default

Quote:
Originally Posted by Pashkuli View Post
Not so fast... You can split items and have multiple independent trim knob on those items.
The trim\gain knob on the TCP and MCP respectively will act as a global trim for all items in the track.
For example on chopping vocal in phrases (separate items) before hitting the track with a compressor.

It's been more than a decade actually and still no trim knob.
I often do it. You double click the track header and adjust any item's gain, they all change by the same amount. The only thing that doesn't work by selecting multiple items is double clicking to bring them back to unity.
Judders is offline   Reply With Quote
Old 12-17-2021, 04:43 PM   #93
Judders
Human being with feelings
 
Join Date: Aug 2014
Posts: 10,551
Default

Quote:
Originally Posted by ashcat_lt View Post
Actually if you have that envelope enabled and visible in its own lane, there IS a knob in in the TCP. I’m not sure that helps much, but...
Is it something you could do with parameter modulation? Stick a gain plugin on unused I/O channels, set it 1:1 with pre or post volume envelope, then add a knob to track controls?
Judders is offline   Reply With Quote
Old 12-17-2021, 04:48 PM   #94
DrFrankencopter
Human being with feelings
 
Join Date: Jan 2011
Location: Ottawa, Canada
Posts: 289
Default

Quote:
Originally Posted by ashcat_lt View Post
Honestly, the “easy mode” for PreFX gain staging of audio items is to select items and use the SWS loudness normalize actions.
I often split items and use item volume as a form of level control prior to starting a mix. It's part of the editing stage for me. I don't think I'd want my items that I just split up all auto leveled.

Maybe the workflow is to let the SWS normalize do its thing, and then split and adjust afterwards. But then again, how would it handle, for example, some takes that were done on say a soft vocal intro relative to the takes that were done on the hard hitting/belted bridge. Pretty sure I wouldn't want both those items at -23 LUFs and I'd end up turning the intro down somewhere else. Alternatively, you could glue all the items into one and then let SWS do it's magic, but if you want to fine tune the results you'd need to re-split them.

That said, I think this approach would work well for continuous takes/items.
__________________
RME TotalMixFX Actions for Reaper here: https://stash.reaper.fm/v/29339/reape...MixOSC_x64.dll
DrFrankencopter is offline   Reply With Quote
Old 12-17-2021, 04:49 PM   #95
DrFrankencopter
Human being with feelings
 
Join Date: Jan 2011
Location: Ottawa, Canada
Posts: 289
Default

Quote:
Originally Posted by ashcat_lt View Post
Actually if you have that envelope enabled and visible in its own lane, there IS a knob in in the TCP. I’m not sure that helps much, but...
Haha...never thought of that. Kinda kludge-ey, but should work.
__________________
RME TotalMixFX Actions for Reaper here: https://stash.reaper.fm/v/29339/reape...MixOSC_x64.dll
DrFrankencopter is offline   Reply With Quote
Old 12-17-2021, 04:51 PM   #96
DrFrankencopter
Human being with feelings
 
Join Date: Jan 2011
Location: Ottawa, Canada
Posts: 289
Default

Quote:
Originally Posted by Pashkuli View Post
Where is that, please? I must have missed it.

But nevertheless a knob or post-FX slot (last in chain) section would be even better! And a TCP\MCP trim knob, of course.
It's in the first section of track Envelopes (where you also find volume/pan/mute)..."Trim". The envelope shows up a a magenta line in the envelope window.

+1 on post FX slot & TCP/MCP trim.
__________________
RME TotalMixFX Actions for Reaper here: https://stash.reaper.fm/v/29339/reape...MixOSC_x64.dll
DrFrankencopter is offline   Reply With Quote
Old 12-18-2021, 10:28 AM   #97
ashcat_lt
Human being with feelings
 
Join Date: Dec 2012
Posts: 6,511
Default

Quote:
Originally Posted by ashcat_lt View Post
...but it would be easy enough to try.
...so I did, and found out I was kinda wrong, but that it's actually better. SWS has loudness normalize for both items and tracks.

If you run the item one, they are all analyzed and Item Volume is adjusted individually. Honestly, I didn't check to see if it would preserve previous settings because that wasn't what we were looking for anyway. Probably doesn't, though.

But if you run the one for Selected Tracks, it analyzes the entire track, and applies a single adjustment to the track's PreFX Volume envelope. Doesn't even touch Item Volume. That makes it almost too easy. Select all tracks, normalize, gain staging complete.
ashcat_lt is offline   Reply With Quote
Old 12-18-2021, 11:07 AM   #98
ashcat_lt
Human being with feelings
 
Join Date: Dec 2012
Posts: 6,511
Default

Quote:
Originally Posted by Judders View Post
Is it something you could do with parameter modulation? Stick a gain plugin on unused I/O channels, set it 1:1 with pre or post volume envelope, then add a knob to track controls?
Sorry I didn't see this earlier, but first I think you'd end up needing something like ReaLearn to actually connect a JS slider to these envelopes because they're not valid targets for PM/Link.

But actually just why in the TF wouldn't you use that plugin slot for an actual gain control and be done with it. You can just as easily write or draw an envelope for that. Make it a visible control in the TCP/MCP. Like, it doesn't help.

Note that I had to look up that normalize stuff because I never use it. I do the item one sometimes when I get client stuff that's all over the place, just to have a starting spot, but kind of only if the static mix needs it.

Otherwise, though, I've always got some point of adjustment in the plugin chain anyway, so I don't stress about it. I personally don't use these kinds of plugins that hide filters and nonlinearities in places that I can't see or touch them. I build them myself out of ReaEQ and ReaComp.

In fact, my Track Trim knob is actually the Threshold of the ReaComp "rail" saturator which is first in about every chain. I literally just bring the limit toward the signal, rather than trying to push the signal toward the limit. I can see right there on the gain reduction meter exactly how much I'm digging into the saturation curve. Of course, I'm also listening, but for rough-sketch-first-pass, I totally just do it by eye. Looking at the peak levels and how close they get to the limit. Average doesn't matter, and VU for this is stupid. With Auto-makeup on, it does pretty much the same thing as adjusting gain before the plugin and leaving the threshold in one place. At least until you start to pull the threshold way down. Auto-makeup doesn't actually work the way I expect it to, and I'm not sure if that's a new thing. It works well enough in most cases, and a little extra manual compensation using the Wet slider won't kill me either.

ReaEQ doesn't have amplitude limits, so I literally don't care what the level is when it gets there. It does have an output gain slider, though, which can be used to compensate levels before going to whatever comes next. That can't be automated, though. Sometimes what I'll do is put in a shelf filter that covers the entire spectrum (high at 20Hz or low at 20KHz), and map the gain of that to whatever as an overall gain control.

Very often ReaComp will also be the last thing in the chain. It's Wet control ultimately becomes my "PostFX" gain, and I adjust it so that it sits where I want on the fader.

But if I really want to be absolutely sure that no single sample will ever come out of this track higher than the limit I've set (plus the fader and whatever of course), ReaComp has to be the very last thing. You don't get to put a filter after it. No anti-aliasing or tone shaping or anything.

Because yes, if you filter after the clipping, you end with overshoots. In fact, any nonlinearity with any time component is pretty much incapable of guaranteeing a hard limit. Believe me I know it can be frustrating. Drove me nuts for years. Frankly it only matters in the (unfortunately more and more common) case where your deliverable absolutely must meet some super rigid specification. Otherwise, what's the issue? Render it. Normalize to 0. Turn it down a bit to allow for ISPs. Move on. But yes, those overshoots can actually lower our integrated loudness. We could get the average that db or so louder if only we didn't have that one sample... So I just shave them off with ReaComp and live with the usually inaudible "artifacts". For the record it happens in analog too, and for the same reasons, but we don't see it because we normally don't have digital meters with sticky clip indicators. It's worse in analog, though, because there's always a filter to fuck up your clipper and basically no good way around it with real world components. So like kinda the fact that plugins model that behavior...

Last edited by ashcat_lt; 12-18-2021 at 11:12 AM.
ashcat_lt is offline   Reply With Quote
Old 12-18-2021, 02:22 PM   #99
Judders
Human being with feelings
 
Join Date: Aug 2014
Posts: 10,551
Default

Quote:
Originally Posted by ashcat_lt View Post
Sorry I didn't see this earlier, but first I think you'd end up needing something like ReaLearn to actually connect a JS slider to these envelopes because they're not valid targets for PM/Link.

But actually just why in the TF wouldn't you use that plugin slot for an actual gain control and be done with it. You can just as easily write or draw an envelope for that. Make it a visible control in the TCP/MCP. Like, it doesn't help.
Sure, I was just trying to think of options for the people in the thread asking for it.
Judders is offline   Reply With Quote
Old 12-18-2021, 06:49 PM   #100
xpander
Human being with feelings
 
xpander's Avatar
 
Join Date: Jun 2007
Location: Terra incognita
Posts: 7,044
Default

Quote:
Originally Posted by Pashkuli View Post
Why they even bother, but I guess "marketing dept" has to come up with something to justify the "painstaking work".

It would be nice to have someone who has worked in "analogue plugins development" to explain in depth, whether such "painstaking work" really is such so to give us 1:1 analogue behaviour.

I know Boz Millar (Boz Digital Labs), Shane McFee (Kazrog) and Thomas (TBProAudio) write in the forum some times, but probably not here.
Fabien Schivre from TDR comments every once in a while in Gearspace, sometimes even about these issues. Him and some other developers have mentioned that we don't have enough computing power to do it, especially realtime. Even simple components could be too much, not to even mention interactions between dozens or hundreds of them. I only build in analog realm, but that answer doesn't surprise me at all, quite the contrary.

Quote:
Originally Posted by Lynx_TWO View Post
I had to lower each vocal track by 12dB to avoid massive distortion even with Reaper's 64-bit processing, so yea, some plugins cannot handle input that's too loud.
Softubes very popular free Saturation Knob was like that, until they updated it just couple of months ago to finally have input and output gain controls. The old version with only one saturation knob and no level adjustments available really needed some extra care to be usable. Depending on material, Saturation knob had a workable full range if average levels were kept somewhere between -20 and -12 dBFS. Levels below that and even with the knob at max there was some saturation but it was not really breaking up hard. Levels higher than that and saturation indicator was already full on with knob at 4.
xpander is offline   Reply With Quote
Old 12-19-2021, 05:07 AM   #101
Pashkuli
Human being with feelings
 
Pashkuli's Avatar
 
Join Date: Jul 2006
Location: United Kingdom, T. Wells
Posts: 2,255
Default

Quote:
Originally Posted by xpander View Post
Fabien Schivre from TDR comments every once in a while... Him and some other developers have mentioned that we don't have enough computing power to do it, especially real time.
It is very well known. The commodity of "digital realm" today is unsurmountable.


For what I would love to see as "analogued" is :
non-linear amplification, compression, tape\tube saturation, EQ roll-off etc. to be effective within certain time span and amplitude (not the whole signal). But that will certainly lead to phasing issues.

For example, compressor with an attack\release (would function as a pseudo-gate), but then only what gets above the threshold to be "analogued" with priority towards Peaks.

That is all.

I've been mingling with some "analogue" channel strip plugins lately (SSL and other brands) but still can not make them work in the above required way. Any help would be appreciated, though.
__________________
♦ video → .: Pashkuli Keyboard :.
♦ instagram → @pashkuli.keyboard
Pashkuli is offline   Reply With Quote
Old 12-21-2021, 05:21 PM   #102
Pashkuli
Human being with feelings
 
Pashkuli's Avatar
 
Join Date: Jul 2006
Location: United Kingdom, T. Wells
Posts: 2,255
Default

There are many people who do not gain properly their input recordings and avoid clipping the Peaks on the way in.
They just record too low and boost in during mixing, not even doing gain-staging, which already would have been useless.

Why? For the beginners and some "gurus" out there, here:

__________________
♦ video → .: Pashkuli Keyboard :.
♦ instagram → @pashkuli.keyboard
Pashkuli is offline   Reply With Quote
Old 12-21-2021, 06:24 PM   #103
White Tie
Pixel Pusher
 
White Tie's Avatar
 
Join Date: Mar 2007
Location: Blighty
Posts: 4,012
Default

Nothing in that video has anything to do with this thread. This is a forum, its not your facebook.
__________________
The House of White Tie
White Tie is offline   Reply With Quote
Old 12-22-2021, 12:26 PM   #104
rMidi
Human being with feelings
 
Join Date: Dec 2016
Posts: 47
Default

Thank you all for your input / opinions.

Though I am not really a 'newbie', I tend to ask my questions in a somewhat naive way which seems to elicit the most and varied responses,
and boy, did I get a lot of very varied opinions and responses. We have certainly been thrown far afield from my original query, but I
do enjoy reading other opinions.

Since my last post I have abandoned the Waves VU meter. I am now using the (free) TBProAudio mvMeter 2.
-- Wave VU Meter did not act as I expected it to.
-- This is hard to believe as a VU meter is one of the simplest plugins to create. Perhaps my issue is'Site Specific' (An unflattering term used by programmers, meaning, 'The user is an idiot').
-- TBProAudio mvMeter 2 acts as expected.

What started me down the VU meter path is my issues with MODO Bass (see comments to Pashkuli below). Since this topic has morphed into purely a 'gain staging' thread I will play along. I am using other plugins for my examples and ignoring my MODO Bass issue.

There seem to be several camps here (Ala PC vs. Mac, Fender vs. Gibson, Global warming vs. denier, ...).
It seems to be more of a religion than anything else.:
-- -18dbfs is not relevant (use your ears, it's a waste of time, emulations of analog hardware in a digital plugin are nonsense).
-- -18dbfs may not be relevant, but it does not hurt. These seem to believe that gain staging is a good practice to start your mix. I am firmly within this camp!

-18dBfs.
I have always gain stage to start a mix. I am not talking about clipping here, but rather am referring to saturation. Besides gain staging tracks / busses as to not overload the master (clipping) I always gain stage between plugins (effects) on individual tracks.
I.e., a simple guitar track:
-- Fader 0.0db.
-- Recording: -18dbfs RMS
-- -18dBfs in, Virtual Tape (Has a a pair of VU Meters on it. Input / Output), -18dBfs out.
-- -18dBfs in, Neve 8048 Console channel (Has VU Meter on it), -18dBfs out.
-- -18dBfs in, Neve 1073 EQ (NO VU Meter), -18dBfsout.
-- -18dBfs in, UREI 1176 Rev A compressor (Has VU Meter on it), -18dBfs put.

The last of these three plugins live within the Slate Digital
'Mix Rack'. When gain staging these plugins, I would simply drop a 'Trimmer' (which has a VU meter) into the mix rack. I would then move it around (before / after plugins) to level everything to 0 dBVU.
Sometimes I may do something in the mix like hit the output of the Virtual Tape really hard. Since my Mix Rack follows the tape machine in the chain, I would simply insert a trimmer before the console and bring it back down to 0 dBVU. This retains the saturation from the tape machine without over driving the console.

I am a big Slate Digital fan.
Note: This statement is not intended to start a Slate Digital Good / Bad flood of posts, it is simply a fact.

Most of the Slate Digital stuff (besides their recent foray into EDM / Hip Hop with things like MO-TT, Murda, MetaTune,..) are digital models of analog hardware.
-- Yes, Steven Slate seems like a used car salesman to me. He is a very slick marketing type, but he does not write the code.
-- On the other hand, I tend to trust what Fabrice Gabriel has to say (probably more so than someone from Waves, perhaps not as much as someone from UAD).

Did they do a good job of modeling the analog hardware? Again, that is a matter of
opinion, let's keep it out of this thread. The real question is can someone digitally model a 1952 hand wired Pultec EQP-1 100% accurately?
Of course not. Can they come close? Maybe.

Whether or not analog emulations are good / accurate is not relevant to me. The fact of the matter is that a significant number of analog modeling plugins are in fact sensitive to the input level. So, gain staging plugins is a sort of a 'know your hardware' thing with me.

What started me down the VU meter path is that I of course utilize non-Slate Digital plugins. Since they cannot live within the 'Mix Rack' I would gain stage them to -18dbFS RMS by disabling all plugins in the chain after the plugin and simply use the Reaper digital track meter. Since I am comfortable with VU meters, I thought I would get a VU meter plugin and use that instead.

I suppose it is also important to point out that I record and mix live instruments save for MODO bass which I use to write the bass lines. I would typically then get a real bass player to do the bass. Sometimes I am left with the MODO bass
for my bass tracks because I cannot find a bass player to do the recording.

I am not doing EDM / Hip Hop, MIDI / Sample driven things which I am sure is a different animal.


In closing, when it comes to gain staging to -18dBFS RMS I will continue to do so.
It does not take me all that long to do and I always prefer a solid / known starting point.

//================================
Pashkuli,

I apologize that I have not responded to some of your queries about MODO Bass.

MODO Bass tends to 'load up'. Especially on Legato slides and it is prevalent with the Rickenbacker model.
It is not unusual for it to swing from -6dBVU to +2dBVU on a legato slide, pitch wheel and sometimes even the initial note on a string other than the string the last note was played on.

Adjusting velocity on a note / series of notes does not fix this.

When this occurs, besides simply being too loud, it will mask my Kick. Yes, I know how to utilize side chain compression to prevent the kick masking but doing
so never sounds good to my ear (cheesy!).

I have tried things like Waves Bass Ridder to level the MODO Bass track, but I have found I need to limit the max/mins to a point (in order to get rid of the load ups) that all the dynamics are then removed from the track. Yes, I could automate Bass Rider to turn it on only at the points where MODO Bass loads up, but this approach is no difference then simple automation in the volume lane for the track, so why should I bother with Bass Rider?

//====================================
White Tie,

If you could learn to contain yourself, be more tactful, and not treat this forum as a place to go and vent your frustrations perhaps I would lend some credence to the things you are saying here. Your attitude (bad) causes me to simply ignore anything you have to say.

I am a C/C++ programmer with over 30 years of experience (I am not a kid. coding is my trade not my hobby) I have done a significant amount of Audio programming. I am not all that impressed with your attempts to somehow 'show off' your knowledge.

I also am very opinionated, believe in what I believe adamantly. My beliefs are based on my technical knowledge. I am very well aware that I could in fact be wrong on all things, you seem to lack this belief / realization about yourself.

I am pretty sure the only 'fact' I have stated in my post is this:
'The fact of the matter is that a significant number of analog modeling plugins are in fact sensitive to the input level.' I cannot even imagine that one could argue with this point.

The rest of my post is simply my workflow and rational behind it. You have already stated that you believe that gain staging is not relevant and is a waste of time. We get that and understand you are of that opinion. No need to repeat yourself.

I am sure you realize that you owe Pashkuli an apology for your 'facebook' comment.

I am not saying that I consider yourself to be a Troll, but others may consider yourself to be so.
//==================================================

Cheers.
rMidi is offline   Reply With Quote
Old 12-22-2021, 01:30 PM   #105
Judders
Human being with feelings
 
Join Date: Aug 2014
Posts: 10,551
Default

Quote:
Originally Posted by rMidi View Post
Thank you all for your input / opinions.

Though I am not really a 'newbie', I tend to ask my questions in a somewhat naive way which seems to elicit the most and varied responses,
and boy, did I get a lot of very varied opinions and responses. We have certainly been thrown far afield from my original query, but I
do enjoy reading other opinions.

Since my last post I have abandoned the Waves VU meter. I am now using the (free) TBProAudio mvMeter 2.
-- Wave VU Meter did not act as I expected it to.
-- This is hard to believe as a VU meter is one of the simplest plugins to create. Perhaps my issue is'Site Specific' (An unflattering term used by programmers, meaning, 'The user is an idiot').
-- TBProAudio mvMeter 2 acts as expected.

What started me down the VU meter path is my issues with MODO Bass (see comments to Pashkuli below). Since this topic has morphed into purely a 'gain staging' thread I will play along. I am using other plugins for my examples and ignoring my MODO Bass issue.

There seem to be several camps here (Ala PC vs. Mac, Fender vs. Gibson, Global warming vs. denier, ...).
It seems to be more of a religion than anything else.:
-- -18dbfs is not relevant (use your ears, it's a waste of time, emulations of analog hardware in a digital plugin are nonsense).
-- -18dbfs may not be relevant, but it does not hurt. These seem to believe that gain staging is a good practice to start your mix. I am firmly within this camp!

-18dBfs.
I have always gain stage to start a mix. I am not talking about clipping here, but rather am referring to saturation. Besides gain staging tracks / busses as to not overload the master (clipping) I always gain stage between plugins (effects) on individual tracks.
I.e., a simple guitar track:
-- Fader 0.0db.
-- Recording: -18dbfs RMS
-- -18dBfs in, Virtual Tape (Has a a pair of VU Meters on it. Input / Output), -18dBfs out.
-- -18dBfs in, Neve 8048 Console channel (Has VU Meter on it), -18dBfs out.
-- -18dBfs in, Neve 1073 EQ (NO VU Meter), -18dBfsout.
-- -18dBfs in, UREI 1176 Rev A compressor (Has VU Meter on it), -18dBfs put.

The last of these three plugins live within the Slate Digital
'Mix Rack'. When gain staging these plugins, I would simply drop a 'Trimmer' (which has a VU meter) into the mix rack. I would then move it around (before / after plugins) to level everything to 0 dBVU.
Sometimes I may do something in the mix like hit the output of the Virtual Tape really hard. Since my Mix Rack follows the tape machine in the chain, I would simply insert a trimmer before the console and bring it back down to 0 dBVU. This retains the saturation from the tape machine without over driving the console.

I am a big Slate Digital fan.
Note: This statement is not intended to start a Slate Digital Good / Bad flood of posts, it is simply a fact.

Most of the Slate Digital stuff (besides their recent foray into EDM / Hip Hop with things like MO-TT, Murda, MetaTune,..) are digital models of analog hardware.
-- Yes, Steven Slate seems like a used car salesman to me. He is a very slick marketing type, but he does not write the code.
-- On the other hand, I tend to trust what Fabrice Gabriel has to say (probably more so than someone from Waves, perhaps not as much as someone from UAD).

Did they do a good job of modeling the analog hardware? Again, that is a matter of
opinion, let's keep it out of this thread. The real question is can someone digitally model a 1952 hand wired Pultec EQP-1 100% accurately?
Of course not. Can they come close? Maybe.

Whether or not analog emulations are good / accurate is not relevant to me. The fact of the matter is that a significant number of analog modeling plugins are in fact sensitive to the input level. So, gain staging plugins is a sort of a 'know your hardware' thing with me.

What started me down the VU meter path is that I of course utilize non-Slate Digital plugins. Since they cannot live within the 'Mix Rack' I would gain stage them to -18dbFS RMS by disabling all plugins in the chain after the plugin and simply use the Reaper digital track meter. Since I am comfortable with VU meters, I thought I would get a VU meter plugin and use that instead.

I suppose it is also important to point out that I record and mix live instruments save for MODO bass which I use to write the bass lines. I would typically then get a real bass player to do the bass. Sometimes I am left with the MODO bass
for my bass tracks because I cannot find a bass player to do the recording.

I am not doing EDM / Hip Hop, MIDI / Sample driven things which I am sure is a different animal.


In closing, when it comes to gain staging to -18dBFS RMS I will continue to do so.
It does not take me all that long to do and I always prefer a solid / known starting point.

//================================
Pashkuli,

I apologize that I have not responded to some of your queries about MODO Bass.

MODO Bass tends to 'load up'. Especially on Legato slides and it is prevalent with the Rickenbacker model.
It is not unusual for it to swing from -6dBVU to +2dBVU on a legato slide, pitch wheel and sometimes even the initial note on a string other than the string the last note was played on.

Adjusting velocity on a note / series of notes does not fix this.

When this occurs, besides simply being too loud, it will mask my Kick. Yes, I know how to utilize side chain compression to prevent the kick masking but doing
so never sounds good to my ear (cheesy!).

I have tried things like Waves Bass Ridder to level the MODO Bass track, but I have found I need to limit the max/mins to a point (in order to get rid of the load ups) that all the dynamics are then removed from the track. Yes, I could automate Bass Rider to turn it on only at the points where MODO Bass loads up, but this approach is no difference then simple automation in the volume lane for the track, so why should I bother with Bass Rider?

//====================================
White Tie,

If you could learn to contain yourself, be more tactful, and not treat this forum as a place to go and vent your frustrations perhaps I would lend some credence to the things you are saying here. Your attitude (bad) causes me to simply ignore anything you have to say.

I am a C/C++ programmer with over 30 years of experience (I am not a kid. coding is my trade not my hobby) I have done a significant amount of Audio programming. I am not all that impressed with your attempts to somehow 'show off' your knowledge.

I also am very opinionated, believe in what I believe adamantly. My beliefs are based on my technical knowledge. I am very well aware that I could in fact be wrong on all things, you seem to lack this belief / realization about yourself.

I am pretty sure the only 'fact' I have stated in my post is this:
'The fact of the matter is that a significant number of analog modeling plugins are in fact sensitive to the input level.' I cannot even imagine that one could argue with this point.

The rest of my post is simply my workflow and rational behind it. You have already stated that you believe that gain staging is not relevant and is a waste of time. We get that and understand you are of that opinion. No need to repeat yourself.

I am sure you realize that you owe Pashkuli an apology for your 'facebook' comment.

I am not saying that I consider yourself to be a Troll, but others may consider yourself to be so.
//==================================================

Cheers.
As I was saying, hitting 0dBVU on a GUI might be visually satisfying, but might not be the best treatment for a sound.

Bob Olhsson, of Motown fame, really opened my ears when he said that he rarely goes anywhere near 0dBVU on plugins because they sound distorted and not as much like the hardware to him. It made me ignore the meters and listen instead. Sometimes it hits 0, often I prefer it some way below.
Judders is offline   Reply With Quote
Old 12-22-2021, 01:39 PM   #106
White Tie
Pixel Pusher
 
White Tie's Avatar
 
Join Date: Mar 2007
Location: Blighty
Posts: 4,012
Default

Quote:
Originally Posted by rMidi View Post
I understand that analog emulations are sort of optimized to operate at -18dbfs.
...is a falsehood, and not the same thing as...

Quote:
Originally Posted by rMidi View Post
a significant number of analog modeling plugins are in fact sensitive to the input level.
since a significant number of all plugins are sensitive to input level, not just analogue modelled ones. That goes without saying. It is the gap between those two statements where the confusion takes root.

----------

If you are happy going through these rituals, good for you, if you're happy you should do as you please. But if you express confusion or uncertainty, I make no apology for trying to lead you out of the clutches of unscrupulous marketing departments. And I will continue to oppose, with vigour, their nonsense because the harm it does, particularly to the peace of mind of beginners, is unnecessary and indefensible.
__________________
The House of White Tie
White Tie is offline   Reply With Quote
Old 12-22-2021, 01:41 PM   #107
rMidi
Human being with feelings
 
Join Date: Dec 2016
Posts: 47
Default

Yes.

To be absolutely clear I am talking about gain staging plugins as a starting point for a mix session, it is not the end product.

Of course, hitting some plugins really hard makes them sound different than hitting them soft. Such actions come in during the mix session, to please the ear.

I think we agree ?
rMidi is offline   Reply With Quote
Old 12-22-2021, 02:01 PM   #108
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 28,744
Default

Quote:
Originally Posted by rMidi View Post

To be absolutely clear I am talking about gain staging plugins as a starting point for a mix session, it is not the end product.
FYI, if these are recorded tracks, and they were "gain staged" anywhere near nominal analog levels on the way in, the wavs are already at the proper level. Test it for yourself, send a "unity" analog signal into Reaper and watch where it shows up on the scale in reaper. Why?...

I don't want to pee on anyone's cornflakes but -18 dBFS' origin (albeit mostly but not completely arbitrary) is about leaving headroom for the digital scale, because the analog scale can exceed zero and the digital scale cannot. Just look at most any credible piece of analog gear and you'll see that it can go somewhere in that +15 to +20 something + range before it clips, you need room for that on the digital scale which is afforded by "pushing it down" by a similar amount aka -18 or -19 or -15 whathaveyou. No magic, it's a loose reference for aligning two different scales.

The only connection to this arbitrary scale/reference adjustment and non-linear plugins is developers trying give consistency to a preset and what it was intended to sound like based on the above (and those who just want to trick consumers into magic sauce). This goes back many years and has been debunked over and over, and not just here. Many get frustrated because this is so old and so debunked it shouldn't even come up in 2021.

So if one does pretty much anything right when they are recording, none of this is needed anyway, it just happens on it's own, and prioritizing this gain thing over your ears (proverbial you) is a red flag toward not being able to actually mix well. If you (again proverbial) can't turn knobs on a non-linear plugin and decide to what you want it to sound like, you probably should give up mixing.
__________________
Music is what feelings sound like.

Last edited by karbomusic; 12-22-2021 at 02:11 PM.
karbomusic is offline   Reply With Quote
Old 12-22-2021, 02:23 PM   #109
Judders
Human being with feelings
 
Join Date: Aug 2014
Posts: 10,551
Default

Quote:
Originally Posted by rMidi View Post
Yes.

To be absolutely clear I am talking about gain staging plugins as a starting point for a mix session, it is not the end product.

Of course, hitting some plugins really hard makes them sound different than hitting them soft. Such actions come in during the mix session, to please the ear.

I think we agree ?
I find this confusing. Why wouldn't you go for what sounds good from the start? Why add a whole load of work just to undo it all afterwards?
Judders is offline   Reply With Quote
Old 12-22-2021, 02:27 PM   #110
Judders
Human being with feelings
 
Join Date: Aug 2014
Posts: 10,551
Default

Regarding ModoBass, or any VI: I find it best to deal with wayward dynamics in the midi editor before resorting to compressors or auto gain plugins.

If a note is too loud, pull the velocity down. If you really want the sound of the higher velocity, but not the level, automate the VI output level.
Judders is offline   Reply With Quote
Old 12-22-2021, 02:59 PM   #111
Phazma
Human being with feelings
 
Join Date: Jun 2019
Posts: 2,188
Default

Why doesn't just everyone mix at whatever levels he/she prefers?

I probably have the most unorthodox workflow possible (basically the opposite of "gain staging"), but never had any quality complaints. I just import my (mastered) references, don't touch their volume and mix so to match the sound as closely as possible. I will of course have peaks above 0dBFS on my master meter but in order to not clip my DA converters I have a -12dB attenuation plugin on my monitoring FX chain. To make sure I don't clip a render I peak-normalize it (which is not necessary if I apply brickwall limiting to the master bus). When using input sensitive plugins I simply adjust their input gain so that they sound good and if necessary compensate using the output gain.

I surely wouldn't recommend my method to beginners as there are plenty of things to be aware (not clipping converters, not clipping files if rendering to fixed point, making sure to properly adjust plugin input gain etc.) but it works well for me and I feel it helps me to more reliably match a mastered reference file and ultimately it proves (at least to me) that there is no need to worry about what people call "gain staging" if you know what is going on with the signal at which stage.
Phazma is online now   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 02:23 PM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, vBulletin Solutions Inc.