Post limiter on a final master?gentleclockdivider wrote: ↑Sat May 23, 2020 10:02 pm In the end it's good to have at least 6-db FS of headroom
Tape simulations for that mastering mojo.
-
- KVRAF
- 4711 posts since 26 Nov, 2015 from Way Downunder
-
- KVRian
- 872 posts since 28 Nov, 2016
what unit of measurement are we talking about. peaks, short-term, momentary, or what?
-
gentleclockdivider gentleclockdivider https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=203660
- KVRAF
- 6112 posts since 22 Mar, 2009 from gent
That's -6db fs , it 's more then enough headroom for further processing /limitingMogwaiBoy wrote: ↑Sat May 23, 2020 10:21 pmPost limiter on a final master?gentleclockdivider wrote: ↑Sat May 23, 2020 10:02 pm In the end it's good to have at least 6-db FS of headroom
In this day and age some people would consider that quit , if you have a listen at what some people upload to their soundcloud ..it's just one big squashed block near 0dbfs
Eyeball exchanging
Soul calibrating ..frequencies
Soul calibrating ..frequencies
-
- KVRian
- 872 posts since 28 Nov, 2016
do people tend to not read replies unless they're directly quoted in them? maybe I should start doing that more.
-
- KVRAF
- 4711 posts since 26 Nov, 2015 from Way Downunder
I'm really not sure. Here's one of the videos that confused me..sleepcircle wrote: ↑Sat May 23, 2020 10:31 pm what unit of measurement are we talking about. peaks, short-term, momentary, or what?
https://www.youtube.com/watch?v=rUjgiGRPAok
-
- KVRAF
- 4711 posts since 26 Nov, 2015 from Way Downunder
For example he sets it to -18 VU and then lowers the Threshold on the limiter until it bounces around 0 VU, and then he says "OK that's a fairly moderate loudness boost. We're not really doing a whole lot of squashing".
?!?
I guess I'm just taking him too literally. There's no squashing/attenuation happening at all, as opposed to not a "whole lot". And he talks about it being fine as a master - really? With peaks around -10db? Then in his last example he makes lots of strange statements about boosting the trim into the limiter instead of pulling the threshold on the limiter down (same thing?), clipping (with the ceiling at -1db you're atleast 1db away from that aren't you?) / "preventing inter-sample peaks" (really, how are you measuring that?).
Anyway, I'm off topic sorry. I appreciate that he's encouraging super-dynamic, quiet mastering. Back to tape sims...
?!?
I guess I'm just taking him too literally. There's no squashing/attenuation happening at all, as opposed to not a "whole lot". And he talks about it being fine as a master - really? With peaks around -10db? Then in his last example he makes lots of strange statements about boosting the trim into the limiter instead of pulling the threshold on the limiter down (same thing?), clipping (with the ceiling at -1db you're atleast 1db away from that aren't you?) / "preventing inter-sample peaks" (really, how are you measuring that?).
Anyway, I'm off topic sorry. I appreciate that he's encouraging super-dynamic, quiet mastering. Back to tape sims...
-
- KVRian
- 872 posts since 28 Nov, 2016
@MogwaiBoy Yeah, that's what I suspected: it's not the absolute maximum peak that the guy's measuring, there, that's the "voltage" (VU) of the signal—a term which is meaningless in purely digital storage, but becomes much more important when any analogue equipment (or analogue-mimicking-plugins) become involved.
Nevertheless, though, people still think of the 'voltage' as a useful way of measuring a signal, because it sort of gives you an idea of the 'organic volume' of the thing, instead of the absolute numerical value of the digital information, which sometimes--
Well, how do I put this…
Like: a very badly processed peak can hit a ridiculously high number for such a short time that the human ear can't even hear it, but it'll show up on the meter as a mysteriously high value, and also peaks which have their phases shifted so that they don't clip through the maximum allowed volume will still sound very loud to the human ear because—if the phase shift is small enough—the ear will hear them 'all at the same time.'
There are many systems for dealing with loudness or signal strength. K-weighting, RMS, DIN ppm, the Fletcher-Munson curves, etc.
Suffice it to say, "VU" is just a convenient system. But the thing is: because a "VU" measures the general volume and 'inertia' of a signal instead of its numeric maximum value, a signal which comes out to 0 on a VU meter calibrated to -10db (or what-have-you) will not actually max out at -10db full scale, it will probably max out much higher than that, a lot closer to the digital -0dbfs.
So aiming for -0VU on a VU meter is good for mixing and stuff because it lets you get an idea of the headroom—the 'breathing room'—your song has left. It doesn't measure the sharp spikes and moments of dynamic interest which make your sound more rhythmical and playful-sounding but, if you purposefully leave yourself that headroom, you know those moments will have room to stick out from the rest of the track instead of being drowned in a wall of equal-volume sound with only changes in frequency defining its musical content.
Now, for mastering, it is good to shoot for—DIGITALLY, like not a VU meter—a value like -1db, -2db, -1.3db, whatever, somewhere in there. It mostly doesn't matter the exact number, and the reason why is this:
For a completely lossless format, you could, in fact, shoot for peaks to end up at -0.0db and the system would be perfectly fine, analogue playback systems often have extra headroom and -0.0db on a CD-ROM track or whatever would not necessarily be the maximum voltage available on the stereo speakers, right?
But in this modern day and age of lossy codecs, there is a new tyrant. With lossy codecs the soundwave's actual SHAPE is not preserved, it is only recreated in a simplified estimation, and it doesn't always match exactly the level that the original file had. If your peaks hit -0.0db in the lossless file, then in the lossy version, the reconstruction math might put them OVER 0, which puts the signal into no-man's land and introduces nasty popping and clicking before the signal can even get converted to an analogue representation for your speakers.
So it's better for your peaks to end up a little under -1db full scale (actual digital peaks, in this case, not VU or RMS, which will be much lower, still) when mastering) because then—when you convert it to AAC or OGG or MP3 or whatever—there won't be any risk of sloppy math resulting in a chopped-up signal at the song's most important moment.
And these days, the desire to get the absolute "loudest song" (besides being a short-sighted psychological trick which makes music less enjoyable to listen to, long-term) can no longer be an issue, because more and more companies are using loudness normalization to make all songs The Same Volume, regardless of how loud they were in-file, so now there is no NEED to mercilessly shave away headroom and compress/clip things more and more.
I hope this explains at least part of the reasoning involved.
There's also inter-sample peaks, but that has to do with the whole reconstructing-a-signal-from-a-bunch-of-digital-points business, and which requires, to explain it, graphs which I do not have ready-to-hand. Anyway, you don't need to worry about that if you shoot for 1db of space between your peaks and absolute digital -0dbfs.
Nevertheless, though, people still think of the 'voltage' as a useful way of measuring a signal, because it sort of gives you an idea of the 'organic volume' of the thing, instead of the absolute numerical value of the digital information, which sometimes--
Well, how do I put this…
Like: a very badly processed peak can hit a ridiculously high number for such a short time that the human ear can't even hear it, but it'll show up on the meter as a mysteriously high value, and also peaks which have their phases shifted so that they don't clip through the maximum allowed volume will still sound very loud to the human ear because—if the phase shift is small enough—the ear will hear them 'all at the same time.'
There are many systems for dealing with loudness or signal strength. K-weighting, RMS, DIN ppm, the Fletcher-Munson curves, etc.
Suffice it to say, "VU" is just a convenient system. But the thing is: because a "VU" measures the general volume and 'inertia' of a signal instead of its numeric maximum value, a signal which comes out to 0 on a VU meter calibrated to -10db (or what-have-you) will not actually max out at -10db full scale, it will probably max out much higher than that, a lot closer to the digital -0dbfs.
So aiming for -0VU on a VU meter is good for mixing and stuff because it lets you get an idea of the headroom—the 'breathing room'—your song has left. It doesn't measure the sharp spikes and moments of dynamic interest which make your sound more rhythmical and playful-sounding but, if you purposefully leave yourself that headroom, you know those moments will have room to stick out from the rest of the track instead of being drowned in a wall of equal-volume sound with only changes in frequency defining its musical content.
Now, for mastering, it is good to shoot for—DIGITALLY, like not a VU meter—a value like -1db, -2db, -1.3db, whatever, somewhere in there. It mostly doesn't matter the exact number, and the reason why is this:
For a completely lossless format, you could, in fact, shoot for peaks to end up at -0.0db and the system would be perfectly fine, analogue playback systems often have extra headroom and -0.0db on a CD-ROM track or whatever would not necessarily be the maximum voltage available on the stereo speakers, right?
But in this modern day and age of lossy codecs, there is
Code: Select all
digital clipping
So it's better for your peaks to end up a little under -1db full scale (actual digital peaks, in this case, not VU or RMS, which will be much lower, still) when mastering) because then—when you convert it to AAC or OGG or MP3 or whatever—there won't be any risk of sloppy math resulting in a chopped-up signal at the song's most important moment.
And these days, the desire to get the absolute "loudest song" (besides being a short-sighted psychological trick which makes music less enjoyable to listen to, long-term) can no longer be an issue, because more and more companies are using loudness normalization to make all songs The Same Volume, regardless of how loud they were in-file, so now there is no NEED to mercilessly shave away headroom and compress/clip things more and more.
I hope this explains at least part of the reasoning involved.
There's also inter-sample peaks, but that has to do with the whole reconstructing-a-signal-from-a-bunch-of-digital-points business, and which requires, to explain it, graphs which I do not have ready-to-hand. Anyway, you don't need to worry about that if you shoot for 1db of space between your peaks and absolute digital -0dbfs.
Last edited by sleepcircle on Mon May 25, 2020 1:29 pm, edited 1 time in total.
-
- KVRAF
- 1742 posts since 9 Jul, 2014 from UK
Nothing at all. Waves seems to be marmite on here, regardless of the plug.
I wonder what happens if I press this button...
-
- KVRian
- 872 posts since 28 Nov, 2016
a cursory review of the conversation about it will reveal what the problem was—i.e. gain-staging issues—and what the solution appears to be
- KVRAF
- 2956 posts since 31 Jan, 2020
Did you listen to the song? It's a lovely example of Kramer tape.ramseysounds wrote: ↑Sun May 24, 2020 12:34 amNothing at all. Waves seems to be marmite on here, regardless of the plug.
-
- KVRer
- 5 posts since 10 Mar, 2020
I like to use the CRTIV Tape Bus by Voxengo and Softube Tape. CRTIV is nice because it has a couple of different "Tape Impulses" which I like to use on vocals (Tape 3 seems to de-harsh vocals nicely). Softube Tape is also great and I find myself using it not only for tape duties but also because I love the High Frequency Trim for boosting on buses and individual elements.
-
- KVRAF
- 1742 posts since 9 Jul, 2014 from UK
Not sure the point you are making? Is it sarcasm?Spring Goose wrote: ↑Sun May 24, 2020 1:11 amDid you listen to the song? It's a lovely example of Kramer tape.ramseysounds wrote: ↑Sun May 24, 2020 12:34 amNothing at all. Waves seems to be marmite on here, regardless of the plug.
I wonder what happens if I press this button...
- KVRAF
- 2956 posts since 31 Jan, 2020
It's not sarcasm. What i meant is contrary what been said here in this thread, the sound of a hot Kramer tape can be quite/very beautiful. I love that Duran Duran song, it's a fantastic finish to an album, and some great drums, and reverent to the lovely sound of the Kramer tape (i think). I assume they (Duran Duran) used the original machine not the Waves plug but i think it does sound basically the same. I recognised it (the tape) because i had made a track with a similar sound:ramseysounds wrote: ↑Sun May 24, 2020 1:28 pmNot sure the point you are making? Is it sarcasm?Spring Goose wrote: ↑Sun May 24, 2020 1:11 amDid you listen to the song? It's a lovely example of Kramer tape.ramseysounds wrote: ↑Sun May 24, 2020 12:34 amNothing at all. Waves seems to be marmite on here, regardless of the plug.
https://soundcloud.com/user-492255322/2 ... ossible-ep
I don't actually own the Waves Kramer Tape. I got the bundle Signature Series Bass & Drums free with another purchase last year, but then after some time the Signature Series Bass & Drums disappeared and instead i had Kramer Master Tape! I gave it a try, and i made the song (above) with it, then i emailed Waves support and they fixed the problem, then a couple of months later the problem reoccurred (the Signature Series Bass & Drums disappeared and instead i had Kramer Master Tape!) but i haven't emailed Waves support again yet.
To cut a long story short (you asked!)
- KVRAF
- 2956 posts since 31 Jan, 2020
Now i'm very tempted to buy the Kramer Tape, the J37 tape, the REDD, the Reel ADT, the BSS DPR-402, and the Brauer Motion, in the latest offer (for a bit more than 100dollar total)ramseysounds wrote: ↑Sun May 24, 2020 1:28 pmNot sure the point you are making? Is it sarcasm?Spring Goose wrote: ↑Sun May 24, 2020 1:11 amDid you listen to the song? It's a lovely example of Kramer tape.ramseysounds wrote: ↑Sun May 24, 2020 12:34 amNothing at all. Waves seems to be marmite on here, regardless of the plug.
(Correction: I just noticed that i do actually own the Kramer Tape. It was included with the Dave Clarke EMP toolbox but i don't remember reading about it)
Last edited by Spring Goose on Sun May 24, 2020 2:41 pm, edited 1 time in total.
- KVRAF
- 2956 posts since 31 Jan, 2020
I hope i made my point this time, that i love the Kramer Tape to bits.ramseysounds wrote: ↑Sun May 24, 2020 1:28 pmNot sure the point you are making? Is it sarcasm?Spring Goose wrote: ↑Sun May 24, 2020 1:11 amDid you listen to the song? It's a lovely example of Kramer tape.ramseysounds wrote: ↑Sun May 24, 2020 12:34 amNothing at all. Waves seems to be marmite on here, regardless of the plug.
Ah man, can't you hear what the (i think) Kramer Tape sounds like on that drumbeat! (i mean before it gets crunchy but also after it gets crunchy!)