Where do you "master" to? (what dB)
- KVRAF
- 5913 posts since 17 Aug, 2004 from Berlin, Germany
-
- KVRAF
- 6740 posts since 25 Mar, 2002 from sheffield, england
These numbers quoted are all totally arbitrary: ISP levels will depend on the nature of the program material, so the only way to determine the appropriate margin is to use an ISP meter, as I demonstrate in the video.
- KVRAF
- 5913 posts since 17 Aug, 2004 from Berlin, Germany
With 4x oversampling in the limiter you will not have intersample peaksIIRs wrote:These numbers quoted are all totally arbitrary: ISP levels will depend on the nature of the program material, so the only way to determine the appropriate margin is to use an ISP meter, as I demonstrate in the video.
| Links-
- KVRAF
- 2973 posts since 10 Sep, 2003 from Karlskoga, Stockholm, Sweden
Thanks IIRs!!
I cant believe i never thought of the curve between samples ..... so obvious
Lookahead, it adds latency, right?
I cant believe i never thought of the curve between samples ..... so obvious
Lookahead, it adds latency, right?
-
- KVRAF
- 6740 posts since 25 Mar, 2002 from sheffield, england
Not entirely true. In Pro-L a combination of 4x oversampling plus a lookahead of at least 0.1ms will limit ISPs to no more than 0.1dB over the threshold. It won't prevent them completely, and they can still be signficant when using clipping type settings with zero lookahead.4damind wrote:With 4x oversampling in the limiter you will not have intersample peaks
-
- KVRAF
- 6740 posts since 25 Mar, 2002 from sheffield, england
No not really: the latency is there already, and adjusting the lookahead time doesn't change it. Switching in one of the oversampling modes will result in a slight increase in latency however.Crackbaby wrote:Lookahead, it adds latency, right?
- KVRAF
- 5913 posts since 17 Aug, 2004 from Berlin, Germany
Try 11ms LookaheadIIRs wrote:Not entirely true. In Pro-L a combination of 4x oversampling plus a lookahead of at least 0.1ms will limit ISPs to no more than 0.1dB over the threshold. It won't prevent them completely, and they can still be signficant when using clipping type settings with zero lookahead.4damind wrote:With 4x oversampling in the limiter you will not have intersample peaks
| Links-
- KVRAF
- 6740 posts since 25 Mar, 2002 from sheffield, england
That's a safer setting of course. But longer lookahead times can sometimes soften the attack of transients: if you're after a really hard hitting master shorter settings will often work better.4damind wrote:Try 11ms Lookahead
-
- KVRAF
- 4692 posts since 28 Jan, 2003 from In these very interwebs
I do it in the same situations that I prefer clipping to limiting. For aggressive and distorted music, pure limiting is too soft.yairhol wrote:Kim, at which situations wouuld you want to peak at 0 dB?
Mind you, these projects often have several stages of saturation and distortion introduced throughout the recording and mixing process. In these cases the final clipping in mastering doesn't sound 'wrong' (like it does for softer smoother music). Instead it actually enhances the distortion that's already in the mix.
If you want a real-world example, take a gander at:
http://punchcardpoet.bandcamp.com
(It's sort of a quiet sneak preview - the official website isn't live yet)
You should be able to hear that for this music, everything is overdriven. It's part of the aesthetic. So it's appropriate to respect and follow that creative direction through all stages of production - including mastering.
Of course, that doesn't mean I run the whole album through a stompbox right before burning a CD. But in mastering (just like every other production stage) there are choices that I make that affect the way the music sounds and feels. For Punch Card Poet, multi band clipping and inter-sample clipping is a deliberate choice.
Hope that makes sense!
-Kim.
-
- KVRAF
- 4692 posts since 28 Jan, 2003 from In these very interwebs
Also, just one more note:
Music is fundamentally a creative endeavour. We should use everything at our disposal (and today, that's a *lot*!) to realise our creative direction and 'vision'. Everyone who walks away from Punch Card Poet does it for reasons completely unrelated to the mastering. No-one who is a fan and buys the album will be disappointed by the mastering.
All this -1/-.7/-.5/-.3/0 business is pretty insignificant in the scheme of things.
-Kim.
Music is fundamentally a creative endeavour. We should use everything at our disposal (and today, that's a *lot*!) to realise our creative direction and 'vision'. Everyone who walks away from Punch Card Poet does it for reasons completely unrelated to the mastering. No-one who is a fan and buys the album will be disappointed by the mastering.
All this -1/-.7/-.5/-.3/0 business is pretty insignificant in the scheme of things.
-Kim.
-
Bronto Scorpio Bronto Scorpio https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=98170
- KVRAF
- 5546 posts since 13 Feb, 2006 from Wiesmoor, Germany
+1Kim Lajoie wrote:Also, just one more note:
Music is fundamentally a creative endeavour. We should use everything at our disposal (and today, that's a *lot*!) to realise our creative direction and 'vision'. Everyone who walks away from Punch Card Poet does it for reasons completely unrelated to the mastering. No-one who is a fan and buys the album will be disappointed by the mastering.
All this -1/-.7/-.5/-.3/0 business is pretty insignificant in the scheme of things.
-Kim.
It really doesn't make much difference
Even if there are ISPs and so on. There are other things to worry about
Cheers
Dennis
- KVRAF
- 5913 posts since 17 Aug, 2004 from Berlin, Germany
Yeah some guys preferring clipping. Best known Infected Mushroom. They using Prism converters and don't using a limiter. But I would not say it's the best sound ever. Sometimes it works, sometimes not.
| Links- KVRAF
- 5913 posts since 17 Aug, 2004 from Berlin, Germany
This is a different story if we need all this things or if there are more important things like thatKim Lajoie wrote:All this -1/-.7/-.5/-.3/0 business is pretty insignificant in the scheme of things.
If I hear some chart music from well known artists, with distorted and over compressed sound then yes! It's completely insignificant if it was limited at -0.1 or -0.7 the sound is always the same crap. But crap or not.. people buy it and love it
| Links-
- KVRAF
- 6740 posts since 25 Mar, 2002 from sheffield, england
I don't really agree with this at all. Yes music is a subjective art form. But of all stages in production, mastering is by far the most scientific and exacting.Bronto Scorpio wrote:+1Kim Lajoie wrote:Also, just one more note:
Music is fundamentally a creative endeavour. We should use everything at our disposal (and today, that's a *lot*!) to realise our creative direction and 'vision'. Everyone who walks away from Punch Card Poet does it for reasons completely unrelated to the mastering. No-one who is a fan and buys the album will be disappointed by the mastering.
All this -1/-.7/-.5/-.3/0 business is pretty insignificant in the scheme of things.
-Kim.
It really doesn't make much difference
Even if there are ISPs and so on. There are other things to worry about
Cheers
Dennis
I can accept that you might have artistic reasons to go for a very loud and distorted master, and to accept high ISP levels as a consequence. But I can't imagine what artistic reasons you would have to aim for high ISP levels, or in other words, to deliberately engineer an album that distorts when played on an iPod, but not when played on a high end CD player. The other way around perhaps...
If the thread was a general one about production techniques your comment might be apposite. But this thread is specifically about how much margin to leave, and the correct answer is "use an ISP meter".
- KVRAF
- 43957 posts since 11 Aug, 2008 from clown world
You know the way you can hear in real time what your Master is going to sound like when it's exported? Is there a real time VST Plugin tool that can allow us to hear what our MP3 Masters will sound like. Hope that makes sense. The idea of 'blindly' converting a WAV file to MP3 seems like a bad one. If you could hear exactly how the frequencies will be affected in real time it would help a lot. You could also avoid Inter-Sample-Peaks Etc.
