Hi,
Here in forum there is nice thread about music spectrum of various genres of music.
Link : Spectrum of Musical Genres
I know cannon shots in 1812 Overture are loud. So it got me curious is there any info on internet/links where Transients/peaks are calculated across various genres of music and mean results are given ?
Thanks and Regards.
Here in forum there is nice thread about music spectrum of various genres of music.
Link : Spectrum of Musical Genres
I know cannon shots in 1812 Overture are loud. So it got me curious is there any info on internet/links where Transients/peaks are calculated across various genres of music and mean results are given ?
Thanks and Regards.
Searched the net. and pardon me for the query. some link : mixing - How loud is 0dB, and what is a good dB to mix to? - Sound Design Stack Exchange
I think Most dynamic loud music may have peak of 0 dB for very short period of time. This dB scale being logarithmic is little confusing. Will search further.
Regards
I think Most dynamic loud music may have peak of 0 dB for very short period of time. This dB scale being logarithmic is little confusing. Will search further.
Regards
The quick and dirty answer:
0db is a measurement of two values, say the difference between the output and input of a preamplifier. 0db being that there is no difference between the two values, in other words unity gain.
db needs to be referenced to something to be meaningful so 0dBm in a telecommunications network would be referenced to 1mW into 600ohms, in the RF world 0dBm is referenced to 1mW into 50ohms. These references are not used for audio or music reproduction.
So, if you are referring to 0db peak, then you will need to have a reference.
There's quiet a bit of info about audio levels and dB's at sengpielaudio.com
0db is a measurement of two values, say the difference between the output and input of a preamplifier. 0db being that there is no difference between the two values, in other words unity gain.
db needs to be referenced to something to be meaningful so 0dBm in a telecommunications network would be referenced to 1mW into 600ohms, in the RF world 0dBm is referenced to 1mW into 50ohms. These references are not used for audio or music reproduction.
So, if you are referring to 0db peak, then you will need to have a reference.
There's quiet a bit of info about audio levels and dB's at sengpielaudio.com
The general measure of transients is dynamic range, expressed as a ratio between the loudest and quietest sounds. A whole lot of albums have been scanned are are catalogued at The Dynamic Range Database
But you should be prepared to be quite disappointed since most modern music has very little, if any, dynamic range at all.
Thanks Douglas.
I did come across dynamic range while searching about average peaks in music. I have 'Brothers in Arms' by Dire straits. And it sounded very nice. The DR Database shows high DR on that album. I forgot to save a link which showed a simple vocal music can have high dynamic range too.
Regards.
I did come across dynamic range while searching about average peaks in music. I have 'Brothers in Arms' by Dire straits. And it sounded very nice. The DR Database shows high DR on that album. I forgot to save a link which showed a simple vocal music can have high dynamic range too.
Regards.
This has recently been researched by Meyersound. They developed a test signal that has the same characteristics as music. The result is based on white noise but with a crest factor that increases with frequency from iirc 1kHz.
Based on a large number of analyses from all genres of music.
More info here.
Jan
Based on a large number of analyses from all genres of music.
More info here.
Jan
At the risk of stretching my limits here... This lack of range is why my music collection ended someplace in the early 1990s. Everything was getting louder and most of it was actually giving me headaches so, at some point a few years ago I just stopped collecting it.
I've been reading up on things like the "Loudness War" and the way modern music is mixed and mastered. In my opinion the big problem is that many modern engineers have missed the boat when it comes to the way you mix digital music.
They've continued shooting for the 0 db mark as an average, as you would on an analog console. But in digital formats 0db is where clipping starts, it's a hard limit, there's no "overhead" so they can't just crank it up like they used to.
If we look at the standard consumer audio levels, the equivalent voltage output (1v peak to peak) actually occurs on CDs and DACs at about -15db... and that's where they should have been mixing.
This video might provide a better insight... YouTube (it's a bit long, but worth seeing)
I'm very interested in sound levels and dynamics, and I can certainly understand and agree that LUFS (in most cases) is very useful for determining perceived loudness. But I have some problems understanding "DR rating", what I experience as a dynamic track and the DR value do not always seem to correlate very well.
Seems like if you just have lots of somewhat varied peaks with absolute silence in between you'd get a great LUFS value + "DR" rating but it would be horrible to listen to.
Seems like if you just have lots of somewhat varied peaks with absolute silence in between you'd get a great LUFS value + "DR" rating but it would be horrible to listen to.
Last edited:
Hi, this is because some mastering engineer are better than others and can make you feel or think there is a lot of dynamic into a track as they effectly reduced the real dynamic range of material.
It is almost always talked about compression but in fact you have a lot of tools at disposal to interact with dynamic range and sometimes it is more relevant to modify the dynamic range from the lowest to the highest (expander) than from the highest to the lowest of the range (compressor).
The way to eq a track play an important role too at perceived loudness and dynamic.
Lufs is an integrated range which was developped at first for broadcast duties iirc.
It integrate on very long time (can't remember exactly how many but sould be in the range of 1minute, search for 'Ebu tech 3341') so yes the situation you describe is realistic.
The point about this was to lower the overall level of compression used during commercial advertisement and produce a kind of levelling of the whole material ( including non commercial contents). This ensure you should not have the excess of 90's andnearly 2000 when you had advert which was almost 6 db louder than 'regular' material.
Iow, it is very nice indicator but not nescessarely dedicated for musical contents.
Something like K- scale is more interesting imho.
Thank you for replying
Yes, indeed. Another aspect is that the DR rating seems to completely ignore which part of the spectrum has most energy, if you'd just use a 100hz high pass on the music, I suspect you could hammer out much more "perceived" sound more optimalized for small systems and still make it sound like crap for "proper" sound systems. Not to mention that it seems to be referenced to absolute reference voltage levels (as usual), so you can have much more high frequency content screaming out of the speakers and still not get into trouble with the LUFS guidelines... :/
I thought LKFS is the same as LUFS? Reading about them they both seem to implement K-Weighting? The weighting thing seems very relevant, but when reading about LKFS/LUFS it is hard to find any mention or examples on what kind of target curves are defined, still seems they are looking mostly at VU levels?
Last edited:
Thank you for replying
Yes, indeed. Another aspect is that the DR rating seems to completely ignore which part of the spectrum has most energy, if you'd just use a 100hz high pass on the music, I suspect you could hammer out much more "perceived" sound more optimalized for small systems and still make it sound like crap for "proper" sound systems. Not to mention that it seems to be referenced to absolute reference voltage levels (as usual), so you can have much more high frequency content screaming out of the speakers and still not get into trouble with the LUFS guidelines... :/
In fact there is part of spectrum which tolerate higher level of compression and other which don't. I must confess i did a lot of bad things until i understood that...
The mid and high tolerate higher compression, bass doesn't. This may seems obvious ( fast attack settings make sense as mid and high peaks are inherently 'faster' and low end is.. mmm low to react! ) but it wasn't to me.
The thing being to keep a way to increase apparent rms value of low end ( without smashing transient) and keep balance coherent with mid and high ( which have gained density as you have them more compressed). Iow find the xover point which suit the material and where to stop processing if you use multiband treatments.
To be honest i don't know about weighting. I don't use LUFS for music. It was not developed for it so... ( that doesn't mean it is not usefull but this is not what i use).
K-metering seems to me sometimes very complex when explained by M.Katz but in fact this easy and adapted to music:
You setup a ref spl level ( depends from your liking and your room* but something between 75dbspl to 85dbspl at listening spot) and from there work with a range which is adapted to your source material : k 20 is 20db of DR (high dynamic materials: classical, jazz, acoutical recording in general), k 14 ( pop, rock, etc), k 12 ( broadcasted material or low DR music like EDM, metal,...).
From there you take a reference track and compare to the material you are working on and here we go.
The point with K scale is to work with vu and not with peaks. In fact there is no real target curves ( for level) except your reference track. That said i discovered that for one to want to sound 'loud' there is a 'target' curve for freq response... which has to be tailored to the track (once you corrected the 'issues' from mix - most of the time room mode).
I hope it helps, not sure if it answer your question.
* the level depend from the room a lot. I think our brain doesn't react the same if room is 'big' or 'small'. The 85dbspl seems too loud to me in small room, it is ok in larger but way too much to me in smaller one, i've had to lower to 78dbspl (but you enter a domain where you need a lot of soundproofing or very quiet location as this doesn't make a lot of difference wrt background noise...). I suppose this is linked to ER delay time and level and a way for our brain to protect our ears... well i suppose.
Last edited:
Looking at traditional instruments and how they're intended to compliment eachother, I think it's fun to run different EQ's and filters on every synth or instrument. Another thing is looking at the gaps between sounds and layering, can be a bit difficult to discern the lower level sounds sometimes, but it's fun to try and follow the non-primary melodies in a mix. Agree with you that compression is often used wrong, one compressor brickwalled on the main mix in addition to another on the drums. Squashed dynamics and screaming mids. Throw in another layer of that awful "soundgoodizer" AFTER you've tweaked the sound to death to guarantee a certain level of annoyance.
Will have to check out the K-scale thing when I have time to sit down with the music again, very interesting.
I've recently made the leap to purchase a license for Acid Pro 10, I thought Acid was great in the 90's, then Sony bought up Sonic Foundry and ruined both Acid and Sound Forge... Now Magix has revived it and added more midi functionality, which it was sorely lacking. Really looking forward to getting into it, haven't really used it since 2002 or something.
Sorry about the OT, I'll have a look at various plugins to implement for improved mastering.
Nevermind my silly questions, it's and interesting and important topic. Seeing as there still seems to be a bit of a problem implementing standards, it's important to figure out what works and what doesn't.
One of the way to make your tunes interesting, with a lot of subliminal movement and sound FAT( layering).
That is why you have to communicate A LOT with the mastering engineer to keep her/him in the way you want your tune to sound like you think it should.
The problem being everyone wants to be louder than one reference which is already too loud... What you describe ( known as 'shredding') can be what is needed to a track, not on another.
For example in this case i suppose this is what had been done ( and it is adapted to the tune, however the vinyl mastering is a bit less 'oppressing' which isn't bad but you loose a bit of the 'wall of sound'):
YouTube
I liked Acid too. A friend of mine used it around same period and we produced some tracks on it and always wondered why no midi. Nice to have alternative to Live and nice to see others in need of midi in this VST times!
About K meter, visit digido and purchase some of the tracks B.Katz list: it will help you a lot to set up things and have 'references' known to sound at equal levels for comparison to your own record collection.
Honor Roll - Digido.com
Some interesting infos:
Monitor Calibration with K-System - Digido.com
One last thing, i know this is difficult to resist doing mastering on his own materials but... you shouldn't.
In fact it is very difficult to 'correct' itself ( on the same room and with the same loudspeakers with the same ears).
I know budget is most always a limitation but most mastering engineer offer a 'test' where they work on one of your track ( following your will) for you to compare. Worth to try and see by yourself if there is a difference. Somes are quite good and not this costly.
Last edited:
I agree.
On most modern music recordings and also decent quality radio, I am completely unable to listen to any of it, on all my headphones, in the car or my three stereo setups. Unless I adjust the bass a tiny bit extra, mids a bit down, and treble a bit further down. There's a lot of music from mid to late 70's I can listen to just fine without any adjustment. Vinyl mastering is slightly more dynamic because it just does not work well to have constant high output on the vinyl grooves. You loose a tiny bit of resolution overall, but the dynamics are usually much better.
Part of the issue is that I have lost a fair bit of my respect towards the professionals, the schools that teach the professionals, the goals they aim for and the methods they use.
Another part of the issue is that the broadcasting companies that strive to implement the new loudness standards, seem to do themselves a disfavor by adding another level compression to sort of "safeguard" against both peak dynamics and to avoid getting too low volume.
If you know of anyone that's actually good at this stuff I am open for suggestions.
Not going to go into defence on my music mastering skills, I'm not a professional by any standard, and my music will never be even remotely popular.
So much moping about too much compression, screaming mids and lack of dynamics. Lots of opinions and tools to analyze all sorts of things, but hardly anyone that actually try doing something about it.
In fact the reason about wider dynamic on vinyl is linked to the fact that if you try to print overcompressed signals you'll run into the risk of destroy the head. And this is very very costly and there is very few people worldwide able to repair it ( afaik the last one is a girl in the US which wanted to retire some years ago. I hope situation changed).
I'm on a different opinion about the resolution of vinyl but it depends of desk, cart and the cleanness of the vinyl.
I can garantee you my technics SL10 with eps310mc is on par with my dolby Lake converters on clean disk.
Overall the higher dynamics as some interesting side effects: i now understand english lyrics on tracks i didn't on cd... they have more 'transparency' to me. And i'am to the point to buy only vinyl and produce my own digital version from them ( not touching dynamic or eq just a tiny bit of 'restoration' in case of surface noise! Crazy situation in my view.
hmm. I've been part of everything you named and...sadly i have to agree. But there is still some good pro. It is sometimes difficult to find them though.
Are talking about 'leveling'? If yes this is one of the only answer to stop the 'loudness war', as it won't give anybenifit to 'sound louder' than your neighbour. And this is very evident when you make comparison to music produced before the loudness war: they sound open, with details and dynamics... 'loudness war' mastered is lifeless, zero finesse and fatiguing short term.
That is one of the whole point about K- metering too: focus on sound quality rather than very high rms level, improving micro and macro dynamic.
It is very style dependent. I know some very good engineer yes but not sure if they are good within your particular style.
One thing though, if you tell them you are not interested into 'sounding louder' and prefer 'sounding good' then they will probably have a large smile on their face and will want to keep you as client forever...
Last edited:
Yes, I'm talking about "leveling". But what I've heard (when watching live concerts) is that even when they try to implement it properly, it sounds kind of low and lacking in dynamics when there's music. And when the presenter is back on the voice still gets a large percentage of the whole level range that the band or orchestra had.
Yes, some of the mediumlow-ish detail level improves a bit, but low level more or less disappears, and peaks are non existent because apparently high volume peaks is what they're trying to avoid.
Sounds reasonably nice, but I end up increasing the volume too much because it's a concert and I want to hear the music. Then the presenter goes back on and I end up turning it down, but just a little bit. If I change channels or watch something on netflix I end up with a minor surprise unless I've already adjusted the volume beforehand.
So it's not so fatigueing anymore, that's a plus. But it still feels "squashed", and I don't think a few peaks are bad, so why are they gone?
Oh, and feel free to point me towards someone you think might be suitable to help me with a bit of mastering.
Last edited:
There's one phenomena I've noticed throughout the years - right when the music level fades at the end of a song, it sounds better in that instant. It's almost like when the level leaves the area of compression / modulation limit, i.e. those operations are out of the way, the sound shines through.
Anyone else have this perception?
FWIW, I'm appalled at the difference in level between when I play a flac file and when I stream from jazzgroove.org which is always louder (which I'm a paying member because I enjoy their programming as much). It's like "c'mon - you have to uplevel the average volume of your digital stream too!?!"
To me, the average listening level should be exactly the same. It's not. Why anyone would further compress, raise the sound level in an internet stream is so far beyond me I cant even imagine! What, if it's perceptibly louder than a competitor, people will subconsciously prefer the louder channel? I HATE it... Yet, try to find an equivalent programming, that doesnt happen to pull this trick...
And of course, there's another stream that's clearly perceptibly lower in level out of Portland OR - I dont have the URL on hand. Average levels, modulation levels - all over the spot - I always thought "digital" would mean "consistent" - forget it!
Anyone else have this perception?
FWIW, I'm appalled at the difference in level between when I play a flac file and when I stream from jazzgroove.org which is always louder (which I'm a paying member because I enjoy their programming as much). It's like "c'mon - you have to uplevel the average volume of your digital stream too!?!"
To me, the average listening level should be exactly the same. It's not. Why anyone would further compress, raise the sound level in an internet stream is so far beyond me I cant even imagine! What, if it's perceptibly louder than a competitor, people will subconsciously prefer the louder channel? I HATE it... Yet, try to find an equivalent programming, that doesnt happen to pull this trick...
And of course, there's another stream that's clearly perceptibly lower in level out of Portland OR - I dont have the URL on hand. Average levels, modulation levels - all over the spot - I always thought "digital" would mean "consistent" - forget it!
Your spot on about perception and cause.
The rest of both your message (Kaffi and JJasniew) are both related to same thing: you talk about broadcasted signal. This can't have the same dynamic range as a disc, flac or a real event.
Why? Mainly because we live in sound polluted world. Everywhere you go there is background noise and most of the time it is loud.
For one to be heard, it needs to be 'louder' than this average noise and so to have a limited dynamic to source material ( or it won't be heard for most of it).
You have to take into account that the variation of source material needs to be 'constrained' too because if you diffuse some classical and thereafter some hiphop stuff there will be an issue. And i don't even talk about spoken words...
There is other things at play too: in the 50's some study showed than the higher the rms level the more hooked the listener is. As for broadcaster the main source of income is advertissement they started to look for ways to augment the average level as it permit to sell advertisement at better price ( the more listeners, the more they get paid).
This was the first step, next one being to do same thing within advertissement... And then the vicious circle begin.
One other point: those treatments used to process the audio is a way to be different than the neighbour station. It gives a 'signature' to material broadcasted. In the 80's and early 90's it was forbidden to 'average' technician to have access to this processing chain (they were the 'secret sauce' so either the gear or parameter had to be kept secret).
Then appeared the digital area. Remember the panic within Major labels when they thought digital copy will be the end of the golden area? Well in a way you face the same thing with digital broadcast: what would happen if you could record your favourite radio station without alteration?
This is most an issue for the big commercial station ( you don't have access to my artist if it can be copyed lossless...) but it then happen to other less bigger station by chain reaction... through standard. ( mp3).
This is why you now have acces to high quality ( whatever it means) streaming through payment. This is another way to keep you under control ( and why the catalogs are split to differents streaming site...).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.