Please note that in this thread I have shown what IS, not what should be. If anyone has problems with the mastering EQ of common recordings, that's another (important) matter. The idea here is to get an idea of what is actually on a wide number of recordings across several different styles.
I've shown what is on typical CD recordings. If someone can show a consistent difference in the same recording on different media, please do so.
And if anyone is interested in the methodology of this survey, please feel free to ask, I would be happy to explain and to help you perform your own survey with these methods or others.
I've shown what is on typical CD recordings. If someone can show a consistent difference in the same recording on different media, please do so.
And if anyone is interested in the methodology of this survey, please feel free to ask, I would be happy to explain and to help you perform your own survey with these methods or others.
> what is actually on a wide number of recordings
Do live music.
I used to. The spectra for "classical" ensembles are correct. Even hiring eight double-bass players won't fill-up the bottom under a mass of fiddles, and few classical instruments have strength above a couple KHz. Contrast with 900-Watt electric bass, screeching guitar distortion, and rings of hard-hammered cymbals in most rock/pop bands.
I would actually expect more droop below 90Hz on the Classical. They may hire more bassists, they may have lusher rooms than I worked in, or the mastering engineer may have touched a knob so it sounds "right" in a small playback space.
Do live music.
I used to. The spectra for "classical" ensembles are correct. Even hiring eight double-bass players won't fill-up the bottom under a mass of fiddles, and few classical instruments have strength above a couple KHz. Contrast with 900-Watt electric bass, screeching guitar distortion, and rings of hard-hammered cymbals in most rock/pop bands.
I would actually expect more droop below 90Hz on the Classical. They may hire more bassists, they may have lusher rooms than I worked in, or the mastering engineer may have touched a knob so it sounds "right" in a small playback space.
Last edited:
Here is a FR plot for the left channel of my system at the listening position:
The center and right channels are similar. The surround channels cut off at about 45 Hz, and the subs cut off at about 14.5 Hz and are crossed at 40 Hz. The rise in FR below 100 Hz was targeted at 1.5 dB/octave.
Each channel has been individually EQed flat using REW (the EQ optimizer facility) and a Xilica XP-8080 loudspeaker processor to within 2 db (excluding room mode dropouts). The front channels are all bi-amped, while the two surrounds are mono-amped but EQ-ed flat.
Chris
The center and right channels are similar. The surround channels cut off at about 45 Hz, and the subs cut off at about 14.5 Hz and are crossed at 40 Hz. The rise in FR below 100 Hz was targeted at 1.5 dB/octave.
Each channel has been individually EQed flat using REW (the EQ optimizer facility) and a Xilica XP-8080 loudspeaker processor to within 2 db (excluding room mode dropouts). The front channels are all bi-amped, while the two surrounds are mono-amped but EQ-ed flat.
Chris
Attachments
Last edited:
I find that Linn classical recordings are pretty much untouched by mastering EQ. Try a few of those...or virtually any 5.1 classical recording. Those pretty much leave the EQ alone, too.
The "I was amazed how flat the [orchestral] response was" comment is only true if the orchestra has sufficient number of double bass players. Many smaller symphonies do not carry enough bass players for reasons of economy of budgets, so the response droops below about 65 Hz (i.e., the fundamental of the cello).
Most orchestral recordings I find rely on the double basses' second harmonic for the listeners to hear the first two strings of the double bass part. This is pretty much an industry practice (with the exceptions noted above).
Try EQing some of your stereo recordings (...not Linn or 5.1...) using a curve similar to the one I showed above (subject to adjustment in the curve slope), then you can form opinions on how it actually sounds and how well it works. Try it first before rejecting the notion.
I think that many here already know the answer--since they have been playing their classical recordings with the "bass" tone control turned slightly to the right for many decades.
Chris
The "I was amazed how flat the [orchestral] response was" comment is only true if the orchestra has sufficient number of double bass players. Many smaller symphonies do not carry enough bass players for reasons of economy of budgets, so the response droops below about 65 Hz (i.e., the fundamental of the cello).
Most orchestral recordings I find rely on the double basses' second harmonic for the listeners to hear the first two strings of the double bass part. This is pretty much an industry practice (with the exceptions noted above).
Try EQing some of your stereo recordings (...not Linn or 5.1...) using a curve similar to the one I showed above (subject to adjustment in the curve slope), then you can form opinions on how it actually sounds and how well it works. Try it first before rejecting the notion.
I think that many here already know the answer--since they have been playing their classical recordings with the "bass" tone control turned slightly to the right for many decades.
Chris
Let's move on. If you wish to read on the demastering process itself, here are a couple of tutorial pdfs:
Demastering Part 1 (What and Why).pdf
Demastering Part 2 (How To).pdf
I'm currently working on Part 3: Advanced Topics, which will be a significantly longer tutorial addressing specific techniques that occur with each track encountered, including genre and instrumentation-specific topics, and specific problems encountered.
Chris
Demastering Part 1 (What and Why).pdf
Demastering Part 2 (How To).pdf
I'm currently working on Part 3: Advanced Topics, which will be a significantly longer tutorial addressing specific techniques that occur with each track encountered, including genre and instrumentation-specific topics, and specific problems encountered.
Chris
The reason I was asking is because it could be that the response at the listening position might be too bright causing a perception of a lack of bass. I (and many others here) like a B&K style curve at the listening position. Note that this is not the FDW response, but a steady state response that includes the power response of the speakers.
You'll note that the bass and lower midrange is quite a bit raised relative to the high frequencies. In most rooms, with the usual furniture and absorption, and usual cone/dome speakers, this is what you would get and most people would find this balance neutral. But, if you have constant directivity speakers combined with less room absoprtion, the total emitted power in the high frequencies may not get absorbed and may cause a perceived lack of bass.
I'm not interested in remastering recordings, however useful that might be. I simply don't have the time to do it. I'll try running one or two classical recordings I have with the EQ you suggested and see what happens! Thanks!
EDIT: The pdf links you provided require registration. Can you just attach them here?
You'll note that the bass and lower midrange is quite a bit raised relative to the high frequencies. In most rooms, with the usual furniture and absorption, and usual cone/dome speakers, this is what you would get and most people would find this balance neutral. But, if you have constant directivity speakers combined with less room absoprtion, the total emitted power in the high frequencies may not get absorbed and may cause a perceived lack of bass.
I'm not interested in remastering recordings, however useful that might be. I simply don't have the time to do it. I'll try running one or two classical recordings I have with the EQ you suggested and see what happens! Thanks!
EDIT: The pdf links you provided require registration. Can you just attach them here?
Attachments
Last edited:
I'm pretty sure that you forgot to ask about the directivity of the loudspeakers that I'm using. Clearly that can change everything that you might have been assuming.
Have you taken the time to build your own loudspeakers? Any other audio components or acoustic treatments? But for the music itself--you have no time for diy? That's an interesting viewpoint.
The techniques that are listed in the two pdf files linked above take less time to use effectively for an entire album than the time you've already spent trying to rationalize not using the technique. It takes me about the same time to do a basic demaster of an album as it takes to listen to it all the way through.
Do you have time to listen actively to music?
Chris
I wonder about that too, glad you asked. I just don't know. Perhaps another set of tracks will give a different response?
ra7: The reason why I responded back to you on the subject of loudspeaker directivity is because your comment is actually 180 degrees out of phase with what really happens using loudspeakers with poor control over directivity at differing frequencies--like direct radiating loudspeakers...
I'll try to keep this short because it's driving off the subject of this thread:
I was invited to visit an anechoic chamber in Hope, AR several years ago (almost 9 years now) where several configurations of loudspeakers were being tested, including demonstrations of walking around the article under test while it was playing music and the chamber door was shut.
Later, after the anechoic test were completed, the loudspeakers were placed in a listening room just adjacent to the chamber, and the same loudspeaker configurations were auditioned while the engineer in charge performed "salt and pepper" EQ adjustments. The loudspeaker configurations that had larger issues with changing directivity at some intermediate frequency band actually sounded louder, not softer. The reason for this: the precedence effect (sometimes called the Haas effect). It was the smaller-mouth horns that were the issue and that created many more problems with in-room EQ and adjustment of placement/near field reflectivity.
Most people don't get the chance to experience that vivid demonstration of directivity control and its effect on loudspeaker acoustic performance in-room...and with loudspeakers/horns that control their polars so well.
So the answer to your question is: directivity makes things much better in terms of monitoring for mastering tasks, not worse. The lack of controlled directivity in each axis creates problems in real rooms...and they're not small problems.
Chris
I'll try to keep this short because it's driving off the subject of this thread:
I was invited to visit an anechoic chamber in Hope, AR several years ago (almost 9 years now) where several configurations of loudspeakers were being tested, including demonstrations of walking around the article under test while it was playing music and the chamber door was shut.
Later, after the anechoic test were completed, the loudspeakers were placed in a listening room just adjacent to the chamber, and the same loudspeaker configurations were auditioned while the engineer in charge performed "salt and pepper" EQ adjustments. The loudspeaker configurations that had larger issues with changing directivity at some intermediate frequency band actually sounded louder, not softer. The reason for this: the precedence effect (sometimes called the Haas effect). It was the smaller-mouth horns that were the issue and that created many more problems with in-room EQ and adjustment of placement/near field reflectivity.
Most people don't get the chance to experience that vivid demonstration of directivity control and its effect on loudspeaker acoustic performance in-room...and with loudspeakers/horns that control their polars so well.
So the answer to your question is: directivity makes things much better in terms of monitoring for mastering tasks, not worse. The lack of controlled directivity in each axis creates problems in real rooms...and they're not small problems.
Chris
I looked at about 90 more opera tracks and could not find the 13Hz peak, so don't know where it came from. This time I did whole albums at a time, and looked at each album individually to get a sense of where the artifact might come from.
What I did find on the 5 albums I processed was a consistent hump at ~3.7 hz. Every recording had it, and it varied by only a few 1/10ths. No idea what that is. It's about 26dB down from the loudest part of the spectrum. Another fun artifact is that almost all opera recordings I've looked at have a peak near 15.8Khz and/or 15.6Khz. The NTSC and PAL horizontal scan rates. Seems that the engineers are watching TV while they mix.
What I did find on the 5 albums I processed was a consistent hump at ~3.7 hz. Every recording had it, and it varied by only a few 1/10ths. No idea what that is. It's about 26dB down from the loudest part of the spectrum. Another fun artifact is that almost all opera recordings I've looked at have a peak near 15.8Khz and/or 15.6Khz. The NTSC and PAL horizontal scan rates. Seems that the engineers are watching TV while they mix.
Chris,
Please don't be offended by my stance that I don't want to re-EQ recordings. I just don't think I'm skilled enough to better than what the sound engineers thought would be the most optimum sounding recording. And no, I really don't have the time. I barely get to listen to music now-a-days. Life gets in the way.
I've built plenty of stuff over the years and I have three different diy speakers (line arrays, horns, dual-concentric cone/dome) in three rooms right now, with another couple sitting on shelves. I have room absorption and diffusion devices as well. I try to manage the power response in the room.
It is amazing that you quoted my first sentence my earlier post, but not the next few. If you had read my post in its entirety you would've noted that I did mention constant directivity speakers and what that could do to the perceived response.
I would still like to see your in-room steady state response, if you have it. Thanks!
Just directivity alone is not going to do anything if you don't control the reflections. Just the same, lack of constant directivity does not create problems on its own, lack of managing the reflections properly does.
It is the combination of directivity (constant, uniformly changing, non-uniformly changing, anything else) together with the room absorption that determines the power response, which ultimately determines the tonal balance you hear. This was the point. If you have constant directivity speakers, which it appears you do, you still have to manage the in-room response, otherwise, it can sound bright, dull, etc. And then, there might be a need to play with tone controls, or remaster recordings. I'm just trying to rule out that it is not the in-room response you are trying to EQ.
I couldn't get to your pdfs last evening. I'll try to do that tonight!
Chris, I've had speakers with well controlled directivity in normal acoustically treated rooms, and open baffle speakers in a hypo-echoic room. Both have threads here. There is no question that the room and speaker together make a huge difference.
I've found that in a normal room, with good speakers, the classic B&K slope that Ra7 posted sounds well balanced. It's well known and well liked across the decades, so this should come as no surprise. However, in a room that was close to anechoic, the B&K curve was too dull for me. I ended up with a much flatter curve, similar to yours above 300Hz. What I could not try, because I didn't have the woofer size, was a low end boost like you have. That might have sounded nice, because the room could certainly take it without trouble. The OBs did play down into the 20s. No one who heard it, including sound engineers, complained about a lack of bass, much to my surprise, because I found it a bit bass shy. They just talked about how clean and non-resonant the bass was. Open baffle and a well damped lava cave will do that.
But that's not the topic of this thread. The main topic of this tread is what is actually on real recordings. And why Pink or Brown noise might not be the best test signal at the top and bottom of the spectrum. It may also illustrate why woofers and tweeters can have such different power ratings without problem.
I've found that in a normal room, with good speakers, the classic B&K slope that Ra7 posted sounds well balanced. It's well known and well liked across the decades, so this should come as no surprise. However, in a room that was close to anechoic, the B&K curve was too dull for me. I ended up with a much flatter curve, similar to yours above 300Hz. What I could not try, because I didn't have the woofer size, was a low end boost like you have. That might have sounded nice, because the room could certainly take it without trouble. The OBs did play down into the 20s. No one who heard it, including sound engineers, complained about a lack of bass, much to my surprise, because I found it a bit bass shy. They just talked about how clean and non-resonant the bass was. Open baffle and a well damped lava cave will do that.
But that's not the topic of this thread. The main topic of this tread is what is actually on real recordings. And why Pink or Brown noise might not be the best test signal at the top and bottom of the spectrum. It may also illustrate why woofers and tweeters can have such different power ratings without problem.
Perhaps the studios that mastered your opera recordings were using monitors that had a FR dip at 3.6 kHz. Looking at these monitor types, it seems many studio monitors have issues with dips at ~3.6 kHz. There are studies on this subject that have been around for years: mastering guys tend to EQ-in the inverse of their monitor's frequency response. It's a commonly found issue.
It's also a frequency that tends to get reinforced around microphones (about 2" half-wave).
Chris
It's also a frequency that tends to get reinforced around microphones (about 2" half-wave).
Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.