View RSS Feed

mitchco

Hear music the way it was intended to be reproduced - part 1

Rating: 2 votes, 5.00 average.
I love music, any kind of music really. As a former recording/mixing engineer/producer for 8 years, and lifetime audio freak, I had the privilege to record, mix, and master a wide variety of music. In this introductory post, we will look at the most important quality of reproducing music called, "timbre". Over a series of posts, the goal is to calibrate your sound system to be the most accurate reproducer of music for your ultimate listening pleasure :-)

In Wikipedia’s definition of timbre, you will see, aside from the technical definition, “In psychoacoustics, timbre is also called tone quality and tone color.” Tone quality is critically important in the reproduction of recorded music.

If you have ever heard live music, (e.g. piano, acoustic guitar, horns, strings, drums, etc.) then you may remember how it sounded. You may also remember when you went home and listened to something similar on your stereo that it did not have the same “tone quality”. Why?

Well, it so happens that another group of folks were also wondering this and produced this outstanding short article on, “Relevant loudspeaker tests in studios in Hi-Fi dealers' demo rooms in the home etc.” Of very particular importance is the frequency response curve in Figure 5. We will come back to that a bit later.

From the article abstract, “The "sound" of a Hi-Fi set is to a great extent room dependent. Very often, the final result is determined by the room rather than by the actual equipment. Fortunately, these influences may readily be measured.”

What the article is describing is musical timbre or tone quality. Unfortunately, the reality is that the tone quality reproduced by your sound system is highly dependent on your listening room. Before becoming a recording engineer, I was in electronics engineering world and as a hobby, built a great deal of speakers, amplifiers and preamps (still do). I also got into room acoustics and managed to get my hands on this wonderful device that revolutionized audio measurement techniques.

The TEF stands for time, energy and frequency. Very quickly you could analyze a room in 3D and determine the rooms “tonal quality” for sound reproduction. Based on that, you could treat the room with “Tube Traps” for bass frequency tuning, absorption materials for dampening overly live rooms, and “diffuser panels” to prevent slap echoes, but not overly dampen the room. I bought every possible book on recording and control room design and room tuning. I will provide a resource list later for those interested.

I had the privilege to observe Chips Davis design and build two multi-million dollar recording studios and control rooms from scratch using his infamous Live End Dead End (LEDE) room design. I then went on to “treat” several recording studios, controls rooms, critical listening rooms at audio dealers, and several private critical listening rooms using the TEF computer and lessons I learned from Chips plus the reference books.

My point in saying all of this is to pass on to you my learning’s to benefit you in your quest for the most tonally accurate sound reproduction system you can achieve with your existing equipment. No, I am not going to suggest you rip up your room or spend thousands or tens of thousands of dollars on acoustical measurement equipment and room treatments. What I am suggesting is that with a few key considerations, and a few bucks, you can make dramatic improvements to the tonal quality of your existing sound system.

Let’s get back to timbre and that B&K article, specifically Figure 5, “Optimum curve for hi fi equipment measured in the actual listening room.” Figure 5 is the key to tonal quality. That curve is the frequency response measured at the listening position. If your sound system measures close to this curve, especially the roll-off, then congratulations, you have achieved tonal perfection! Once you have heard a sound system that is calibrated to this curve, then you will understand exactly what I mean. Everything sounds “right” and all of a sudden the depth soundstage magically appears.

There is good reason for this curve, affectionately called the B&K house curve. In the recording studio world, in the control room, there will most always be a set of speakers that are tuned or calibrated to the B&K house curve. Why? Because it most accurately reproduces instruments that sound tonally correct. I.e. it has the best timbre. Additionally, when mixing engineers move from one studio to the next and listen to their mix downs, with this curve, it will have the same tone quality it had in the previous studio. Consistency is the key.

My wife, who is not an audiophile and puts ups with my tape measures and swept sine waves once commented, “I was in the garage and even there it sounded like someone is playing the piano in our living room.” That is near perfect timbre.

So the first step in understanding whether your sound system is tonally correct or at least as best as it can be, is to measure the frequency response at the listening position in your listening room and compare it to the B&K house curve. In my next post, I will show you how to do that without breaking the bank.

Happy listening!

Mitch

Updated 05-09-2012 at 01:27 AM by mitchco

Categories
Personal Blogs

Comments

  1. The Computer Audiophile's Avatar
    Hi Mitch - I look forward to reading your posts.
  2. rancew's Avatar
    I'm really looking forward to following this.



    The right timbre is what makes the recording sound like "music". It's much harder to achieve in a system than anything else IME.



    Bring on Part 2...
  3. tmokbel's Avatar
    Hi Mitch,



    This is wonderful and i'm really looking forward to reading and learining from your posts. I know that my music can sound so much better than what I'm currently hearing due to the fact that my living room has the acoustic qualities of a subway tunnel!!

    I know of a few guys here in Singapore that have spent thousands of dollars on room treatment only to find that they've gone from one problem to another due to a lack of understanding and actual measurements/readings of their rooms.

    I think even the most basic of correct knowledge and treatment

    will go a long way. Bring it on Mitch...
  4. esldude's Avatar
    Should be good info here. I read where Peter Walker of Quad suggested something near flat response to 2 khz and a 3 dB/decade slope after that. That isn't all that much different than your #5 curve. When digital software like CoolEdit was out and I could play with curves on files off CD's I did a number of things one of which was make the Walker curve and the Academy curve (used in theaters). The Walker curve more often than not was an improvement.



    More recently to fix room problems I have used a Tact. Tact has some stock EQ's they suggest and most are some variation on the #5 curve. Differences are in how deep in the low end your speaker goes and down to personal preference. I usually tweak curves for my personal preference off of the stock Tact curves. It is surprising how an octave or two lifted or dipped just a bit can alter soundstaging, depth and character of the music. Tact RoomEQ is very effective. I wouldn't say it fixes everything, but it does a world of good to many room and speaker response problems. This is where I think the next big advancements in home audio are to be found. In fixing speaker anomalies and room response problems. I believe the source material is of higher fidelity than it has ever been. At least the better electronics up to the speakers are of very high quality and darn near to fully transparent. Fixing the sound in the room at the listener's ears is where lots of good can now be done. Very high fidelity right to the listener is possible like never before in history. After that, I think artificially sweetening sound quality for improved subjective enjoyment is the next frontier. Something I think some LP playback gear and vacuum tube gear happen to do by accident. This means going beyond simply fidelity to the source (like it was ever simple) and enriching it beyond reality.
  5. rancew's Avatar
    I think of timbre ("tone quality") as a distinct quality separate from frequency distribution ("tonal balance") which DSP room correction devices like the TacT address.



    I've had a couple of TacT devices in my system over the years. Achieving a "perfect" (or preferred) room correction curve in terms of frequency distribution does not directly address the actual timbre -- the sort of "inner quality" of sound: things like spatial cues, attack, decay which create the larger qualities of sound we refer to with terms like warm, dry, hard, etc. These qualities are really a combination of room characteristics, as Mitch refers to, and equipment -- these days to include software. And of course the actual source, with recordings themselves having varied timbres...
  6. esldude's Avatar
    I find making those little minor bumps here and there in the Tact target response curves can do those things you refer to as not being frequency related. These adjustments would be a half octave to maybe two octaves wide, and usually between .2 dB and 1 dB. You can alter apparent depth, spaciousness, timbre and other things related to overall balance once you first get the basic target curve dialed in. It also unfortunately will vary some from recording to recording. So maximum timbre or whatever your prime desired result is can never be one set perfect thing. You have to reach a good compromise for the kinds of things you like and the type of music you listen to most of the time. Plus whatever changes are needed for a particular room/speaker combo.



    It is worth remembering the amp puts out a varying voltage. Speakers only see that signal assuming the amp can supply to current not to be affected by the load. Timbre or any other musical quality resides within the varying voltage as that is the only thing communicated to your speakers. By altering target curves you are altering that signal and with enough precision and control it will make those changes. Tact software doesn't quite offer everything you might desire of course, but it goes a long way in that direction. Even then the room doesn't get completely compensated for by the room correction or other curves. So having a good room with some treatment to work with a given speaker will certainly help.



    There are other ways to alter timbre that involve altering the original signal beyond just frequency. Distortion, compression, signal/noise ratios, all have an effect on perceived timbre. And yet in the end all you are altering is the varying voltage fed to your amp and onto your speakers with the goal of it arriving at your ears with the qualities you desire. Many of these work like they do because of the ear/brain reaction to masking effects.
  7. rancew's Avatar
    Yes, I certainly agree and have experienced the fact that frequency response correction of a system (e.g. DSP correction a la TacT) can alter the timbre. But typically not a wholesale change.



    For instance, you may have a system that sounds cold, analytical and thin because of the nature of the electronics, speakers, and (yes) even cables. By correcting a null in the mid-bass region you can add a bit more warmth to the sound. Other corrections may affect things like spatial cues to some degree. But the overall timbre is still going to be on the cold and analytical side.



    Anyway, I'm curious to hear Mitch's approach to room correction...
  8. AudioExplorations's Avatar
    I have recently purchased a measurement mic and preamp and measured my room with the goal to flatten out the sub 100Hz area where I had some severe room mode issues. So far the results have been far better than expected, with digital EQ I have been able to flatten the response within +/- 3dB down to 20Hz which has also greatly improved the RT60 (reverberation decay). The improvement in the bass response is absolutely night and day.



    I had not heard about the B&K house curve, so look forward to learning about this and the measurement methodology/workflow followed! Great blog!
  9. mitchco's Avatar
    Would love to hear more on how you measured and the results. Got any graphs to post?



    Happy listening!



    Mitch
  10. Mike Gillespie's Avatar
    Mitch, great post. Any chance of part 2 soon?
  11. mitchco's Avatar
    Hi Mike, Part 2 thru Part 6 (conclusion) are on my blog at:


    http://www.computeraudiophile.com/blogs/8177-mitchco



    Cheers, Mitch
    Updated 05-09-2012 at 01:30 AM by mitchco