Q&A # 51 - February 17, 1998

Staff

Q I'd like to know what your opinion is about listening to CDs with Dolby Pro Logic on? What are the pros and the cons?

A There are many CDs that are recorded with Pro Logic in mind. What I have found, however, is that I just have to switch the Pro Logic in and out to see if I like the effect. Usually, Pro Logic works fine with large ensembles, such as orchestras, but sounds very strange with small groups, such as trios, and especially unsuitable with single instrument recordings, such as piano. Also, anytime you put additional circuitry in the signal path, such as a Pro Logic chip, some deterioration in the sound occurs. Very often, you might find that plain vanilla two channel stereo sounds best.

A couple of readers wrote about my comments last week on the DSP soundfields that can be applied to audio. Apparently, there may be soundfields that are based on L-shaped rooms, and so, the sound in the rear might be louder on one side than on the other. Personally, I think this is an inappropriate way to design soundfields, since it results in unbalanced surround sound, but then, auditoriums, jazz clubs, etc., are not necessarily balanced either. If readers would like to send us their comments on finding some strangely balanced DSP soundfields that they have found in their receivers, we will post them. Send the brand name, model number, and name of the soundfield.

Q The Toshiba TP61G90 specs say it has 800 lines. Can I run a DVD player through a line doubler (such as the Extron Lancia) and
improve the picture noticeably? Also, what are the benefits of progressive scan vs. interlaced?

A The resolution story on TVs is very confusing. All NTSC TVs are set up to view an image that has 525 scanning lines that are arranged horizontally on the screen, top to bottom (625 in Europe, called the PAL System). This is called the "Vertical Resolution", and there really are not 525 lines visible, but rather, there are only 480, because some lines are lost as the scanning beam goes from the bottom back to the top in order to begin another scan (this takes a finite time, and the lines that would have otherwise been put onto the screen during this time are lost). But it is still called 525. Over the past few years, TV manufacturers began listing a specification on the number of vertical lines that could potentially be displayed if the signal source has them. This is called the "Horizontal Resolution", and that is the 800 your Toshiba is referring to. However, no current NTSC signal has that many vertical lines. VHS has 240, broadcast programming has 330, LD has 420, and DVD has 480 vertical lines of horizontal resolution (all of them have 525 horizontal lines of vertical resolution, because that number is constant). Recently, I counted, as best I could, the number of RGB phosphor groups across a TV screen left to right. It came out around 500, which is the maximum horizontal resolution of our current signal capabilities. So, I don't know where all these marketing specs come from. Lastly, a line doubler works on the horizontal scanning lines, looking at what is in the line above and the line below, and filling in between as a "best guess" as to what the signal processor thinks was probably there in the original video scene (the original NTSC image is interlaced, where half the 525 lines are shown as a "field", then the other half as another "field", at 1/60 second per field. The two fields constitute one "frame" that is 1/30 second in length). The picture looks better with line doubling, depending on the quality of the line doubler. The final image is, after all, half computer generated. The best image is progressive line scan, where all the image is original, and you see all 525 lines at one time (one frame), rather than two successive fields of 262 lines each. In order to use a line doubler, the TV has to be able to scan at twice the normal 15.75 kHz NTSC rate, i.e., 31.5 kHz. The number 15.75 kHz comes from multiplying 30 frames per second times 525 lines per frame. Most TVs right now can't scan at 31.5 kHz, but some of the new models can, and most front projectors can. DTVs will be able to scan at multiple frequencies, much like our computer monitors can.

Q I was wondering if you could shed some light on how to properly set up a home theater system to output reference theatrical dB
levels with typical amplification power of about 100W/Ch and speakers of relatively high efficiency (90dB/W/M). How loud should sound be? And does this loudness have to be set differently for various frequency ranges (e.g., subwoofer levels)?

A I feel that theaters have the sound system too loud. The reason they have it so loud is the same reason A/V stores crank the volume when they are trying to sell you a system. It's impressive, and leaves you with a "Wow" sensation. Unfortunately, sound levels above 80 dB or so can damage your hearing over time. Constant tinnitus (ringing in the ears) is a sign that damage has been done. So, I would say first that you should not aim for theater SPLs because it is not good for your ears, and secondly, 100 w/ch is not enough to do it anyway, at least with 90 dB/w/m speakers. If you try to turn the system up really loud to reproduce the theater sound levels, your amplfiers will go into clipping, and that could damage your speakers along with your ears. What I do is set the overall system loudness to a comfortable level, but I turn the subwoofer level up a little extra with the low pass set to about 50 Hz. Sometimes I may have to increase the volume of the center channel a bit so the dialogue is clear. But the overall level is such that I can talk to someone in the room if necessary. Since the five channels are all set to modest level, my ears quickly adjust, and I get the full effect of surround sound, without hearing loss, but the room still shudders just enough to let me have the "Wow". Try this and see if you like it. The amplifiers and speakers will produce less distortion this way too, so it actually sounds cleaner.

Q My question centers on some speakers I have been seeing (i.e., Definitive Tech., Linn, etc.) that have multiple binding posts for bi- and even tri-wiring and/or amping. How would you go about bi-amping a speaker? I am not familiar with any preamp/processors that have multiple line outs for all channels, so there obviously must be another way. Does a crossover-containing amp such as NHT's subwoofer amps have to be used? Is it really worth all the aggravation (not to mention expense) of doing so?

A Bi-wiring is a controversial subject. To bi-wire, you need speakers that have a separate set of binding posts for the tweeter and for the woofer. With regular wiring (non bi-wiring), there is a metal strip that connects the + post of the tweeter to the + post of the woofer, and a strip that connects the - post of the tweeter to the - post of the woofer. First, you remove these strips. Then, connect two sets of speaker cables to the left channel outputs and two sets to the right channel outputs of the amplifier. One set of cables from the left channel output goes to the tweeter for the left speaker, and the other set from the left channel goes to the woofer for the left speaker. Do the same for the right channel.

I have not found any improvement with bi-wiring, but my own tests were made using high performance cables. There may be a difference when using zip cord. Tri-wiring would work the same way, but you would need three sets of cables for each channel. Bi-amping, on the other hand, is much different. In this case, the metal strips on the speakers are removed as before, but a separate power amplifier has to be used for the tweeter and woofer of each speaker. The most amplifier-efficient way of doing this is to use an active crossover in between the preamplifier and the power amplifiers, so that only the high frequencies are sent through the amplifier that is driving the tweeter, and only the low frequencies are sent through the power amplifier driving the woofer. The crossover networks in the speakers should be bypassed with this technique, but it necessitates a very carefully programmed active crossover, and one of the flexible units such as made by Rane. A Y-connector can be used between the preamp and power amplifiers to split the preamp signal so that you can connect the preamp output to several power amplifiers. You can also tri-amp if the speakers have three sets of binding posts for such use. Bi-amping and tri-amping are worthwhile if you have a set of speakers that you want absolute control over, but under most circumstances, it is really not worth the trouble.

Q I have a bet that audio reproduction from CDs is "better" than reproduction from vinyl LPs. This is dependent on the definition of
"better", quality of stylus for LPs, and sampling/lasers for CDs. But, all else being equal, which medium has the greater range of frequency storage, S/N, etc? Thank you for your response!!

A This is a question that has been asked ever since CDs first hit the shelves, and may never really be definitively answered. Because CDs use a sampling frequency of 44.1 kHz, the frequency response of the CD has to be kept below 1/2 of this. So, at best, it is 22.05 kHz. LPs on the other hand, can have frequencies well above 22 kHz, even though most people can't hear them. As to noise, LPs have more surface noise, such as clicks and pops, as well as a rubbing sound from stylus against plastic. CDs have quantization noise and other artifacts. Neither is perfect. Overall, I would say that CDs have less "noise" and potentially greater dynamic range. I used to think that LPs sounded better, but now I am more comfortable with CDs. LPs seem to have more "body" to them, but some of this is probably a mechanical artifact. The new 96 kHz/24 bit CD will be fantastic when and if it gets going. To me, the best possible CD would be one that has a sampling frequency of about 500 kHz and 24 bit words. That would allow the signal to be handled with no filters. Of course, that is just my opinion. There are plenty of people out there who feel the best improvements would come from other types of changes.

Q I would like to improve the overall sound of my system by improving my room with acoustic treatment. The problem is that I don't know which brand, or type, are the best to be used with my existing speakers. I have a pair of Legacy FOCUS speakers. It has a total of 7 drivers, two tweeters, two kevlar midbass, two woofers and a REAR in-phase firing woofer. Does a typical bass trap "absorb" all the sound from the rear woofer, therefore reducing the amount of bass/performance of my speaker? Do you generally recommend for against bass traps/acoustic treatments on bipolar design? Is a speaker with one rear firing woofer classified as bipolar?

Thanks for the best AV site on the planet.

A The room treatment won't depend so much on the speaker as it will on the room itself. Secondly, absorption devices work better on the mids and highs than they do on the bass. So, whether the speaker is bipolar or dipolar (your speaker would be quasi, since there are two on the front and one on the back), the mids and highs would be partially absorbed by treatments onto which the tweeters and midrange speakers project, but not so much the bass. You can put bass traps in all the corners, and this will make a difference no matter which way the woofers are facing. For the mids and highs, pretend you were standing where the speakers are and would be making a billiards shot bouncing off the wall or ceiling to put the ball in the place where you sit to listen. Put the wall treatment on that reflection position. A company called Acoustic Sciences Corporation makes good room treatments, but there are also many other nice brands. Each has their own technology. You can make your own too. Simply hanging an egg crate mattress pad ($30 at linen stores) on the wall helps a great deal.

� Copyright 1998 Secrets of Home Theater & High Fidelity
Return to Q&A Index.